From 47846c9b0c10808d9337d2e7d09361f3e0a0a71a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 25 Nov 2009 17:46:19 +0100 Subject: mac80211: reduce reliance on netdev For bluetooth 3, we will most likely not have a netdev for a virtual interface (sdata), so prepare for that by reducing the reliance on having a netdev. This patch moves the name and address fields into the sdata struct and uses them from there all over. Some work is needed to keep them sync'ed, but that's not a lot of work and in slow paths anyway. In doing so, this also reduces the number of pointer dereferences in many places, because of things like sdata->dev->dev_addr becoming sdata->vif.addr. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 2aff4906b2a..e94cc526b0f 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -659,12 +659,14 @@ struct ieee80211_conf { * @type: type of this virtual interface * @bss_conf: BSS configuration for this interface, either our own * or the BSS we're associated to + * @addr: address of this interface * @drv_priv: data area for driver use, will always be aligned to * sizeof(void *). */ struct ieee80211_vif { enum nl80211_iftype type; struct ieee80211_bss_conf bss_conf; + u8 addr[ETH_ALEN]; /* must be last */ u8 drv_priv[0] __attribute__((__aligned__(sizeof(void *)))); }; -- cgit v1.2.3-70-g09d2 From eaf85ca7fecb218fc41ff57c1642ca73b097aabb Mon Sep 17 00:00:00 2001 From: Zhu Yi Date: Tue, 1 Dec 2009 10:18:37 +0800 Subject: wireless: add ieee80211_amsdu_to_8023s Move the A-MSDU handling code from mac80211 to cfg80211 so that more drivers can use it. The new created function ieee80211_amsdu_to_8023s converts an A-MSDU frame to a list of 802.3 frames. Cc: Johannes Berg Signed-off-by: Zhu Yi Signed-off-by: John W. Linville --- include/net/cfg80211.h | 22 +++++++++- net/mac80211/rx.c | 106 ++++++++++--------------------------------------- net/wireless/util.c | 99 ++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 137 insertions(+), 90 deletions(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 0884b9a0f77..542a477a94d 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1578,7 +1578,7 @@ unsigned int ieee80211_hdrlen(__le16 fc); * @addr: the device MAC address * @iftype: the virtual interface type */ -int ieee80211_data_to_8023(struct sk_buff *skb, u8 *addr, +int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, enum nl80211_iftype iftype); /** @@ -1589,9 +1589,27 @@ int ieee80211_data_to_8023(struct sk_buff *skb, u8 *addr, * @bssid: the network bssid (used only for iftype STATION and ADHOC) * @qos: build 802.11 QoS data frame */ -int ieee80211_data_from_8023(struct sk_buff *skb, u8 *addr, +int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, enum nl80211_iftype iftype, u8 *bssid, bool qos); +/** + * ieee80211_amsdu_to_8023s - decode an IEEE 802.11n A-MSDU frame + * + * Decode an IEEE 802.11n A-MSDU frame and convert it to a list of + * 802.3 frames. The @list will be empty if the decode fails. The + * @skb is consumed after the function returns. + * + * @skb: The input IEEE 802.11n A-MSDU frame. + * @list: The output list of 802.3 frames. It must be allocated and + * initialized by by the caller. + * @addr: The device MAC address. + * @iftype: The device interface type. + * @extra_headroom: The hardware extra headroom for SKBs in the @list. + */ +void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, + const u8 *addr, enum nl80211_iftype iftype, + const unsigned int extra_headroom); + /** * cfg80211_classify8021d - determine the 802.1p/1d tag for a data frame * @skb: the data frame diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index dbfd684e3e2..a182e423109 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1541,16 +1541,10 @@ static ieee80211_rx_result debug_noinline ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) { struct net_device *dev = rx->sdata->dev; - struct ieee80211_local *local = rx->local; - u16 ethertype; - u8 *payload; - struct sk_buff *skb = rx->skb, *frame = NULL; + struct sk_buff *skb = rx->skb; struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; __le16 fc = hdr->frame_control; - const struct ethhdr *eth; - int remaining, err; - u8 dst[ETH_ALEN]; - u8 src[ETH_ALEN]; + struct sk_buff_head frame_list; if (unlikely(!ieee80211_is_data(fc))) return RX_CONTINUE; @@ -1561,94 +1555,34 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) if (!(rx->flags & IEEE80211_RX_AMSDU)) return RX_CONTINUE; - err = __ieee80211_data_to_8023(rx); - if (unlikely(err)) + if (ieee80211_has_a4(hdr->frame_control) && + rx->sdata->vif.type == NL80211_IFTYPE_AP_VLAN && + !rx->sdata->u.vlan.sta) return RX_DROP_UNUSABLE; - skb->dev = dev; - - dev->stats.rx_packets++; - dev->stats.rx_bytes += skb->len; - - /* skip the wrapping header */ - eth = (struct ethhdr *) skb_pull(skb, sizeof(struct ethhdr)); - if (!eth) + if (is_multicast_ether_addr(hdr->addr1) && + ((rx->sdata->vif.type == NL80211_IFTYPE_AP_VLAN && + rx->sdata->u.vlan.sta) || + (rx->sdata->vif.type == NL80211_IFTYPE_STATION && + rx->sdata->u.mgd.use_4addr))) return RX_DROP_UNUSABLE; - while (skb != frame) { - u8 padding; - __be16 len = eth->h_proto; - unsigned int subframe_len = sizeof(struct ethhdr) + ntohs(len); - - remaining = skb->len; - memcpy(dst, eth->h_dest, ETH_ALEN); - memcpy(src, eth->h_source, ETH_ALEN); + skb->dev = dev; + __skb_queue_head_init(&frame_list); - padding = ((4 - subframe_len) & 0x3); - /* the last MSDU has no padding */ - if (subframe_len > remaining) - return RX_DROP_UNUSABLE; + ieee80211_amsdu_to_8023s(skb, &frame_list, dev->dev_addr, + rx->sdata->vif.type, + rx->local->hw.extra_tx_headroom); - skb_pull(skb, sizeof(struct ethhdr)); - /* if last subframe reuse skb */ - if (remaining <= subframe_len + padding) - frame = skb; - else { - /* - * Allocate and reserve two bytes more for payload - * alignment since sizeof(struct ethhdr) is 14. - */ - frame = dev_alloc_skb( - ALIGN(local->hw.extra_tx_headroom, 4) + - subframe_len + 2); - - if (frame == NULL) - return RX_DROP_UNUSABLE; - - skb_reserve(frame, - ALIGN(local->hw.extra_tx_headroom, 4) + - sizeof(struct ethhdr) + 2); - memcpy(skb_put(frame, ntohs(len)), skb->data, - ntohs(len)); - - eth = (struct ethhdr *) skb_pull(skb, ntohs(len) + - padding); - if (!eth) { - dev_kfree_skb(frame); - return RX_DROP_UNUSABLE; - } - } - - skb_reset_network_header(frame); - frame->dev = dev; - frame->priority = skb->priority; - rx->skb = frame; - - payload = frame->data; - ethertype = (payload[6] << 8) | payload[7]; - - if (likely((compare_ether_addr(payload, rfc1042_header) == 0 && - ethertype != ETH_P_AARP && ethertype != ETH_P_IPX) || - compare_ether_addr(payload, - bridge_tunnel_header) == 0)) { - /* remove RFC1042 or Bridge-Tunnel - * encapsulation and replace EtherType */ - skb_pull(frame, 6); - memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN); - memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN); - } else { - memcpy(skb_push(frame, sizeof(__be16)), - &len, sizeof(__be16)); - memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN); - memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN); - } + while (!skb_queue_empty(&frame_list)) { + rx->skb = __skb_dequeue(&frame_list); if (!ieee80211_frame_allowed(rx, fc)) { - if (skb == frame) /* last frame */ - return RX_DROP_UNUSABLE; - dev_kfree_skb(frame); + dev_kfree_skb(rx->skb); continue; } + dev->stats.rx_packets++; + dev->stats.rx_bytes += rx->skb->len; ieee80211_deliver_skb(rx); } diff --git a/net/wireless/util.c b/net/wireless/util.c index a3c841a255d..23557c1d0a9 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -285,7 +285,7 @@ static int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr) } } -int ieee80211_data_to_8023(struct sk_buff *skb, u8 *addr, +int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, enum nl80211_iftype iftype) { struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; @@ -383,7 +383,7 @@ int ieee80211_data_to_8023(struct sk_buff *skb, u8 *addr, } EXPORT_SYMBOL(ieee80211_data_to_8023); -int ieee80211_data_from_8023(struct sk_buff *skb, u8 *addr, +int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, enum nl80211_iftype iftype, u8 *bssid, bool qos) { struct ieee80211_hdr hdr; @@ -497,6 +497,101 @@ int ieee80211_data_from_8023(struct sk_buff *skb, u8 *addr, } EXPORT_SYMBOL(ieee80211_data_from_8023); + +void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, + const u8 *addr, enum nl80211_iftype iftype, + const unsigned int extra_headroom) +{ + struct sk_buff *frame = NULL; + u16 ethertype; + u8 *payload; + const struct ethhdr *eth; + int remaining, err; + u8 dst[ETH_ALEN], src[ETH_ALEN]; + + err = ieee80211_data_to_8023(skb, addr, iftype); + if (err) + goto out; + + /* skip the wrapping header */ + eth = (struct ethhdr *) skb_pull(skb, sizeof(struct ethhdr)); + if (!eth) + goto out; + + while (skb != frame) { + u8 padding; + __be16 len = eth->h_proto; + unsigned int subframe_len = sizeof(struct ethhdr) + ntohs(len); + + remaining = skb->len; + memcpy(dst, eth->h_dest, ETH_ALEN); + memcpy(src, eth->h_source, ETH_ALEN); + + padding = (4 - subframe_len) & 0x3; + /* the last MSDU has no padding */ + if (subframe_len > remaining) + goto purge; + + skb_pull(skb, sizeof(struct ethhdr)); + /* reuse skb for the last subframe */ + if (remaining <= subframe_len + padding) + frame = skb; + else { + unsigned int hlen = ALIGN(extra_headroom, 4); + /* + * Allocate and reserve two bytes more for payload + * alignment since sizeof(struct ethhdr) is 14. + */ + frame = dev_alloc_skb(hlen + subframe_len + 2); + if (!frame) + goto purge; + + skb_reserve(frame, hlen + sizeof(struct ethhdr) + 2); + memcpy(skb_put(frame, ntohs(len)), skb->data, + ntohs(len)); + + eth = (struct ethhdr *)skb_pull(skb, ntohs(len) + + padding); + if (!eth) { + dev_kfree_skb(frame); + goto purge; + } + } + + skb_reset_network_header(frame); + frame->dev = skb->dev; + frame->priority = skb->priority; + + payload = frame->data; + ethertype = (payload[6] << 8) | payload[7]; + + if (likely((compare_ether_addr(payload, rfc1042_header) == 0 && + ethertype != ETH_P_AARP && ethertype != ETH_P_IPX) || + compare_ether_addr(payload, + bridge_tunnel_header) == 0)) { + /* remove RFC1042 or Bridge-Tunnel + * encapsulation and replace EtherType */ + skb_pull(frame, 6); + memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN); + memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN); + } else { + memcpy(skb_push(frame, sizeof(__be16)), &len, + sizeof(__be16)); + memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN); + memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN); + } + __skb_queue_tail(list, frame); + } + + return; + + purge: + __skb_queue_purge(list); + out: + dev_kfree_skb(skb); +} +EXPORT_SYMBOL(ieee80211_amsdu_to_8023s); + /* Given a data frame determine the 802.1p/1d tag to use. */ unsigned int cfg80211_classify8021d(struct sk_buff *skb) { -- cgit v1.2.3-70-g09d2 From 0f78231bffb868a30e8533aace142213266bb811 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 1 Dec 2009 13:37:02 +0100 Subject: mac80211: enable spatial multiplexing powersave Enable spatial multiplexing in mac80211 by telling the driver what to do and, where necessary, sending action frames to the AP to update the requested SMPS mode. Also includes a trivial implementation for hwsim that just logs the requested mode. For now, the userspace interface is in debugfs only, and let you toggle the requested mode at any time. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/mac80211_hwsim.c | 24 ++++++-- include/linux/ieee80211.h | 25 +++++++- include/net/mac80211.h | 59 ++++++++++++++++++ net/mac80211/cfg.c | 49 +++++++++++++++ net/mac80211/debugfs_netdev.c | 111 +++++++++++++++++++++++++++++++++- net/mac80211/driver-trace.h | 2 + net/mac80211/ht.c | 47 ++++++++++++++ net/mac80211/ieee80211_i.h | 14 +++++ net/mac80211/main.c | 24 ++++++++ net/mac80211/mlme.c | 63 +++++++++++++++++-- net/mac80211/status.c | 38 ++++++++++++ net/mac80211/util.c | 74 +++++++++++++++++++++++ 12 files changed, 518 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 88e41176e7f..92c669ebb35 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -618,12 +618,26 @@ static int mac80211_hwsim_config(struct ieee80211_hw *hw, u32 changed) { struct mac80211_hwsim_data *data = hw->priv; struct ieee80211_conf *conf = &hw->conf; - - printk(KERN_DEBUG "%s:%s (freq=%d idle=%d ps=%d)\n", + static const char *chantypes[4] = { + [NL80211_CHAN_NO_HT] = "noht", + [NL80211_CHAN_HT20] = "ht20", + [NL80211_CHAN_HT40MINUS] = "ht40-", + [NL80211_CHAN_HT40PLUS] = "ht40+", + }; + static const char *smps_modes[IEEE80211_SMPS_NUM_MODES] = { + [IEEE80211_SMPS_AUTOMATIC] = "auto", + [IEEE80211_SMPS_OFF] = "off", + [IEEE80211_SMPS_STATIC] = "static", + [IEEE80211_SMPS_DYNAMIC] = "dynamic", + }; + + printk(KERN_DEBUG "%s:%s (freq=%d/%s idle=%d ps=%d smps=%s)\n", wiphy_name(hw->wiphy), __func__, conf->channel->center_freq, + chantypes[conf->channel_type], !!(conf->flags & IEEE80211_CONF_IDLE), - !!(conf->flags & IEEE80211_CONF_PS)); + !!(conf->flags & IEEE80211_CONF_PS), + smps_modes[conf->smps_mode]); data->idle = !!(conf->flags & IEEE80211_CONF_IDLE); @@ -1082,7 +1096,9 @@ static int __init init_mac80211_hwsim(void) BIT(NL80211_IFTYPE_MESH_POINT); hw->flags = IEEE80211_HW_MFP_CAPABLE | - IEEE80211_HW_SIGNAL_DBM; + IEEE80211_HW_SIGNAL_DBM | + IEEE80211_HW_SUPPORTS_STATIC_SMPS | + IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS; /* ask mac80211 to reserve space for magic */ hw->vif_data_size = sizeof(struct hwsim_vif_priv); diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index d9724a28c0c..e8d43d0ff2c 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -707,6 +707,10 @@ struct ieee80211_mgmt { u8 action; u8 trans_id[WLAN_SA_QUERY_TR_ID_LEN]; } __attribute__ ((packed)) sa_query; + struct { + u8 action; + u8 smps_control; + } __attribute__ ((packed)) ht_smps; } u; } __attribute__ ((packed)) action; } u; @@ -824,6 +828,7 @@ struct ieee80211_ht_cap { #define IEEE80211_HT_CAP_LDPC_CODING 0x0001 #define IEEE80211_HT_CAP_SUP_WIDTH_20_40 0x0002 #define IEEE80211_HT_CAP_SM_PS 0x000C +#define IEEE80211_HT_CAP_SM_PS_SHIFT 2 #define IEEE80211_HT_CAP_GRN_FLD 0x0010 #define IEEE80211_HT_CAP_SGI_20 0x0020 #define IEEE80211_HT_CAP_SGI_40 0x0040 @@ -839,6 +844,7 @@ struct ieee80211_ht_cap { /* 802.11n HT capability AMPDU settings (for ampdu_params_info) */ #define IEEE80211_HT_AMPDU_PARM_FACTOR 0x03 #define IEEE80211_HT_AMPDU_PARM_DENSITY 0x1C +#define IEEE80211_HT_AMPDU_PARM_DENSITY_SHIFT 2 /* * Maximum length of AMPDU that the STA can receive. @@ -922,12 +928,17 @@ struct ieee80211_ht_info { #define IEEE80211_MAX_AMPDU_BUF 0x40 -/* Spatial Multiplexing Power Save Modes */ +/* Spatial Multiplexing Power Save Modes (for capability) */ #define WLAN_HT_CAP_SM_PS_STATIC 0 #define WLAN_HT_CAP_SM_PS_DYNAMIC 1 #define WLAN_HT_CAP_SM_PS_INVALID 2 #define WLAN_HT_CAP_SM_PS_DISABLED 3 +/* for SM power control field lower two bits */ +#define WLAN_HT_SMPS_CONTROL_DISABLED 0 +#define WLAN_HT_SMPS_CONTROL_STATIC 1 +#define WLAN_HT_SMPS_CONTROL_DYNAMIC 3 + /* Authentication algorithms */ #define WLAN_AUTH_OPEN 0 #define WLAN_AUTH_SHARED_KEY 1 @@ -1150,6 +1161,18 @@ enum ieee80211_spectrum_mgmt_actioncode { WLAN_ACTION_SPCT_CHL_SWITCH = 4, }; +/* HT action codes */ +enum ieee80211_ht_actioncode { + WLAN_HT_ACTION_NOTIFY_CHANWIDTH = 0, + WLAN_HT_ACTION_SMPS = 1, + WLAN_HT_ACTION_PSMP = 2, + WLAN_HT_ACTION_PCO_PHASE = 3, + WLAN_HT_ACTION_CSI = 4, + WLAN_HT_ACTION_NONCOMPRESSED_BF = 5, + WLAN_HT_ACTION_COMPRESSED_BF = 6, + WLAN_HT_ACTION_ASEL_IDX_FEEDBACK = 7, +}; + /* Security key length */ enum ieee80211_key_len { WLAN_KEY_LEN_WEP40 = 5, diff --git a/include/net/mac80211.h b/include/net/mac80211.h index e94cc526b0f..e6b6bf81d5b 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -597,8 +597,10 @@ enum ieee80211_conf_flags { * @IEEE80211_CONF_CHANGE_CHANNEL: the channel/channel_type changed * @IEEE80211_CONF_CHANGE_RETRY_LIMITS: retry limits changed * @IEEE80211_CONF_CHANGE_IDLE: Idle flag changed + * @IEEE80211_CONF_CHANGE_SMPS: Spatial multiplexing powersave mode changed */ enum ieee80211_conf_changed { + IEEE80211_CONF_CHANGE_SMPS = BIT(1), IEEE80211_CONF_CHANGE_LISTEN_INTERVAL = BIT(2), IEEE80211_CONF_CHANGE_MONITOR = BIT(3), IEEE80211_CONF_CHANGE_PS = BIT(4), @@ -608,6 +610,21 @@ enum ieee80211_conf_changed { IEEE80211_CONF_CHANGE_IDLE = BIT(8), }; +/** + * enum ieee80211_smps_mode - spatial multiplexing power save mode + * + * @ + */ +enum ieee80211_smps_mode { + IEEE80211_SMPS_AUTOMATIC, + IEEE80211_SMPS_OFF, + IEEE80211_SMPS_STATIC, + IEEE80211_SMPS_DYNAMIC, + + /* keep last */ + IEEE80211_SMPS_NUM_MODES, +}; + /** * struct ieee80211_conf - configuration of the device * @@ -636,6 +653,10 @@ enum ieee80211_conf_changed { * @short_frame_max_tx_count: Maximum number of transmissions for a "short" * frame, called "dot11ShortRetryLimit" in 802.11, but actually means the * number of transmissions not the number of retries + * + * @smps_mode: spatial multiplexing powersave mode; note that + * %IEEE80211_SMPS_STATIC is used when the device is not + * configured for an HT channel */ struct ieee80211_conf { u32 flags; @@ -648,6 +669,7 @@ struct ieee80211_conf { struct ieee80211_channel *channel; enum nl80211_channel_type channel_type; + enum ieee80211_smps_mode smps_mode; }; /** @@ -930,6 +952,16 @@ enum ieee80211_tkip_key_type { * @IEEE80211_HW_BEACON_FILTER: * Hardware supports dropping of irrelevant beacon frames to * avoid waking up cpu. + * + * @IEEE80211_HW_SUPPORTS_STATIC_SMPS: + * Hardware supports static spatial multiplexing powersave, + * ie. can turn off all but one chain even on HT connections + * that should be using more chains. + * + * @IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS: + * Hardware supports dynamic spatial multiplexing powersave, + * ie. can turn off all but one chain and then wake the rest + * up as required after, for example, rts/cts handshake. */ enum ieee80211_hw_flags { IEEE80211_HW_HAS_RATE_CONTROL = 1<<0, @@ -947,6 +979,8 @@ enum ieee80211_hw_flags { IEEE80211_HW_SUPPORTS_DYNAMIC_PS = 1<<12, IEEE80211_HW_MFP_CAPABLE = 1<<13, IEEE80211_HW_BEACON_FILTER = 1<<14, + IEEE80211_HW_SUPPORTS_STATIC_SMPS = 1<<15, + IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS = 1<<16, }; /** @@ -1214,6 +1248,31 @@ ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw, * signal strength threshold checking. */ +/** + * DOC: Spatial multiplexing power save + * + * SMPS (Spatial multiplexing power save) is a mechanism to conserve + * power in an 802.11n implementation. For details on the mechanism + * and rationale, please refer to 802.11 (as amended by 802.11n-2009) + * "11.2.3 SM power save". + * + * The mac80211 implementation is capable of sending action frames + * to update the AP about the station's SMPS mode, and will instruct + * the driver to enter the specific mode. It will also announce the + * requested SMPS mode during the association handshake. Hardware + * support for this feature is required, and can be indicated by + * hardware flags. + * + * The default mode will be "automatic", which nl80211/cfg80211 + * defines to be dynamic SMPS in (regular) powersave, and SMPS + * turned off otherwise. + * + * To support this feature, the driver must set the appropriate + * hardware support flags, and handle the SMPS flag to the config() + * operation. It will then with this mechanism be instructed to + * enter the requested SMPS mode while associated to an HT AP. + */ + /** * DOC: Frame filtering * diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index fcfa1bf776a..8c35418d1c9 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1318,6 +1318,50 @@ static int ieee80211_testmode_cmd(struct wiphy *wiphy, void *data, int len) } #endif +int __ieee80211_request_smps(struct ieee80211_sub_if_data *sdata, + enum ieee80211_smps_mode smps_mode) +{ + const u8 *ap; + enum ieee80211_smps_mode old_req; + int err; + + old_req = sdata->u.mgd.req_smps; + sdata->u.mgd.req_smps = smps_mode; + + if (old_req == smps_mode && + smps_mode != IEEE80211_SMPS_AUTOMATIC) + return 0; + + /* + * If not associated, or current association is not an HT + * association, there's no need to send an action frame. + */ + if (!sdata->u.mgd.associated || + sdata->local->oper_channel_type == NL80211_CHAN_NO_HT) { + mutex_lock(&sdata->local->iflist_mtx); + ieee80211_recalc_smps(sdata->local, sdata); + mutex_unlock(&sdata->local->iflist_mtx); + return 0; + } + + ap = sdata->u.mgd.associated->cbss.bssid; + + if (smps_mode == IEEE80211_SMPS_AUTOMATIC) { + if (sdata->u.mgd.powersave) + smps_mode = IEEE80211_SMPS_DYNAMIC; + else + smps_mode = IEEE80211_SMPS_OFF; + } + + /* send SM PS frame to AP */ + err = ieee80211_send_smps_action(sdata, smps_mode, + ap, ap); + if (err) + sdata->u.mgd.req_smps = old_req; + + return err; +} + static int ieee80211_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev, bool enabled, int timeout) { @@ -1335,6 +1379,11 @@ static int ieee80211_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev, sdata->u.mgd.powersave = enabled; conf->dynamic_ps_timeout = timeout; + /* no change, but if automatic follow powersave */ + mutex_lock(&sdata->u.mgd.mtx); + __ieee80211_request_smps(sdata, sdata->u.mgd.req_smps); + mutex_unlock(&sdata->u.mgd.mtx); + if (local->hw.flags & IEEE80211_HW_SUPPORTS_DYNAMIC_PS) ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_PS); diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 5d9c797635a..35598350388 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -41,6 +41,30 @@ static ssize_t ieee80211_if_read( return ret; } +static ssize_t ieee80211_if_write( + struct ieee80211_sub_if_data *sdata, + const char __user *userbuf, + size_t count, loff_t *ppos, + ssize_t (*write)(struct ieee80211_sub_if_data *, const char *, int)) +{ + u8 *buf; + ssize_t ret = -ENODEV; + + buf = kzalloc(count, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + if (copy_from_user(buf, userbuf, count)) + return -EFAULT; + + rtnl_lock(); + if (sdata->dev->reg_state == NETREG_REGISTERED) + ret = (*write)(sdata, buf, count); + rtnl_unlock(); + + return ret; +} + #define IEEE80211_IF_FMT(name, field, format_string) \ static ssize_t ieee80211_if_fmt_##name( \ const struct ieee80211_sub_if_data *sdata, char *buf, \ @@ -71,7 +95,7 @@ static ssize_t ieee80211_if_fmt_##name( \ return scnprintf(buf, buflen, "%pM\n", sdata->field); \ } -#define __IEEE80211_IF_FILE(name) \ +#define __IEEE80211_IF_FILE(name, _write) \ static ssize_t ieee80211_if_read_##name(struct file *file, \ char __user *userbuf, \ size_t count, loff_t *ppos) \ @@ -82,12 +106,24 @@ static ssize_t ieee80211_if_read_##name(struct file *file, \ } \ static const struct file_operations name##_ops = { \ .read = ieee80211_if_read_##name, \ + .write = (_write), \ .open = mac80211_open_file_generic, \ } +#define __IEEE80211_IF_FILE_W(name) \ +static ssize_t ieee80211_if_write_##name(struct file *file, \ + const char __user *userbuf, \ + size_t count, loff_t *ppos) \ +{ \ + return ieee80211_if_write(file->private_data, userbuf, count, \ + ppos, ieee80211_if_parse_##name); \ +} \ +__IEEE80211_IF_FILE(name, ieee80211_if_write_##name) + + #define IEEE80211_IF_FILE(name, field, format) \ IEEE80211_IF_FMT_##format(name, field) \ - __IEEE80211_IF_FILE(name) + __IEEE80211_IF_FILE(name, NULL) /* common attributes */ IEEE80211_IF_FILE(drop_unencrypted, drop_unencrypted, DEC); @@ -99,6 +135,70 @@ IEEE80211_IF_FILE(bssid, u.mgd.bssid, MAC); IEEE80211_IF_FILE(aid, u.mgd.aid, DEC); IEEE80211_IF_FILE(capab, u.mgd.capab, HEX); +static int ieee80211_set_smps(struct ieee80211_sub_if_data *sdata, + enum ieee80211_smps_mode smps_mode) +{ + struct ieee80211_local *local = sdata->local; + int err; + + if (!(local->hw.flags & IEEE80211_HW_SUPPORTS_STATIC_SMPS) && + smps_mode == IEEE80211_SMPS_STATIC) + return -EINVAL; + + /* auto should be dynamic if in PS mode */ + if (!(local->hw.flags & IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS) && + (smps_mode == IEEE80211_SMPS_DYNAMIC || + smps_mode == IEEE80211_SMPS_AUTOMATIC)) + return -EINVAL; + + /* supported only on managed interfaces for now */ + if (sdata->vif.type != NL80211_IFTYPE_STATION) + return -EOPNOTSUPP; + + mutex_lock(&local->iflist_mtx); + err = __ieee80211_request_smps(sdata, smps_mode); + mutex_unlock(&local->iflist_mtx); + + return err; +} + +static const char *smps_modes[IEEE80211_SMPS_NUM_MODES] = { + [IEEE80211_SMPS_AUTOMATIC] = "auto", + [IEEE80211_SMPS_OFF] = "off", + [IEEE80211_SMPS_STATIC] = "static", + [IEEE80211_SMPS_DYNAMIC] = "dynamic", +}; + +static ssize_t ieee80211_if_fmt_smps(const struct ieee80211_sub_if_data *sdata, + char *buf, int buflen) +{ + if (sdata->vif.type != NL80211_IFTYPE_STATION) + return -EOPNOTSUPP; + + return snprintf(buf, buflen, "request: %s\nused: %s\n", + smps_modes[sdata->u.mgd.req_smps], + smps_modes[sdata->u.mgd.ap_smps]); +} + +static ssize_t ieee80211_if_parse_smps(struct ieee80211_sub_if_data *sdata, + const char *buf, int buflen) +{ + enum ieee80211_smps_mode mode; + + for (mode = 0; mode < IEEE80211_SMPS_NUM_MODES; mode++) { + if (strncmp(buf, smps_modes[mode], buflen) == 0) { + int err = ieee80211_set_smps(sdata, mode); + if (!err) + return buflen; + return err; + } + } + + return -EINVAL; +} + +__IEEE80211_IF_FILE_W(smps); + /* AP attributes */ IEEE80211_IF_FILE(num_sta_ps, u.ap.num_sta_ps, ATOMIC); IEEE80211_IF_FILE(dtim_count, u.ap.dtim_count, DEC); @@ -109,7 +209,7 @@ static ssize_t ieee80211_if_fmt_num_buffered_multicast( return scnprintf(buf, buflen, "%u\n", skb_queue_len(&sdata->u.ap.ps_bc_buf)); } -__IEEE80211_IF_FILE(num_buffered_multicast); +__IEEE80211_IF_FILE(num_buffered_multicast, NULL); /* WDS attributes */ IEEE80211_IF_FILE(peer, u.wds.remote_addr, MAC); @@ -158,6 +258,10 @@ IEEE80211_IF_FILE(dot11MeshHWMPRootMode, debugfs_create_file(#name, 0400, sdata->debugfs.dir, \ sdata, &name##_ops); +#define DEBUGFS_ADD_MODE(name, mode) \ + debugfs_create_file(#name, mode, sdata->debugfs.dir, \ + sdata, &name##_ops); + static void add_sta_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, sta); @@ -167,6 +271,7 @@ static void add_sta_files(struct ieee80211_sub_if_data *sdata) DEBUGFS_ADD(bssid, sta); DEBUGFS_ADD(aid, sta); DEBUGFS_ADD(capab, sta); + DEBUGFS_ADD_MODE(smps, 0600); } static void add_ap_files(struct ieee80211_sub_if_data *sdata) diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index ee2d19a25ce..7a849b92016 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -140,6 +140,7 @@ TRACE_EVENT(drv_config, __field(u8, short_frame_max_tx_count) __field(int, center_freq) __field(int, channel_type) + __field(int, smps) ), TP_fast_assign( @@ -155,6 +156,7 @@ TRACE_EVENT(drv_config, __entry->short_frame_max_tx_count = local->hw.conf.short_frame_max_tx_count; __entry->center_freq = local->hw.conf.channel->center_freq; __entry->channel_type = local->hw.conf.channel_type; + __entry->smps = local->hw.conf.smps_mode; ), TP_printk( diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c index 45ebd062a2f..63b8f86b7f1 100644 --- a/net/mac80211/ht.c +++ b/net/mac80211/ht.c @@ -166,3 +166,50 @@ void ieee80211_process_delba(struct ieee80211_sub_if_data *sdata, spin_unlock_bh(&sta->lock); } } + +int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, + enum ieee80211_smps_mode smps, const u8 *da, + const u8 *bssid) +{ + struct ieee80211_local *local = sdata->local; + struct sk_buff *skb; + struct ieee80211_mgmt *action_frame; + + /* 27 = header + category + action + smps mode */ + skb = dev_alloc_skb(27 + local->hw.extra_tx_headroom); + if (!skb) + return -ENOMEM; + + skb_reserve(skb, local->hw.extra_tx_headroom); + action_frame = (void *)skb_put(skb, 27); + memcpy(action_frame->da, da, ETH_ALEN); + memcpy(action_frame->sa, sdata->dev->dev_addr, ETH_ALEN); + memcpy(action_frame->bssid, bssid, ETH_ALEN); + action_frame->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION); + action_frame->u.action.category = WLAN_CATEGORY_HT; + action_frame->u.action.u.ht_smps.action = WLAN_HT_ACTION_SMPS; + switch (smps) { + case IEEE80211_SMPS_AUTOMATIC: + case IEEE80211_SMPS_NUM_MODES: + WARN_ON(1); + case IEEE80211_SMPS_OFF: + action_frame->u.action.u.ht_smps.smps_control = + WLAN_HT_SMPS_CONTROL_DISABLED; + break; + case IEEE80211_SMPS_STATIC: + action_frame->u.action.u.ht_smps.smps_control = + WLAN_HT_SMPS_CONTROL_STATIC; + break; + case IEEE80211_SMPS_DYNAMIC: + action_frame->u.action.u.ht_smps.smps_control = + WLAN_HT_SMPS_CONTROL_DYNAMIC; + break; + } + + /* we'll do more on status of this frame */ + IEEE80211_SKB_CB(skb)->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS; + ieee80211_tx_skb(sdata, skb); + + return 0; +} diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 178e329f925..e63aecbddfb 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -297,6 +297,8 @@ struct ieee80211_if_managed { unsigned long timers_running; /* used for quiesce/restart */ bool powersave; /* powersave requested for this iface */ + enum ieee80211_smps_mode req_smps, /* requested smps mode */ + ap_smps; /* smps mode AP thinks we're in */ unsigned long request; @@ -587,6 +589,9 @@ struct ieee80211_local { /* used for uploading changed mc list */ struct work_struct reconfig_filter; + /* used to reconfigure hardware SM PS */ + struct work_struct recalc_smps; + /* aggregated multicast list */ struct dev_addr_list *mc_list; int mc_count; @@ -760,6 +765,8 @@ struct ieee80211_local { int user_power_level; /* in dBm */ int power_constr_level; /* in dBm */ + enum ieee80211_smps_mode smps_mode; + struct work_struct restart_work; #ifdef CONFIG_MAC80211_DEBUGFS @@ -978,6 +985,9 @@ void ieee80211_send_bar(struct ieee80211_sub_if_data *sdata, u8 *ra, u16 tid, u1 void ieee80211_send_delba(struct ieee80211_sub_if_data *sdata, const u8 *da, u16 tid, u16 initiator, u16 reason_code); +int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, + enum ieee80211_smps_mode smps, const u8 *da, + const u8 *bssid); void ieee80211_sta_stop_rx_ba_session(struct ieee80211_sub_if_data *sdata, u8 *da, u16 tid, u16 initiator, u16 reason); @@ -1088,6 +1098,10 @@ void ieee80211_sta_def_wmm_params(struct ieee80211_sub_if_data *sdata, u32 ieee80211_sta_get_rates(struct ieee80211_local *local, struct ieee802_11_elems *elems, enum ieee80211_band band); +int __ieee80211_request_smps(struct ieee80211_sub_if_data *sdata, + enum ieee80211_smps_mode smps_mode); +void ieee80211_recalc_smps(struct ieee80211_local *local, + struct ieee80211_sub_if_data *forsdata); #ifdef CONFIG_MAC80211_NOINLINE #define debug_noinline noinline diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 98320a94c27..e1293e8ed83 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -113,6 +113,18 @@ int ieee80211_hw_config(struct ieee80211_local *local, u32 changed) changed |= IEEE80211_CONF_CHANGE_CHANNEL; } + if (!conf_is_ht(&local->hw.conf)) { + /* + * mac80211.h documents that this is only valid + * when the channel is set to an HT type, and + * that otherwise STATIC is used. + */ + local->hw.conf.smps_mode = IEEE80211_SMPS_STATIC; + } else if (local->hw.conf.smps_mode != local->smps_mode) { + local->hw.conf.smps_mode = local->smps_mode; + changed |= IEEE80211_CONF_CHANGE_SMPS; + } + if (scan_chan) power = chan->max_power; else @@ -297,6 +309,16 @@ void ieee80211_restart_hw(struct ieee80211_hw *hw) } EXPORT_SYMBOL(ieee80211_restart_hw); +static void ieee80211_recalc_smps_work(struct work_struct *work) +{ + struct ieee80211_local *local = + container_of(work, struct ieee80211_local, recalc_smps); + + mutex_lock(&local->iflist_mtx); + ieee80211_recalc_smps(local, NULL); + mutex_unlock(&local->iflist_mtx); +} + struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, const struct ieee80211_ops *ops) { @@ -370,6 +392,8 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, INIT_WORK(&local->restart_work, ieee80211_restart_work); INIT_WORK(&local->reconfig_filter, ieee80211_reconfig_filter); + INIT_WORK(&local->recalc_smps, ieee80211_recalc_smps_work); + local->smps_mode = IEEE80211_SMPS_OFF; INIT_WORK(&local->dynamic_ps_enable_work, ieee80211_dynamic_ps_enable_work); diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index cd5dcc3d8c2..0a762a9ba4d 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -398,6 +398,8 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, __le16 tmp; u32 flags = local->hw.conf.channel->flags; + /* determine capability flags */ + switch (ht_info->ht_param & IEEE80211_HT_PARAM_CHA_SEC_OFFSET) { case IEEE80211_HT_PARAM_CHA_SEC_ABOVE: if (flags & IEEE80211_CHAN_NO_HT40PLUS) { @@ -413,17 +415,64 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, break; } - tmp = cpu_to_le16(cap); - pos = skb_put(skb, sizeof(struct ieee80211_ht_cap)+2); + /* set SM PS mode properly */ + cap &= ~IEEE80211_HT_CAP_SM_PS; + /* new association always uses requested smps mode */ + if (ifmgd->req_smps == IEEE80211_SMPS_AUTOMATIC) { + if (ifmgd->powersave) + ifmgd->ap_smps = IEEE80211_SMPS_DYNAMIC; + else + ifmgd->ap_smps = IEEE80211_SMPS_OFF; + } else + ifmgd->ap_smps = ifmgd->req_smps; + + switch (ifmgd->ap_smps) { + case IEEE80211_SMPS_AUTOMATIC: + case IEEE80211_SMPS_NUM_MODES: + WARN_ON(1); + case IEEE80211_SMPS_OFF: + cap |= WLAN_HT_CAP_SM_PS_DISABLED << + IEEE80211_HT_CAP_SM_PS_SHIFT; + break; + case IEEE80211_SMPS_STATIC: + cap |= WLAN_HT_CAP_SM_PS_STATIC << + IEEE80211_HT_CAP_SM_PS_SHIFT; + break; + case IEEE80211_SMPS_DYNAMIC: + cap |= WLAN_HT_CAP_SM_PS_DYNAMIC << + IEEE80211_HT_CAP_SM_PS_SHIFT; + break; + } + + /* reserve and fill IE */ + + pos = skb_put(skb, sizeof(struct ieee80211_ht_cap) + 2); *pos++ = WLAN_EID_HT_CAPABILITY; *pos++ = sizeof(struct ieee80211_ht_cap); memset(pos, 0, sizeof(struct ieee80211_ht_cap)); + + /* capability flags */ + tmp = cpu_to_le16(cap); memcpy(pos, &tmp, sizeof(u16)); pos += sizeof(u16); - /* TODO: needs a define here for << 2 */ + + /* AMPDU parameters */ *pos++ = sband->ht_cap.ampdu_factor | - (sband->ht_cap.ampdu_density << 2); + (sband->ht_cap.ampdu_density << + IEEE80211_HT_AMPDU_PARM_DENSITY_SHIFT); + + /* MCS set */ memcpy(pos, &sband->ht_cap.mcs, sizeof(sband->ht_cap.mcs)); + pos += sizeof(sband->ht_cap.mcs); + + /* extended capabilities */ + pos += sizeof(__le16); + + /* BF capabilities */ + pos += sizeof(__le32); + + /* antenna selection */ + pos += sizeof(u8); } IEEE80211_SKB_CB(skb)->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT; @@ -932,6 +981,7 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, mutex_lock(&local->iflist_mtx); ieee80211_recalc_ps(local, -1); + ieee80211_recalc_smps(local, sdata); mutex_unlock(&local->iflist_mtx); netif_start_queue(sdata->dev); @@ -2327,6 +2377,11 @@ void ieee80211_sta_setup_sdata(struct ieee80211_sub_if_data *sdata) ifmgd->flags |= IEEE80211_STA_WMM_ENABLED; mutex_init(&ifmgd->mtx); + + if (sdata->local->hw.flags & IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS) + ifmgd->req_smps = IEEE80211_SMPS_AUTOMATIC; + else + ifmgd->req_smps = IEEE80211_SMPS_OFF; } /* scan finished notification */ diff --git a/net/mac80211/status.c b/net/mac80211/status.c index b4608f11a40..0c0850d37dd 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -134,6 +134,40 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, dev_kfree_skb(skb); } +static void ieee80211_frame_acked(struct sta_info *sta, struct sk_buff *skb) +{ + struct ieee80211_mgmt *mgmt = (void *) skb->data; + struct ieee80211_local *local = sta->local; + struct ieee80211_sub_if_data *sdata = sta->sdata; + + if (ieee80211_is_action(mgmt->frame_control) && + sdata->vif.type == NL80211_IFTYPE_STATION && + mgmt->u.action.category == WLAN_CATEGORY_HT && + mgmt->u.action.u.ht_smps.action == WLAN_HT_ACTION_SMPS) { + /* + * This update looks racy, but isn't -- if we come + * here we've definitely got a station that we're + * talking to, and on a managed interface that can + * only be the AP. And the only other place updating + * this variable is before we're associated. + */ + switch (mgmt->u.action.u.ht_smps.smps_control) { + case WLAN_HT_SMPS_CONTROL_DYNAMIC: + sta->sdata->u.mgd.ap_smps = IEEE80211_SMPS_DYNAMIC; + break; + case WLAN_HT_SMPS_CONTROL_STATIC: + sta->sdata->u.mgd.ap_smps = IEEE80211_SMPS_STATIC; + break; + case WLAN_HT_SMPS_CONTROL_DISABLED: + default: /* shouldn't happen since we don't send that */ + sta->sdata->u.mgd.ap_smps = IEEE80211_SMPS_OFF; + break; + } + + ieee80211_queue_work(&local->hw, &local->recalc_smps); + } +} + void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) { struct sk_buff *skb2; @@ -210,6 +244,10 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) rate_control_tx_status(local, sband, sta, skb); if (ieee80211_vif_is_mesh(&sta->sdata->vif)) ieee80211s_update_metric(local, sta, skb); + + if (!(info->flags & IEEE80211_TX_CTL_INJECTED) && + (info->flags & IEEE80211_TX_STAT_ACK)) + ieee80211_frame_acked(sta, skb); } rcu_read_unlock(); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index d54dbe8e09b..086ef6257b4 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -1170,3 +1170,77 @@ int ieee80211_reconfig(struct ieee80211_local *local) return 0; } +static int check_mgd_smps(struct ieee80211_if_managed *ifmgd, + enum ieee80211_smps_mode *smps_mode) +{ + if (ifmgd->associated) { + *smps_mode = ifmgd->ap_smps; + + if (*smps_mode == IEEE80211_SMPS_AUTOMATIC) { + if (ifmgd->powersave) + *smps_mode = IEEE80211_SMPS_DYNAMIC; + else + *smps_mode = IEEE80211_SMPS_OFF; + } + + return 1; + } + + return 0; +} + +/* must hold iflist_mtx */ +void ieee80211_recalc_smps(struct ieee80211_local *local, + struct ieee80211_sub_if_data *forsdata) +{ + struct ieee80211_sub_if_data *sdata; + enum ieee80211_smps_mode smps_mode = IEEE80211_SMPS_OFF; + int count = 0; + + if (forsdata) + WARN_ON(!mutex_is_locked(&forsdata->u.mgd.mtx)); + + WARN_ON(!mutex_is_locked(&local->iflist_mtx)); + + /* + * This function could be improved to handle multiple + * interfaces better, but right now it makes any + * non-station interfaces force SM PS to be turned + * off. If there are multiple station interfaces it + * could also use the best possible mode, e.g. if + * one is in static and the other in dynamic then + * dynamic is ok. + */ + + list_for_each_entry(sdata, &local->interfaces, list) { + if (!netif_running(sdata->dev)) + continue; + if (sdata->vif.type != NL80211_IFTYPE_STATION) + goto set; + if (sdata != forsdata) { + /* + * This nested is ok -- we are holding the iflist_mtx + * so can't get here twice or so. But it's required + * since normally we acquire it first and then the + * iflist_mtx. + */ + mutex_lock_nested(&sdata->u.mgd.mtx, SINGLE_DEPTH_NESTING); + count += check_mgd_smps(&sdata->u.mgd, &smps_mode); + mutex_unlock(&sdata->u.mgd.mtx); + } else + count += check_mgd_smps(&sdata->u.mgd, &smps_mode); + + if (count > 1) { + smps_mode = IEEE80211_SMPS_OFF; + break; + } + } + + if (smps_mode == local->smps_mode) + return; + + set: + local->smps_mode = smps_mode; + /* changed flag is auto-detected for this */ + ieee80211_hw_config(local, 0); +} -- cgit v1.2.3-70-g09d2 From 9da3e068142ec7856b2f13261dcf0660fad32b61 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Mon, 7 Dec 2009 15:57:50 -0500 Subject: mac80211: only bother printing highest data rate on debugfs if its set IEEE-802.11n spec says the RX highest data rate field does not specify the highest supported RX data rate if its not set. Ignore it if not set then. Refer to section 7.3.56.4 Cc: johannes@sipsolutions.net Signed-off-by: Luis R. Rodriguez Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 5 ++++- net/mac80211/debugfs_sta.c | 12 ++++++++++-- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index e8d43d0ff2c..098bedcde9b 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -775,7 +775,10 @@ struct ieee80211_bar { /** * struct ieee80211_mcs_info - MCS information * @rx_mask: RX mask - * @rx_highest: highest supported RX rate + * @rx_highest: highest supported RX rate. If set represents + * the highest supported RX data rate in units of 1 Mbps. + * If this field is 0 this value should not be used to + * consider the highest RX data rate supported. * @tx_params: TX parameters */ struct ieee80211_mcs_info { diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index c833b6ce990..0d4a759ba72 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -218,11 +218,19 @@ static ssize_t sta_ht_capa_read(struct file *file, char __user *userbuf, p += scnprintf(p, sizeof(buf)+buf-p, "ampdu factor/density: %d/%d\n", htc->ampdu_factor, htc->ampdu_density); p += scnprintf(p, sizeof(buf)+buf-p, "MCS mask:"); + for (i = 0; i < IEEE80211_HT_MCS_MASK_LEN; i++) p += scnprintf(p, sizeof(buf)+buf-p, " %.2x", htc->mcs.rx_mask[i]); - p += scnprintf(p, sizeof(buf)+buf-p, "\nMCS rx highest: %d\n", - le16_to_cpu(htc->mcs.rx_highest)); + p += scnprintf(p, sizeof(buf)+buf-p, "\n"); + + /* If not set this is meaningless */ + if (le16_to_cpu(htc->mcs.rx_highest)) { + p += scnprintf(p, sizeof(buf)+buf-p, + "MCS rx highest: %d Mbps\n", + le16_to_cpu(htc->mcs.rx_highest)); + } + p += scnprintf(p, sizeof(buf)+buf-p, "MCS tx params: %x\n", htc->mcs.tx_params); } -- cgit v1.2.3-70-g09d2 From 12d50c46dc0f7fd2e625c4befaa5fa5740a7a594 Mon Sep 17 00:00:00 2001 From: Krishna Kumar Date: Tue, 8 Dec 2009 22:26:13 +0000 Subject: tcp: Remove check in __tcp_push_pending_frames MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit tcp_push checks tcp_send_head and calls __tcp_push_pending_frames, which again checks tcp_send_head, and this unnecessary check is done for every other caller of __tcp_push_pending_frames. Remove tcp_send_head check in __tcp_push_pending_frames and add the check to tcp_push_pending_frames. Other functions call __tcp_push_pending_frames only when tcp_send_head would evaluate to true. Signed-off-by: Krishna Kumar Acked-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/net/tcp.h | 16 +++++++++------- net/ipv4/tcp_output.c | 5 ----- 2 files changed, 9 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 34f5cc24d90..185e22baecb 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -856,13 +856,6 @@ static inline void tcp_check_probe_timer(struct sock *sk) icsk->icsk_rto, TCP_RTO_MAX); } -static inline void tcp_push_pending_frames(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - __tcp_push_pending_frames(sk, tcp_current_mss(sk), tp->nonagle); -} - static inline void tcp_init_wl(struct tcp_sock *tp, u32 seq) { tp->snd_wl1 = seq; @@ -1342,6 +1335,15 @@ static inline int tcp_write_queue_empty(struct sock *sk) return skb_queue_empty(&sk->sk_write_queue); } +static inline void tcp_push_pending_frames(struct sock *sk) +{ + if (tcp_send_head(sk)) { + struct tcp_sock *tp = tcp_sk(sk); + + __tcp_push_pending_frames(sk, tcp_current_mss(sk), tp->nonagle); + } +} + /* Start sequence of the highest skb with SACKed bit, valid only if * sacked > 0 or when the caller has ensured validity by itself. */ diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 383ce237640..12b2af36eab 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1794,11 +1794,6 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, int nonagle) { - struct sk_buff *skb = tcp_send_head(sk); - - if (!skb) - return; - /* If we are closed, the bytes will have to remain here. * In time closedown will finish, we empty the write queue and * all will be happy. -- cgit v1.2.3-70-g09d2 From 31d12926e37291970dd4f6e9940df3897766a81d Mon Sep 17 00:00:00 2001 From: laurent chavey Date: Tue, 15 Dec 2009 11:15:28 +0000 Subject: net: Add rtnetlink init_rcvwnd to set the TCP initial receive window Add rtnetlink init_rcvwnd to set the TCP initial receive window size advertised by passive and active TCP connections. The current Linux TCP implementation limits the advertised TCP initial receive window to the one prescribed by slow start. For short lived TCP connections used for transaction type of traffic (i.e. http requests), bounding the advertised TCP initial receive window results in increased latency to complete the transaction. Support for setting initial congestion window is already supported using rtnetlink init_cwnd, but the feature is useless without the ability to set a larger TCP initial receive window. The rtnetlink init_rcvwnd allows increasing the TCP initial receive window, allowing TCP connection to advertise larger TCP receive window than the ones bounded by slow start. Signed-off-by: Laurent Chavey Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 2 ++ include/net/dst.h | 2 -- include/net/tcp.h | 3 ++- net/ipv4/syncookies.c | 3 ++- net/ipv4/tcp_output.c | 17 +++++++++++++---- net/ipv6/syncookies.c | 3 ++- 6 files changed, 21 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 05330fc5b43..9590364fe8b 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -362,6 +362,8 @@ enum { #define RTAX_FEATURES RTAX_FEATURES RTAX_RTO_MIN, #define RTAX_RTO_MIN RTAX_RTO_MIN + RTAX_INITRWND, +#define RTAX_INITRWND RTAX_INITRWND __RTAX_MAX }; diff --git a/include/net/dst.h b/include/net/dst.h index 39c4a5963e1..ce078cda6b7 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -83,8 +83,6 @@ struct dst_entry { * (L1_CACHE_SIZE would be too much) */ #ifdef CONFIG_64BIT - long __pad_to_align_refcnt[2]; -#else long __pad_to_align_refcnt[1]; #endif /* diff --git a/include/net/tcp.h b/include/net/tcp.h index 185e22baecb..788c99f9859 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -965,7 +965,8 @@ static inline void tcp_sack_reset(struct tcp_options_received *rx_opt) /* Determine a window scaling and initial window to offer. */ extern void tcp_select_initial_window(int __space, __u32 mss, __u32 *rcv_wnd, __u32 *window_clamp, - int wscale_ok, __u8 *rcv_wscale); + int wscale_ok, __u8 *rcv_wscale, + __u32 init_rcv_wnd); static inline int tcp_win_from_space(int space) { diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 66fd80ef247..5c24db4a3c9 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -358,7 +358,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, tcp_select_initial_window(tcp_full_space(sk), req->mss, &req->rcv_wnd, &req->window_clamp, - ireq->wscale_ok, &rcv_wscale); + ireq->wscale_ok, &rcv_wscale, + dst_metric(&rt->u.dst, RTAX_INITRWND)); ireq->rcv_wscale = rcv_wscale; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 12b2af36eab..4a1605d3f90 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -183,7 +183,8 @@ static inline void tcp_event_ack_sent(struct sock *sk, unsigned int pkts) */ void tcp_select_initial_window(int __space, __u32 mss, __u32 *rcv_wnd, __u32 *window_clamp, - int wscale_ok, __u8 *rcv_wscale) + int wscale_ok, __u8 *rcv_wscale, + __u32 init_rcv_wnd) { unsigned int space = (__space < 0 ? 0 : __space); @@ -232,7 +233,13 @@ void tcp_select_initial_window(int __space, __u32 mss, init_cwnd = 2; else if (mss > 1460) init_cwnd = 3; - if (*rcv_wnd > init_cwnd * mss) + /* when initializing use the value from init_rcv_wnd + * rather than the default from above + */ + if (init_rcv_wnd && + (*rcv_wnd > init_rcv_wnd * mss)) + *rcv_wnd = init_rcv_wnd * mss; + else if (*rcv_wnd > init_cwnd * mss) *rcv_wnd = init_cwnd * mss; } @@ -2417,7 +2424,8 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, &req->rcv_wnd, &req->window_clamp, ireq->wscale_ok, - &rcv_wscale); + &rcv_wscale, + dst_metric(dst, RTAX_INITRWND)); ireq->rcv_wscale = rcv_wscale; } @@ -2544,7 +2552,8 @@ static void tcp_connect_init(struct sock *sk) &tp->rcv_wnd, &tp->window_clamp, sysctl_tcp_window_scaling, - &rcv_wscale); + &rcv_wscale, + dst_metric(dst, RTAX_INITRWND)); tp->rx_opt.rcv_wscale = rcv_wscale; tp->rcv_ssthresh = tp->rcv_wnd; diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 7208a06576c..34d1f0690d7 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -269,7 +269,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) req->window_clamp = tp->window_clamp ? :dst_metric(dst, RTAX_WINDOW); tcp_select_initial_window(tcp_full_space(sk), req->mss, &req->rcv_wnd, &req->window_clamp, - ireq->wscale_ok, &rcv_wscale); + ireq->wscale_ok, &rcv_wscale, + dst_metric(dst, RTAX_INITRWND)); ireq->rcv_wscale = rcv_wscale; -- cgit v1.2.3-70-g09d2 From e5cd6fe391aa8c93560bb7ffdfe334cf4d0a02e4 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sat, 26 Dec 2009 11:51:00 +0000 Subject: llc: add support for LLC_OPT_PKTINFO Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/linux/llc.h | 7 +++++++ include/net/llc_conn.h | 1 + net/llc/af_llc.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 37 insertions(+) (limited to 'include') diff --git a/include/linux/llc.h b/include/linux/llc.h index 7733585603f..ad7074ba81a 100644 --- a/include/linux/llc.h +++ b/include/linux/llc.h @@ -36,6 +36,7 @@ enum llc_sockopts { LLC_OPT_BUSY_TMR_EXP, /* busy state expire time (secs). */ LLC_OPT_TX_WIN, /* tx window size. */ LLC_OPT_RX_WIN, /* rx window size. */ + LLC_OPT_PKTINFO, /* ancillary packet information. */ LLC_OPT_MAX }; @@ -70,6 +71,12 @@ enum llc_sockopts { #define LLC_SAP_RM 0xD4 /* Resource Management */ #define LLC_SAP_GLOBAL 0xFF /* Global SAP. */ +struct llc_pktinfo { + int lpi_ifindex; + unsigned char lpi_sap; + unsigned char lpi_mac[IFHWADDRLEN]; +}; + #ifdef __KERNEL__ #define LLC_SAP_DYN_START 0xC0 #define LLC_SAP_DYN_STOP 0xDE diff --git a/include/net/llc_conn.h b/include/net/llc_conn.h index e2374e34989..fe982fd94c4 100644 --- a/include/net/llc_conn.h +++ b/include/net/llc_conn.h @@ -76,6 +76,7 @@ struct llc_sock { u32 rx_pdu_hdr; /* used for saving header of last pdu received and caused sending FRMR. Used for resending FRMR */ + u32 cmsg_flags; }; static inline struct llc_sock *llc_sk(const struct sock *sk) diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 3a66546cad0..ac691fe0807 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -47,6 +47,10 @@ static int llc_ui_wait_for_busy_core(struct sock *sk, long timeout); #define dprintk(args...) #endif +/* Maybe we'll add some more in the future. */ +#define LLC_CMSG_PKTINFO 1 + + /** * llc_ui_next_link_no - return the next unused link number for a sap * @sap: Address of sap to get link number from. @@ -591,6 +595,20 @@ static int llc_wait_data(struct sock *sk, long timeo) return rc; } +static void llc_cmsg_rcv(struct msghdr *msg, struct sk_buff *skb) +{ + struct llc_sock *llc = llc_sk(skb->sk); + + if (llc->cmsg_flags & LLC_CMSG_PKTINFO) { + struct llc_pktinfo info; + + info.lpi_ifindex = llc_sk(skb->sk)->dev->ifindex; + llc_pdu_decode_dsap(skb, &info.lpi_sap); + llc_pdu_decode_da(skb, info.lpi_mac); + put_cmsg(msg, SOL_LLC, LLC_OPT_PKTINFO, sizeof(info), &info); + } +} + /** * llc_ui_accept - accept a new incoming connection. * @sock: Socket which connections arrive on. @@ -812,6 +830,8 @@ copy_uaddr: memcpy(uaddr, llc_ui_skb_cb(skb), sizeof(*uaddr)); msg->msg_namelen = sizeof(*uaddr); } + if (llc_sk(sk)->cmsg_flags) + llc_cmsg_rcv(msg, skb); goto out; } @@ -1030,6 +1050,12 @@ static int llc_ui_setsockopt(struct socket *sock, int level, int optname, goto out; llc->rw = opt; break; + case LLC_OPT_PKTINFO: + if (opt) + llc->cmsg_flags |= LLC_CMSG_PKTINFO; + else + llc->cmsg_flags &= ~LLC_CMSG_PKTINFO; + break; default: rc = -ENOPROTOOPT; goto out; @@ -1083,6 +1109,9 @@ static int llc_ui_getsockopt(struct socket *sock, int level, int optname, val = llc->k; break; case LLC_OPT_RX_WIN: val = llc->rw; break; + case LLC_OPT_PKTINFO: + val = (llc->cmsg_flags & LLC_CMSG_PKTINFO) != 0; + break; default: rc = -ENOPROTOOPT; goto out; -- cgit v1.2.3-70-g09d2 From b76f5a8427ac2928c07fa4ff2144bb8db072c240 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sat, 26 Dec 2009 11:51:02 +0000 Subject: llc: convert the socket list to RCU locking For the reclamation phase we use the SLAB_DESTROY_BY_RCU mechanism, which require some extra checks in the lookup code: a) If the current socket was released, reallocated & inserted in another list it will short circuit the iteration for the current list, thus we need to restart the lookup. b) If the current socket was released, reallocated & inserted in the same list we just need to recheck it matches the look-up criteria and if not we can skip to the next element. In this case there is no need to restart the lookup, since sockets are inserted at the start of the list and the worst that will happen is that we will iterate throught some of the list elements more then once. Note that the /proc and multicast delivery was not yet converted to RCU, it still uses spinlocks for protection. Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/net/llc.h | 7 ++-- net/llc/af_llc.c | 1 + net/llc/llc_conn.c | 93 ++++++++++++++++++++++++++++++++++++------------------ net/llc/llc_core.c | 5 +-- net/llc/llc_proc.c | 22 ++++++------- net/llc/llc_sap.c | 66 ++++++++++++++++++++++++-------------- 6 files changed, 123 insertions(+), 71 deletions(-) (limited to 'include') diff --git a/include/net/llc.h b/include/net/llc.h index 7940da1606e..1559cf10e87 100644 --- a/include/net/llc.h +++ b/include/net/llc.h @@ -16,6 +16,7 @@ #include #include #include +#include #include @@ -53,10 +54,8 @@ struct llc_sap { struct net_device *orig_dev); struct llc_addr laddr; struct list_head node; - struct { - rwlock_t lock; - struct hlist_head list; - } sk_list; + spinlock_t sk_lock; + struct hlist_nulls_head sk_list; }; #define LLC_DEST_INVALID 0 /* Invalid LLC PDU type */ diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index c4d1a1da813..f49f3dd6fbd 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -140,6 +140,7 @@ static struct proto llc_proto = { .name = "LLC", .owner = THIS_MODULE, .obj_size = sizeof(struct llc_sock), + .slab_flags = SLAB_DESTROY_BY_RCU, }; /** diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index c6bab39b018..77bb3816655 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -468,6 +468,19 @@ static int llc_exec_conn_trans_actions(struct sock *sk, return rc; } +static inline bool llc_estab_match(const struct llc_sap *sap, + const struct llc_addr *daddr, + const struct llc_addr *laddr, + const struct sock *sk) +{ + struct llc_sock *llc = llc_sk(sk); + + return llc->laddr.lsap == laddr->lsap && + llc->daddr.lsap == daddr->lsap && + llc_mac_match(llc->laddr.mac, laddr->mac) && + llc_mac_match(llc->daddr.mac, daddr->mac); +} + /** * __llc_lookup_established - Finds connection for the remote/local sap/mac * @sap: SAP @@ -484,23 +497,26 @@ static struct sock *__llc_lookup_established(struct llc_sap *sap, struct llc_addr *laddr) { struct sock *rc; - struct hlist_node *node; - - read_lock(&sap->sk_list.lock); - sk_for_each(rc, node, &sap->sk_list.list) { - struct llc_sock *llc = llc_sk(rc); - - if (llc->laddr.lsap == laddr->lsap && - llc->daddr.lsap == daddr->lsap && - llc_mac_match(llc->laddr.mac, laddr->mac) && - llc_mac_match(llc->daddr.mac, daddr->mac)) { - sock_hold(rc); + struct hlist_nulls_node *node; + + rcu_read_lock(); +again: + sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + if (llc_estab_match(sap, daddr, laddr, rc)) { + /* Extra checks required by SLAB_DESTROY_BY_RCU */ + if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) + goto again; + if (unlikely(llc_sk(rc)->sap != sap || + !llc_estab_match(sap, daddr, laddr, rc))) { + sock_put(rc); + continue; + } goto found; } } rc = NULL; found: - read_unlock(&sap->sk_list.lock); + rcu_read_unlock(); return rc; } @@ -516,6 +532,18 @@ struct sock *llc_lookup_established(struct llc_sap *sap, return sk; } +static inline bool llc_listener_match(const struct llc_sap *sap, + const struct llc_addr *laddr, + const struct sock *sk) +{ + struct llc_sock *llc = llc_sk(sk); + + return sk->sk_type == SOCK_STREAM && sk->sk_state == TCP_LISTEN && + llc->laddr.lsap == laddr->lsap && + (llc_mac_match(llc->laddr.mac, laddr->mac) || + llc_mac_null(llc->laddr.mac)); +} + /** * llc_lookup_listener - Finds listener for local MAC + SAP * @sap: SAP @@ -530,23 +558,26 @@ static struct sock *llc_lookup_listener(struct llc_sap *sap, struct llc_addr *laddr) { struct sock *rc; - struct hlist_node *node; - - read_lock(&sap->sk_list.lock); - sk_for_each(rc, node, &sap->sk_list.list) { - struct llc_sock *llc = llc_sk(rc); - - if (rc->sk_type == SOCK_STREAM && rc->sk_state == TCP_LISTEN && - llc->laddr.lsap == laddr->lsap && - (llc_mac_match(llc->laddr.mac, laddr->mac) || - llc_mac_null(llc->laddr.mac))) { - sock_hold(rc); + struct hlist_nulls_node *node; + + rcu_read_lock(); +again: + sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + if (llc_listener_match(sap, laddr, rc)) { + /* Extra checks required by SLAB_DESTROY_BY_RCU */ + if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) + goto again; + if (unlikely(llc_sk(rc)->sap != sap || + !llc_listener_match(sap, laddr, rc))) { + sock_put(rc); + continue; + } goto found; } } rc = NULL; found: - read_unlock(&sap->sk_list.lock); + rcu_read_unlock(); return rc; } @@ -652,10 +683,10 @@ static int llc_find_offset(int state, int ev_type) void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) { llc_sap_hold(sap); - write_lock_bh(&sap->sk_list.lock); + spin_lock_bh(&sap->sk_lock); llc_sk(sk)->sap = sap; - sk_add_node(sk, &sap->sk_list.list); - write_unlock_bh(&sap->sk_list.lock); + sk_nulls_add_node_rcu(sk, &sap->sk_list); + spin_unlock_bh(&sap->sk_lock); } /** @@ -663,14 +694,14 @@ void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) * @sap: SAP * @sk: socket * - * This function removes a connection from sk_list.list of a SAP if + * This function removes a connection from sk_list of a SAP if * the connection was in this list. */ void llc_sap_remove_socket(struct llc_sap *sap, struct sock *sk) { - write_lock_bh(&sap->sk_list.lock); - sk_del_node_init(sk); - write_unlock_bh(&sap->sk_list.lock); + spin_lock_bh(&sap->sk_lock); + sk_nulls_del_node_init_rcu(sk); + spin_unlock_bh(&sap->sk_lock); llc_sap_put(sap); } diff --git a/net/llc/llc_core.c b/net/llc/llc_core.c index ff4c0ab96a6..5276b972207 100644 --- a/net/llc/llc_core.c +++ b/net/llc/llc_core.c @@ -37,7 +37,8 @@ static struct llc_sap *llc_sap_alloc(void) if (sap) { /* sap->laddr.mac - leave as a null, it's filled by bind */ sap->state = LLC_SAP_STATE_ACTIVE; - rwlock_init(&sap->sk_list.lock); + spin_lock_init(&sap->sk_lock); + INIT_HLIST_NULLS_HEAD(&sap->sk_list, 0); atomic_set(&sap->refcnt, 1); } return sap; @@ -142,7 +143,7 @@ out: */ void llc_sap_close(struct llc_sap *sap) { - WARN_ON(!hlist_empty(&sap->sk_list.list)); + WARN_ON(!hlist_nulls_empty(&sap->sk_list)); llc_del_sap(sap); kfree(sap); } diff --git a/net/llc/llc_proc.c b/net/llc/llc_proc.c index be47ac427f6..6b3d033b323 100644 --- a/net/llc/llc_proc.c +++ b/net/llc/llc_proc.c @@ -34,19 +34,19 @@ static struct sock *llc_get_sk_idx(loff_t pos) { struct list_head *sap_entry; struct llc_sap *sap; - struct hlist_node *node; + struct hlist_nulls_node *node; struct sock *sk = NULL; list_for_each(sap_entry, &llc_sap_list) { sap = list_entry(sap_entry, struct llc_sap, node); - read_lock_bh(&sap->sk_list.lock); - sk_for_each(sk, node, &sap->sk_list.list) { + spin_lock_bh(&sap->sk_lock); + sk_nulls_for_each(sk, node, &sap->sk_list) { if (!pos) goto found; --pos; } - read_unlock_bh(&sap->sk_list.lock); + spin_unlock_bh(&sap->sk_lock); } sk = NULL; found: @@ -73,25 +73,25 @@ static void *llc_seq_next(struct seq_file *seq, void *v, loff_t *pos) goto out; } sk = v; - next = sk_next(sk); + next = sk_nulls_next(sk); if (next) { sk = next; goto out; } llc = llc_sk(sk); sap = llc->sap; - read_unlock_bh(&sap->sk_list.lock); + spin_unlock_bh(&sap->sk_lock); sk = NULL; for (;;) { if (sap->node.next == &llc_sap_list) break; sap = list_entry(sap->node.next, struct llc_sap, node); - read_lock_bh(&sap->sk_list.lock); - if (!hlist_empty(&sap->sk_list.list)) { - sk = sk_head(&sap->sk_list.list); + spin_lock_bh(&sap->sk_lock); + if (!hlist_nulls_empty(&sap->sk_list)) { + sk = sk_nulls_head(&sap->sk_list); break; } - read_unlock_bh(&sap->sk_list.lock); + spin_unlock_bh(&sap->sk_lock); } out: return sk; @@ -104,7 +104,7 @@ static void llc_seq_stop(struct seq_file *seq, void *v) struct llc_sock *llc = llc_sk(sk); struct llc_sap *sap = llc->sap; - read_unlock_bh(&sap->sk_list.lock); + spin_unlock_bh(&sap->sk_lock); } read_unlock_bh(&llc_sap_list_lock); } diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c index 008de1fc42c..39760d013ce 100644 --- a/net/llc/llc_sap.c +++ b/net/llc/llc_sap.c @@ -297,6 +297,17 @@ static void llc_sap_rcv(struct llc_sap *sap, struct sk_buff *skb, llc_sap_state_process(sap, skb); } +static inline bool llc_dgram_match(const struct llc_sap *sap, + const struct llc_addr *laddr, + const struct sock *sk) +{ + struct llc_sock *llc = llc_sk(sk); + + return sk->sk_type == SOCK_DGRAM && + llc->laddr.lsap == laddr->lsap && + llc_mac_match(llc->laddr.mac, laddr->mac); +} + /** * llc_lookup_dgram - Finds dgram socket for the local sap/mac * @sap: SAP @@ -309,25 +320,41 @@ static struct sock *llc_lookup_dgram(struct llc_sap *sap, const struct llc_addr *laddr) { struct sock *rc; - struct hlist_node *node; - - read_lock_bh(&sap->sk_list.lock); - sk_for_each(rc, node, &sap->sk_list.list) { - struct llc_sock *llc = llc_sk(rc); - - if (rc->sk_type == SOCK_DGRAM && - llc->laddr.lsap == laddr->lsap && - llc_mac_match(llc->laddr.mac, laddr->mac)) { - sock_hold(rc); + struct hlist_nulls_node *node; + + rcu_read_lock_bh(); +again: + sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + if (llc_dgram_match(sap, laddr, rc)) { + /* Extra checks required by SLAB_DESTROY_BY_RCU */ + if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) + goto again; + if (unlikely(llc_sk(rc)->sap != sap || + !llc_dgram_match(sap, laddr, rc))) { + sock_put(rc); + continue; + } goto found; } } rc = NULL; found: - read_unlock_bh(&sap->sk_list.lock); + rcu_read_unlock_bh(); return rc; } +static inline bool llc_mcast_match(const struct llc_sap *sap, + const struct llc_addr *laddr, + const struct sk_buff *skb, + const struct sock *sk) +{ + struct llc_sock *llc = llc_sk(sk); + + return sk->sk_type == SOCK_DGRAM && + llc->laddr.lsap == laddr->lsap && + llc->dev == skb->dev; +} + /** * llc_sap_mcast - Deliver multicast PDU's to all matching datagram sockets. * @sap: SAP @@ -341,20 +368,13 @@ static void llc_sap_mcast(struct llc_sap *sap, struct sk_buff *skb) { struct sock *sk; - struct hlist_node *node; + struct hlist_nulls_node *node; - read_lock_bh(&sap->sk_list.lock); - sk_for_each(sk, node, &sap->sk_list.list) { - struct llc_sock *llc = llc_sk(sk); + spin_lock_bh(&sap->sk_lock); + sk_nulls_for_each_rcu(sk, node, &sap->sk_list) { struct sk_buff *skb1; - if (sk->sk_type != SOCK_DGRAM) - continue; - - if (llc->laddr.lsap != laddr->lsap) - continue; - - if (llc->dev != skb->dev) + if (!llc_mcast_match(sap, laddr, skb, sk)) continue; skb1 = skb_clone(skb, GFP_ATOMIC); @@ -365,7 +385,7 @@ static void llc_sap_mcast(struct llc_sap *sap, llc_sap_rcv(sap, skb1, sk); sock_put(sk); } - read_unlock_bh(&sap->sk_list.lock); + spin_unlock_bh(&sap->sk_lock); } -- cgit v1.2.3-70-g09d2 From 6d2e3ea284463d5ab34e9cf2a41d0b8627b95d02 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sat, 26 Dec 2009 11:51:04 +0000 Subject: llc: use a device based hash table to speed up multicast delivery This patch adds a per SAP device based hash table to solve the multicast delivery scalability issue when we have large number of interfaces and a large number of sockets bound to the same SAP. Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/net/llc.h | 11 +++++++++++ include/net/llc_conn.h | 1 + net/llc/llc_conn.c | 10 +++++++++- net/llc/llc_sap.c | 8 ++++++-- 4 files changed, 27 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/llc.h b/include/net/llc.h index 1559cf10e87..dbef5917905 100644 --- a/include/net/llc.h +++ b/include/net/llc.h @@ -32,6 +32,9 @@ struct llc_addr { #define LLC_SAP_STATE_INACTIVE 1 #define LLC_SAP_STATE_ACTIVE 2 +#define LLC_SK_DEV_HASH_BITS 6 +#define LLC_SK_DEV_HASH_ENTRIES (1<sk_dev_hash[ifindex % LLC_SK_DEV_HASH_ENTRIES]; +} + + #define LLC_DEST_INVALID 0 /* Invalid LLC PDU type */ #define LLC_DEST_SAP 1 /* Type 1 goes here */ #define LLC_DEST_CONN 2 /* Type 2 goes here */ diff --git a/include/net/llc_conn.h b/include/net/llc_conn.h index fe982fd94c4..2f97d8ddce9 100644 --- a/include/net/llc_conn.h +++ b/include/net/llc_conn.h @@ -77,6 +77,7 @@ struct llc_sock { received and caused sending FRMR. Used for resending FRMR */ u32 cmsg_flags; + struct hlist_node dev_hash_node; }; static inline struct llc_sock *llc_sk(const struct sock *sk) diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 77bb3816655..10cdfe2db83 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -682,10 +682,15 @@ static int llc_find_offset(int state, int ev_type) */ void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) { + struct llc_sock *llc = llc_sk(sk); + struct hlist_head *dev_hb = llc_sk_dev_hash(sap, llc->dev->ifindex); + llc_sap_hold(sap); - spin_lock_bh(&sap->sk_lock); llc_sk(sk)->sap = sap; + + spin_lock_bh(&sap->sk_lock); sk_nulls_add_node_rcu(sk, &sap->sk_list); + hlist_add_head(&llc->dev_hash_node, dev_hb); spin_unlock_bh(&sap->sk_lock); } @@ -699,8 +704,11 @@ void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) */ void llc_sap_remove_socket(struct llc_sap *sap, struct sock *sk) { + struct llc_sock *llc = llc_sk(sk); + spin_lock_bh(&sap->sk_lock); sk_nulls_del_node_init_rcu(sk); + hlist_del(&llc->dev_hash_node); spin_unlock_bh(&sap->sk_lock); llc_sap_put(sap); } diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c index 94790e60d07..94cb706f6cc 100644 --- a/net/llc/llc_sap.c +++ b/net/llc/llc_sap.c @@ -387,10 +387,14 @@ static void llc_sap_mcast(struct llc_sap *sap, { int i = 0, count = 256 / sizeof(struct sock *); struct sock *sk, *stack[count]; - struct hlist_nulls_node *node; + struct hlist_node *node; + struct llc_sock *llc; + struct hlist_head *dev_hb = llc_sk_dev_hash(sap, skb->dev->ifindex); spin_lock_bh(&sap->sk_lock); - sk_nulls_for_each_rcu(sk, node, &sap->sk_list) { + hlist_for_each_entry(llc, node, dev_hb, dev_hash_node) { + + sk = &llc->sk; if (!llc_mcast_match(sap, laddr, skb, sk)) continue; -- cgit v1.2.3-70-g09d2 From 52d58aef5ee460fedd7f250f05e79081019f2c79 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sat, 26 Dec 2009 11:51:05 +0000 Subject: llc: replace the socket list with a local address based hash For the cases where a lot of interfaces are used in conjunction with a lot of LLC sockets bound to the same SAP, the iteration of the socket list becomes prohibitively expensive. Replacing the list with a a local address based hash significantly improves the bind and listener lookup operations as well as the datagram delivery. Connected sockets delivery is also improved, but this patch does not address the case where we have lots of sockets with the same local address connected to different remote addresses. In order to keep the socket sanity checks alive and fast a socket counter was added to the SAP structure. Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/net/llc.h | 21 +++++++++++++++- net/llc/llc_conn.c | 70 +++++++++++++++++++++++++++++++++++++++--------------- net/llc/llc_core.c | 6 +++-- net/llc/llc_proc.c | 44 +++++++++++++++++++++++----------- net/llc/llc_sap.c | 11 ++++++++- 5 files changed, 115 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/include/net/llc.h b/include/net/llc.h index dbef5917905..709a9b37e23 100644 --- a/include/net/llc.h +++ b/include/net/llc.h @@ -17,6 +17,8 @@ #include #include #include +#include +#include #include @@ -35,6 +37,9 @@ struct llc_addr { #define LLC_SK_DEV_HASH_BITS 6 #define LLC_SK_DEV_HASH_ENTRIES (1<sk_dev_hash[ifindex % LLC_SK_DEV_HASH_ENTRIES]; } +static inline +u32 llc_sk_laddr_hashfn(struct llc_sap *sap, const struct llc_addr *laddr) +{ + return hash_32(jhash(laddr->mac, sizeof(laddr->mac), 0), + LLC_SK_LADDR_HASH_BITS); +} + +static inline +struct hlist_nulls_head *llc_sk_laddr_hash(struct llc_sap *sap, + const struct llc_addr *laddr) +{ + return &sap->sk_laddr_hash[llc_sk_laddr_hashfn(sap, laddr)]; +} #define LLC_DEST_INVALID 0 /* Invalid LLC PDU type */ #define LLC_DEST_SAP 1 /* Type 1 goes here */ diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 10cdfe2db83..a8dde9b010d 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -498,10 +498,12 @@ static struct sock *__llc_lookup_established(struct llc_sap *sap, { struct sock *rc; struct hlist_nulls_node *node; + int slot = llc_sk_laddr_hashfn(sap, laddr); + struct hlist_nulls_head *laddr_hb = &sap->sk_laddr_hash[slot]; rcu_read_lock(); again: - sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + sk_nulls_for_each_rcu(rc, node, laddr_hb) { if (llc_estab_match(sap, daddr, laddr, rc)) { /* Extra checks required by SLAB_DESTROY_BY_RCU */ if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) @@ -515,6 +517,13 @@ again: } } rc = NULL; + /* + * if the nulls value we got at the end of this lookup is + * not the expected one, we must restart lookup. + * We probably met an item that was moved to another chain. + */ + if (unlikely(get_nulls_value(node) != slot)) + goto again; found: rcu_read_unlock(); return rc; @@ -540,29 +549,20 @@ static inline bool llc_listener_match(const struct llc_sap *sap, return sk->sk_type == SOCK_STREAM && sk->sk_state == TCP_LISTEN && llc->laddr.lsap == laddr->lsap && - (llc_mac_match(llc->laddr.mac, laddr->mac) || - llc_mac_null(llc->laddr.mac)); + llc_mac_match(llc->laddr.mac, laddr->mac); } -/** - * llc_lookup_listener - Finds listener for local MAC + SAP - * @sap: SAP - * @laddr: address of local LLC (MAC + SAP) - * - * Search connection list of the SAP and finds connection listening on - * local mac, and local sap. Returns pointer for parent socket found, - * %NULL otherwise. - * Caller has to make sure local_bh is disabled. - */ -static struct sock *llc_lookup_listener(struct llc_sap *sap, - struct llc_addr *laddr) +static struct sock *__llc_lookup_listener(struct llc_sap *sap, + struct llc_addr *laddr) { struct sock *rc; struct hlist_nulls_node *node; + int slot = llc_sk_laddr_hashfn(sap, laddr); + struct hlist_nulls_head *laddr_hb = &sap->sk_laddr_hash[slot]; rcu_read_lock(); again: - sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + sk_nulls_for_each_rcu(rc, node, laddr_hb) { if (llc_listener_match(sap, laddr, rc)) { /* Extra checks required by SLAB_DESTROY_BY_RCU */ if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) @@ -576,11 +576,40 @@ again: } } rc = NULL; + /* + * if the nulls value we got at the end of this lookup is + * not the expected one, we must restart lookup. + * We probably met an item that was moved to another chain. + */ + if (unlikely(get_nulls_value(node) != slot)) + goto again; found: rcu_read_unlock(); return rc; } +/** + * llc_lookup_listener - Finds listener for local MAC + SAP + * @sap: SAP + * @laddr: address of local LLC (MAC + SAP) + * + * Search connection list of the SAP and finds connection listening on + * local mac, and local sap. Returns pointer for parent socket found, + * %NULL otherwise. + * Caller has to make sure local_bh is disabled. + */ +static struct sock *llc_lookup_listener(struct llc_sap *sap, + struct llc_addr *laddr) +{ + static struct llc_addr null_addr; + struct sock *rc = __llc_lookup_listener(sap, laddr); + + if (!rc) + rc = __llc_lookup_listener(sap, &null_addr); + + return rc; +} + static struct sock *__llc_lookup(struct llc_sap *sap, struct llc_addr *daddr, struct llc_addr *laddr) @@ -678,18 +707,20 @@ static int llc_find_offset(int state, int ev_type) * @sap: SAP * @sk: socket * - * This function adds a socket to sk_list of a SAP. + * This function adds a socket to the hash tables of a SAP. */ void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) { struct llc_sock *llc = llc_sk(sk); struct hlist_head *dev_hb = llc_sk_dev_hash(sap, llc->dev->ifindex); + struct hlist_nulls_head *laddr_hb = llc_sk_laddr_hash(sap, &llc->laddr); llc_sap_hold(sap); llc_sk(sk)->sap = sap; spin_lock_bh(&sap->sk_lock); - sk_nulls_add_node_rcu(sk, &sap->sk_list); + sap->sk_count++; + sk_nulls_add_node_rcu(sk, laddr_hb); hlist_add_head(&llc->dev_hash_node, dev_hb); spin_unlock_bh(&sap->sk_lock); } @@ -699,7 +730,7 @@ void llc_sap_add_socket(struct llc_sap *sap, struct sock *sk) * @sap: SAP * @sk: socket * - * This function removes a connection from sk_list of a SAP if + * This function removes a connection from the hash tables of a SAP if * the connection was in this list. */ void llc_sap_remove_socket(struct llc_sap *sap, struct sock *sk) @@ -709,6 +740,7 @@ void llc_sap_remove_socket(struct llc_sap *sap, struct sock *sk) spin_lock_bh(&sap->sk_lock); sk_nulls_del_node_init_rcu(sk); hlist_del(&llc->dev_hash_node); + sap->sk_count--; spin_unlock_bh(&sap->sk_lock); llc_sap_put(sap); } diff --git a/net/llc/llc_core.c b/net/llc/llc_core.c index 5276b972207..0c9ef8bc765 100644 --- a/net/llc/llc_core.c +++ b/net/llc/llc_core.c @@ -33,12 +33,14 @@ DEFINE_RWLOCK(llc_sap_list_lock); static struct llc_sap *llc_sap_alloc(void) { struct llc_sap *sap = kzalloc(sizeof(*sap), GFP_ATOMIC); + int i; if (sap) { /* sap->laddr.mac - leave as a null, it's filled by bind */ sap->state = LLC_SAP_STATE_ACTIVE; spin_lock_init(&sap->sk_lock); - INIT_HLIST_NULLS_HEAD(&sap->sk_list, 0); + for (i = 0; i < LLC_SK_LADDR_HASH_ENTRIES; i++) + INIT_HLIST_NULLS_HEAD(&sap->sk_laddr_hash[i], i); atomic_set(&sap->refcnt, 1); } return sap; @@ -143,7 +145,7 @@ out: */ void llc_sap_close(struct llc_sap *sap) { - WARN_ON(!hlist_nulls_empty(&sap->sk_list)); + WARN_ON(sap->sk_count); llc_del_sap(sap); kfree(sap); } diff --git a/net/llc/llc_proc.c b/net/llc/llc_proc.c index 6b3d033b323..09dec630720 100644 --- a/net/llc/llc_proc.c +++ b/net/llc/llc_proc.c @@ -34,17 +34,22 @@ static struct sock *llc_get_sk_idx(loff_t pos) { struct list_head *sap_entry; struct llc_sap *sap; - struct hlist_nulls_node *node; struct sock *sk = NULL; + int i; list_for_each(sap_entry, &llc_sap_list) { sap = list_entry(sap_entry, struct llc_sap, node); spin_lock_bh(&sap->sk_lock); - sk_nulls_for_each(sk, node, &sap->sk_list) { - if (!pos) - goto found; - --pos; + for (i = 0; i < LLC_SK_LADDR_HASH_ENTRIES; i++) { + struct hlist_nulls_head *head = &sap->sk_laddr_hash[i]; + struct hlist_nulls_node *node; + + sk_nulls_for_each(sk, node, head) { + if (!pos) + goto found; /* keep the lock */ + --pos; + } } spin_unlock_bh(&sap->sk_lock); } @@ -61,6 +66,19 @@ static void *llc_seq_start(struct seq_file *seq, loff_t *pos) return l ? llc_get_sk_idx(--l) : SEQ_START_TOKEN; } +static struct sock *laddr_hash_next(struct llc_sap *sap, int bucket) +{ + struct hlist_nulls_node *node; + struct sock *sk = NULL; + + while (++bucket < LLC_SK_LADDR_HASH_ENTRIES) + sk_nulls_for_each(sk, node, &sap->sk_laddr_hash[bucket]) + goto out; + +out: + return sk; +} + static void *llc_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct sock* sk, *next; @@ -80,17 +98,15 @@ static void *llc_seq_next(struct seq_file *seq, void *v, loff_t *pos) } llc = llc_sk(sk); sap = llc->sap; + sk = laddr_hash_next(sap, llc_sk_laddr_hashfn(sap, &llc->laddr)); + if (sk) + goto out; spin_unlock_bh(&sap->sk_lock); - sk = NULL; - for (;;) { - if (sap->node.next == &llc_sap_list) - break; - sap = list_entry(sap->node.next, struct llc_sap, node); + list_for_each_entry_continue(sap, &llc_sap_list, node) { spin_lock_bh(&sap->sk_lock); - if (!hlist_nulls_empty(&sap->sk_list)) { - sk = sk_nulls_head(&sap->sk_list); - break; - } + sk = laddr_hash_next(sap, -1); + if (sk) + break; /* keep the lock */ spin_unlock_bh(&sap->sk_lock); } out: diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c index 94cb706f6cc..ad6e6e1cf22 100644 --- a/net/llc/llc_sap.c +++ b/net/llc/llc_sap.c @@ -321,10 +321,12 @@ static struct sock *llc_lookup_dgram(struct llc_sap *sap, { struct sock *rc; struct hlist_nulls_node *node; + int slot = llc_sk_laddr_hashfn(sap, laddr); + struct hlist_nulls_head *laddr_hb = &sap->sk_laddr_hash[slot]; rcu_read_lock_bh(); again: - sk_nulls_for_each_rcu(rc, node, &sap->sk_list) { + sk_nulls_for_each_rcu(rc, node, laddr_hb) { if (llc_dgram_match(sap, laddr, rc)) { /* Extra checks required by SLAB_DESTROY_BY_RCU */ if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt))) @@ -338,6 +340,13 @@ again: } } rc = NULL; + /* + * if the nulls value we got at the end of this lookup is + * not the expected one, we must restart lookup. + * We probably met an item that was moved to another chain. + */ + if (unlikely(get_nulls_value(node) != slot)) + goto again; found: rcu_read_unlock_bh(); return rc; -- cgit v1.2.3-70-g09d2 From 8beb9ab6c2df203e8d68cb1f48cf42604a6bed86 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sat, 26 Dec 2009 11:51:06 +0000 Subject: llc: convert llc_sap_list to RCU Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/net/llc.h | 2 +- net/llc/llc_core.c | 46 ++++++++++++++-------------------------------- net/llc/llc_proc.c | 11 ++++------- 3 files changed, 19 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/net/llc.h b/include/net/llc.h index 709a9b37e23..5503b74ab17 100644 --- a/include/net/llc.h +++ b/include/net/llc.h @@ -93,7 +93,7 @@ struct hlist_nulls_head *llc_sk_laddr_hash(struct llc_sap *sap, #define LLC_DEST_CONN 2 /* Type 2 goes here */ extern struct list_head llc_sap_list; -extern rwlock_t llc_sap_list_lock; +extern spinlock_t llc_sap_list_lock; extern int llc_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev); diff --git a/net/llc/llc_core.c b/net/llc/llc_core.c index 0c9ef8bc765..78167e81dfe 100644 --- a/net/llc/llc_core.c +++ b/net/llc/llc_core.c @@ -23,7 +23,7 @@ #include LIST_HEAD(llc_sap_list); -DEFINE_RWLOCK(llc_sap_list_lock); +DEFINE_SPINLOCK(llc_sap_list_lock); /** * llc_sap_alloc - allocates and initializes sap. @@ -46,30 +46,6 @@ static struct llc_sap *llc_sap_alloc(void) return sap; } -/** - * llc_add_sap - add sap to station list - * @sap: Address of the sap - * - * Adds a sap to the LLC's station sap list. - */ -static void llc_add_sap(struct llc_sap *sap) -{ - list_add_tail(&sap->node, &llc_sap_list); -} - -/** - * llc_del_sap - del sap from station list - * @sap: Address of the sap - * - * Removes a sap to the LLC's station sap list. - */ -static void llc_del_sap(struct llc_sap *sap) -{ - write_lock_bh(&llc_sap_list_lock); - list_del(&sap->node); - write_unlock_bh(&llc_sap_list_lock); -} - static struct llc_sap *__llc_sap_find(unsigned char sap_value) { struct llc_sap* sap; @@ -93,13 +69,13 @@ out: */ struct llc_sap *llc_sap_find(unsigned char sap_value) { - struct llc_sap* sap; + struct llc_sap *sap; - read_lock_bh(&llc_sap_list_lock); + rcu_read_lock_bh(); sap = __llc_sap_find(sap_value); if (sap) llc_sap_hold(sap); - read_unlock_bh(&llc_sap_list_lock); + rcu_read_unlock_bh(); return sap; } @@ -120,7 +96,7 @@ struct llc_sap *llc_sap_open(unsigned char lsap, { struct llc_sap *sap = NULL; - write_lock_bh(&llc_sap_list_lock); + spin_lock_bh(&llc_sap_list_lock); if (__llc_sap_find(lsap)) /* SAP already exists */ goto out; sap = llc_sap_alloc(); @@ -128,9 +104,9 @@ struct llc_sap *llc_sap_open(unsigned char lsap, goto out; sap->laddr.lsap = lsap; sap->rcv_func = func; - llc_add_sap(sap); + list_add_tail_rcu(&sap->node, &llc_sap_list); out: - write_unlock_bh(&llc_sap_list_lock); + spin_unlock_bh(&llc_sap_list_lock); return sap; } @@ -146,7 +122,13 @@ out: void llc_sap_close(struct llc_sap *sap) { WARN_ON(sap->sk_count); - llc_del_sap(sap); + + spin_lock_bh(&llc_sap_list_lock); + list_del_rcu(&sap->node); + spin_unlock_bh(&llc_sap_list_lock); + + synchronize_rcu(); + kfree(sap); } diff --git a/net/llc/llc_proc.c b/net/llc/llc_proc.c index 09dec630720..7af1ff2d1f1 100644 --- a/net/llc/llc_proc.c +++ b/net/llc/llc_proc.c @@ -32,14 +32,11 @@ static void llc_ui_format_mac(struct seq_file *seq, u8 *addr) static struct sock *llc_get_sk_idx(loff_t pos) { - struct list_head *sap_entry; struct llc_sap *sap; struct sock *sk = NULL; int i; - list_for_each(sap_entry, &llc_sap_list) { - sap = list_entry(sap_entry, struct llc_sap, node); - + list_for_each_entry_rcu(sap, &llc_sap_list, node) { spin_lock_bh(&sap->sk_lock); for (i = 0; i < LLC_SK_LADDR_HASH_ENTRIES; i++) { struct hlist_nulls_head *head = &sap->sk_laddr_hash[i]; @@ -62,7 +59,7 @@ static void *llc_seq_start(struct seq_file *seq, loff_t *pos) { loff_t l = *pos; - read_lock_bh(&llc_sap_list_lock); + rcu_read_lock_bh(); return l ? llc_get_sk_idx(--l) : SEQ_START_TOKEN; } @@ -102,7 +99,7 @@ static void *llc_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (sk) goto out; spin_unlock_bh(&sap->sk_lock); - list_for_each_entry_continue(sap, &llc_sap_list, node) { + list_for_each_entry_continue_rcu(sap, &llc_sap_list, node) { spin_lock_bh(&sap->sk_lock); sk = laddr_hash_next(sap, -1); if (sk) @@ -122,7 +119,7 @@ static void llc_seq_stop(struct seq_file *seq, void *v) spin_unlock_bh(&sap->sk_lock); } - read_unlock_bh(&llc_sap_list_lock); + rcu_read_unlock_bh(); } static int llc_seq_socket_show(struct seq_file *seq, void *v) -- cgit v1.2.3-70-g09d2 From a80f7c0b088187c8471b441d461e937991870661 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 23 Dec 2009 13:15:32 +0100 Subject: mac80211: introduce flush operation We've long lacked a good confirmation that frames have really gone out, e.g. before going off-channel for a scan. Add a flush() operation that drivers can implement to provide that confirmation, and use it in a few places: * before scanning sends the nullfunc frames * after scanning sends the nullfunc frames, if any * when going idle, to send any pending frames Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/mac80211_hwsim.c | 11 +++++++++++ include/net/mac80211.h | 5 +++++ net/mac80211/driver-ops.h | 7 +++++++ net/mac80211/driver-trace.h | 21 +++++++++++++++++++++ net/mac80211/iface.c | 2 ++ net/mac80211/scan.c | 13 +++++++++++-- 6 files changed, 57 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 718a5f198c3..4dee69a38c1 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -896,6 +896,16 @@ static int mac80211_hwsim_ampdu_action(struct ieee80211_hw *hw, return 0; } +static void mac80211_hwsim_flush(struct ieee80211_hw *hw, bool drop) +{ + /* + * In this special case, there's nothing we need to + * do because hwsim does transmission synchronously. + * In the future, when it does transmissions via + * userspace, we may need to do something. + */ +} + static const struct ieee80211_ops mac80211_hwsim_ops = { @@ -912,6 +922,7 @@ static const struct ieee80211_ops mac80211_hwsim_ops = .conf_tx = mac80211_hwsim_conf_tx, CFG80211_TESTMODE_CMD(mac80211_hwsim_testmode_cmd) .ampdu_action = mac80211_hwsim_ampdu_action, + .flush = mac80211_hwsim_flush, }; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 494ac69ff47..77ea34b0328 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1545,6 +1545,10 @@ enum ieee80211_ampdu_mlme_action { * and need to call wiphy_rfkill_set_hw_state() in the callback. * * @testmode_cmd: Implement a cfg80211 test mode command. + * + * @flush: Flush all pending frames from the hardware queue, making sure + * that the hardware queues are empty. If the parameter @drop is set + * to %true, pending frames may be dropped. */ struct ieee80211_ops { int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb); @@ -1601,6 +1605,7 @@ struct ieee80211_ops { #ifdef CONFIG_NL80211_TESTMODE int (*testmode_cmd)(struct ieee80211_hw *hw, void *data, int len); #endif + void (*flush)(struct ieee80211_hw *hw, bool drop); }; /** diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 727e4cf7b8a..cbe133bcdf3 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -259,4 +259,11 @@ static inline void drv_rfkill_poll(struct ieee80211_local *local) if (local->ops->rfkill_poll) local->ops->rfkill_poll(&local->hw); } + +static inline void drv_flush(struct ieee80211_local *local, bool drop) +{ + trace_drv_flush(local, drop); + if (local->ops->flush) + local->ops->flush(&local->hw, drop); +} #endif /* __MAC80211_DRIVER_OPS */ diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index 7a849b92016..977cc7528bc 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -690,6 +690,27 @@ TRACE_EVENT(drv_ampdu_action, LOCAL_PR_ARG, VIF_PR_ARG, STA_PR_ARG, __entry->action, __entry->tid, __entry->ret ) ); + +TRACE_EVENT(drv_flush, + TP_PROTO(struct ieee80211_local *local, bool drop), + + TP_ARGS(local, drop), + + TP_STRUCT__entry( + LOCAL_ENTRY + __field(bool, drop) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + __entry->drop = drop; + ), + + TP_printk( + LOCAL_PR_FMT " drop:%d", + LOCAL_PR_ARG, __entry->drop + ) +); #endif /* !__MAC80211_DRIVER_TRACE || TRACE_HEADER_MULTI_READ */ #undef TRACE_INCLUDE_PATH diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 1ceca14331d..389dc8d880f 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -917,6 +917,8 @@ static u32 ieee80211_idle_on(struct ieee80211_local *local) wiphy_name(local->hw.wiphy)); #endif + drv_flush(local, false); + local->hw.conf.flags |= IEEE80211_CONF_IDLE; return IEEE80211_CONF_CHANGE_IDLE; } diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index ae183005652..d98c45e5528 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -418,9 +418,10 @@ static int ieee80211_start_sw_scan(struct ieee80211_local *local) local->next_scan_state = SCAN_DECISION; local->scan_channel_idx = 0; + drv_flush(local, false); + ieee80211_configure_filter(local); - /* TODO: start scan as soon as all nullfunc frames are ACKed */ ieee80211_queue_delayed_work(&local->hw, &local->scan_work, IEEE80211_CHANNEL_TIME); @@ -584,8 +585,16 @@ static void ieee80211_scan_state_leave_oper_channel(struct ieee80211_local *loca __set_bit(SCAN_OFF_CHANNEL, &local->scanning); + /* + * What if the nullfunc frames didn't arrive? + */ + drv_flush(local, false); + if (local->ops->flush) + *next_delay = 0; + else + *next_delay = HZ / 10; + /* advance to the next channel to be scanned */ - *next_delay = HZ / 10; local->next_scan_state = SCAN_SET_CHANNEL; } -- cgit v1.2.3-70-g09d2 From 8e664fb3fd2b04e3ac5fad7f046000ba54e0e275 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 23 Dec 2009 13:15:38 +0100 Subject: mac80211: split up and insert custom IEs correctly Currently, we insert all user-specified IEs before the HT IE for association, and after the HT IE for probe requests. For association, that's correct only if the user-specified IEs are RSN only, incorrect in all other cases including WPA. Change this to split apart the user-specified IEs in two places for association: before the HT IE (e.g. RSN), after the HT IE (generally empty right now I think?) and after WMM (all other vendor-specific IEs). For probes, split the IEs in different places to be correct according to the spec. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 39 ++++++++++--- net/mac80211/ieee80211_i.h | 4 ++ net/mac80211/util.c | 134 ++++++++++++++++++++++++++++++++++++++------- net/mac80211/work.c | 43 ++++++++++++--- 4 files changed, 184 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index d62edc7df3a..aeea282bd2f 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1085,12 +1085,12 @@ enum ieee80211_eid { WLAN_EID_TIM = 5, WLAN_EID_IBSS_PARAMS = 6, WLAN_EID_CHALLENGE = 16, - /* 802.11d */ + WLAN_EID_COUNTRY = 7, WLAN_EID_HP_PARAMS = 8, WLAN_EID_HP_TABLE = 9, WLAN_EID_REQUEST = 10, - /* 802.11e */ + WLAN_EID_QBSS_LOAD = 11, WLAN_EID_EDCA_PARAM_SET = 12, WLAN_EID_TSPEC = 13, @@ -1113,7 +1113,7 @@ enum ieee80211_eid { WLAN_EID_PREP = 69, WLAN_EID_PERR = 70, WLAN_EID_RANN = 49, /* compatible with FreeBSD */ - /* 802.11h */ + WLAN_EID_PWR_CONSTRAINT = 32, WLAN_EID_PWR_CAPABILITY = 33, WLAN_EID_TPC_REQUEST = 34, @@ -1124,20 +1124,41 @@ enum ieee80211_eid { WLAN_EID_MEASURE_REPORT = 39, WLAN_EID_QUIET = 40, WLAN_EID_IBSS_DFS = 41, - /* 802.11g */ + WLAN_EID_ERP_INFO = 42, WLAN_EID_EXT_SUPP_RATES = 50, - /* 802.11n */ + WLAN_EID_HT_CAPABILITY = 45, WLAN_EID_HT_INFORMATION = 61, - /* 802.11i */ + WLAN_EID_RSN = 48, - WLAN_EID_TIMEOUT_INTERVAL = 56, - WLAN_EID_MMIE = 76 /* 802.11w */, + WLAN_EID_MMIE = 76, WLAN_EID_WPA = 221, WLAN_EID_GENERIC = 221, WLAN_EID_VENDOR_SPECIFIC = 221, - WLAN_EID_QOS_PARAMETER = 222 + WLAN_EID_QOS_PARAMETER = 222, + + WLAN_EID_AP_CHAN_REPORT = 51, + WLAN_EID_NEIGHBOR_REPORT = 52, + WLAN_EID_RCPI = 53, + WLAN_EID_BSS_AVG_ACCESS_DELAY = 63, + WLAN_EID_ANTENNA_INFO = 64, + WLAN_EID_RSNI = 65, + WLAN_EID_MEASUREMENT_PILOT_TX_INFO = 66, + WLAN_EID_BSS_AVAILABLE_CAPACITY = 67, + WLAN_EID_BSS_AC_ACCESS_DELAY = 68, + WLAN_EID_RRM_ENABLED_CAPABILITIES = 70, + WLAN_EID_MULTIPLE_BSSID = 71, + + WLAN_EID_MOBILITY_DOMAIN = 54, + WLAN_EID_FAST_BSS_TRANSITION = 55, + WLAN_EID_TIMEOUT_INTERVAL = 56, + WLAN_EID_RIC_DATA = 57, + WLAN_EID_RIC_DESCRIPTOR = 75, + + WLAN_EID_DSE_REGISTERED_LOCATION = 58, + WLAN_EID_SUPPORTED_REGULATORY_CLASSES = 59, + WLAN_EID_EXT_CHANSWITCH_ANN = 60, }; /* Action category code */ diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 97b6076b492..6ea4ffbf84d 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1148,6 +1148,10 @@ int __ieee80211_request_smps(struct ieee80211_sub_if_data *sdata, void ieee80211_recalc_smps(struct ieee80211_local *local, struct ieee80211_sub_if_data *forsdata); +size_t ieee80211_ie_split(const u8 *ies, size_t ielen, + const u8 *ids, int n_ids, size_t offset); +size_t ieee80211_ie_split_vendor(const u8 *ies, size_t ielen, size_t offset); + /* internal work items */ void ieee80211_work_init(struct ieee80211_local *local); void ieee80211_add_work(struct ieee80211_work *wk); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 5ffe9e831b6..1fdb80ff924 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -881,30 +881,66 @@ int ieee80211_build_preq_ies(struct ieee80211_local *local, u8 *buffer, enum ieee80211_band band) { struct ieee80211_supported_band *sband; - u8 *pos, *supp_rates_len, *esupp_rates_len = NULL; - int i; + u8 *pos; + size_t offset = 0, noffset; + int supp_rates_len, i; sband = local->hw.wiphy->bands[band]; pos = buffer; + supp_rates_len = min_t(int, sband->n_bitrates, 8); + *pos++ = WLAN_EID_SUPP_RATES; - supp_rates_len = pos; - *pos++ = 0; - - for (i = 0; i < sband->n_bitrates; i++) { - struct ieee80211_rate *rate = &sband->bitrates[i]; - - if (esupp_rates_len) { - *esupp_rates_len += 1; - } else if (*supp_rates_len == 8) { - *pos++ = WLAN_EID_EXT_SUPP_RATES; - esupp_rates_len = pos; - *pos++ = 1; - } else - *supp_rates_len += 1; + *pos++ = supp_rates_len; - *pos++ = rate->bitrate / 5; + for (i = 0; i < supp_rates_len; i++) { + int rate = sband->bitrates[i].bitrate; + *pos++ = (u8) (rate / 5); + } + + /* insert "request information" if in custom IEs */ + if (ie && ie_len) { + static const u8 before_extrates[] = { + WLAN_EID_SSID, + WLAN_EID_SUPP_RATES, + WLAN_EID_REQUEST, + }; + noffset = ieee80211_ie_split(ie, ie_len, + before_extrates, + ARRAY_SIZE(before_extrates), + offset); + memcpy(pos, ie + offset, noffset - offset); + pos += noffset - offset; + offset = noffset; + } + + if (sband->n_bitrates > i) { + *pos++ = WLAN_EID_EXT_SUPP_RATES; + *pos++ = sband->n_bitrates - i; + + for (; i < sband->n_bitrates; i++) { + int rate = sband->bitrates[i].bitrate; + *pos++ = (u8) (rate / 5); + } + } + + /* insert custom IEs that go before HT */ + if (ie && ie_len) { + static const u8 before_ht[] = { + WLAN_EID_SSID, + WLAN_EID_SUPP_RATES, + WLAN_EID_REQUEST, + WLAN_EID_EXT_SUPP_RATES, + WLAN_EID_DS_PARAMS, + WLAN_EID_SUPPORTED_REGULATORY_CLASSES, + }; + noffset = ieee80211_ie_split(ie, ie_len, + before_ht, ARRAY_SIZE(before_ht), + offset); + memcpy(pos, ie + offset, noffset - offset); + pos += noffset - offset; + offset = noffset; } if (sband->ht_cap.ht_supported) { @@ -936,9 +972,11 @@ int ieee80211_build_preq_ies(struct ieee80211_local *local, u8 *buffer, * that calculates local->scan_ies_len. */ - if (ie) { - memcpy(pos, ie, ie_len); - pos += ie_len; + /* add any remaining custom IEs */ + if (ie && ie_len) { + noffset = ie_len; + memcpy(pos, ie + offset, noffset - offset); + pos += noffset - offset; } return pos - buffer; @@ -1252,3 +1290,59 @@ void ieee80211_recalc_smps(struct ieee80211_local *local, /* changed flag is auto-detected for this */ ieee80211_hw_config(local, 0); } + +static bool ieee80211_id_in_list(const u8 *ids, int n_ids, u8 id) +{ + int i; + + for (i = 0; i < n_ids; i++) + if (ids[i] == id) + return true; + return false; +} + +/** + * ieee80211_ie_split - split an IE buffer according to ordering + * + * @ies: the IE buffer + * @ielen: the length of the IE buffer + * @ids: an array with element IDs that are allowed before + * the split + * @n_ids: the size of the element ID array + * @offset: offset where to start splitting in the buffer + * + * This function splits an IE buffer by updating the @offset + * variable to point to the location where the buffer should be + * split. + * + * It assumes that the given IE buffer is well-formed, this + * has to be guaranteed by the caller! + * + * It also assumes that the IEs in the buffer are ordered + * correctly, if not the result of using this function will not + * be ordered correctly either, i.e. it does no reordering. + * + * The function returns the offset where the next part of the + * buffer starts, which may be @ielen if the entire (remainder) + * of the buffer should be used. + */ +size_t ieee80211_ie_split(const u8 *ies, size_t ielen, + const u8 *ids, int n_ids, size_t offset) +{ + size_t pos = offset; + + while (pos < ielen && ieee80211_id_in_list(ids, n_ids, ies[pos])) + pos += 2 + ies[pos + 1]; + + return pos; +} + +size_t ieee80211_ie_split_vendor(const u8 *ies, size_t ielen, size_t offset) +{ + size_t pos = offset; + + while (pos < ielen && ies[pos] != WLAN_EID_VENDOR_SPECIFIC) + pos += 2 + ies[pos + 1]; + + return pos; +} diff --git a/net/mac80211/work.c b/net/mac80211/work.c index c03c22d5bca..affdd10b67a 100644 --- a/net/mac80211/work.c +++ b/net/mac80211/work.c @@ -204,6 +204,7 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, struct ieee80211_mgmt *mgmt; u8 *pos; const u8 *ies; + size_t offset = 0, noffset; int i, len, count, rates_len, supp_rates_len; u16 capab; struct ieee80211_supported_band *sband; @@ -337,14 +338,26 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, } } - /* - * XXX: These IEs could contain (vendor-specified) - * IEs that belong after HT -- the buffer may - * need to be split up. - */ + /* if present, add any custom IEs that go before HT */ if (wk->ie_len && wk->ie) { - pos = skb_put(skb, wk->ie_len); - memcpy(pos, wk->ie, wk->ie_len); + static const u8 before_ht[] = { + WLAN_EID_SSID, + WLAN_EID_SUPP_RATES, + WLAN_EID_EXT_SUPP_RATES, + WLAN_EID_PWR_CAPABILITY, + WLAN_EID_SUPPORTED_CHANNELS, + WLAN_EID_RSN, + WLAN_EID_QOS_CAPA, + WLAN_EID_RRM_ENABLED_CAPABILITIES, + WLAN_EID_MOBILITY_DOMAIN, + WLAN_EID_SUPPORTED_REGULATORY_CLASSES, + }; + noffset = ieee80211_ie_split(wk->ie, wk->ie_len, + before_ht, ARRAY_SIZE(before_ht), + offset); + pos = skb_put(skb, noffset - offset); + memcpy(pos, wk->ie + offset, noffset - offset); + offset = noffset; } if (wk->assoc.use_11n && wk->assoc.wmm_used && @@ -352,6 +365,15 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, ieee80211_add_ht_ie(skb, wk->assoc.ht_information_ie, sband, wk->chan, wk->assoc.smps); + /* if present, add any custom non-vendor IEs that go after HT */ + if (wk->ie_len && wk->ie) { + noffset = ieee80211_ie_split_vendor(wk->ie, wk->ie_len, + offset); + pos = skb_put(skb, noffset - offset); + memcpy(pos, wk->ie + offset, noffset - offset); + offset = noffset; + } + if (wk->assoc.wmm_used && local->hw.queues >= 4) { pos = skb_put(skb, 9); *pos++ = WLAN_EID_VENDOR_SPECIFIC; @@ -365,6 +387,13 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, *pos++ = 0; } + /* add any remaining custom (i.e. vendor specific here) IEs */ + if (wk->ie_len && wk->ie) { + noffset = wk->ie_len; + pos = skb_put(skb, noffset - offset); + memcpy(pos, wk->ie + offset, noffset - offset); + } + IEEE80211_SKB_CB(skb)->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT; ieee80211_tx_skb(sdata, skb); } -- cgit v1.2.3-70-g09d2 From 9588bbd5529461a3dacd435bf239c84c3508f569 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Wed, 23 Dec 2009 13:15:41 +0100 Subject: cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 36 ++++++++ include/net/cfg80211.h | 47 +++++++++++ net/wireless/chan.c | 41 +++++---- net/wireless/core.h | 3 + net/wireless/mlme.c | 27 ++++++ net/wireless/nl80211.c | 218 +++++++++++++++++++++++++++++++++++++++++++++++- net/wireless/nl80211.h | 11 +++ 7 files changed, 368 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index da8ea2e1927..2bfbe88837e 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -270,6 +270,31 @@ * @NL80211_CMD_SET_WIPHY_NETNS: Set a wiphy's netns. Note that all devices * associated with this wiphy must be down and will follow. * + * @NL80211_CMD_REMAIN_ON_CHANNEL: Request to remain awake on the specified + * channel for the specified amount of time. This can be used to do + * off-channel operations like transmit a Public Action frame and wait for + * a response while being associated to an AP on another channel. + * %NL80211_ATTR_WIPHY or %NL80211_ATTR_IFINDEX is used to specify which + * radio is used. %NL80211_ATTR_WIPHY_FREQ is used to specify the + * frequency for the operation and %NL80211_ATTR_WIPHY_CHANNEL_TYPE may be + * optionally used to specify additional channel parameters. + * %NL80211_ATTR_DURATION is used to specify the duration in milliseconds + * to remain on the channel. This command is also used as an event to + * notify when the requested duration starts (it may take a while for the + * driver to schedule this time due to other concurrent needs for the + * radio). + * When called, this operation returns a cookie (%NL80211_ATTR_COOKIE) + * that will be included with any events pertaining to this request; + * the cookie is also used to cancel the request. + * @NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL: This command can be used to cancel a + * pending remain-on-channel duration if the desired operation has been + * completed prior to expiration of the originally requested duration. + * %NL80211_ATTR_WIPHY or %NL80211_ATTR_IFINDEX is used to specify the + * radio. The %NL80211_ATTR_COOKIE attribute must be given as well to + * uniquely identify the request. + * This command is also used as an event to notify when a requested + * remain-on-channel duration has expired. + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -353,6 +378,9 @@ enum nl80211_commands { NL80211_CMD_DEL_PMKSA, NL80211_CMD_FLUSH_PMKSA, + NL80211_CMD_REMAIN_ON_CHANNEL, + NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -606,6 +634,10 @@ enum nl80211_commands { * @NL80211_ATTR_MAX_NUM_PMKIDS: maximum number of PMKIDs a firmware can * cache, a wiphy attribute. * + * @NL80211_ATTR_DURATION: Duration of an operation in milliseconds, u32. + * + * @NL80211_ATTR_COOKIE: Generic 64-bit cookie to identify objects. + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -743,6 +775,10 @@ enum nl80211_attrs { NL80211_ATTR_PMKID, NL80211_ATTR_MAX_NUM_PMKIDS, + NL80211_ATTR_DURATION, + + NL80211_ATTR_COOKIE, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 542a477a94d..b66beb05205 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -988,6 +988,15 @@ struct cfg80211_pmksa { * * @dump_survey: get site survey information. * + * @remain_on_channel: Request the driver to remain awake on the specified + * channel for the specified duration to complete an off-channel + * operation (e.g., public action frame exchange). When the driver is + * ready on the requested channel, it must indicate this with an event + * notification by calling cfg80211_ready_on_channel(). + * @cancel_remain_on_channel: Cancel an on-going remain-on-channel operation. + * This allows the operation to be terminated prior to timeout based on + * the duration value. + * * @testmode_cmd: run a test mode command * * @set_pmksa: Cache a PMKID for a BSSID. This is mostly useful for fullmac @@ -1123,6 +1132,16 @@ struct cfg80211_ops { struct cfg80211_pmksa *pmksa); int (*flush_pmksa)(struct wiphy *wiphy, struct net_device *netdev); + int (*remain_on_channel)(struct wiphy *wiphy, + struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, + u64 *cookie); + int (*cancel_remain_on_channel)(struct wiphy *wiphy, + struct net_device *dev, + u64 cookie); + /* some temporary stuff to finish wext */ int (*set_power_mgmt)(struct wiphy *wiphy, struct net_device *dev, bool enabled, int timeout); @@ -2147,5 +2166,33 @@ void cfg80211_roamed(struct net_device *dev, const u8 *bssid, void cfg80211_disconnected(struct net_device *dev, u16 reason, u8 *ie, size_t ie_len, gfp_t gfp); +/** + * cfg80211_ready_on_channel - notification of remain_on_channel start + * @dev: network device + * @cookie: the request cookie + * @chan: The current channel (from remain_on_channel request) + * @channel_type: Channel type + * @duration: Duration in milliseconds that the driver intents to remain on the + * channel + * @gfp: allocation flags + */ +void cfg80211_ready_on_channel(struct net_device *dev, u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, gfp_t gfp); + +/** + * cfg80211_remain_on_channel_expired - remain_on_channel duration expired + * @dev: network device + * @cookie: the request cookie + * @chan: The current channel (from remain_on_channel request) + * @channel_type: Channel type + * @gfp: allocation flags + */ +void cfg80211_remain_on_channel_expired(struct net_device *dev, + u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + gfp_t gfp); #endif /* __NET_CFG80211_H */ diff --git a/net/wireless/chan.c b/net/wireless/chan.c index a46ac6c9b36..bf1737fc9a7 100644 --- a/net/wireless/chan.c +++ b/net/wireless/chan.c @@ -41,44 +41,57 @@ rdev_fixed_channel(struct cfg80211_registered_device *rdev, return result; } -int rdev_set_freq(struct cfg80211_registered_device *rdev, - struct wireless_dev *for_wdev, +struct ieee80211_channel * +rdev_freq_to_chan(struct cfg80211_registered_device *rdev, int freq, enum nl80211_channel_type channel_type) { struct ieee80211_channel *chan; struct ieee80211_sta_ht_cap *ht_cap; - int result; - - if (rdev_fixed_channel(rdev, for_wdev)) - return -EBUSY; - - if (!rdev->ops->set_channel) - return -EOPNOTSUPP; chan = ieee80211_get_channel(&rdev->wiphy, freq); /* Primary channel not allowed */ if (!chan || chan->flags & IEEE80211_CHAN_DISABLED) - return -EINVAL; + return NULL; if (channel_type == NL80211_CHAN_HT40MINUS && chan->flags & IEEE80211_CHAN_NO_HT40MINUS) - return -EINVAL; + return NULL; else if (channel_type == NL80211_CHAN_HT40PLUS && chan->flags & IEEE80211_CHAN_NO_HT40PLUS) - return -EINVAL; + return NULL; ht_cap = &rdev->wiphy.bands[chan->band]->ht_cap; if (channel_type != NL80211_CHAN_NO_HT) { if (!ht_cap->ht_supported) - return -EINVAL; + return NULL; if (!(ht_cap->cap & IEEE80211_HT_CAP_SUP_WIDTH_20_40) || ht_cap->cap & IEEE80211_HT_CAP_40MHZ_INTOLERANT) - return -EINVAL; + return NULL; } + return chan; +} + +int rdev_set_freq(struct cfg80211_registered_device *rdev, + struct wireless_dev *for_wdev, + int freq, enum nl80211_channel_type channel_type) +{ + struct ieee80211_channel *chan; + int result; + + if (rdev_fixed_channel(rdev, for_wdev)) + return -EBUSY; + + if (!rdev->ops->set_channel) + return -EOPNOTSUPP; + + chan = rdev_freq_to_chan(rdev, freq, channel_type); + if (!chan) + return -EINVAL; + result = rdev->ops->set_channel(&rdev->wiphy, chan, channel_type); if (result) return result; diff --git a/net/wireless/core.h b/net/wireless/core.h index 35b71212714..30ec95f05b5 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -374,6 +374,9 @@ void cfg80211_process_rdev_events(struct cfg80211_registered_device *rdev); struct ieee80211_channel * rdev_fixed_channel(struct cfg80211_registered_device *rdev, struct wireless_dev *for_wdev); +struct ieee80211_channel * +rdev_freq_to_chan(struct cfg80211_registered_device *rdev, + int freq, enum nl80211_channel_type channel_type); int rdev_set_freq(struct cfg80211_registered_device *rdev, struct wireless_dev *for_wdev, int freq, enum nl80211_channel_type channel_type); diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index acaeaa784d6..11f6469b3f9 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -680,3 +680,30 @@ void cfg80211_mlme_down(struct cfg80211_registered_device *rdev, } } } + +void cfg80211_ready_on_channel(struct net_device *dev, u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, gfp_t gfp) +{ + struct wiphy *wiphy = dev->ieee80211_ptr->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + + nl80211_send_remain_on_channel(rdev, dev, cookie, chan, channel_type, + duration, gfp); +} +EXPORT_SYMBOL(cfg80211_ready_on_channel); + +void cfg80211_remain_on_channel_expired(struct net_device *dev, + u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + gfp_t gfp) +{ + struct wiphy *wiphy = dev->ieee80211_ptr->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + + nl80211_send_remain_on_channel_cancel(rdev, dev, cookie, chan, + channel_type, gfp); +} +EXPORT_SYMBOL(cfg80211_remain_on_channel_expired); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 60f854377f9..ff857f10cb8 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -141,6 +141,8 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { [NL80211_ATTR_4ADDR] = { .type = NLA_U8 }, [NL80211_ATTR_PMKID] = { .type = NLA_BINARY, .len = WLAN_PMKID_LEN }, + [NL80211_ATTR_DURATION] = { .type = NLA_U32 }, + [NL80211_ATTR_COOKIE] = { .type = NLA_U64 }, }; /* policy for the attributes */ @@ -569,6 +571,7 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, CMD(set_pmksa, SET_PMKSA); CMD(del_pmksa, DEL_PMKSA); CMD(flush_pmksa, FLUSH_PMKSA); + CMD(remain_on_channel, REMAIN_ON_CHANNEL); if (dev->wiphy.flags & WIPHY_FLAG_NETNS_OK) { i++; NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS); @@ -4283,6 +4286,143 @@ static int nl80211_flush_pmksa(struct sk_buff *skb, struct genl_info *info) } +static int nl80211_remain_on_channel(struct sk_buff *skb, + struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct net_device *dev; + struct ieee80211_channel *chan; + struct sk_buff *msg; + void *hdr; + u64 cookie; + enum nl80211_channel_type channel_type = NL80211_CHAN_NO_HT; + u32 freq, duration; + int err; + + if (!info->attrs[NL80211_ATTR_WIPHY_FREQ] || + !info->attrs[NL80211_ATTR_DURATION]) + return -EINVAL; + + duration = nla_get_u32(info->attrs[NL80211_ATTR_DURATION]); + + /* + * We should be on that channel for at least one jiffie, + * and more than 5 seconds seems excessive. + */ + if (!duration || !msecs_to_jiffies(duration) || duration > 5000) + return -EINVAL; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + if (!rdev->ops->remain_on_channel) { + err = -EOPNOTSUPP; + goto out; + } + + if (!netif_running(dev)) { + err = -ENETDOWN; + goto out; + } + + if (info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) { + channel_type = nla_get_u32( + info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]); + if (channel_type != NL80211_CHAN_NO_HT && + channel_type != NL80211_CHAN_HT20 && + channel_type != NL80211_CHAN_HT40PLUS && + channel_type != NL80211_CHAN_HT40MINUS) + err = -EINVAL; + goto out; + } + + freq = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ]); + chan = rdev_freq_to_chan(rdev, freq, channel_type); + if (chan == NULL) { + err = -EINVAL; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + err = -ENOMEM; + goto out; + } + + hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, + NL80211_CMD_REMAIN_ON_CHANNEL); + + if (IS_ERR(hdr)) { + err = PTR_ERR(hdr); + goto free_msg; + } + + err = rdev->ops->remain_on_channel(&rdev->wiphy, dev, chan, + channel_type, duration, &cookie); + + if (err) + goto free_msg; + + NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); + + genlmsg_end(msg, hdr); + err = genlmsg_reply(msg, info); + goto out; + + nla_put_failure: + err = -ENOBUFS; + free_msg: + nlmsg_free(msg); + out: + cfg80211_unlock_rdev(rdev); + dev_put(dev); + unlock_rtnl: + rtnl_unlock(); + return err; +} + +static int nl80211_cancel_remain_on_channel(struct sk_buff *skb, + struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct net_device *dev; + u64 cookie; + int err; + + if (!info->attrs[NL80211_ATTR_COOKIE]) + return -EINVAL; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + if (!rdev->ops->cancel_remain_on_channel) { + err = -EOPNOTSUPP; + goto out; + } + + if (!netif_running(dev)) { + err = -ENETDOWN; + goto out; + } + + cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]); + + err = rdev->ops->cancel_remain_on_channel(&rdev->wiphy, dev, cookie); + + out: + cfg80211_unlock_rdev(rdev); + dev_put(dev); + unlock_rtnl: + rtnl_unlock(); + return err; +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -4545,8 +4685,20 @@ static struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, - + { + .cmd = NL80211_CMD_REMAIN_ON_CHANNEL, + .doit = nl80211_remain_on_channel, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, + .doit = nl80211_cancel_remain_on_channel, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, }; + static struct genl_multicast_group nl80211_mlme_mcgrp = { .name = "mlme", }; @@ -5134,6 +5286,70 @@ nla_put_failure: nlmsg_free(msg); } +static void nl80211_send_remain_on_chan_event( + int cmd, struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, gfp_t gfp) +{ + struct sk_buff *msg; + void *hdr; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp); + if (!msg) + return; + + hdr = nl80211hdr_put(msg, 0, 0, 0, cmd); + if (!hdr) { + nlmsg_free(msg); + return; + } + + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx); + NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex); + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY_FREQ, chan->center_freq); + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY_CHANNEL_TYPE, channel_type); + NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); + + if (cmd == NL80211_CMD_REMAIN_ON_CHANNEL) + NLA_PUT_U32(msg, NL80211_ATTR_DURATION, duration); + + if (genlmsg_end(msg, hdr) < 0) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0, + nl80211_mlme_mcgrp.id, gfp); + return; + + nla_put_failure: + genlmsg_cancel(msg, hdr); + nlmsg_free(msg); +} + +void nl80211_send_remain_on_channel(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, gfp_t gfp) +{ + nl80211_send_remain_on_chan_event(NL80211_CMD_REMAIN_ON_CHANNEL, + rdev, netdev, cookie, chan, + channel_type, duration, gfp); +} + +void nl80211_send_remain_on_channel_cancel( + struct cfg80211_registered_device *rdev, struct net_device *netdev, + u64 cookie, struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, gfp_t gfp) +{ + nl80211_send_remain_on_chan_event(NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, + rdev, netdev, cookie, chan, + channel_type, 0, gfp); +} + /* initialisation/exit functions */ int nl80211_init(void) diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index 44cc2a76a1b..a5e2de419b7 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -59,4 +59,15 @@ void nl80211_send_ibss_bssid(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, gfp_t gfp); +void nl80211_send_remain_on_channel(struct cfg80211_registered_device *rdev, + struct net_device *netdev, + u64 cookie, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + unsigned int duration, gfp_t gfp); +void nl80211_send_remain_on_channel_cancel( + struct cfg80211_registered_device *rdev, struct net_device *netdev, + u64 cookie, struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, gfp_t gfp); + #endif /* __NET_WIRELESS_NL80211_H */ -- cgit v1.2.3-70-g09d2 From 98b6218388e345064c3f2d3c161383a18274c638 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 23 Dec 2009 13:15:44 +0100 Subject: mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/cfg80211.h | 12 ++++++++++++ net/mac80211/sta_info.c | 5 +++++ net/wireless/mlme.c | 10 ++++++++++ net/wireless/nl80211.c | 21 ++++++++++++++++++++- net/wireless/nl80211.h | 4 ++++ 5 files changed, 51 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index b66beb05205..add79930f47 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2195,4 +2195,16 @@ void cfg80211_remain_on_channel_expired(struct net_device *dev, enum nl80211_channel_type channel_type, gfp_t gfp); + +/** + * cfg80211_new_sta - notify userspace about station + * + * @dev: the netdev + * @mac_addr: the station's address + * @sinfo: the station information + * @gfp: allocation flags + */ +void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr, + struct station_info *sinfo, gfp_t gfp); + #endif /* __NET_CFG80211_H */ diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index f039e761aec..47da552ce8a 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -359,6 +359,7 @@ int sta_info_insert(struct sta_info *sta) { struct ieee80211_local *local = sta->local; struct ieee80211_sub_if_data *sdata = sta->sdata; + struct station_info sinfo; unsigned long flags; int err = 0; @@ -408,6 +409,10 @@ int sta_info_insert(struct sta_info *sta) spin_unlock_irqrestore(&local->sta_lock, flags); + sinfo.filled = 0; + sinfo.generation = local->sta_generation; + cfg80211_new_sta(sdata->dev, sta->sta.addr, &sinfo, GFP_ATOMIC); + #ifdef CONFIG_MAC80211_DEBUGFS /* * Debugfs entry adding might sleep, so schedule process diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index 11f6469b3f9..3dba19e1727 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -707,3 +707,13 @@ void cfg80211_remain_on_channel_expired(struct net_device *dev, channel_type, gfp); } EXPORT_SYMBOL(cfg80211_remain_on_channel_expired); + +void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr, + struct station_info *sinfo, gfp_t gfp) +{ + struct wiphy *wiphy = dev->ieee80211_ptr->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + + nl80211_send_sta_event(rdev, dev, mac_addr, sinfo, gfp); +} +EXPORT_SYMBOL(cfg80211_new_sta); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index ff857f10cb8..e3bee3cecdf 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1642,7 +1642,7 @@ static int parse_station_flags(struct genl_info *info, static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq, int flags, struct net_device *dev, - u8 *mac_addr, struct station_info *sinfo) + const u8 *mac_addr, struct station_info *sinfo) { void *hdr; struct nlattr *sinfoattr, *txrate; @@ -5350,6 +5350,25 @@ void nl80211_send_remain_on_channel_cancel( channel_type, 0, gfp); } +void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, + struct net_device *dev, const u8 *mac_addr, + struct station_info *sinfo, gfp_t gfp) +{ + struct sk_buff *msg; + + msg = nlmsg_new(NLMSG_GOODSIZE, gfp); + if (!msg) + return; + + if (nl80211_send_station(msg, 0, 0, 0, dev, mac_addr, sinfo) < 0) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0, + nl80211_mlme_mcgrp.id, gfp); +} + /* initialisation/exit functions */ int nl80211_init(void) diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index a5e2de419b7..14855b8fb43 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -70,4 +70,8 @@ void nl80211_send_remain_on_channel_cancel( u64 cookie, struct ieee80211_channel *chan, enum nl80211_channel_type channel_type, gfp_t gfp); +void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, + struct net_device *dev, const u8 *mac_addr, + struct station_info *sinfo, gfp_t gfp); + #endif /* __NET_WIRELESS_NL80211_H */ -- cgit v1.2.3-70-g09d2 From 1ed32e4fc8cfc9656cc1101e7f9617d485fcbe7b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 23 Dec 2009 13:15:45 +0100 Subject: mac80211: remove struct ieee80211_if_init_conf All its members (vif, mac_addr, type) are now available in the vif struct directly, so we can pass that instead of the conf struct. I generated this patch (except the mac80211 and header file changes) with this semantic patch: @@ identifier conf, fn, hw; type tp; @@ tp fn(struct ieee80211_hw *hw, -struct ieee80211_if_init_conf *conf) +struct ieee80211_vif *vif) { <... ( -conf->type +vif->type | -conf->mac_addr +vif->addr | -conf->vif +vif ) ...> } Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/adm8211.c | 12 +++---- drivers/net/wireless/at76c50x-usb.c | 6 ++-- drivers/net/wireless/ath/ar9170/main.c | 8 ++--- drivers/net/wireless/ath/ath5k/base.c | 18 +++++----- drivers/net/wireless/ath/ath9k/main.c | 28 +++++++-------- drivers/net/wireless/b43/main.c | 26 +++++++------- drivers/net/wireless/b43legacy/main.c | 24 ++++++------- drivers/net/wireless/iwlwifi/iwl-core.c | 20 +++++------ drivers/net/wireless/iwlwifi/iwl-core.h | 4 +-- drivers/net/wireless/libertas_tf/main.c | 10 +++--- drivers/net/wireless/mac80211_hwsim.c | 18 +++++----- drivers/net/wireless/mwl8k.c | 14 ++++---- drivers/net/wireless/p54/main.c | 12 +++---- drivers/net/wireless/rt2x00/rt2x00.h | 4 +-- drivers/net/wireless/rt2x00/rt2x00mac.c | 26 +++++++------- drivers/net/wireless/rtl818x/rtl8180_dev.c | 12 +++---- drivers/net/wireless/rtl818x/rtl8187_dev.c | 10 +++--- drivers/net/wireless/wl12xx/wl1251_main.c | 14 ++++---- drivers/net/wireless/wl12xx/wl1271_main.c | 10 +++--- drivers/net/wireless/zd1211rw/zd_mac.c | 10 +++--- include/net/mac80211.h | 57 ++++++++---------------------- net/mac80211/driver-ops.h | 12 +++---- net/mac80211/iface.c | 14 ++------ net/mac80211/pm.c | 6 +--- net/mac80211/util.c | 9 ++--- 25 files changed, 170 insertions(+), 214 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c index 39410016b4f..e1f04bb437e 100644 --- a/drivers/net/wireless/adm8211.c +++ b/drivers/net/wireless/adm8211.c @@ -1400,15 +1400,15 @@ static void adm8211_configure_filter(struct ieee80211_hw *dev, } static int adm8211_add_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct adm8211_priv *priv = dev->priv; if (priv->mode != NL80211_IFTYPE_MONITOR) return -EOPNOTSUPP; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: - priv->mode = conf->type; + priv->mode = vif->type; break; default: return -EOPNOTSUPP; @@ -1416,8 +1416,8 @@ static int adm8211_add_interface(struct ieee80211_hw *dev, ADM8211_IDLE(); - ADM8211_CSR_WRITE(PAR0, le32_to_cpu(*(__le32 *)conf->mac_addr)); - ADM8211_CSR_WRITE(PAR1, le16_to_cpu(*(__le16 *)(conf->mac_addr + 4))); + ADM8211_CSR_WRITE(PAR0, le32_to_cpu(*(__le32 *)vif->addr)); + ADM8211_CSR_WRITE(PAR1, le16_to_cpu(*(__le16 *)(vif->addr + 4))); adm8211_update_mode(dev); @@ -1427,7 +1427,7 @@ static int adm8211_add_interface(struct ieee80211_hw *dev, } static void adm8211_remove_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct adm8211_priv *priv = dev->priv; priv->mode = NL80211_IFTYPE_MONITOR; diff --git a/drivers/net/wireless/at76c50x-usb.c b/drivers/net/wireless/at76c50x-usb.c index 2517364d3eb..0fb419936df 100644 --- a/drivers/net/wireless/at76c50x-usb.c +++ b/drivers/net/wireless/at76c50x-usb.c @@ -1789,7 +1789,7 @@ static void at76_mac80211_stop(struct ieee80211_hw *hw) } static int at76_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct at76_priv *priv = hw->priv; int ret = 0; @@ -1798,7 +1798,7 @@ static int at76_add_interface(struct ieee80211_hw *hw, mutex_lock(&priv->mtx); - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: priv->iw_mode = IW_MODE_INFRA; break; @@ -1814,7 +1814,7 @@ exit: } static void at76_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { at76_dbg(DBG_MAC80211, "%s()", __func__); } diff --git a/drivers/net/wireless/ath/ar9170/main.c b/drivers/net/wireless/ath/ar9170/main.c index 20f04ab2b13..4d27f7f67c7 100644 --- a/drivers/net/wireless/ath/ar9170/main.c +++ b/drivers/net/wireless/ath/ar9170/main.c @@ -1939,7 +1939,7 @@ err_free: } static int ar9170_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ar9170 *ar = hw->priv; struct ath_common *common = &ar->common; @@ -1952,8 +1952,8 @@ static int ar9170_op_add_interface(struct ieee80211_hw *hw, goto unlock; } - ar->vif = conf->vif; - memcpy(common->macaddr, conf->mac_addr, ETH_ALEN); + ar->vif = vif; + memcpy(common->macaddr, vif->addr, ETH_ALEN); if (modparam_nohwcrypt || (ar->vif->type != NL80211_IFTYPE_STATION)) { ar->rx_software_decryption = true; @@ -1973,7 +1973,7 @@ unlock: } static void ar9170_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ar9170 *ar = hw->priv; diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c index a4c086f069b..20c7e5b450a 100644 --- a/drivers/net/wireless/ath/ath5k/base.c +++ b/drivers/net/wireless/ath/ath5k/base.c @@ -225,9 +225,9 @@ static int ath5k_reset_wake(struct ath5k_softc *sc); static int ath5k_start(struct ieee80211_hw *hw); static void ath5k_stop(struct ieee80211_hw *hw); static int ath5k_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); static void ath5k_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); static int ath5k_config(struct ieee80211_hw *hw, u32 changed); static u64 ath5k_prepare_multicast(struct ieee80211_hw *hw, int mc_count, struct dev_addr_list *mc_list); @@ -2785,7 +2785,7 @@ static void ath5k_stop(struct ieee80211_hw *hw) } static int ath5k_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ath5k_softc *sc = hw->priv; int ret; @@ -2796,22 +2796,22 @@ static int ath5k_add_interface(struct ieee80211_hw *hw, goto end; } - sc->vif = conf->vif; + sc->vif = vif; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_STATION: case NL80211_IFTYPE_ADHOC: case NL80211_IFTYPE_MESH_POINT: case NL80211_IFTYPE_MONITOR: - sc->opmode = conf->type; + sc->opmode = vif->type; break; default: ret = -EOPNOTSUPP; goto end; } - ath5k_hw_set_lladdr(sc->ah, conf->mac_addr); + ath5k_hw_set_lladdr(sc->ah, vif->addr); ath5k_mode_setup(sc); ret = 0; @@ -2822,13 +2822,13 @@ end: static void ath5k_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ath5k_softc *sc = hw->priv; u8 mac[ETH_ALEN] = {}; mutex_lock(&sc->lock); - if (sc->vif != conf->vif) + if (sc->vif != vif) goto end; ath5k_hw_set_lladdr(sc->ah, mac); diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 3f5b887d0fc..446bd23756e 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -2534,12 +2534,12 @@ static void ath9k_stop(struct ieee80211_hw *hw) } static int ath9k_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ath_wiphy *aphy = hw->priv; struct ath_softc *sc = aphy->sc; struct ath_common *common = ath9k_hw_common(sc->sc_ah); - struct ath_vif *avp = (void *)conf->vif->drv_priv; + struct ath_vif *avp = (void *)vif->drv_priv; enum nl80211_iftype ic_opmode = NL80211_IFTYPE_UNSPECIFIED; int ret = 0; @@ -2551,7 +2551,7 @@ static int ath9k_add_interface(struct ieee80211_hw *hw, goto out; } - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: ic_opmode = NL80211_IFTYPE_STATION; break; @@ -2562,11 +2562,11 @@ static int ath9k_add_interface(struct ieee80211_hw *hw, ret = -ENOBUFS; goto out; } - ic_opmode = conf->type; + ic_opmode = vif->type; break; default: ath_print(common, ATH_DBG_FATAL, - "Interface type %d not yet supported\n", conf->type); + "Interface type %d not yet supported\n", vif->type); ret = -EOPNOTSUPP; goto out; } @@ -2598,18 +2598,18 @@ static int ath9k_add_interface(struct ieee80211_hw *hw, * Enable MIB interrupts when there are hardware phy counters. * Note we only do this (at the moment) for station mode. */ - if ((conf->type == NL80211_IFTYPE_STATION) || - (conf->type == NL80211_IFTYPE_ADHOC) || - (conf->type == NL80211_IFTYPE_MESH_POINT)) { + if ((vif->type == NL80211_IFTYPE_STATION) || + (vif->type == NL80211_IFTYPE_ADHOC) || + (vif->type == NL80211_IFTYPE_MESH_POINT)) { sc->imask |= ATH9K_INT_MIB; sc->imask |= ATH9K_INT_TSFOOR; } ath9k_hw_set_interrupts(sc->sc_ah, sc->imask); - if (conf->type == NL80211_IFTYPE_AP || - conf->type == NL80211_IFTYPE_ADHOC || - conf->type == NL80211_IFTYPE_MONITOR) + if (vif->type == NL80211_IFTYPE_AP || + vif->type == NL80211_IFTYPE_ADHOC || + vif->type == NL80211_IFTYPE_MONITOR) ath_start_ani(common); out: @@ -2618,12 +2618,12 @@ out: } static void ath9k_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct ath_wiphy *aphy = hw->priv; struct ath_softc *sc = aphy->sc; struct ath_common *common = ath9k_hw_common(sc->sc_ah); - struct ath_vif *avp = (void *)conf->vif->drv_priv; + struct ath_vif *avp = (void *)vif->drv_priv; int i; ath_print(common, ATH_DBG_CONFIG, "Detach Interface\n"); @@ -2644,7 +2644,7 @@ static void ath9k_remove_interface(struct ieee80211_hw *hw, sc->sc_flags &= ~SC_OP_BEACONS; for (i = 0; i < ARRAY_SIZE(sc->beacon.bslot); i++) { - if (sc->beacon.bslot[i] == conf->vif) { + if (sc->beacon.bslot[i] == vif) { printk(KERN_DEBUG "%s: vif had allocated beacon " "slot\n", __func__); sc->beacon.bslot[i] = NULL; diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c index b0b5ce95000..6634a77fc76 100644 --- a/drivers/net/wireless/b43/main.c +++ b/drivers/net/wireless/b43/main.c @@ -4391,7 +4391,7 @@ err_busdown: } static int b43_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct b43_wl *wl = hw_to_b43_wl(hw); struct b43_wldev *dev; @@ -4399,24 +4399,24 @@ static int b43_op_add_interface(struct ieee80211_hw *hw, /* TODO: allow WDS/AP devices to coexist */ - if (conf->type != NL80211_IFTYPE_AP && - conf->type != NL80211_IFTYPE_MESH_POINT && - conf->type != NL80211_IFTYPE_STATION && - conf->type != NL80211_IFTYPE_WDS && - conf->type != NL80211_IFTYPE_ADHOC) + if (vif->type != NL80211_IFTYPE_AP && + vif->type != NL80211_IFTYPE_MESH_POINT && + vif->type != NL80211_IFTYPE_STATION && + vif->type != NL80211_IFTYPE_WDS && + vif->type != NL80211_IFTYPE_ADHOC) return -EOPNOTSUPP; mutex_lock(&wl->mutex); if (wl->operating) goto out_mutex_unlock; - b43dbg(wl, "Adding Interface type %d\n", conf->type); + b43dbg(wl, "Adding Interface type %d\n", vif->type); dev = wl->current_dev; wl->operating = 1; - wl->vif = conf->vif; - wl->if_type = conf->type; - memcpy(wl->mac_addr, conf->mac_addr, ETH_ALEN); + wl->vif = vif; + wl->if_type = vif->type; + memcpy(wl->mac_addr, vif->addr, ETH_ALEN); b43_adjust_opmode(dev); b43_set_pretbtt(dev); @@ -4431,17 +4431,17 @@ static int b43_op_add_interface(struct ieee80211_hw *hw, } static void b43_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct b43_wl *wl = hw_to_b43_wl(hw); struct b43_wldev *dev = wl->current_dev; - b43dbg(wl, "Removing Interface type %d\n", conf->type); + b43dbg(wl, "Removing Interface type %d\n", vif->type); mutex_lock(&wl->mutex); B43_WARN_ON(!wl->operating); - B43_WARN_ON(wl->vif != conf->vif); + B43_WARN_ON(wl->vif != vif); wl->vif = NULL; wl->operating = 0; diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c index ab6a18c2e9d..494017e4fcc 100644 --- a/drivers/net/wireless/b43legacy/main.c +++ b/drivers/net/wireless/b43legacy/main.c @@ -3361,7 +3361,7 @@ err_kfree_lo_control: } static int b43legacy_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw); struct b43legacy_wldev *dev; @@ -3370,23 +3370,23 @@ static int b43legacy_op_add_interface(struct ieee80211_hw *hw, /* TODO: allow WDS/AP devices to coexist */ - if (conf->type != NL80211_IFTYPE_AP && - conf->type != NL80211_IFTYPE_STATION && - conf->type != NL80211_IFTYPE_WDS && - conf->type != NL80211_IFTYPE_ADHOC) + if (vif->type != NL80211_IFTYPE_AP && + vif->type != NL80211_IFTYPE_STATION && + vif->type != NL80211_IFTYPE_WDS && + vif->type != NL80211_IFTYPE_ADHOC) return -EOPNOTSUPP; mutex_lock(&wl->mutex); if (wl->operating) goto out_mutex_unlock; - b43legacydbg(wl, "Adding Interface type %d\n", conf->type); + b43legacydbg(wl, "Adding Interface type %d\n", vif->type); dev = wl->current_dev; wl->operating = 1; - wl->vif = conf->vif; - wl->if_type = conf->type; - memcpy(wl->mac_addr, conf->mac_addr, ETH_ALEN); + wl->vif = vif; + wl->if_type = vif->type; + memcpy(wl->mac_addr, vif->addr, ETH_ALEN); spin_lock_irqsave(&wl->irq_lock, flags); b43legacy_adjust_opmode(dev); @@ -3403,18 +3403,18 @@ static int b43legacy_op_add_interface(struct ieee80211_hw *hw, } static void b43legacy_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw); struct b43legacy_wldev *dev = wl->current_dev; unsigned long flags; - b43legacydbg(wl, "Removing Interface type %d\n", conf->type); + b43legacydbg(wl, "Removing Interface type %d\n", vif->type); mutex_lock(&wl->mutex); B43legacy_WARN_ON(!wl->operating); - B43legacy_WARN_ON(wl->vif != conf->vif); + B43legacy_WARN_ON(wl->vif != vif); wl->vif = NULL; wl->operating = 0; diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c index e3b96b48b7f..14f482960d7 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.c +++ b/drivers/net/wireless/iwlwifi/iwl-core.c @@ -2584,12 +2584,12 @@ int iwl_set_mode(struct iwl_priv *priv, int mode) EXPORT_SYMBOL(iwl_set_mode); int iwl_mac_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct iwl_priv *priv = hw->priv; unsigned long flags; - IWL_DEBUG_MAC80211(priv, "enter: type %d\n", conf->type); + IWL_DEBUG_MAC80211(priv, "enter: type %d\n", vif->type); if (priv->vif) { IWL_DEBUG_MAC80211(priv, "leave - vif != NULL\n"); @@ -2597,19 +2597,19 @@ int iwl_mac_add_interface(struct ieee80211_hw *hw, } spin_lock_irqsave(&priv->lock, flags); - priv->vif = conf->vif; - priv->iw_mode = conf->type; + priv->vif = vif; + priv->iw_mode = vif->type; spin_unlock_irqrestore(&priv->lock, flags); mutex_lock(&priv->mutex); - if (conf->mac_addr) { - IWL_DEBUG_MAC80211(priv, "Set %pM\n", conf->mac_addr); - memcpy(priv->mac_addr, conf->mac_addr, ETH_ALEN); + if (vif->addr) { + IWL_DEBUG_MAC80211(priv, "Set %pM\n", vif->addr); + memcpy(priv->mac_addr, vif->addr, ETH_ALEN); } - if (iwl_set_mode(priv, conf->type) == -EAGAIN) + if (iwl_set_mode(priv, vif->type) == -EAGAIN) /* we are not ready, will run again when ready */ set_bit(STATUS_MODE_PENDING, &priv->status); @@ -2621,7 +2621,7 @@ int iwl_mac_add_interface(struct ieee80211_hw *hw, EXPORT_SYMBOL(iwl_mac_add_interface); void iwl_mac_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct iwl_priv *priv = hw->priv; @@ -2634,7 +2634,7 @@ void iwl_mac_remove_interface(struct ieee80211_hw *hw, priv->staging_rxon.filter_flags &= ~RXON_FILTER_ASSOC_MSK; iwlcore_commit_rxon(priv); } - if (priv->vif == conf->vif) { + if (priv->vif == vif) { priv->vif = NULL; memset(priv->bssid, 0, ETH_ALEN); } diff --git a/drivers/net/wireless/iwlwifi/iwl-core.h b/drivers/net/wireless/iwlwifi/iwl-core.h index f7acbb32900..1728f961dcb 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.h +++ b/drivers/net/wireless/iwlwifi/iwl-core.h @@ -332,9 +332,9 @@ int iwl_mac_beacon_update(struct ieee80211_hw *hw, struct sk_buff *skb); int iwl_commit_rxon(struct iwl_priv *priv); int iwl_set_mode(struct iwl_priv *priv, int mode); int iwl_mac_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); void iwl_mac_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); int iwl_mac_config(struct ieee80211_hw *hw, u32 changed); void iwl_config_ap(struct iwl_priv *priv); int iwl_mac_get_tx_stats(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/libertas_tf/main.c b/drivers/net/wireless/libertas_tf/main.c index 019431d2f8a..15a367680f5 100644 --- a/drivers/net/wireless/libertas_tf/main.c +++ b/drivers/net/wireless/libertas_tf/main.c @@ -318,14 +318,14 @@ static void lbtf_op_stop(struct ieee80211_hw *hw) } static int lbtf_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct lbtf_private *priv = hw->priv; if (priv->vif != NULL) return -EOPNOTSUPP; - priv->vif = conf->vif; - switch (conf->type) { + priv->vif = vif; + switch (vif->type) { case NL80211_IFTYPE_MESH_POINT: case NL80211_IFTYPE_AP: lbtf_set_mode(priv, LBTF_AP_MODE); @@ -337,12 +337,12 @@ static int lbtf_op_add_interface(struct ieee80211_hw *hw, priv->vif = NULL; return -EOPNOTSUPP; } - lbtf_set_mac_address(priv, (u8 *) conf->mac_addr); + lbtf_set_mac_address(priv, (u8 *) vif->addr); return 0; } static void lbtf_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct lbtf_private *priv = hw->priv; diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 4dee69a38c1..84df3fcf37b 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -584,24 +584,24 @@ static void mac80211_hwsim_stop(struct ieee80211_hw *hw) static int mac80211_hwsim_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { printk(KERN_DEBUG "%s:%s (type=%d mac_addr=%pM)\n", - wiphy_name(hw->wiphy), __func__, conf->type, - conf->mac_addr); - hwsim_set_magic(conf->vif); + wiphy_name(hw->wiphy), __func__, vif->type, + vif->addr); + hwsim_set_magic(vif); return 0; } static void mac80211_hwsim_remove_interface( - struct ieee80211_hw *hw, struct ieee80211_if_init_conf *conf) + struct ieee80211_hw *hw, struct ieee80211_vif *vif) { printk(KERN_DEBUG "%s:%s (type=%d mac_addr=%pM)\n", - wiphy_name(hw->wiphy), __func__, conf->type, - conf->mac_addr); - hwsim_check_magic(conf->vif); - hwsim_clear_magic(conf->vif); + wiphy_name(hw->wiphy), __func__, vif->type, + vif->addr); + hwsim_check_magic(vif); + hwsim_clear_magic(vif); } diff --git a/drivers/net/wireless/mwl8k.c b/drivers/net/wireless/mwl8k.c index 97a95decf9a..c1c6ecd0c5b 100644 --- a/drivers/net/wireless/mwl8k.c +++ b/drivers/net/wireless/mwl8k.c @@ -2835,7 +2835,7 @@ static void mwl8k_stop(struct ieee80211_hw *hw) } static int mwl8k_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct mwl8k_priv *priv = hw->priv; struct mwl8k_vif *mwl8k_vif; @@ -2849,7 +2849,7 @@ static int mwl8k_add_interface(struct ieee80211_hw *hw, /* * We only support managed interfaces for now. */ - if (conf->type != NL80211_IFTYPE_STATION) + if (vif->type != NL80211_IFTYPE_STATION) return -EINVAL; /* @@ -2865,24 +2865,24 @@ static int mwl8k_add_interface(struct ieee80211_hw *hw, } /* Clean out driver private area */ - mwl8k_vif = MWL8K_VIF(conf->vif); + mwl8k_vif = MWL8K_VIF(vif); memset(mwl8k_vif, 0, sizeof(*mwl8k_vif)); /* Set and save the mac address */ - mwl8k_cmd_set_mac_addr(hw, conf->mac_addr); - memcpy(mwl8k_vif->mac_addr, conf->mac_addr, ETH_ALEN); + mwl8k_cmd_set_mac_addr(hw, vif->addr); + memcpy(mwl8k_vif->mac_addr, vif->addr, ETH_ALEN); /* Set Initial sequence number to zero */ mwl8k_vif->seqno = 0; - priv->vif = conf->vif; + priv->vif = vif; priv->current_channel = NULL; return 0; } static void mwl8k_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct mwl8k_priv *priv = hw->priv; diff --git a/drivers/net/wireless/p54/main.c b/drivers/net/wireless/p54/main.c index 18012dbfb45..26428e4c9c6 100644 --- a/drivers/net/wireless/p54/main.c +++ b/drivers/net/wireless/p54/main.c @@ -216,7 +216,7 @@ static void p54_stop(struct ieee80211_hw *dev) } static int p54_add_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct p54_common *priv = dev->priv; @@ -226,28 +226,28 @@ static int p54_add_interface(struct ieee80211_hw *dev, return -EOPNOTSUPP; } - priv->vif = conf->vif; + priv->vif = vif; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: case NL80211_IFTYPE_ADHOC: case NL80211_IFTYPE_AP: case NL80211_IFTYPE_MESH_POINT: - priv->mode = conf->type; + priv->mode = vif->type; break; default: mutex_unlock(&priv->conf_mutex); return -EOPNOTSUPP; } - memcpy(priv->mac_addr, conf->mac_addr, ETH_ALEN); + memcpy(priv->mac_addr, vif->addr, ETH_ALEN); p54_setup_mac(priv); mutex_unlock(&priv->conf_mutex); return 0; } static void p54_remove_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct p54_common *priv = dev->priv; diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h index a664a999b2b..b4c6e0a6d7e 100644 --- a/drivers/net/wireless/rt2x00/rt2x00.h +++ b/drivers/net/wireless/rt2x00/rt2x00.h @@ -1019,9 +1019,9 @@ int rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb); int rt2x00mac_start(struct ieee80211_hw *hw); void rt2x00mac_stop(struct ieee80211_hw *hw); int rt2x00mac_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); void rt2x00mac_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); int rt2x00mac_config(struct ieee80211_hw *hw, u32 changed); void rt2x00mac_configure_filter(struct ieee80211_hw *hw, unsigned int changed_flags, diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c index de549c244ed..00f1f939f1b 100644 --- a/drivers/net/wireless/rt2x00/rt2x00mac.c +++ b/drivers/net/wireless/rt2x00/rt2x00mac.c @@ -187,10 +187,10 @@ void rt2x00mac_stop(struct ieee80211_hw *hw) EXPORT_SYMBOL_GPL(rt2x00mac_stop); int rt2x00mac_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rt2x00_dev *rt2x00dev = hw->priv; - struct rt2x00_intf *intf = vif_to_intf(conf->vif); + struct rt2x00_intf *intf = vif_to_intf(vif); struct data_queue *queue = rt2x00queue_get_queue(rt2x00dev, QID_BEACON); struct queue_entry *entry = NULL; unsigned int i; @@ -203,7 +203,7 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw, !test_bit(DEVICE_STATE_STARTED, &rt2x00dev->flags)) return -ENODEV; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_AP: /* * We don't support mixed combinations of @@ -263,7 +263,7 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw, * increase interface count and start initialization. */ - if (conf->type == NL80211_IFTYPE_AP) + if (vif->type == NL80211_IFTYPE_AP) rt2x00dev->intf_ap_count++; else rt2x00dev->intf_sta_count++; @@ -273,16 +273,16 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw, mutex_init(&intf->beacon_skb_mutex); intf->beacon = entry; - if (conf->type == NL80211_IFTYPE_AP) - memcpy(&intf->bssid, conf->mac_addr, ETH_ALEN); - memcpy(&intf->mac, conf->mac_addr, ETH_ALEN); + if (vif->type == NL80211_IFTYPE_AP) + memcpy(&intf->bssid, vif->addr, ETH_ALEN); + memcpy(&intf->mac, vif->addr, ETH_ALEN); /* * The MAC adddress must be configured after the device * has been initialized. Otherwise the device can reset * the MAC registers. */ - rt2x00lib_config_intf(rt2x00dev, intf, conf->type, intf->mac, NULL); + rt2x00lib_config_intf(rt2x00dev, intf, vif->type, intf->mac, NULL); /* * Some filters depend on the current working mode. We can force @@ -296,10 +296,10 @@ int rt2x00mac_add_interface(struct ieee80211_hw *hw, EXPORT_SYMBOL_GPL(rt2x00mac_add_interface); void rt2x00mac_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rt2x00_dev *rt2x00dev = hw->priv; - struct rt2x00_intf *intf = vif_to_intf(conf->vif); + struct rt2x00_intf *intf = vif_to_intf(vif); /* * Don't allow interfaces to be remove while @@ -307,11 +307,11 @@ void rt2x00mac_remove_interface(struct ieee80211_hw *hw, * no interface is present. */ if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags) || - (conf->type == NL80211_IFTYPE_AP && !rt2x00dev->intf_ap_count) || - (conf->type != NL80211_IFTYPE_AP && !rt2x00dev->intf_sta_count)) + (vif->type == NL80211_IFTYPE_AP && !rt2x00dev->intf_ap_count) || + (vif->type != NL80211_IFTYPE_AP && !rt2x00dev->intf_sta_count)) return; - if (conf->type == NL80211_IFTYPE_AP) + if (vif->type == NL80211_IFTYPE_AP) rt2x00dev->intf_ap_count--; else rt2x00dev->intf_sta_count--; diff --git a/drivers/net/wireless/rtl818x/rtl8180_dev.c b/drivers/net/wireless/rtl818x/rtl8180_dev.c index f01f1ef9e3b..5a2b7199f5d 100644 --- a/drivers/net/wireless/rtl818x/rtl8180_dev.c +++ b/drivers/net/wireless/rtl818x/rtl8180_dev.c @@ -652,7 +652,7 @@ static void rtl8180_stop(struct ieee80211_hw *dev) } static int rtl8180_add_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rtl8180_priv *priv = dev->priv; @@ -662,27 +662,27 @@ static int rtl8180_add_interface(struct ieee80211_hw *dev, if (priv->vif) return -EBUSY; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: break; default: return -EOPNOTSUPP; } - priv->vif = conf->vif; + priv->vif = vif; rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG); rtl818x_iowrite32(priv, (__le32 __iomem *)&priv->map->MAC[0], - le32_to_cpu(*(__le32 *)conf->mac_addr)); + le32_to_cpu(*(__le32 *)vif->addr)); rtl818x_iowrite16(priv, (__le16 __iomem *)&priv->map->MAC[4], - le16_to_cpu(*(__le16 *)(conf->mac_addr + 4))); + le16_to_cpu(*(__le16 *)(vif->addr + 4))); rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL); return 0; } static void rtl8180_remove_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rtl8180_priv *priv = dev->priv; priv->vif = NULL; diff --git a/drivers/net/wireless/rtl818x/rtl8187_dev.c b/drivers/net/wireless/rtl818x/rtl8187_dev.c index 1cb0eff4622..f336c63053c 100644 --- a/drivers/net/wireless/rtl818x/rtl8187_dev.c +++ b/drivers/net/wireless/rtl818x/rtl8187_dev.c @@ -1018,7 +1018,7 @@ static void rtl8187_stop(struct ieee80211_hw *dev) } static int rtl8187_add_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rtl8187_priv *priv = dev->priv; int i; @@ -1028,7 +1028,7 @@ static int rtl8187_add_interface(struct ieee80211_hw *dev, if (priv->vif) goto exit; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: break; default: @@ -1036,12 +1036,12 @@ static int rtl8187_add_interface(struct ieee80211_hw *dev, } ret = 0; - priv->vif = conf->vif; + priv->vif = vif; rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_CONFIG); for (i = 0; i < ETH_ALEN; i++) rtl818x_iowrite8(priv, &priv->map->MAC[i], - ((u8 *)conf->mac_addr)[i]); + ((u8 *)vif->addr)[i]); rtl818x_iowrite8(priv, &priv->map->EEPROM_CMD, RTL818X_EEPROM_CMD_NORMAL); exit: @@ -1050,7 +1050,7 @@ exit: } static void rtl8187_remove_interface(struct ieee80211_hw *dev, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct rtl8187_priv *priv = dev->priv; mutex_lock(&priv->conf_mutex); diff --git a/drivers/net/wireless/wl12xx/wl1251_main.c b/drivers/net/wireless/wl12xx/wl1251_main.c index 6aeffbe9e40..4e373f3dbc4 100644 --- a/drivers/net/wireless/wl12xx/wl1251_main.c +++ b/drivers/net/wireless/wl12xx/wl1251_main.c @@ -511,13 +511,13 @@ static void wl1251_op_stop(struct ieee80211_hw *hw) } static int wl1251_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct wl1251 *wl = hw->priv; int ret = 0; wl1251_debug(DEBUG_MAC80211, "mac80211 add interface type %d mac %pM", - conf->type, conf->mac_addr); + vif->type, vif->addr); mutex_lock(&wl->mutex); if (wl->vif) { @@ -525,9 +525,9 @@ static int wl1251_op_add_interface(struct ieee80211_hw *hw, goto out; } - wl->vif = conf->vif; + wl->vif = vif; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: wl->bss_type = BSS_TYPE_STA_BSS; break; @@ -539,8 +539,8 @@ static int wl1251_op_add_interface(struct ieee80211_hw *hw, goto out; } - if (memcmp(wl->mac_addr, conf->mac_addr, ETH_ALEN)) { - memcpy(wl->mac_addr, conf->mac_addr, ETH_ALEN); + if (memcmp(wl->mac_addr, vif->addr, ETH_ALEN)) { + memcpy(wl->mac_addr, vif->addr, ETH_ALEN); SET_IEEE80211_PERM_ADDR(wl->hw, wl->mac_addr); ret = wl1251_acx_station_id(wl); if (ret < 0) @@ -553,7 +553,7 @@ out: } static void wl1251_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct wl1251 *wl = hw->priv; diff --git a/drivers/net/wireless/wl12xx/wl1271_main.c b/drivers/net/wireless/wl12xx/wl1271_main.c index 4a997381a8d..e4867b895c4 100644 --- a/drivers/net/wireless/wl12xx/wl1271_main.c +++ b/drivers/net/wireless/wl12xx/wl1271_main.c @@ -1039,13 +1039,13 @@ static void wl1271_op_stop(struct ieee80211_hw *hw) } static int wl1271_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct wl1271 *wl = hw->priv; int ret = 0; wl1271_debug(DEBUG_MAC80211, "mac80211 add interface type %d mac %pM", - conf->type, conf->mac_addr); + vif->type, vif->addr); mutex_lock(&wl->mutex); if (wl->vif) { @@ -1053,9 +1053,9 @@ static int wl1271_op_add_interface(struct ieee80211_hw *hw, goto out; } - wl->vif = conf->vif; + wl->vif = vif; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_STATION: wl->bss_type = BSS_TYPE_STA_BSS; break; @@ -1075,7 +1075,7 @@ out: } static void wl1271_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct wl1271 *wl = hw->priv; diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 8a243732c51..c4f41d0016c 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -872,7 +872,7 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length) } static int zd_op_add_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct zd_mac *mac = zd_hw_mac(hw); @@ -880,22 +880,22 @@ static int zd_op_add_interface(struct ieee80211_hw *hw, if (mac->type != NL80211_IFTYPE_UNSPECIFIED) return -EOPNOTSUPP; - switch (conf->type) { + switch (vif->type) { case NL80211_IFTYPE_MONITOR: case NL80211_IFTYPE_MESH_POINT: case NL80211_IFTYPE_STATION: case NL80211_IFTYPE_ADHOC: - mac->type = conf->type; + mac->type = vif->type; break; default: return -EOPNOTSUPP; } - return zd_write_mac_addr(&mac->chip, conf->mac_addr); + return zd_write_mac_addr(&mac->chip, vif->addr); } static void zd_op_remove_interface(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { struct zd_mac *mac = zd_hw_mac(hw); mac->type = NL80211_IFTYPE_UNSPECIFIED; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 77ea34b0328..08d41357dcb 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -701,33 +701,6 @@ static inline bool ieee80211_vif_is_mesh(struct ieee80211_vif *vif) return false; } -/** - * struct ieee80211_if_init_conf - initial configuration of an interface - * - * @vif: pointer to a driver-use per-interface structure. The pointer - * itself is also used for various functions including - * ieee80211_beacon_get() and ieee80211_get_buffered_bc(). - * @type: one of &enum nl80211_iftype constants. Determines the type of - * added/removed interface. - * @mac_addr: pointer to MAC address of the interface. This pointer is valid - * until the interface is removed (i.e. it cannot be used after - * remove_interface() callback was called for this interface). - * - * This structure is used in add_interface() and remove_interface() - * callbacks of &struct ieee80211_hw. - * - * When you allow multiple interfaces to be added to your PHY, take care - * that the hardware can actually handle multiple MAC addresses. However, - * also take care that when there's no interface left with mac_addr != %NULL - * you remove the MAC address from the device to avoid acknowledging packets - * in pure monitor mode. - */ -struct ieee80211_if_init_conf { - enum nl80211_iftype type; - struct ieee80211_vif *vif; - void *mac_addr; -}; - /** * enum ieee80211_key_alg - key algorithm * @ALG_WEP: WEP40 or WEP104 @@ -1555,9 +1528,9 @@ struct ieee80211_ops { int (*start)(struct ieee80211_hw *hw); void (*stop)(struct ieee80211_hw *hw); int (*add_interface)(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); void (*remove_interface)(struct ieee80211_hw *hw, - struct ieee80211_if_init_conf *conf); + struct ieee80211_vif *vif); int (*config)(struct ieee80211_hw *hw, u32 changed); void (*bss_info_changed)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, @@ -1845,7 +1818,7 @@ void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw, /** * ieee80211_beacon_get_tim - beacon generation function * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @tim_offset: pointer to variable that will receive the TIM IE offset. * Set to 0 if invalid (in non-AP modes). * @tim_length: pointer to variable that will receive the TIM IE length, @@ -1873,7 +1846,7 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, /** * ieee80211_beacon_get - beacon generation function * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * * See ieee80211_beacon_get_tim(). */ @@ -1886,7 +1859,7 @@ static inline struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, /** * ieee80211_rts_get - RTS frame generation function * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @frame: pointer to the frame that is going to be protected by the RTS. * @frame_len: the frame length (in octets). * @frame_txctl: &struct ieee80211_tx_info of the frame. @@ -1905,7 +1878,7 @@ void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif, /** * ieee80211_rts_duration - Get the duration field for an RTS frame * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @frame_len: the length of the frame that is going to be protected by the RTS. * @frame_txctl: &struct ieee80211_tx_info of the frame. * @@ -1920,7 +1893,7 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw, /** * ieee80211_ctstoself_get - CTS-to-self frame generation function * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @frame: pointer to the frame that is going to be protected by the CTS-to-self. * @frame_len: the frame length (in octets). * @frame_txctl: &struct ieee80211_tx_info of the frame. @@ -1940,7 +1913,7 @@ void ieee80211_ctstoself_get(struct ieee80211_hw *hw, /** * ieee80211_ctstoself_duration - Get the duration field for a CTS-to-self frame * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @frame_len: the length of the frame that is going to be protected by the CTS-to-self. * @frame_txctl: &struct ieee80211_tx_info of the frame. * @@ -1956,7 +1929,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, /** * ieee80211_generic_frame_duration - Calculate the duration field for a frame * @hw: pointer obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @frame_len: the length of the frame. * @rate: the rate at which the frame is going to be transmitted. * @@ -1971,7 +1944,7 @@ __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, /** * ieee80211_get_buffered_bc - accessing buffered broadcast and multicast frames * @hw: pointer as obtained from ieee80211_alloc_hw(). - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * * Function for accessing buffered broadcast and multicast frames. If * hardware/firmware does not implement buffering of broadcast/multicast @@ -2139,7 +2112,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *sta, u16 tid); /** * ieee80211_start_tx_ba_cb - low level driver ready to aggregate. - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf + * @vif: &struct ieee80211_vif pointer from the add_interface callback * @ra: receiver address of the BA session recipient. * @tid: the TID to BA on. * @@ -2150,7 +2123,7 @@ void ieee80211_start_tx_ba_cb(struct ieee80211_vif *vif, u8 *ra, u16 tid); /** * ieee80211_start_tx_ba_cb_irqsafe - low level driver ready to aggregate. - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf + * @vif: &struct ieee80211_vif pointer from the add_interface callback * @ra: receiver address of the BA session recipient. * @tid: the TID to BA on. * @@ -2178,7 +2151,7 @@ int ieee80211_stop_tx_ba_session(struct ieee80211_sta *sta, u16 tid, /** * ieee80211_stop_tx_ba_cb - low level driver ready to stop aggregate. - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf + * @vif: &struct ieee80211_vif pointer from the add_interface callback * @ra: receiver address of the BA session recipient. * @tid: the desired TID to BA on. * @@ -2189,7 +2162,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_vif *vif, u8 *ra, u8 tid); /** * ieee80211_stop_tx_ba_cb_irqsafe - low level driver ready to stop aggregate. - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf + * @vif: &struct ieee80211_vif pointer from the add_interface callback * @ra: receiver address of the BA session recipient. * @tid: the desired TID to BA on. * @@ -2268,7 +2241,7 @@ void ieee80211_sta_block_awake(struct ieee80211_hw *hw, /** * ieee80211_beacon_loss - inform hardware does not receive beacons * - * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. + * @vif: &struct ieee80211_vif pointer from the add_interface callback. * * When beacon filtering is enabled with IEEE80211_HW_BEACON_FILTERING and * IEEE80211_CONF_PS is set, the driver needs to inform whenever the diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index cbe133bcdf3..bc7c8f55487 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -36,18 +36,18 @@ static inline void drv_stop(struct ieee80211_local *local) } static inline int drv_add_interface(struct ieee80211_local *local, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { - int ret = local->ops->add_interface(&local->hw, conf); - trace_drv_add_interface(local, vif_to_sdata(conf->vif), ret); + int ret = local->ops->add_interface(&local->hw, vif); + trace_drv_add_interface(local, vif_to_sdata(vif), ret); return ret; } static inline void drv_remove_interface(struct ieee80211_local *local, - struct ieee80211_if_init_conf *conf) + struct ieee80211_vif *vif) { - local->ops->remove_interface(&local->hw, conf); - trace_drv_remove_interface(local, vif_to_sdata(conf->vif)); + local->ops->remove_interface(&local->hw, vif); + trace_drv_remove_interface(local, vif_to_sdata(vif)); } static inline int drv_config(struct ieee80211_local *local, u32 changed) diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 7d410f15281..00a1f4ccdaf 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -96,7 +96,6 @@ static int ieee80211_open(struct net_device *dev) struct ieee80211_sub_if_data *nsdata; struct ieee80211_local *local = sdata->local; struct sta_info *sta; - struct ieee80211_if_init_conf conf; u32 changed = 0; int res; u32 hw_reconf_flags = 0; @@ -248,10 +247,7 @@ static int ieee80211_open(struct net_device *dev) ieee80211_configure_filter(local); break; default: - conf.vif = &sdata->vif; - conf.type = sdata->vif.type; - conf.mac_addr = sdata->vif.addr; - res = drv_add_interface(local, &conf); + res = drv_add_interface(local, &sdata->vif); if (res) goto err_stop; @@ -334,7 +330,7 @@ static int ieee80211_open(struct net_device *dev) return 0; err_del_interface: - drv_remove_interface(local, &conf); + drv_remove_interface(local, &sdata->vif); err_stop: if (!local->open_count) drv_stop(local); @@ -349,7 +345,6 @@ static int ieee80211_stop(struct net_device *dev) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; - struct ieee80211_if_init_conf conf; struct sta_info *sta; unsigned long flags; struct sk_buff *skb, *tmp; @@ -533,12 +528,9 @@ static int ieee80211_stop(struct net_device *dev) BSS_CHANGED_BEACON_ENABLED); } - conf.vif = &sdata->vif; - conf.type = sdata->vif.type; - conf.mac_addr = sdata->vif.addr; /* disable all keys for as long as this netdev is down */ ieee80211_disable_keys(sdata); - drv_remove_interface(local, &conf); + drv_remove_interface(local, &sdata->vif); } sdata->bss = NULL; diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index 913dc7e3b29..47f818959ad 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -10,7 +10,6 @@ int __ieee80211_suspend(struct ieee80211_hw *hw) { struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata; - struct ieee80211_if_init_conf conf; struct sta_info *sta; unsigned long flags; @@ -100,10 +99,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw) ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_BEACON_ENABLED); - conf.vif = &sdata->vif; - conf.type = sdata->vif.type; - conf.mac_addr = sdata->vif.addr; - drv_remove_interface(local, &conf); + drv_remove_interface(local, &sdata->vif); } /* stop hardware - this must stop RX */ diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 1fdb80ff924..4b930308b1f 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -1075,7 +1075,6 @@ int ieee80211_reconfig(struct ieee80211_local *local) { struct ieee80211_hw *hw = &local->hw; struct ieee80211_sub_if_data *sdata; - struct ieee80211_if_init_conf conf; struct sta_info *sta; unsigned long flags; int res; @@ -1094,12 +1093,8 @@ int ieee80211_reconfig(struct ieee80211_local *local) list_for_each_entry(sdata, &local->interfaces, list) { if (sdata->vif.type != NL80211_IFTYPE_AP_VLAN && sdata->vif.type != NL80211_IFTYPE_MONITOR && - ieee80211_sdata_running(sdata)) { - conf.vif = &sdata->vif; - conf.type = sdata->vif.type; - conf.mac_addr = sdata->vif.addr; - res = drv_add_interface(local, &conf); - } + ieee80211_sdata_running(sdata)) + res = drv_add_interface(local, &sdata->vif); } /* add STAs back */ -- cgit v1.2.3-70-g09d2 From e1781ed33a8809c58ad6c3b6d432d656446efa43 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Wed, 23 Dec 2009 13:15:47 +0100 Subject: mac80211: annotate sleeping driver ops To make it easier to notice cases of calling sleeping ops in atomic context, annotate driver-ops.h with appropiate might_sleep() calls. At the same time, also document in mac80211.h the op functions with missing contexts. mac80211 doesn't seem to use get_tx_stats anywhere currently. Just to be on the safe side, I documented it to be atomic, but hopefully the op can be removed in the future. Compile-tested only. Signed-off-by: Kalle Valo Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 43 +++++++++++++++++++++++---------- net/mac80211/driver-ops.h | 60 +++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 87 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 08d41357dcb..5ee666ae4c8 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1383,7 +1383,7 @@ enum ieee80211_ampdu_mlme_action { * When the device is started it should not have a MAC address * to avoid acknowledging frames before a non-monitor device * is added. - * Must be implemented. + * Must be implemented and can sleep. * * @stop: Called after last netdevice attached to the hardware * is disabled. This should turn off the hardware (at least @@ -1391,7 +1391,7 @@ enum ieee80211_ampdu_mlme_action { * May be called right after add_interface if that rejects * an interface. If you added any work onto the mac80211 workqueue * you should ensure to cancel it on this callback. - * Must be implemented. + * Must be implemented and can sleep. * * @add_interface: Called when a netdevice attached to the hardware is * enabled. Because it is not called for monitor mode devices, @start @@ -1401,7 +1401,7 @@ enum ieee80211_ampdu_mlme_action { * interface is given in the conf parameter. * The callback may refuse to add an interface by returning a * negative error code (which will be seen in userspace.) - * Must be implemented. + * Must be implemented and can sleep. * * @remove_interface: Notifies a driver that an interface is going down. * The @stop callback is called after this if it is the last interface @@ -1410,19 +1410,20 @@ enum ieee80211_ampdu_mlme_action { * must be cleared so the device no longer acknowledges packets, * the mac_addr member of the conf structure is, however, set to the * MAC address of the device going away. - * Hence, this callback must be implemented. + * Hence, this callback must be implemented. It can sleep. * * @config: Handler for configuration requests. IEEE 802.11 code calls this * function to change hardware configuration, e.g., channel. * This function should never fail but returns a negative error code - * if it does. + * if it does. The callback can sleep. * * @bss_info_changed: Handler for configuration requests related to BSS * parameters that may vary during BSS's lifespan, and may affect low * level driver (e.g. assoc/disassoc status, erp parameters). * This function should not be used if no BSS has been set, unless * for association indication. The @changed parameter indicates which - * of the bss parameters has changed when a call is made. + * of the bss parameters has changed when a call is made. The callback + * can sleep. * * @prepare_multicast: Prepare for multicast filter configuration. * This callback is optional, and its return value is passed @@ -1430,20 +1431,22 @@ enum ieee80211_ampdu_mlme_action { * * @configure_filter: Configure the device's RX filter. * See the section "Frame filtering" for more information. - * This callback must be implemented. + * This callback must be implemented and can sleep. * * @set_tim: Set TIM bit. mac80211 calls this function when a TIM bit * must be set or cleared for a given STA. Must be atomic. * * @set_key: See the section "Hardware crypto acceleration" - * This callback can sleep, and is only called between add_interface - * and remove_interface calls, i.e. while the given virtual interface + * This callback is only called between add_interface and + * remove_interface calls, i.e. while the given virtual interface * is enabled. * Returns a negative error code if the key can't be added. + * The callback can sleep. * * @update_tkip_key: See the section "Hardware crypto acceleration" * This callback will be called in the context of Rx. Called for drivers * which set IEEE80211_KEY_FLAG_TKIP_REQ_RX_P1_KEY. + * The callback can sleep. * * @hw_scan: Ask the hardware to service the scan request, no need to start * the scan state machine in stack. The scan must honour the channel @@ -1457,21 +1460,28 @@ enum ieee80211_ampdu_mlme_action { * When the scan finishes, ieee80211_scan_completed() must be called; * note that it also must be called when the scan cannot finish due to * any error unless this callback returned a negative error code. + * The callback can sleep. * * @sw_scan_start: Notifier function that is called just before a software scan * is started. Can be NULL, if the driver doesn't need this notification. + * The callback can sleep. * - * @sw_scan_complete: Notifier function that is called just after a software scan - * finished. Can be NULL, if the driver doesn't need this notification. + * @sw_scan_complete: Notifier function that is called just after a + * software scan finished. Can be NULL, if the driver doesn't need + * this notification. + * The callback can sleep. * * @get_stats: Return low-level statistics. * Returns zero if statistics are available. + * The callback can sleep. * * @get_tkip_seq: If your device implements TKIP encryption in hardware this * callback should be provided to read the TKIP transmit IVs (both IV32 * and IV16) for the given key from hardware. + * The callback must be atomic. * * @set_rts_threshold: Configuration of RTS threshold (if device needs it) + * The callback can sleep. * * @sta_notify: Notifies low level driver about addition, removal or power * state transition of an associated station, AP, IBSS/WDS/mesh peer etc. @@ -1480,30 +1490,36 @@ enum ieee80211_ampdu_mlme_action { * @conf_tx: Configure TX queue parameters (EDCF (aifs, cw_min, cw_max), * bursting) for a hardware TX queue. * Returns a negative error code on failure. + * The callback can sleep. * * @get_tx_stats: Get statistics of the current TX queue status. This is used * to get number of currently queued packets (queue length), maximum queue * size (limit), and total number of packets sent using each TX queue * (count). The 'stats' pointer points to an array that has hw->queues * items. + * The callback must be atomic. * * @get_tsf: Get the current TSF timer value from firmware/hardware. Currently, * this is only used for IBSS mode BSSID merging and debugging. Is not a * required function. + * The callback can sleep. * * @set_tsf: Set the TSF timer to the specified value in the firmware/hardware. * Currently, this is only used for IBSS mode debugging. Is not a * required function. + * The callback can sleep. * * @reset_tsf: Reset the TSF timer and allow firmware/hardware to synchronize * with other STAs in the IBSS. This is only used in IBSS mode. This * function is optional if the firmware/hardware takes full care of * TSF synchronization. + * The callback can sleep. * * @tx_last_beacon: Determine whether the last IBSS beacon was sent by us. * This is needed only for IBSS mode and the result of this function is * used to determine whether to reply to Probe Requests. * Returns non-zero if this device sent the last beacon. + * The callback can sleep. * * @ampdu_action: Perform a certain A-MPDU action * The RA/TID combination determines the destination and TID we want @@ -1512,16 +1528,19 @@ enum ieee80211_ampdu_mlme_action { * is the first frame we expect to perform the action on. Notice * that TX/RX_STOP can pass NULL for this parameter. * Returns a negative error code on failure. + * The callback must be atomic. * * @rfkill_poll: Poll rfkill hardware state. If you need this, you also * need to set wiphy->rfkill_poll to %true before registration, * and need to call wiphy_rfkill_set_hw_state() in the callback. + * The callback can sleep. * * @testmode_cmd: Implement a cfg80211 test mode command. + * The callback can sleep. * * @flush: Flush all pending frames from the hardware queue, making sure * that the hardware queues are empty. If the parameter @drop is set - * to %true, pending frames may be dropped. + * to %true, pending frames may be dropped. The callback can sleep. */ struct ieee80211_ops { int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index bc7c8f55487..8757ea73d54 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -14,6 +14,8 @@ static inline int drv_start(struct ieee80211_local *local) { int ret; + might_sleep(); + local->started = true; smp_mb(); ret = local->ops->start(&local->hw); @@ -23,6 +25,8 @@ static inline int drv_start(struct ieee80211_local *local) static inline void drv_stop(struct ieee80211_local *local) { + might_sleep(); + local->ops->stop(&local->hw); trace_drv_stop(local); @@ -38,7 +42,11 @@ static inline void drv_stop(struct ieee80211_local *local) static inline int drv_add_interface(struct ieee80211_local *local, struct ieee80211_vif *vif) { - int ret = local->ops->add_interface(&local->hw, vif); + int ret; + + might_sleep(); + + ret = local->ops->add_interface(&local->hw, vif); trace_drv_add_interface(local, vif_to_sdata(vif), ret); return ret; } @@ -46,13 +54,19 @@ static inline int drv_add_interface(struct ieee80211_local *local, static inline void drv_remove_interface(struct ieee80211_local *local, struct ieee80211_vif *vif) { + might_sleep(); + local->ops->remove_interface(&local->hw, vif); trace_drv_remove_interface(local, vif_to_sdata(vif)); } static inline int drv_config(struct ieee80211_local *local, u32 changed) { - int ret = local->ops->config(&local->hw, changed); + int ret; + + might_sleep(); + + ret = local->ops->config(&local->hw, changed); trace_drv_config(local, changed, ret); return ret; } @@ -62,6 +76,8 @@ static inline void drv_bss_info_changed(struct ieee80211_local *local, struct ieee80211_bss_conf *info, u32 changed) { + might_sleep(); + if (local->ops->bss_info_changed) local->ops->bss_info_changed(&local->hw, &sdata->vif, info, changed); trace_drv_bss_info_changed(local, sdata, info, changed); @@ -111,7 +127,11 @@ static inline int drv_set_key(struct ieee80211_local *local, struct ieee80211_sta *sta, struct ieee80211_key_conf *key) { - int ret = local->ops->set_key(&local->hw, cmd, &sdata->vif, sta, key); + int ret; + + might_sleep(); + + ret = local->ops->set_key(&local->hw, cmd, &sdata->vif, sta, key); trace_drv_set_key(local, cmd, sdata, sta, key, ret); return ret; } @@ -121,6 +141,8 @@ static inline void drv_update_tkip_key(struct ieee80211_local *local, const u8 *address, u32 iv32, u16 *phase1key) { + might_sleep(); + if (local->ops->update_tkip_key) local->ops->update_tkip_key(&local->hw, conf, address, iv32, phase1key); @@ -130,13 +152,19 @@ static inline void drv_update_tkip_key(struct ieee80211_local *local, static inline int drv_hw_scan(struct ieee80211_local *local, struct cfg80211_scan_request *req) { - int ret = local->ops->hw_scan(&local->hw, req); + int ret; + + might_sleep(); + + ret = local->ops->hw_scan(&local->hw, req); trace_drv_hw_scan(local, req, ret); return ret; } static inline void drv_sw_scan_start(struct ieee80211_local *local) { + might_sleep(); + if (local->ops->sw_scan_start) local->ops->sw_scan_start(&local->hw); trace_drv_sw_scan_start(local); @@ -144,6 +172,8 @@ static inline void drv_sw_scan_start(struct ieee80211_local *local) static inline void drv_sw_scan_complete(struct ieee80211_local *local) { + might_sleep(); + if (local->ops->sw_scan_complete) local->ops->sw_scan_complete(&local->hw); trace_drv_sw_scan_complete(local); @@ -154,6 +184,8 @@ static inline int drv_get_stats(struct ieee80211_local *local, { int ret = -EOPNOTSUPP; + might_sleep(); + if (local->ops->get_stats) ret = local->ops->get_stats(&local->hw, stats); trace_drv_get_stats(local, stats, ret); @@ -173,6 +205,9 @@ static inline int drv_set_rts_threshold(struct ieee80211_local *local, u32 value) { int ret = 0; + + might_sleep(); + if (local->ops->set_rts_threshold) ret = local->ops->set_rts_threshold(&local->hw, value); trace_drv_set_rts_threshold(local, value, ret); @@ -193,6 +228,9 @@ static inline int drv_conf_tx(struct ieee80211_local *local, u16 queue, const struct ieee80211_tx_queue_params *params) { int ret = -EOPNOTSUPP; + + might_sleep(); + if (local->ops->conf_tx) ret = local->ops->conf_tx(&local->hw, queue, params); trace_drv_conf_tx(local, queue, params, ret); @@ -210,6 +248,9 @@ static inline int drv_get_tx_stats(struct ieee80211_local *local, static inline u64 drv_get_tsf(struct ieee80211_local *local) { u64 ret = -1ULL; + + might_sleep(); + if (local->ops->get_tsf) ret = local->ops->get_tsf(&local->hw); trace_drv_get_tsf(local, ret); @@ -218,6 +259,8 @@ static inline u64 drv_get_tsf(struct ieee80211_local *local) static inline void drv_set_tsf(struct ieee80211_local *local, u64 tsf) { + might_sleep(); + if (local->ops->set_tsf) local->ops->set_tsf(&local->hw, tsf); trace_drv_set_tsf(local, tsf); @@ -225,6 +268,8 @@ static inline void drv_set_tsf(struct ieee80211_local *local, u64 tsf) static inline void drv_reset_tsf(struct ieee80211_local *local) { + might_sleep(); + if (local->ops->reset_tsf) local->ops->reset_tsf(&local->hw); trace_drv_reset_tsf(local); @@ -233,6 +278,9 @@ static inline void drv_reset_tsf(struct ieee80211_local *local) static inline int drv_tx_last_beacon(struct ieee80211_local *local) { int ret = 1; + + might_sleep(); + if (local->ops->tx_last_beacon) ret = local->ops->tx_last_beacon(&local->hw); trace_drv_tx_last_beacon(local, ret); @@ -256,12 +304,16 @@ static inline int drv_ampdu_action(struct ieee80211_local *local, static inline void drv_rfkill_poll(struct ieee80211_local *local) { + might_sleep(); + if (local->ops->rfkill_poll) local->ops->rfkill_poll(&local->hw); } static inline void drv_flush(struct ieee80211_local *local, bool drop) { + might_sleep(); + trace_drv_flush(local, drop); if (local->ops->flush) local->ops->flush(&local->hw, drop); -- cgit v1.2.3-70-g09d2 From c1c5523dd1517250cac8b15a4acbc237c24a67d4 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Wed, 23 Dec 2009 01:27:48 +0000 Subject: can/netlink: add CAN_CTRLMODE_ONE_SHOT This patch adds the flag CAN_CTRLMODE_ONE_SHOT. It is used as mask or flag in the "struct can_ctrlmode". It allows userspace via netlink to set a CAN controller into the special "one-shot" mode. In this mode, if supported by the CAN controller, it tries only once to deliver a CAN frame and aborts it if an error (e.g.: arbitration lost) happens. Signed-off-by: Marc Kleine-Budde Acked-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- include/linux/can/netlink.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/can/netlink.h b/include/linux/can/netlink.h index 9ecbb7871c0..c818335fbb1 100644 --- a/include/linux/can/netlink.h +++ b/include/linux/can/netlink.h @@ -80,6 +80,7 @@ struct can_ctrlmode { #define CAN_CTRLMODE_LOOPBACK 0x1 /* Loopback mode */ #define CAN_CTRLMODE_LISTENONLY 0x2 /* Listen-only mode */ #define CAN_CTRLMODE_3_SAMPLES 0x4 /* Triple sampling mode */ +#define CAN_CTRLMODE_ONE_SHOT 0x8 /* One-Shot mode */ /* * CAN device statistics -- cgit v1.2.3-70-g09d2 From 6f7edb4881bf51300060e89915926e070ace8c4d Mon Sep 17 00:00:00 2001 From: "Catalin(ux) M. BOIE" Date: Tue, 5 Jan 2010 05:50:24 +0100 Subject: IPVS: Allow boot time change of hash size I was very frustrated about the fact that I have to recompile the kernel to change the hash size. So, I created this patch. If IPVS is built-in you can append ip_vs.conn_tab_bits=?? to kernel command line, or, if you built IPVS as modules, you can add options ip_vs conn_tab_bits=??. To keep everything backward compatible, you still can select the size at compile time, and that will be used as default. It has been about a year since this patch was originally posted and subsequently dropped on the basis of insufficient test data. Mark Bergsma has provided the following test results which seem to strongly support the need for larger hash table sizes: We do however run into the same problem with the default setting (212 = 4096 entries), as most of our LVS balancers handle around a million connections/SLAB entries at any point in time (around 100-150 kpps load). With only 4096 hash table entries this implies that each entry consists of a linked list of 256 connections *on average*. To provide some statistics, I did an oprofile run on an 2.6.31 kernel, with both the default 4096 table size, and the same kernel recompiled with IP_VS_CONN_TAB_BITS set to 18 (218 = 262144 entries). I built a quick test setup with a part of Wikimedia/Wikipedia's live traffic mirrored by the switch to the test host. With the default setting, at ~ 120 kpps packet load we saw a typical %si CPU usage of around 30-35%, and oprofile reported a hot spot in ip_vs_conn_in_get: samples % image name app name symbol name 1719761 42.3741 ip_vs.ko ip_vs.ko ip_vs_conn_in_get 302577 7.4554 bnx2 bnx2 /bnx2 181984 4.4840 vmlinux vmlinux __ticket_spin_lock 128636 3.1695 vmlinux vmlinux ip_route_input 74345 1.8318 ip_vs.ko ip_vs.ko ip_vs_conn_out_get 68482 1.6874 vmlinux vmlinux mwait_idle After loading the recompiled kernel with 218 entries, %si CPU usage dropped in half to around 12-18%, and oprofile looks much healthier, with only 7% spent in ip_vs_conn_in_get: samples % image name app name symbol name 265641 14.4616 bnx2 bnx2 /bnx2 143251 7.7986 vmlinux vmlinux __ticket_spin_lock 140661 7.6576 ip_vs.ko ip_vs.ko ip_vs_conn_in_get 94364 5.1372 vmlinux vmlinux mwait_idle 86267 4.6964 vmlinux vmlinux ip_route_input [ horms@verge.net.au: trivial up-port and minor style fixes ] Signed-off-by: Catalin(ux) M. BOIE Cc: Mark Bergsma Signed-off-by: Simon Horman Signed-off-by: Patrick McHardy --- include/net/ip_vs.h | 16 +++++----------- net/netfilter/ipvs/Kconfig | 4 ++++ net/netfilter/ipvs/ip_vs_conn.c | 42 ++++++++++++++++++++++++++++++----------- net/netfilter/ipvs/ip_vs_ctl.c | 8 ++++---- 4 files changed, 44 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index 8dc3296b7be..a816c37417b 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -26,6 +26,11 @@ #include /* for struct ipv6hdr */ #include /* for ipv6_addr_copy */ + +/* Connections' size value needed by ip_vs_ctl.c */ +extern int ip_vs_conn_tab_size; + + struct ip_vs_iphdr { int len; __u8 protocol; @@ -592,17 +597,6 @@ extern void ip_vs_init_hash_table(struct list_head *table, int rows); * (from ip_vs_conn.c) */ -/* - * IPVS connection entry hash table - */ -#ifndef CONFIG_IP_VS_TAB_BITS -#define CONFIG_IP_VS_TAB_BITS 12 -#endif - -#define IP_VS_CONN_TAB_BITS CONFIG_IP_VS_TAB_BITS -#define IP_VS_CONN_TAB_SIZE (1 << IP_VS_CONN_TAB_BITS) -#define IP_VS_CONN_TAB_MASK (IP_VS_CONN_TAB_SIZE - 1) - enum { IP_VS_DIR_INPUT = 0, IP_VS_DIR_OUTPUT, diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig index 79a69805221..c71e543d254 100644 --- a/net/netfilter/ipvs/Kconfig +++ b/net/netfilter/ipvs/Kconfig @@ -68,6 +68,10 @@ config IP_VS_TAB_BITS each hash entry uses 8 bytes, so you can estimate how much memory is needed for your box. + You can overwrite this number setting conn_tab_bits module parameter + or by appending ip_vs.conn_tab_bits=? to the kernel command line + if IP VS was compiled built-in. + comment "IPVS transport protocol load balancing support" config IP_VS_PROTO_TCP diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 27c30cf933d..60bb41a8d8d 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -40,6 +40,21 @@ #include +#ifndef CONFIG_IP_VS_TAB_BITS +#define CONFIG_IP_VS_TAB_BITS 12 +#endif + +/* + * Connection hash size. Default is what was selected at compile time. +*/ +int ip_vs_conn_tab_bits = CONFIG_IP_VS_TAB_BITS; +module_param_named(conn_tab_bits, ip_vs_conn_tab_bits, int, 0444); +MODULE_PARM_DESC(conn_tab_bits, "Set connections' hash size"); + +/* size and mask values */ +int ip_vs_conn_tab_size; +int ip_vs_conn_tab_mask; + /* * Connection hash table: for input and output packets lookups of IPVS */ @@ -125,11 +140,11 @@ static unsigned int ip_vs_conn_hashkey(int af, unsigned proto, if (af == AF_INET6) return jhash_3words(jhash(addr, 16, ip_vs_conn_rnd), (__force u32)port, proto, ip_vs_conn_rnd) - & IP_VS_CONN_TAB_MASK; + & ip_vs_conn_tab_mask; #endif return jhash_3words((__force u32)addr->ip, (__force u32)port, proto, ip_vs_conn_rnd) - & IP_VS_CONN_TAB_MASK; + & ip_vs_conn_tab_mask; } @@ -760,7 +775,7 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) int idx; struct ip_vs_conn *cp; - for(idx = 0; idx < IP_VS_CONN_TAB_SIZE; idx++) { + for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { ct_read_lock_bh(idx); list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { if (pos-- == 0) { @@ -797,7 +812,7 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) idx = l - ip_vs_conn_tab; ct_read_unlock_bh(idx); - while (++idx < IP_VS_CONN_TAB_SIZE) { + while (++idx < ip_vs_conn_tab_size) { ct_read_lock_bh(idx); list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { seq->private = &ip_vs_conn_tab[idx]; @@ -976,8 +991,8 @@ void ip_vs_random_dropentry(void) /* * Randomly scan 1/32 of the whole table every second */ - for (idx = 0; idx < (IP_VS_CONN_TAB_SIZE>>5); idx++) { - unsigned hash = net_random() & IP_VS_CONN_TAB_MASK; + for (idx = 0; idx < (ip_vs_conn_tab_size>>5); idx++) { + unsigned hash = net_random() & ip_vs_conn_tab_mask; /* * Lock is actually needed in this loop. @@ -1029,7 +1044,7 @@ static void ip_vs_conn_flush(void) struct ip_vs_conn *cp; flush_again: - for (idx=0; idx Date: Mon, 4 Jan 2010 02:02:47 +0000 Subject: Phonet: zero-copy aligned GPRS RX MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Newer Nokia cellular modems can use aligned payload for their GPRS pipe. Signed-off-by: Rémi Denis-Courmont Signed-off-by: David S. Miller --- include/net/phonet/pep.h | 1 + net/phonet/pep-gprs.c | 4 ++-- net/phonet/pep.c | 3 +++ 3 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/phonet/pep.h b/include/net/phonet/pep.h index 4c61cdce4e5..c37162547a9 100644 --- a/include/net/phonet/pep.h +++ b/include/net/phonet/pep.h @@ -77,6 +77,7 @@ static inline struct pnpipehdr *pnp_hdr(struct sk_buff *skb) enum { PNS_PIPE_DATA = 0x20, + PNS_PIPE_ALIGNED_DATA, PNS_PEP_CONNECT_REQ = 0x40, PNS_PEP_CONNECT_RESP, diff --git a/net/phonet/pep-gprs.c b/net/phonet/pep-gprs.c index d183509d3fa..d01208968c8 100644 --- a/net/phonet/pep-gprs.c +++ b/net/phonet/pep-gprs.c @@ -96,11 +96,11 @@ static int gprs_recv(struct gprs_dev *gp, struct sk_buff *skb) goto drop; } - if (likely(skb_headroom(skb) & 3)) { + if (skb_headroom(skb) & 3) { struct sk_buff *rskb, *fs; int flen = 0; - /* Phonet Pipe data header is misaligned (3 bytes), + /* Phonet Pipe data header may be misaligned (3 bytes), * so wrap the IP packet as a single fragment of an head-less * socket buffer. The network stack will pull what it needs, * but at least, the whole IP payload is not memcpy'd. */ diff --git a/net/phonet/pep.c b/net/phonet/pep.c index b6356f3832f..e23e30907d3 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -354,6 +354,9 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb) queue = &pn->ctrlreq_queue; goto queue; + case PNS_PIPE_ALIGNED_DATA: + __skb_pull(skb, 1); + /* fall through */ case PNS_PIPE_DATA: __skb_pull(skb, 3); /* Pipe data header */ if (!pn_flow_safe(pn->rx_fc)) { -- cgit v1.2.3-70-g09d2 From fea93ecef619b5779ca6984568517b1685079b05 Mon Sep 17 00:00:00 2001 From: Rémi Denis-Courmont Date: Mon, 4 Jan 2010 02:02:48 +0000 Subject: Phonet: zero-copy GPRS TX MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Send aligned pipe payload if requested to do so. Then, the socket buffer needs not be fragmented anymore. Signed-off-by: Rémi Denis-Courmont Signed-off-by: David S. Miller --- include/net/phonet/pep.h | 2 ++ net/phonet/pep.c | 16 ++++++++++++++-- 2 files changed, 16 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/phonet/pep.h b/include/net/phonet/pep.h index c37162547a9..35672b1cf44 100644 --- a/include/net/phonet/pep.h +++ b/include/net/phonet/pep.h @@ -44,6 +44,7 @@ struct pep_sock { u8 rx_fc; /* RX flow control */ u8 tx_fc; /* TX flow control */ u8 init_enable; /* auto-enable at creation */ + u8 aligned; }; static inline struct pep_sock *pep_sk(struct sock *sk) @@ -139,6 +140,7 @@ enum { PN_PIPE_SB_NEGOTIATED_FC, PN_PIPE_SB_REQUIRED_FC_TX, PN_PIPE_SB_PREFERRED_FC_RX, + PN_PIPE_SB_ALIGNED_DATA, }; /* Phonet pipe flow control models */ diff --git a/net/phonet/pep.c b/net/phonet/pep.c index e23e30907d3..72db27e3fc0 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -444,6 +444,7 @@ static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb) struct sockaddr_pn dst; u16 peer_type; u8 pipe_handle, enabled, n_sb; + u8 aligned = 0; if (!pskb_pull(skb, sizeof(*hdr) + 4)) return -EINVAL; @@ -482,6 +483,9 @@ static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb) return -EINVAL; peer_type = (peer_type & 0xff00) | data[0]; break; + case PN_PIPE_SB_ALIGNED_DATA: + aligned = data[0] != 0; + break; } n_sb--; } @@ -513,6 +517,7 @@ static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb) newpn->rx_credits = 0; newpn->rx_fc = newpn->tx_fc = PN_LEGACY_FLOW_CONTROL; newpn->init_enable = enabled; + newpn->aligned = aligned; BUG_ON(!skb_queue_empty(&newsk->sk_receive_queue)); skb_queue_head(&newsk->sk_receive_queue, skb); @@ -832,11 +837,15 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb) return -ENOBUFS; } - skb_push(skb, 3); + skb_push(skb, 3 + pn->aligned); skb_reset_transport_header(skb); ph = pnp_hdr(skb); ph->utid = 0; - ph->message_id = PNS_PIPE_DATA; + if (pn->aligned) { + ph->message_id = PNS_PIPE_ALIGNED_DATA; + ph->data[0] = 0; /* padding */ + } else + ph->message_id = PNS_PIPE_DATA; ph->pipe_handle = pn->pipe_handle; return pn_skb_send(sk, skb, &pipe_srv); @@ -930,6 +939,9 @@ int pep_write(struct sock *sk, struct sk_buff *skb) struct sk_buff *rskb, *fs; int flen = 0; + if (pep_sk(sk)->aligned) + return pipe_skb_send(sk, skb); + rskb = alloc_skb(MAX_PNPIPE_HEADER, GFP_ATOMIC); if (!rskb) { kfree_skb(skb); -- cgit v1.2.3-70-g09d2 From 65324144b50bc7022cc9b6ca8f4a536a957019e3 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Tue, 5 Jan 2010 05:50:47 +0000 Subject: net: RFC3069, private VLAN proxy arp support This is to be used together with switch technologies, like RFC3069, that where the individual ports are not allowed to communicate with each other, but they are allowed to talk to the upstream router. As described in RFC 3069, it is possible to allow these hosts to communicate through the upstream router by proxy_arp'ing. This patch basically allow proxy arp replies back to the same interface (from which the ARP request/solicitation was received). Tunable per device via proc "proxy_arp_pvlan": /proc/sys/net/ipv4/conf/*/proxy_arp_pvlan This switch technology is known by different vendor names: - In RFC 3069 it is called VLAN Aggregation. - Cisco and Allied Telesyn call it Private VLAN. - Hewlett-Packard call it Source-Port filtering or port-isolation. - Ericsson call it MAC-Forced Forwarding (RFC Draft). Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 19 +++++++++++++ include/linux/inetdevice.h | 1 + include/linux/sysctl.h | 1 + net/ipv4/arp.c | 52 +++++++++++++++++++++++++++++++--- net/ipv4/devinet.c | 1 + net/ipv4/route.c | 7 ++++- 6 files changed, 76 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 006b39dec87..c532884f4fe 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -692,6 +692,25 @@ proxy_arp - BOOLEAN conf/{all,interface}/proxy_arp is set to TRUE, it will be disabled otherwise +proxy_arp_pvlan - BOOLEAN + Private VLAN proxy arp. + Basically allow proxy arp replies back to the same interface + (from which the ARP request/solicitation was received). + + This is done to support (ethernet) switch features, like RFC + 3069, where the individual ports are NOT allowed to + communicate with each other, but they are allowed to talk to + the upstream router. As described in RFC 3069, it is possible + to allow these hosts to communicate through the upstream + router by proxy_arp'ing. Don't need to be used together with + proxy_arp. + + This technology is known by different names: + In RFC 3069 it is called VLAN Aggregation. + Cisco and Allied Telesyn call it Private VLAN. + Hewlett-Packard call it Source-Port filtering or port-isolation. + Ericsson call it MAC-Forced Forwarding (RFC Draft). + shared_media - BOOLEAN Send(router) or accept(host) RFC1620 shared media redirects. Overrides ip_secure_redirects. diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 699e85c01a4..9a8c57467d3 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -88,6 +88,7 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev) #define IN_DEV_LOG_MARTIANS(in_dev) IN_DEV_ORCONF((in_dev), LOG_MARTIANS) #define IN_DEV_PROXY_ARP(in_dev) IN_DEV_ORCONF((in_dev), PROXY_ARP) +#define IN_DEV_PROXY_ARP_PVLAN(in_dev) IN_DEV_CONF_GET(in_dev, PROXY_ARP_PVLAN) #define IN_DEV_SHARED_MEDIA(in_dev) IN_DEV_ORCONF((in_dev), SHARED_MEDIA) #define IN_DEV_TX_REDIRECTS(in_dev) IN_DEV_ORCONF((in_dev), SEND_REDIRECTS) #define IN_DEV_SEC_REDIRECTS(in_dev) IN_DEV_ORCONF((in_dev), \ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 877ba039e6a..24ff7e3a0d5 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -482,6 +482,7 @@ enum NET_IPV4_CONF_ARP_ACCEPT=21, NET_IPV4_CONF_ARP_NOTIFY=22, NET_IPV4_CONF_ACCEPT_LOCAL=23, + NET_IPV4_CONF_PROXY_ARP_PVLAN=24, __NET_IPV4_CONF_MAX }; diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index c95cd93acf2..078709233bc 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -70,6 +70,7 @@ * bonding can change the skb before * sending (e.g. insert 8021q tag). * Harald Welte : convert to make use of jenkins hash + * Jesper D. Brouer: Proxy ARP PVLAN RFC 3069 support. */ #include @@ -524,12 +525,15 @@ int arp_bind_neighbour(struct dst_entry *dst) /* * Check if we can use proxy ARP for this path */ - -static inline int arp_fwd_proxy(struct in_device *in_dev, struct rtable *rt) +static inline int arp_fwd_proxy(struct in_device *in_dev, + struct net_device *dev, struct rtable *rt) { struct in_device *out_dev; int imi, omi = -1; + if (rt->u.dst.dev == dev) + return 0; + if (!IN_DEV_PROXY_ARP(in_dev)) return 0; @@ -547,6 +551,43 @@ static inline int arp_fwd_proxy(struct in_device *in_dev, struct rtable *rt) return (omi != imi && omi != -1); } +/* + * Check for RFC3069 proxy arp private VLAN (allow to send back to same dev) + * + * RFC3069 supports proxy arp replies back to the same interface. This + * is done to support (ethernet) switch features, like RFC 3069, where + * the individual ports are not allowed to communicate with each + * other, BUT they are allowed to talk to the upstream router. As + * described in RFC 3069, it is possible to allow these hosts to + * communicate through the upstream router, by proxy_arp'ing. + * + * RFC 3069: "VLAN Aggregation for Efficient IP Address Allocation" + * + * This technology is known by different names: + * In RFC 3069 it is called VLAN Aggregation. + * Cisco and Allied Telesyn call it Private VLAN. + * Hewlett-Packard call it Source-Port filtering or port-isolation. + * Ericsson call it MAC-Forced Forwarding (RFC Draft). + * + */ +static inline int arp_fwd_pvlan(struct in_device *in_dev, + struct net_device *dev, struct rtable *rt, + __be32 sip, __be32 tip) +{ + /* Private VLAN is only concerned about the same ethernet segment */ + if (rt->u.dst.dev != dev) + return 0; + + /* Don't reply on self probes (often done by windowz boxes)*/ + if (sip == tip) + return 0; + + if (IN_DEV_PROXY_ARP_PVLAN(in_dev)) + return 1; + else + return 0; +} + /* * Interface to link layer: send routine and receive handler. */ @@ -833,8 +874,11 @@ static int arp_process(struct sk_buff *skb) } goto out; } else if (IN_DEV_FORWARD(in_dev)) { - if (addr_type == RTN_UNICAST && rt->u.dst.dev != dev && - (arp_fwd_proxy(in_dev, rt) || pneigh_lookup(&arp_tbl, net, &tip, dev, 0))) { + if (addr_type == RTN_UNICAST && + (arp_fwd_proxy(in_dev, dev, rt) || + arp_fwd_pvlan(in_dev, dev, rt, sip, tip) || + pneigh_lookup(&arp_tbl, net, &tip, dev, 0))) + { n = neigh_event_ns(&arp_tbl, sha, &sip, dev); if (n) neigh_release(n); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 5cdbc102a41..0715f4cac39 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1407,6 +1407,7 @@ static struct devinet_sysctl_table { DEVINET_SYSCTL_RW_ENTRY(ARP_IGNORE, "arp_ignore"), DEVINET_SYSCTL_RW_ENTRY(ARP_ACCEPT, "arp_accept"), DEVINET_SYSCTL_RW_ENTRY(ARP_NOTIFY, "arp_notify"), + DEVINET_SYSCTL_RW_ENTRY(PROXY_ARP_PVLAN, "proxy_arp_pvlan"), DEVINET_SYSCTL_FLUSHING_ENTRY(NOXFRM, "disable_xfrm"), DEVINET_SYSCTL_FLUSHING_ENTRY(NOPOLICY, "disable_policy"), diff --git a/net/ipv4/route.c b/net/ipv4/route.c index e446496f564..1cc339441e7 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1988,8 +1988,13 @@ static int __mkroute_input(struct sk_buff *skb, if (skb->protocol != htons(ETH_P_IP)) { /* Not IP (i.e. ARP). Do not create route, if it is * invalid for proxy arp. DNAT routes are always valid. + * + * Proxy arp feature have been extended to allow, ARP + * replies back to the same interface, to support + * Private VLAN switch technologies. See arp.c. */ - if (out_dev == in_dev) { + if (out_dev == in_dev && + IN_DEV_PROXY_ARP_PVLAN(in_dev) == 0) { err = -EINVAL; goto cleanup; } -- cgit v1.2.3-70-g09d2 From 3c9732c06879d85f2fdf7ec69198c1d78da42a98 Mon Sep 17 00:00:00 2001 From: Giuseppe CAVALLARO Date: Wed, 6 Jan 2010 23:07:13 +0000 Subject: stmmac: add the new Header file for stmmac platform data Signed-off-by: Giuseppe Cavallaro Signed-off-by: David S. Miller --- include/linux/stmmac.h | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+) create mode 100644 include/linux/stmmac.h (limited to 'include') diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h new file mode 100644 index 00000000000..32bfd1a8a48 --- /dev/null +++ b/include/linux/stmmac.h @@ -0,0 +1,53 @@ +/******************************************************************************* + + Header file for stmmac platform data + + Copyright (C) 2009 STMicroelectronics Ltd + + This program is free software; you can redistribute it and/or modify it + under the terms and conditions of the GNU General Public License, + version 2, as published by the Free Software Foundation. + + This program is distributed in the hope it will be useful, but WITHOUT + ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + more details. + + You should have received a copy of the GNU General Public License along with + this program; if not, write to the Free Software Foundation, Inc., + 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. + + The full GNU General Public License is included in this distribution in + the file called "COPYING". + + Author: Giuseppe Cavallaro +*******************************************************************************/ + +#ifndef __STMMAC_PLATFORM_DATA +#define __STMMAC_PLATFORM_DATA + +/* platfrom data for platfrom device structure's platfrom_data field */ + +/* Private data for the STM on-board ethernet driver */ +struct plat_stmmacenet_data { + int bus_id; + int pbl; + int has_gmac; + void (*fix_mac_speed)(void *priv, unsigned int speed); + void (*bus_setup)(unsigned long ioaddr); +#ifdef CONFIG_STM_DRIVERS + struct stm_pad_config *pad_config; +#endif + void *bsp_priv; +}; + +struct plat_stmmacphy_data { + int bus_id; + int phy_addr; + unsigned int phy_mask; + int interface; + int (*phy_reset)(void *priv); + void *priv; +}; +#endif + -- cgit v1.2.3-70-g09d2 From d218d11133d888f9745802146a50255a4781d37a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 11 Jan 2010 16:28:01 -0800 Subject: tcp: Generalized TTL Security Mechanism This patch adds the kernel portions needed to implement RFC 5082 Generalized TTL Security Mechanism (GTSM). It is a lightweight security measure against forged packets causing DoS attacks (for BGP). This is already implemented the same way in BSD kernels. For the necessary Quagga patch http://www.gossamer-threads.com/lists/quagga/dev/17389 Description from Cisco http://www.cisco.com/en/US/docs/ios/12_3t/12_3t7/feature/guide/gt_btsh.html It does add one byte to each socket structure, but I did a little rearrangement to reuse a hole (on 64 bit), but it does grow the structure on 32 bit This should be documented on ip(4) man page and the Glibc in.h file also needs update. IPV6_MINHOPLIMIT should also be added (although BSD doesn't support that). Only TCP is supported, but could also be added to UDP, DCCP, SCTP if desired. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/in.h | 2 ++ include/net/inet_sock.h | 4 +++- net/ipv4/ip_sockglue.c | 14 +++++++++++++- net/ipv4/tcp_ipv4.c | 3 +++ 4 files changed, 21 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/in.h b/include/linux/in.h index b615649db12..583c76f9c30 100644 --- a/include/linux/in.h +++ b/include/linux/in.h @@ -84,6 +84,8 @@ struct in_addr { #define IP_ORIGDSTADDR 20 #define IP_RECVORIGDSTADDR IP_ORIGDSTADDR +#define IP_MINTTL 21 + /* IP_MTU_DISCOVER values */ #define IP_PMTUDISC_DONT 0 /* Never send DF frames */ #define IP_PMTUDISC_WANT 1 /* Use per route hints */ diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index bd4c53f75ac..83fd34437cf 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -122,10 +122,12 @@ struct inet_sock { __be32 inet_saddr; __s16 uc_ttl; __u16 cmsg_flags; - struct ip_options *opt; __be16 inet_sport; __u16 inet_id; + + struct ip_options *opt; __u8 tos; + __u8 min_ttl; __u8 mc_ttl; __u8 pmtudisc; __u8 recverr:1, diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index cafad9baff0..644dc43a55d 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -451,7 +451,8 @@ static int do_ip_setsockopt(struct sock *sk, int level, (1<transparent = !!val; break; + case IP_MINTTL: + if (optlen < 1) + goto e_inval; + if (val < 0 || val > 255) + goto e_inval; + inet->min_ttl = val; + break; + default: err = -ENOPROTOOPT; break; @@ -1198,6 +1207,9 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_TRANSPARENT: val = inet->transparent; break; + case IP_MINTTL: + val = inet->min_ttl; + break; default: release_sock(sk); return -ENOPROTOOPT; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 65b8ebfd078..382f667238e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1649,6 +1649,9 @@ int tcp_v4_rcv(struct sk_buff *skb) if (!sk) goto no_tcp_socket; + if (iph->ttl < inet_sk(sk)->min_ttl) + goto discard_and_relse; + process: if (sk->sk_state == TCP_TIME_WAIT) goto do_time_wait; -- cgit v1.2.3-70-g09d2 From 3ccd4c6167d3b39d52631767ebbf8b5677c5855d Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 12 Jan 2010 02:00:46 -0800 Subject: can: Unify droping of invalid tx skbs and netdev stats To prevent the CAN drivers to operate on invalid socketbuffers the skbs are now checked and silently dropped at the xmit-function consistently. Also the netdev stats are consistently using the CAN data length code (dlc) for [rx|tx]_bytes now. Signed-off-by: Oliver Hartkopp Acked-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/at91_can.c | 3 +++ drivers/net/can/bfin_can.c | 3 +++ drivers/net/can/mcp251x.c | 6 +----- drivers/net/can/mscan/mscan.c | 5 +---- drivers/net/can/sja1000/sja1000.c | 3 +++ drivers/net/can/ti_hecc.c | 4 +++- drivers/net/can/usb/ems_usb.c | 3 +++ drivers/net/can/vcan.c | 12 +++++++++--- include/linux/can/dev.h | 15 +++++++++++++++ 9 files changed, 41 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c index 166cc7e579c..f7287497ba6 100644 --- a/drivers/net/can/at91_can.c +++ b/drivers/net/can/at91_can.c @@ -342,6 +342,9 @@ static netdev_tx_t at91_start_xmit(struct sk_buff *skb, struct net_device *dev) unsigned int mb, prio; u32 reg_mid, reg_mcr; + if (can_dropped_invalid_skb(dev, skb)) + return NETDEV_TX_OK; + mb = get_tx_next_mb(priv); prio = get_tx_next_prio(priv); diff --git a/drivers/net/can/bfin_can.c b/drivers/net/can/bfin_can.c index 0ec1524523c..7e1926e79e9 100644 --- a/drivers/net/can/bfin_can.c +++ b/drivers/net/can/bfin_can.c @@ -318,6 +318,9 @@ static int bfin_can_start_xmit(struct sk_buff *skb, struct net_device *dev) u16 val; int i; + if (can_dropped_invalid_skb(dev, skb)) + return NETDEV_TX_OK; + netif_stop_queue(dev); /* fill id */ diff --git a/drivers/net/can/mcp251x.c b/drivers/net/can/mcp251x.c index 1a72ca066a1..afa2fa45fed 100644 --- a/drivers/net/can/mcp251x.c +++ b/drivers/net/can/mcp251x.c @@ -494,12 +494,8 @@ static netdev_tx_t mcp251x_hard_start_xmit(struct sk_buff *skb, return NETDEV_TX_BUSY; } - if (skb->len != sizeof(struct can_frame)) { - dev_err(&spi->dev, "dropping packet - bad length\n"); - dev_kfree_skb(skb); - net->stats.tx_dropped++; + if (can_dropped_invalid_skb(net, skb)) return NETDEV_TX_OK; - } netif_stop_queue(net); priv->tx_skb = skb; diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 500d18918bd..40827c128b6 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -204,11 +204,8 @@ static netdev_tx_t mscan_start_xmit(struct sk_buff *skb, struct net_device *dev) int i, rtr, buf_id; u32 can_id; - if (skb->len != sizeof(*frame) || frame->can_dlc > 8) { - kfree_skb(skb); - dev->stats.tx_dropped++; + if (can_dropped_invalid_skb(dev, skb)) return NETDEV_TX_OK; - } out_8(®s->cantier, 0); diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index 542a4f7255b..345304d779b 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -249,6 +249,9 @@ static netdev_tx_t sja1000_start_xmit(struct sk_buff *skb, uint8_t dreg; int i; + if (can_dropped_invalid_skb(dev, skb)) + return NETDEV_TX_OK; + netif_stop_queue(dev); fi = dlc = cf->can_dlc; diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index 5c993c2da52..7d370e32a7a 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -477,6 +477,9 @@ static netdev_tx_t ti_hecc_xmit(struct sk_buff *skb, struct net_device *ndev) u32 mbxno, mbx_mask, data; unsigned long flags; + if (can_dropped_invalid_skb(ndev, skb)) + return NETDEV_TX_OK; + mbxno = get_tx_head_mb(priv); mbx_mask = BIT(mbxno); spin_lock_irqsave(&priv->mbx_lock, flags); @@ -491,7 +494,6 @@ static netdev_tx_t ti_hecc_xmit(struct sk_buff *skb, struct net_device *ndev) spin_unlock_irqrestore(&priv->mbx_lock, flags); /* Prepare mailbox for transmission */ - data = min_t(u8, cf->can_dlc, 8); if (cf->can_id & CAN_RTR_FLAG) /* Remote transmission request */ data |= HECC_CANMCF_RTR; data |= get_tx_head_prio(priv) << 8; diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index efbb05c71bf..ddb17e25665 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -767,6 +767,9 @@ static netdev_tx_t ems_usb_start_xmit(struct sk_buff *skb, struct net_device *ne size_t size = CPC_HEADER_SIZE + CPC_MSG_HEADER_LEN + sizeof(struct cpc_can_msg); + if (can_dropped_invalid_skb(netdev, skb)) + return NETDEV_TX_OK; + /* create a URB, and a buffer for it, and copy the data to the URB */ urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { diff --git a/drivers/net/can/vcan.c b/drivers/net/can/vcan.c index 80ac5631398..d124d837ae5 100644 --- a/drivers/net/can/vcan.c +++ b/drivers/net/can/vcan.c @@ -47,6 +47,7 @@ #include #include #include +#include #include static __initdata const char banner[] = @@ -70,10 +71,11 @@ MODULE_PARM_DESC(echo, "Echo sent frames (for testing). Default: 0 (Off)"); static void vcan_rx(struct sk_buff *skb, struct net_device *dev) { + struct can_frame *cf = (struct can_frame *)skb->data; struct net_device_stats *stats = &dev->stats; stats->rx_packets++; - stats->rx_bytes += skb->len; + stats->rx_bytes += cf->can_dlc; skb->protocol = htons(ETH_P_CAN); skb->pkt_type = PACKET_BROADCAST; @@ -85,11 +87,15 @@ static void vcan_rx(struct sk_buff *skb, struct net_device *dev) static netdev_tx_t vcan_tx(struct sk_buff *skb, struct net_device *dev) { + struct can_frame *cf = (struct can_frame *)skb->data; struct net_device_stats *stats = &dev->stats; int loop; + if (can_dropped_invalid_skb(dev, skb)) + return NETDEV_TX_OK; + stats->tx_packets++; - stats->tx_bytes += skb->len; + stats->tx_bytes += cf->can_dlc; /* set flag whether this packet has to be looped back */ loop = skb->pkt_type == PACKET_LOOPBACK; @@ -103,7 +109,7 @@ static netdev_tx_t vcan_tx(struct sk_buff *skb, struct net_device *dev) * CAN core already did the echo for us */ stats->rx_packets++; - stats->rx_bytes += skb->len; + stats->rx_bytes += cf->can_dlc; } kfree_skb(skb); return NETDEV_TX_OK; diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 3db7767d2a1..7e7c98a3e90 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -60,6 +60,21 @@ struct can_priv { */ #define get_can_dlc(i) (min_t(__u8, (i), 8)) +/* Drop a given socketbuffer if it does not contain a valid CAN frame. */ +static inline int can_dropped_invalid_skb(struct net_device *dev, + struct sk_buff *skb) +{ + const struct can_frame *cf = (struct can_frame *)skb->data; + + if (unlikely(skb->len != sizeof(*cf) || cf->can_dlc > 8)) { + kfree_skb(skb); + dev->stats.tx_dropped++; + return 1; + } + + return 0; +} + struct net_device *alloc_candev(int sizeof_priv, unsigned int echo_skb_max); void free_candev(struct net_device *dev); -- cgit v1.2.3-70-g09d2 From 81077e82c3f591578625805dd6464a27a9ff56ec Mon Sep 17 00:00:00 2001 From: Lukáš Turek <8an@praha12.net> Date: Mon, 21 Dec 2009 22:50:47 +0100 Subject: nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville --- include/linux/nl80211.h | 4 ++++ include/net/cfg80211.h | 2 ++ net/wireless/core.c | 1 + net/wireless/nl80211.c | 15 +++++++++++++++ 4 files changed, 22 insertions(+) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 2bfbe88837e..d4c556de717 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -430,6 +430,8 @@ enum nl80211_commands { * @NL80211_ATTR_WIPHY_RTS_THRESHOLD: RTS threshold (TX frames with length * larger than or equal to this use RTS/CTS handshake); allowed range: * 0..65536, disable with (u32)-1; dot11RTSThreshold; u32 + * @NL80211_ATTR_WIPHY_COVERAGE_CLASS: Coverage Class as defined by IEEE 802.11 + * section 7.3.2.9; dot11CoverageClass; u8 * * @NL80211_ATTR_IFINDEX: network interface index of the device to operate on * @NL80211_ATTR_IFNAME: network interface name @@ -779,6 +781,8 @@ enum nl80211_attrs { NL80211_ATTR_COOKIE, + NL80211_ATTR_WIPHY_COVERAGE_CLASS, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index add79930f47..a8d5d04314b 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -837,6 +837,7 @@ enum wiphy_params_flags { WIPHY_PARAM_RETRY_LONG = 1 << 1, WIPHY_PARAM_FRAG_THRESHOLD = 1 << 2, WIPHY_PARAM_RTS_THRESHOLD = 1 << 3, + WIPHY_PARAM_COVERAGE_CLASS = 1 << 4, }; /** @@ -1236,6 +1237,7 @@ struct wiphy { u8 retry_long; u32 frag_threshold; u32 rts_threshold; + u8 coverage_class; char fw_version[ETHTOOL_BUSINFO_LEN]; u32 hw_version; diff --git a/net/wireless/core.c b/net/wireless/core.c index c2a2c563d21..0a545bb6ed0 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -402,6 +402,7 @@ struct wiphy *wiphy_new(const struct cfg80211_ops *ops, int sizeof_priv) rdev->wiphy.retry_long = 4; rdev->wiphy.frag_threshold = (u32) -1; rdev->wiphy.rts_threshold = (u32) -1; + rdev->wiphy.coverage_class = 0; return &rdev->wiphy; } diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index e3bee3cecdf..c09fbcd278f 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -69,6 +69,7 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { [NL80211_ATTR_WIPHY_RETRY_LONG] = { .type = NLA_U8 }, [NL80211_ATTR_WIPHY_FRAG_THRESHOLD] = { .type = NLA_U32 }, [NL80211_ATTR_WIPHY_RTS_THRESHOLD] = { .type = NLA_U32 }, + [NL80211_ATTR_WIPHY_COVERAGE_CLASS] = { .type = NLA_U8 }, [NL80211_ATTR_IFTYPE] = { .type = NLA_U32 }, [NL80211_ATTR_IFINDEX] = { .type = NLA_U32 }, @@ -444,6 +445,8 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, dev->wiphy.frag_threshold); NLA_PUT_U32(msg, NL80211_ATTR_WIPHY_RTS_THRESHOLD, dev->wiphy.rts_threshold); + NLA_PUT_U8(msg, NL80211_ATTR_WIPHY_COVERAGE_CLASS, + dev->wiphy.coverage_class); NLA_PUT_U8(msg, NL80211_ATTR_MAX_NUM_SCAN_SSIDS, dev->wiphy.max_scan_ssids); @@ -684,6 +687,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) u32 changed; u8 retry_short = 0, retry_long = 0; u32 frag_threshold = 0, rts_threshold = 0; + u8 coverage_class = 0; rtnl_lock(); @@ -806,9 +810,16 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) changed |= WIPHY_PARAM_RTS_THRESHOLD; } + if (info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]) { + coverage_class = nla_get_u8( + info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]); + changed |= WIPHY_PARAM_COVERAGE_CLASS; + } + if (changed) { u8 old_retry_short, old_retry_long; u32 old_frag_threshold, old_rts_threshold; + u8 old_coverage_class; if (!rdev->ops->set_wiphy_params) { result = -EOPNOTSUPP; @@ -819,6 +830,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) old_retry_long = rdev->wiphy.retry_long; old_frag_threshold = rdev->wiphy.frag_threshold; old_rts_threshold = rdev->wiphy.rts_threshold; + old_coverage_class = rdev->wiphy.coverage_class; if (changed & WIPHY_PARAM_RETRY_SHORT) rdev->wiphy.retry_short = retry_short; @@ -828,6 +840,8 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) rdev->wiphy.frag_threshold = frag_threshold; if (changed & WIPHY_PARAM_RTS_THRESHOLD) rdev->wiphy.rts_threshold = rts_threshold; + if (changed & WIPHY_PARAM_COVERAGE_CLASS) + rdev->wiphy.coverage_class = coverage_class; result = rdev->ops->set_wiphy_params(&rdev->wiphy, changed); if (result) { @@ -835,6 +849,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) rdev->wiphy.retry_long = old_retry_long; rdev->wiphy.frag_threshold = old_frag_threshold; rdev->wiphy.rts_threshold = old_rts_threshold; + rdev->wiphy.coverage_class = old_coverage_class; } } -- cgit v1.2.3-70-g09d2 From 310bc676e314e92c18257bfc916951879451ee32 Mon Sep 17 00:00:00 2001 From: Lukáš Turek <8an@praha12.net> Date: Mon, 21 Dec 2009 22:50:48 +0100 Subject: mac80211: Add new callback set_coverage_class Mac80211 callback to driver set_coverage_class() sets slot time and ACK timeout for given IEEE 802.11 coverage class. The callback is optional, but it's essential for long distance links. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville --- include/net/mac80211.h | 5 +++++ net/mac80211/cfg.c | 7 +++++++ net/mac80211/driver-ops.h | 15 +++++++++++++++ net/mac80211/driver-trace.h | 23 +++++++++++++++++++++++ 4 files changed, 50 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f073a2a5057..ad4b70034e7 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1533,6 +1533,10 @@ enum ieee80211_ampdu_mlme_action { * and need to call wiphy_rfkill_set_hw_state() in the callback. * The callback can sleep. * + * @set_coverage_class: Set slot time for given coverage class as specified + * in IEEE 802.11-2007 section 17.3.8.6 and modify ACK timeout + * accordingly. This callback is not required and may sleep. + * * @testmode_cmd: Implement a cfg80211 test mode command. * The callback can sleep. * @@ -1592,6 +1596,7 @@ struct ieee80211_ops { struct ieee80211_sta *sta, u16 tid, u16 *ssn); void (*rfkill_poll)(struct ieee80211_hw *hw); + void (*set_coverage_class)(struct ieee80211_hw *hw, u8 coverage_class); #ifdef CONFIG_NL80211_TESTMODE int (*testmode_cmd)(struct ieee80211_hw *hw, void *data, int len); #endif diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 2e5e841e9b7..976014c5e74 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1230,6 +1230,13 @@ static int ieee80211_set_wiphy_params(struct wiphy *wiphy, u32 changed) struct ieee80211_local *local = wiphy_priv(wiphy); int err; + if (changed & WIPHY_PARAM_COVERAGE_CLASS) { + err = drv_set_coverage_class(local, wiphy->coverage_class); + + if (err) + return err; + } + if (changed & WIPHY_PARAM_RTS_THRESHOLD) { err = drv_set_rts_threshold(local, wiphy->rts_threshold); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 8757ea73d54..de91d39e027 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -214,6 +214,21 @@ static inline int drv_set_rts_threshold(struct ieee80211_local *local, return ret; } +static inline int drv_set_coverage_class(struct ieee80211_local *local, + u8 value) +{ + int ret = 0; + might_sleep(); + + if (local->ops->set_coverage_class) + local->ops->set_coverage_class(&local->hw, value); + else + ret = -EOPNOTSUPP; + + trace_drv_set_coverage_class(local, value, ret); + return ret; +} + static inline void drv_sta_notify(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, enum sta_notify_cmd cmd, diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index 977cc7528bc..0ea258123b8 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -491,6 +491,29 @@ TRACE_EVENT(drv_set_rts_threshold, ) ); +TRACE_EVENT(drv_set_coverage_class, + TP_PROTO(struct ieee80211_local *local, u8 value, int ret), + + TP_ARGS(local, value, ret), + + TP_STRUCT__entry( + LOCAL_ENTRY + __field(u8, value) + __field(int, ret) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + __entry->ret = ret; + __entry->value = value; + ), + + TP_printk( + LOCAL_PR_FMT " value:%d ret:%d", + LOCAL_PR_ARG, __entry->value, __entry->ret + ) +); + TRACE_EVENT(drv_sta_notify, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, -- cgit v1.2.3-70-g09d2 From e00cfce0cb2a397859607bf515c6de9ce064b64a Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Tue, 29 Dec 2009 12:59:19 +0200 Subject: mac80211: Select lowest rate based on basic rate set in AP mode If the basic rate set is configured to not include the lowest rate (e.g., basic rate set = 6, 12, 24 Mbps in IEEE 802.11g mode), the AP should not send out broadcast frames at 1 Mbps. This type of configuration can be used to optimize channel usage in cases where there is no need for backwards compatibility with IEEE 802.11b-only devices. In AP mode, mac80211 was unconditionally using the lowest rate for Beacon frames and similarly, with all rate control algorithms that use rate_control_send_low(), the lowest rate ended up being used for all broadcast frames (and all unicast frames that are sent before association). Change this to take into account the basic rate configuration in AP mode, i.e., use the lowest rate in the basic rate set instead of the lowest supported rate when selecting the rate. Signed-off-by: Jouni Malinen Signed-off-by: John W. Linville --- include/net/mac80211.h | 2 ++ net/mac80211/rate.c | 25 +++++++++++++++++++++++++ net/mac80211/tx.c | 24 +++++++++++++----------- 3 files changed, 40 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index ad4b70034e7..39c516c352b 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2299,6 +2299,7 @@ enum rate_control_changed { * @max_rate_idx: user-requested maximum rate (not MCS for now) * @skb: the skb that will be transmitted, the control information in it needs * to be filled in + * @ap: whether this frame is sent out in AP mode */ struct ieee80211_tx_rate_control { struct ieee80211_hw *hw; @@ -2308,6 +2309,7 @@ struct ieee80211_tx_rate_control { struct ieee80211_tx_rate reported_rate; bool rts, short_preamble; u8 max_rate_idx; + bool ap; }; struct rate_control_ops { diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index b9007f80cb9..6349e7f4dca 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -207,6 +207,27 @@ static bool rc_no_data_or_no_ack(struct ieee80211_tx_rate_control *txrc) return ((info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc)); } +static void rc_send_low_broadcast(s8 *idx, u32 basic_rates, u8 max_rate_idx) +{ + u8 i; + + if (basic_rates == 0) + return; /* assume basic rates unknown and accept rate */ + if (*idx < 0) + return; + if (basic_rates & (1 << *idx)) + return; /* selected rate is a basic rate */ + + for (i = *idx + 1; i <= max_rate_idx; i++) { + if (basic_rates & (1 << i)) { + *idx = i; + return; + } + } + + /* could not find a basic rate; use original selection */ +} + bool rate_control_send_low(struct ieee80211_sta *sta, void *priv_sta, struct ieee80211_tx_rate_control *txrc) @@ -218,6 +239,10 @@ bool rate_control_send_low(struct ieee80211_sta *sta, info->control.rates[0].count = (info->flags & IEEE80211_TX_CTL_NO_ACK) ? 1 : txrc->hw->max_rate_tries; + if (!sta && txrc->ap) + rc_send_low_broadcast(&info->control.rates[0].idx, + txrc->bss_conf->basic_rates, + txrc->sband->n_bitrates); return true; } return false; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 140da4a7f13..4961168f509 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -520,6 +520,7 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx) txrc.skb = tx->skb; txrc.reported_rate.idx = -1; txrc.max_rate_idx = tx->sdata->max_ratectrl_rateidx; + txrc.ap = tx->sdata->vif.type == NL80211_IFTYPE_AP; /* set up RTS protection if desired */ if (len > tx->local->hw.wiphy->rts_threshold) { @@ -2060,6 +2061,7 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, struct beacon_data *beacon; struct ieee80211_supported_band *sband; enum ieee80211_band band = local->hw.conf.channel->band; + struct ieee80211_tx_rate_control txrc; sband = local->hw.wiphy->bands[band]; @@ -2167,21 +2169,21 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, info = IEEE80211_SKB_CB(skb); info->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT; + info->flags |= IEEE80211_TX_CTL_NO_ACK; info->band = band; - /* - * XXX: For now, always use the lowest rate - */ - info->control.rates[0].idx = 0; - info->control.rates[0].count = 1; - info->control.rates[1].idx = -1; - info->control.rates[2].idx = -1; - info->control.rates[3].idx = -1; - info->control.rates[4].idx = -1; - BUILD_BUG_ON(IEEE80211_TX_MAX_RATES != 5); + + memset(&txrc, 0, sizeof(txrc)); + txrc.hw = hw; + txrc.sband = sband; + txrc.bss_conf = &sdata->vif.bss_conf; + txrc.skb = skb; + txrc.reported_rate.idx = -1; + txrc.max_rate_idx = sdata->max_ratectrl_rateidx; + txrc.ap = true; + rate_control_get_rate(sdata, NULL, &txrc); info->control.vif = vif; - info->flags |= IEEE80211_TX_CTL_NO_ACK; info->flags |= IEEE80211_TX_CTL_CLEAR_PS_FILT; info->flags |= IEEE80211_TX_CTL_ASSIGN_SEQ; out: -- cgit v1.2.3-70-g09d2 From 37eb0b164cf9fa9f70c8500926f5cde7c652f48e Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Wed, 6 Jan 2010 13:09:08 +0200 Subject: cfg80211/mac80211: Use more generic bitrate mask for rate control Extend struct cfg80211_bitrate_mask to actually use a bitfield mask instead of just a single fixed or maximum rate index. This change itself does not modify the behavior (except for debugfs files), but it prepares cfg80211 and mac80211 for a new nl80211 command for setting which rates can be used in TX rate control. Since frames are now going through the rate control algorithm unconditionally, the internal IEEE80211_TX_INTFL_RCALGO flag can now be removed. The RC implementations can use the rate_idx_mask value to optimize their behavior if only a single rate is enabled. The old max_rate_idx in struct ieee80211_tx_rate_control is maintained (but commented as deprecated) for backwards compatibility with existing RC implementations. Once these implementations have been updated to use the more generic rate_idx_mask, the max_rate_idx value can be removed. Signed-off-by: Jouni Malinen Signed-off-by: John W. Linville --- include/net/cfg80211.h | 13 ++------- include/net/mac80211.h | 8 +++--- net/mac80211/cfg.c | 32 +++------------------- net/mac80211/debugfs_netdev.c | 22 ++++++++------- net/mac80211/ieee80211_i.h | 4 +-- net/mac80211/iface.c | 8 ++++-- net/mac80211/rate.c | 63 +++++++++++++++++++++++++++++++++++-------- net/mac80211/rate.h | 5 +--- net/mac80211/tx.c | 12 +++++++-- net/wireless/wext-compat.c | 34 ++++++++++++++++++++--- 10 files changed, 122 insertions(+), 79 deletions(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index a8d5d04314b..22e062afb5a 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -857,20 +857,11 @@ enum tx_power_setting { * cfg80211_bitrate_mask - masks for bitrate control */ struct cfg80211_bitrate_mask { -/* - * As discussed in Berlin, this struct really - * should look like this: - struct { u32 legacy; - u8 mcs[IEEE80211_HT_MCS_MASK_LEN]; + /* TODO: add support for masking MCS rates; e.g.: */ + /* u8 mcs[IEEE80211_HT_MCS_MASK_LEN]; */ } control[IEEE80211_NUM_BANDS]; - - * Since we can always fix in-kernel users, let's keep - * it simpler for now: - */ - u32 fixed; /* fixed bitrate, 0 == not fixed */ - u32 maxrate; /* in kbps, 0 == no limit */ }; /** * struct cfg80211_pmksa - PMK Security Association diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 39c516c352b..7e5af6d90b9 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -255,9 +255,6 @@ struct ieee80211_bss_conf { * @IEEE80211_TX_CTL_RATE_CTRL_PROBE: internal to mac80211, can be * set by rate control algorithms to indicate probe rate, will * be cleared for fragmented frames (except on the last fragment) - * @IEEE80211_TX_INTFL_RCALGO: mac80211 internal flag, do not test or - * set this flag in the driver; indicates that the rate control - * algorithm was used and should be notified of TX status * @IEEE80211_TX_INTFL_NEED_TXPROCESSING: completely internal to mac80211, * used to indicate that a pending frame requires TX processing before * it can be sent out. @@ -287,7 +284,6 @@ enum mac80211_tx_control_flags { IEEE80211_TX_STAT_AMPDU = BIT(10), IEEE80211_TX_STAT_AMPDU_NO_BACK = BIT(11), IEEE80211_TX_CTL_RATE_CTRL_PROBE = BIT(12), - IEEE80211_TX_INTFL_RCALGO = BIT(13), IEEE80211_TX_INTFL_NEED_TXPROCESSING = BIT(14), IEEE80211_TX_INTFL_RETRIED = BIT(15), IEEE80211_TX_INTFL_DONT_ENCRYPT = BIT(16), @@ -2297,6 +2293,9 @@ enum rate_control_changed { * @short_preamble: whether mac80211 will request short-preamble transmission * if the selected rate supports it * @max_rate_idx: user-requested maximum rate (not MCS for now) + * (deprecated; this will be removed once drivers get updated to use + * rate_idx_mask) + * @rate_idx_mask: user-requested rate mask (not MCS for now) * @skb: the skb that will be transmitted, the control information in it needs * to be filled in * @ap: whether this frame is sent out in AP mode @@ -2309,6 +2308,7 @@ struct ieee80211_tx_rate_control { struct ieee80211_tx_rate reported_rate; bool rts, short_preamble; u8 max_rate_idx; + u32 rate_idx_mask; bool ap; }; diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 976014c5e74..e5dda6fb8df 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1406,8 +1406,6 @@ static int ieee80211_set_bitrate_mask(struct wiphy *wiphy, struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); int i; - u32 target_rate; - struct ieee80211_supported_band *sband; /* * This _could_ be supported by providing a hook for @@ -1417,35 +1415,11 @@ static int ieee80211_set_bitrate_mask(struct wiphy *wiphy, if (local->hw.flags & IEEE80211_HW_HAS_RATE_CONTROL) return -EOPNOTSUPP; - sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; - - /* - * target_rate = -1, rate->fixed = 0 means auto only, so use all rates - * target_rate = X, rate->fixed = 1 means only rate X - * target_rate = X, rate->fixed = 0 means all rates <= X - */ - sdata->max_ratectrl_rateidx = -1; - sdata->force_unicast_rateidx = -1; - if (mask->fixed) - target_rate = mask->fixed / 100; - else if (mask->maxrate) - target_rate = mask->maxrate / 100; - else - return 0; + for (i = 0; i < IEEE80211_NUM_BANDS; i++) + sdata->rc_rateidx_mask[i] = mask->control[i].legacy; - for (i = 0; i< sband->n_bitrates; i++) { - if (target_rate != sband->bitrates[i].bitrate) - continue; - - /* requested bitrate found */ - sdata->max_ratectrl_rateidx = i; - if (mask->fixed) - sdata->force_unicast_rateidx = i; - return 0; - } - - return -EINVAL; + return 0; } static int ieee80211_remain_on_channel(struct wiphy *wiphy, diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 59f6e3bcbd0..1481049f0f7 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -127,8 +127,10 @@ __IEEE80211_IF_FILE(name, ieee80211_if_write_##name) /* common attributes */ IEEE80211_IF_FILE(drop_unencrypted, drop_unencrypted, DEC); -IEEE80211_IF_FILE(force_unicast_rateidx, force_unicast_rateidx, DEC); -IEEE80211_IF_FILE(max_ratectrl_rateidx, max_ratectrl_rateidx, DEC); +IEEE80211_IF_FILE(rc_rateidx_mask_2ghz, rc_rateidx_mask[IEEE80211_BAND_2GHZ], + HEX); +IEEE80211_IF_FILE(rc_rateidx_mask_5ghz, rc_rateidx_mask[IEEE80211_BAND_5GHZ], + HEX); /* STA attributes */ IEEE80211_IF_FILE(bssid, u.mgd.bssid, MAC); @@ -264,8 +266,8 @@ IEEE80211_IF_FILE(dot11MeshHWMPRootMode, static void add_sta_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, sta); - DEBUGFS_ADD(force_unicast_rateidx, sta); - DEBUGFS_ADD(max_ratectrl_rateidx, sta); + DEBUGFS_ADD(rc_rateidx_mask_2ghz, sta); + DEBUGFS_ADD(rc_rateidx_mask_5ghz, sta); DEBUGFS_ADD(bssid, sta); DEBUGFS_ADD(aid, sta); @@ -275,8 +277,8 @@ static void add_sta_files(struct ieee80211_sub_if_data *sdata) static void add_ap_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, ap); - DEBUGFS_ADD(force_unicast_rateidx, ap); - DEBUGFS_ADD(max_ratectrl_rateidx, ap); + DEBUGFS_ADD(rc_rateidx_mask_2ghz, ap); + DEBUGFS_ADD(rc_rateidx_mask_5ghz, ap); DEBUGFS_ADD(num_sta_ps, ap); DEBUGFS_ADD(dtim_count, ap); @@ -286,8 +288,8 @@ static void add_ap_files(struct ieee80211_sub_if_data *sdata) static void add_wds_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, wds); - DEBUGFS_ADD(force_unicast_rateidx, wds); - DEBUGFS_ADD(max_ratectrl_rateidx, wds); + DEBUGFS_ADD(rc_rateidx_mask_2ghz, wds); + DEBUGFS_ADD(rc_rateidx_mask_5ghz, wds); DEBUGFS_ADD(peer, wds); } @@ -295,8 +297,8 @@ static void add_wds_files(struct ieee80211_sub_if_data *sdata) static void add_vlan_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, vlan); - DEBUGFS_ADD(force_unicast_rateidx, vlan); - DEBUGFS_ADD(max_ratectrl_rateidx, vlan); + DEBUGFS_ADD(rc_rateidx_mask_2ghz, vlan); + DEBUGFS_ADD(rc_rateidx_mask_5ghz, vlan); } static void add_monitor_files(struct ieee80211_sub_if_data *sdata) diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index a27921ee6e6..3e4ac3f3085 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -494,8 +494,8 @@ struct ieee80211_sub_if_data { */ struct ieee80211_if_ap *bss; - int force_unicast_rateidx; /* forced TX rateidx for unicast frames */ - int max_ratectrl_rateidx; /* max TX rateidx for rate control */ + /* bitmap of allowed (non-MCS) rate indexes for rate control */ + u32 rc_rateidx_mask[IEEE80211_NUM_BANDS]; union { struct ieee80211_if_ap ap; diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 264a6c975f8..fe140bf033f 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -856,8 +856,12 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name, INIT_LIST_HEAD(&sdata->key_list); - sdata->force_unicast_rateidx = -1; - sdata->max_ratectrl_rateidx = -1; + for (i = 0; i < IEEE80211_NUM_BANDS; i++) { + struct ieee80211_supported_band *sband; + sband = local->hw.wiphy->bands[i]; + sdata->rc_rateidx_mask[i] = + sband ? (1 << sband->n_bitrates) - 1 : 0; + } /* setup type-dependent data */ ieee80211_setup_sdata(sdata, type); diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c index 6349e7f4dca..c74b7c85403 100644 --- a/net/mac80211/rate.c +++ b/net/mac80211/rate.c @@ -249,6 +249,38 @@ bool rate_control_send_low(struct ieee80211_sta *sta, } EXPORT_SYMBOL(rate_control_send_low); +static void rate_idx_match_mask(struct ieee80211_tx_rate *rate, + int n_bitrates, u32 mask) +{ + int j; + + /* See whether the selected rate or anything below it is allowed. */ + for (j = rate->idx; j >= 0; j--) { + if (mask & (1 << j)) { + /* Okay, found a suitable rate. Use it. */ + rate->idx = j; + return; + } + } + + /* Try to find a higher rate that would be allowed */ + for (j = rate->idx + 1; j < n_bitrates; j++) { + if (mask & (1 << j)) { + /* Okay, found a suitable rate. Use it. */ + rate->idx = j; + return; + } + } + + /* + * Uh.. No suitable rate exists. This should not really happen with + * sane TX rate mask configurations. However, should someone manage to + * configure supported rates and TX rate mask in incompatible way, + * allow the frame to be transmitted with whatever the rate control + * selected. + */ +} + void rate_control_get_rate(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, struct ieee80211_tx_rate_control *txrc) @@ -258,6 +290,7 @@ void rate_control_get_rate(struct ieee80211_sub_if_data *sdata, struct ieee80211_sta *ista = NULL; struct ieee80211_tx_info *info = IEEE80211_SKB_CB(txrc->skb); int i; + u32 mask; if (sta) { ista = &sta->sta; @@ -270,23 +303,31 @@ void rate_control_get_rate(struct ieee80211_sub_if_data *sdata, info->control.rates[i].count = 1; } - if (sta && sdata->force_unicast_rateidx > -1) { - info->control.rates[0].idx = sdata->force_unicast_rateidx; - } else { - ref->ops->get_rate(ref->priv, ista, priv_sta, txrc); - info->flags |= IEEE80211_TX_INTFL_RCALGO; - } + ref->ops->get_rate(ref->priv, ista, priv_sta, txrc); /* - * try to enforce the maximum rate the user wanted + * Try to enforce the rateidx mask the user wanted. skip this if the + * default mask (allow all rates) is used to save some processing for + * the common case. */ - if (sdata->max_ratectrl_rateidx > -1) + mask = sdata->rc_rateidx_mask[info->band]; + if (mask != (1 << txrc->sband->n_bitrates) - 1) { + if (sta) { + /* Filter out rates that the STA does not support */ + mask &= sta->sta.supp_rates[info->band]; + } + /* + * Make sure the rate index selected for each TX rate is + * included in the configured mask and change the rate indexes + * if needed. + */ for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) { + /* Rate masking supports only legacy rates for now */ if (info->control.rates[i].flags & IEEE80211_TX_RC_MCS) continue; - info->control.rates[i].idx = - min_t(s8, info->control.rates[i].idx, - sdata->max_ratectrl_rateidx); + rate_idx_match_mask(&info->control.rates[i], + txrc->sband->n_bitrates, mask); + } } BUG_ON(info->control.rates[0].idx < 0); diff --git a/net/mac80211/rate.h b/net/mac80211/rate.h index cb9bd1f65e2..669dddd4052 100644 --- a/net/mac80211/rate.h +++ b/net/mac80211/rate.h @@ -44,10 +44,7 @@ static inline void rate_control_tx_status(struct ieee80211_local *local, struct rate_control_ref *ref = local->rate_ctrl; struct ieee80211_sta *ista = &sta->sta; void *priv_sta = sta->rate_ctrl_priv; - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - - if (likely(info->flags & IEEE80211_TX_INTFL_RCALGO)) - ref->ops->tx_status(ref->priv, sband, ista, priv_sta, skb); + ref->ops->tx_status(ref->priv, sband, ista, priv_sta, skb); } diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 4961168f509..d3a44812f8b 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -519,7 +519,11 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx) txrc.bss_conf = &tx->sdata->vif.bss_conf; txrc.skb = tx->skb; txrc.reported_rate.idx = -1; - txrc.max_rate_idx = tx->sdata->max_ratectrl_rateidx; + txrc.rate_idx_mask = tx->sdata->rc_rateidx_mask[tx->channel->band]; + if (txrc.rate_idx_mask == (1 << sband->n_bitrates) - 1) + txrc.max_rate_idx = -1; + else + txrc.max_rate_idx = fls(txrc.rate_idx_mask) - 1; txrc.ap = tx->sdata->vif.type == NL80211_IFTYPE_AP; /* set up RTS protection if desired */ @@ -2178,7 +2182,11 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, txrc.bss_conf = &sdata->vif.bss_conf; txrc.skb = skb; txrc.reported_rate.idx = -1; - txrc.max_rate_idx = sdata->max_ratectrl_rateidx; + txrc.rate_idx_mask = sdata->rc_rateidx_mask[band]; + if (txrc.rate_idx_mask == (1 << sband->n_bitrates) - 1) + txrc.max_rate_idx = -1; + else + txrc.max_rate_idx = fls(txrc.rate_idx_mask) - 1; txrc.ap = true; rate_control_get_rate(sdata, NULL, &txrc); diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index 4198243a3df..966d2f01bea 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -1204,21 +1204,47 @@ int cfg80211_wext_siwrate(struct net_device *dev, struct wireless_dev *wdev = dev->ieee80211_ptr; struct cfg80211_registered_device *rdev = wiphy_to_dev(wdev->wiphy); struct cfg80211_bitrate_mask mask; + u32 fixed, maxrate; + struct ieee80211_supported_band *sband; + int band, ridx; + bool match = false; if (!rdev->ops->set_bitrate_mask) return -EOPNOTSUPP; - mask.fixed = 0; - mask.maxrate = 0; + memset(&mask, 0, sizeof(mask)); + fixed = 0; + maxrate = 0; if (rate->value < 0) { /* nothing */ } else if (rate->fixed) { - mask.fixed = rate->value / 1000; /* kbps */ + fixed = rate->value / 100000; } else { - mask.maxrate = rate->value / 1000; /* kbps */ + maxrate = rate->value / 100000; + } + + for (band = 0; band < IEEE80211_NUM_BANDS; band++) { + sband = wdev->wiphy->bands[band]; + if (sband == NULL) + continue; + for (ridx = 0; ridx < sband->n_bitrates; ridx++) { + struct ieee80211_rate *srate = &sband->bitrates[ridx]; + if (fixed == srate->bitrate) { + mask.control[band].legacy = 1 << ridx; + match = true; + break; + } + if (srate->bitrate <= maxrate) { + mask.control[band].legacy |= 1 << ridx; + match = true; + } + } } + if (!match) + return -EINVAL; + return rdev->ops->set_bitrate_mask(wdev->wiphy, dev, NULL, &mask); } EXPORT_SYMBOL_GPL(cfg80211_wext_siwrate); -- cgit v1.2.3-70-g09d2 From 13ae75b103e07304a34ab40c9136e9f53e06475c Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Tue, 29 Dec 2009 12:59:45 +0200 Subject: nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 44 +++++++++++++++++++ include/net/cfg80211.h | 4 +- net/wireless/nl80211.c | 111 ++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 157 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index d4c556de717..7a1c8c145b2 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -295,6 +295,10 @@ * This command is also used as an event to notify when a requested * remain-on-channel duration has expired. * + * @NL80211_CMD_SET_TX_BITRATE_MASK: Set the mask of rates to be used in TX + * rate selection. %NL80211_ATTR_IFINDEX is used to specify the interface + * and @NL80211_ATTR_TX_RATES the set of allowed rates. + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -381,6 +385,8 @@ enum nl80211_commands { NL80211_CMD_REMAIN_ON_CHANNEL, NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, + NL80211_CMD_SET_TX_BITRATE_MASK, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -640,6 +646,13 @@ enum nl80211_commands { * * @NL80211_ATTR_COOKIE: Generic 64-bit cookie to identify objects. * + * @NL80211_ATTR_TX_RATES: Nested set of attributes + * (enum nl80211_tx_rate_attributes) describing TX rates per band. The + * enum nl80211_band value is used as the index (nla_type() of the nested + * data. If a band is not included, it will be configured to allow all + * rates based on negotiated supported rates information. This attribute + * is used with %NL80211_CMD_SET_TX_BITRATE_MASK. + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -783,6 +796,8 @@ enum nl80211_attrs { NL80211_ATTR_WIPHY_COVERAGE_CLASS, + NL80211_ATTR_TX_RATES, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -1482,4 +1497,33 @@ enum nl80211_key_attributes { NL80211_KEY_MAX = __NL80211_KEY_AFTER_LAST - 1 }; +/** + * enum nl80211_tx_rate_attributes - TX rate set attributes + * @__NL80211_TXRATE_INVALID: invalid + * @NL80211_TXRATE_LEGACY: Legacy (non-MCS) rates allowed for TX rate selection + * in an array of rates as defined in IEEE 802.11 7.3.2.2 (u8 values with + * 1 = 500 kbps) but without the IE length restriction (at most + * %NL80211_MAX_SUPP_RATES in a single array). + * @__NL80211_TXRATE_AFTER_LAST: internal + * @NL80211_TXRATE_MAX: highest TX rate attribute + */ +enum nl80211_tx_rate_attributes { + __NL80211_TXRATE_INVALID, + NL80211_TXRATE_LEGACY, + + /* keep last */ + __NL80211_TXRATE_AFTER_LAST, + NL80211_TXRATE_MAX = __NL80211_TXRATE_AFTER_LAST - 1 +}; + +/** + * enum nl80211_band - Frequency band + * @NL80211_BAND_2GHZ - 2.4 GHz ISM band + * @NL80211_BAND_5GHZ - around 5 GHz band (4.9 - 5.7 GHz) + */ +enum nl80211_band { + NL80211_BAND_2GHZ, + NL80211_BAND_5GHZ, +}; + #endif /* __LINUX_NL80211_H */ diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 22e062afb5a..0d734413b5f 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -39,8 +39,8 @@ * @IEEE80211_BAND_5GHZ: around 5GHz band (4.9-5.7) */ enum ieee80211_band { - IEEE80211_BAND_2GHZ, - IEEE80211_BAND_5GHZ, + IEEE80211_BAND_2GHZ = NL80211_BAND_2GHZ, + IEEE80211_BAND_5GHZ = NL80211_BAND_5GHZ, /* keep last */ IEEE80211_NUM_BANDS diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c09fbcd278f..b804062e017 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -144,6 +144,7 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { .len = WLAN_PMKID_LEN }, [NL80211_ATTR_DURATION] = { .type = NLA_U32 }, [NL80211_ATTR_COOKIE] = { .type = NLA_U64 }, + [NL80211_ATTR_TX_RATES] = { .type = NLA_NESTED }, }; /* policy for the attributes */ @@ -575,6 +576,7 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, CMD(del_pmksa, DEL_PMKSA); CMD(flush_pmksa, FLUSH_PMKSA); CMD(remain_on_channel, REMAIN_ON_CHANNEL); + CMD(set_bitrate_mask, SET_TX_BITRATE_MASK); if (dev->wiphy.flags & WIPHY_FLAG_NETNS_OK) { i++; NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS); @@ -4438,6 +4440,109 @@ static int nl80211_cancel_remain_on_channel(struct sk_buff *skb, return err; } +static u32 rateset_to_mask(struct ieee80211_supported_band *sband, + u8 *rates, u8 rates_len) +{ + u8 i; + u32 mask = 0; + + for (i = 0; i < rates_len; i++) { + int rate = (rates[i] & 0x7f) * 5; + int ridx; + for (ridx = 0; ridx < sband->n_bitrates; ridx++) { + struct ieee80211_rate *srate = + &sband->bitrates[ridx]; + if (rate == srate->bitrate) { + mask |= 1 << ridx; + break; + } + } + if (ridx == sband->n_bitrates) + return 0; /* rate not found */ + } + + return mask; +} + +static struct nla_policy +nl80211_txattr_policy[NL80211_TXRATE_MAX + 1] __read_mostly = { + [NL80211_TXRATE_LEGACY] = { .type = NLA_BINARY, + .len = NL80211_MAX_SUPP_RATES }, +}; + +static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb, + struct genl_info *info) +{ + struct nlattr *tb[NL80211_TXRATE_MAX + 1]; + struct cfg80211_registered_device *rdev; + struct cfg80211_bitrate_mask mask; + int err, rem, i; + struct net_device *dev; + struct nlattr *tx_rates; + struct ieee80211_supported_band *sband; + + if (info->attrs[NL80211_ATTR_TX_RATES] == NULL) + return -EINVAL; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + if (!rdev->ops->set_bitrate_mask) { + err = -EOPNOTSUPP; + goto unlock; + } + + memset(&mask, 0, sizeof(mask)); + /* Default to all rates enabled */ + for (i = 0; i < IEEE80211_NUM_BANDS; i++) { + sband = rdev->wiphy.bands[i]; + mask.control[i].legacy = + sband ? (1 << sband->n_bitrates) - 1 : 0; + } + + /* + * The nested attribute uses enum nl80211_band as the index. This maps + * directly to the enum ieee80211_band values used in cfg80211. + */ + nla_for_each_nested(tx_rates, info->attrs[NL80211_ATTR_TX_RATES], rem) + { + enum ieee80211_band band = nla_type(tx_rates); + if (band < 0 || band >= IEEE80211_NUM_BANDS) { + err = -EINVAL; + goto unlock; + } + sband = rdev->wiphy.bands[band]; + if (sband == NULL) { + err = -EINVAL; + goto unlock; + } + nla_parse(tb, NL80211_TXRATE_MAX, nla_data(tx_rates), + nla_len(tx_rates), nl80211_txattr_policy); + if (tb[NL80211_TXRATE_LEGACY]) { + mask.control[band].legacy = rateset_to_mask( + sband, + nla_data(tb[NL80211_TXRATE_LEGACY]), + nla_len(tb[NL80211_TXRATE_LEGACY])); + if (mask.control[band].legacy == 0) { + err = -EINVAL; + goto unlock; + } + } + } + + err = rdev->ops->set_bitrate_mask(&rdev->wiphy, dev, NULL, &mask); + + unlock: + dev_put(dev); + cfg80211_unlock_rdev(rdev); + unlock_rtnl: + rtnl_unlock(); + return err; +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -4712,6 +4817,12 @@ static struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, + { + .cmd = NL80211_CMD_SET_TX_BITRATE_MASK, + .doit = nl80211_set_tx_bitrate_mask, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, }; static struct genl_multicast_group nl80211_mlme_mcgrp = { -- cgit v1.2.3-70-g09d2 From 7044cc565b45a898c140fb185174a66f2d68a163 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Tue, 5 Jan 2010 20:16:19 +0200 Subject: mac80211: add functions to create PS Poll and Nullfunc templates Some hardware, for example wl1251 and wl1271, handle the transmission of power save related frames in hardware, but the driver is responsible for creating the templates. It's better to create the templates in mac80211, that way all drivers can benefit from this. Add two new functions, ieee80211_pspoll_get() and ieee80211_nullfunc_get() which drivers need to call to get the frame. Drivers are also responsible for updating the templates after each association. Also new struct ieee80211_hdr_3addr is added to ieee80211.h to make it easy to calculate length of the Nullfunc frame. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 9 ++++++ include/net/mac80211.h | 30 ++++++++++++++++++ net/mac80211/tx.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 117 insertions(+) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index aeea282bd2f..602c0692c3f 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -130,6 +130,15 @@ struct ieee80211_hdr { u8 addr4[6]; } __attribute__ ((packed)); +struct ieee80211_hdr_3addr { + __le16 frame_control; + __le16 duration_id; + u8 addr1[6]; + u8 addr2[6]; + u8 addr3[6]; + __le16 seq_ctrl; +} __attribute__ ((packed)); + /** * ieee80211_has_tods - check if IEEE80211_FCTL_TODS is set * @fc: frame control bytes in little-endian byteorder diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 7e5af6d90b9..75f46e26ad6 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1874,6 +1874,36 @@ static inline struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, return ieee80211_beacon_get_tim(hw, vif, NULL, NULL); } +/** + * ieee80211_pspoll_get - retrieve a PS Poll template + * @hw: pointer obtained from ieee80211_alloc_hw(). + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * + * Creates a PS Poll a template which can, for example, uploaded to + * hardware. The template must be updated after association so that correct + * AID, BSSID and MAC address is used. + * + * Note: Caller (or hardware) is responsible for setting the + * &IEEE80211_FCTL_PM bit. + */ +struct sk_buff *ieee80211_pspoll_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif); + +/** + * ieee80211_nullfunc_get - retrieve a nullfunc template + * @hw: pointer obtained from ieee80211_alloc_hw(). + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * + * Creates a Nullfunc template which can, for example, uploaded to + * hardware. The template must be updated after association so that correct + * BSSID and address is used. + * + * Note: Caller (or hardware) is responsible for setting the + * &IEEE80211_FCTL_PM bit as well as Duration and Sequence Control fields. + */ +struct sk_buff *ieee80211_nullfunc_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif); + /** * ieee80211_rts_get - RTS frame generation function * @hw: pointer obtained from ieee80211_alloc_hw(). diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index d3a44812f8b..055b45b146d 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2200,6 +2200,84 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, } EXPORT_SYMBOL(ieee80211_beacon_get_tim); +struct sk_buff *ieee80211_pspoll_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif) +{ + struct ieee80211_sub_if_data *sdata; + struct ieee80211_if_managed *ifmgd; + struct ieee80211_pspoll *pspoll; + struct ieee80211_local *local; + struct sk_buff *skb; + + if (WARN_ON(vif->type != NL80211_IFTYPE_STATION)) + return NULL; + + sdata = vif_to_sdata(vif); + ifmgd = &sdata->u.mgd; + local = sdata->local; + + skb = dev_alloc_skb(local->hw.extra_tx_headroom + sizeof(*pspoll)); + if (!skb) { + printk(KERN_DEBUG "%s: failed to allocate buffer for " + "pspoll template\n", sdata->name); + return NULL; + } + skb_reserve(skb, local->hw.extra_tx_headroom); + + pspoll = (struct ieee80211_pspoll *) skb_put(skb, sizeof(*pspoll)); + memset(pspoll, 0, sizeof(*pspoll)); + pspoll->frame_control = cpu_to_le16(IEEE80211_FTYPE_CTL | + IEEE80211_STYPE_PSPOLL); + pspoll->aid = cpu_to_le16(ifmgd->aid); + + /* aid in PS-Poll has its two MSBs each set to 1 */ + pspoll->aid |= cpu_to_le16(1 << 15 | 1 << 14); + + memcpy(pspoll->bssid, ifmgd->bssid, ETH_ALEN); + memcpy(pspoll->ta, vif->addr, ETH_ALEN); + + return skb; +} +EXPORT_SYMBOL(ieee80211_pspoll_get); + +struct sk_buff *ieee80211_nullfunc_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif) +{ + struct ieee80211_hdr_3addr *nullfunc; + struct ieee80211_sub_if_data *sdata; + struct ieee80211_if_managed *ifmgd; + struct ieee80211_local *local; + struct sk_buff *skb; + + if (WARN_ON(vif->type != NL80211_IFTYPE_STATION)) + return NULL; + + sdata = vif_to_sdata(vif); + ifmgd = &sdata->u.mgd; + local = sdata->local; + + skb = dev_alloc_skb(local->hw.extra_tx_headroom + sizeof(*nullfunc)); + if (!skb) { + printk(KERN_DEBUG "%s: failed to allocate buffer for nullfunc " + "template\n", sdata->name); + return NULL; + } + skb_reserve(skb, local->hw.extra_tx_headroom); + + nullfunc = (struct ieee80211_hdr_3addr *) skb_put(skb, + sizeof(*nullfunc)); + memset(nullfunc, 0, sizeof(*nullfunc)); + nullfunc->frame_control = cpu_to_le16(IEEE80211_FTYPE_DATA | + IEEE80211_STYPE_NULLFUNC | + IEEE80211_FCTL_TODS); + memcpy(nullfunc->addr1, ifmgd->bssid, ETH_ALEN); + memcpy(nullfunc->addr2, vif->addr, ETH_ALEN); + memcpy(nullfunc->addr3, ifmgd->bssid, ETH_ALEN); + + return skb; +} +EXPORT_SYMBOL(ieee80211_nullfunc_get); + void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif, const void *frame, size_t frame_len, const struct ieee80211_tx_info *frame_txctl, -- cgit v1.2.3-70-g09d2 From 05e54ea6cce400ac34528d705179b45244f61074 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Tue, 5 Jan 2010 20:16:38 +0200 Subject: mac80211: create Probe Request template Certain type of hardware, for example wl1251 and wl1271, need a template for the Probe Request. Create a function ieee80211_probereq_get() which creates the template and drivers send it to hardware. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/net/mac80211.h | 17 +++++++++++++++++ net/mac80211/tx.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 67 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 75f46e26ad6..e1e73c6abef 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1904,6 +1904,23 @@ struct sk_buff *ieee80211_pspoll_get(struct ieee80211_hw *hw, struct sk_buff *ieee80211_nullfunc_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif); +/** + * ieee80211_probereq_get - retrieve a Probe Request template + * @hw: pointer obtained from ieee80211_alloc_hw(). + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * @ssid: SSID buffer + * @ssid_len: length of SSID + * @ie: buffer containing all IEs except SSID for the template + * @ie_len: length of the IE buffer + * + * Creates a Probe Request template which can, for example, be uploaded to + * hardware. + */ +struct sk_buff *ieee80211_probereq_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + const u8 *ssid, size_t ssid_len, + const u8 *ie, size_t ie_len); + /** * ieee80211_rts_get - RTS frame generation function * @hw: pointer obtained from ieee80211_alloc_hw(). diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 055b45b146d..0661e696a1d 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2278,6 +2278,56 @@ struct sk_buff *ieee80211_nullfunc_get(struct ieee80211_hw *hw, } EXPORT_SYMBOL(ieee80211_nullfunc_get); +struct sk_buff *ieee80211_probereq_get(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + const u8 *ssid, size_t ssid_len, + const u8 *ie, size_t ie_len) +{ + struct ieee80211_sub_if_data *sdata; + struct ieee80211_local *local; + struct ieee80211_hdr_3addr *hdr; + struct sk_buff *skb; + size_t ie_ssid_len; + u8 *pos; + + sdata = vif_to_sdata(vif); + local = sdata->local; + ie_ssid_len = 2 + ssid_len; + + skb = dev_alloc_skb(local->hw.extra_tx_headroom + sizeof(*hdr) + + ie_ssid_len + ie_len); + if (!skb) { + printk(KERN_DEBUG "%s: failed to allocate buffer for probe " + "request template\n", sdata->name); + return NULL; + } + + skb_reserve(skb, local->hw.extra_tx_headroom); + + hdr = (struct ieee80211_hdr_3addr *) skb_put(skb, sizeof(*hdr)); + memset(hdr, 0, sizeof(*hdr)); + hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_PROBE_REQ); + memset(hdr->addr1, 0xff, ETH_ALEN); + memcpy(hdr->addr2, vif->addr, ETH_ALEN); + memset(hdr->addr3, 0xff, ETH_ALEN); + + pos = skb_put(skb, ie_ssid_len); + *pos++ = WLAN_EID_SSID; + *pos++ = ssid_len; + if (ssid) + memcpy(pos, ssid, ssid_len); + pos += ssid_len; + + if (ie) { + pos = skb_put(skb, ie_len); + memcpy(pos, ie, ie_len); + } + + return skb; +} +EXPORT_SYMBOL(ieee80211_probereq_get); + void ieee80211_rts_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif, const void *frame, size_t frame_len, const struct ieee80211_tx_info *frame_txctl, -- cgit v1.2.3-70-g09d2 From 34a6eddbabd704b3c7dae9362234552267573be2 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Wed, 6 Jan 2010 16:19:24 +0200 Subject: cfg80211: Store IEs from both Beacon and Probe Response frames Store information elements from Beacon and Probe Response frames in separate buffers to allow both sets to be made available through nl80211. This allows user space applications to get access to IEs from Beacon frames even if we have received Probe Response frames from the BSS. Previously, the IEs from Probe Response frames would have overridden the IEs from Beacon frames. This feature is of somewhat limited use since most protocols include the same (or extended) information in Probe Response frames. However, there are couple of exceptions where the IEs from Beacon frames could be of some use: TIM IE is only included in Beacon frames (and it would be needed to figure out the DTIM period used in the BSS) and at least some implementations of Wireless Provisioning Services seem to include the full IE only in Beacon frames). The new BSS attribute for scan results is added to allow both the IE sets to be delivered. This is done in a way that maintains the previously used behavior for applications that are not aware of the new NL80211_BSS_BEACON_IES attribute. Signed-off-by: Jouni Malinen Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 10 +++- include/net/cfg80211.h | 12 ++++- net/wireless/core.h | 3 +- net/wireless/nl80211.c | 4 ++ net/wireless/scan.c | 120 ++++++++++++++++++++++++++++++++++++------------ 5 files changed, 116 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 7a1c8c145b2..127a7301576 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -1378,13 +1378,20 @@ enum nl80211_channel_type { * @NL80211_BSS_BEACON_INTERVAL: beacon interval of the (I)BSS (u16) * @NL80211_BSS_CAPABILITY: capability field (CPU order, u16) * @NL80211_BSS_INFORMATION_ELEMENTS: binary attribute containing the - * raw information elements from the probe response/beacon (bin) + * raw information elements from the probe response/beacon (bin); + * if the %NL80211_BSS_BEACON_IES attribute is present, the IEs here are + * from a Probe Response frame; otherwise they are from a Beacon frame. + * However, if the driver does not indicate the source of the IEs, these + * IEs may be from either frame subtype. * @NL80211_BSS_SIGNAL_MBM: signal strength of probe response/beacon * in mBm (100 * dBm) (s32) * @NL80211_BSS_SIGNAL_UNSPEC: signal strength of the probe response/beacon * in unspecified units, scaled to 0..100 (u8) * @NL80211_BSS_STATUS: status, if this BSS is "used" * @NL80211_BSS_SEEN_MS_AGO: age of this BSS entry in ms + * @NL80211_BSS_BEACON_IES: binary attribute containing the raw information + * elements from a Beacon frame (bin); not present if no Beacon frame has + * yet been received * @__NL80211_BSS_AFTER_LAST: internal * @NL80211_BSS_MAX: highest BSS attribute */ @@ -1400,6 +1407,7 @@ enum nl80211_bss { NL80211_BSS_SIGNAL_UNSPEC, NL80211_BSS_STATUS, NL80211_BSS_SEEN_MS_AGO, + NL80211_BSS_BEACON_IES, /* keep last */ __NL80211_BSS_AFTER_LAST, diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 0d734413b5f..2af52704e67 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -626,8 +626,14 @@ enum cfg80211_signal_type { * @beacon_interval: the beacon interval as from the frame * @capability: the capability field in host byte order * @information_elements: the information elements (Note that there - * is no guarantee that these are well-formed!) + * is no guarantee that these are well-formed!); this is a pointer to + * either the beacon_ies or proberesp_ies depending on whether Probe + * Response frame has been received * @len_information_elements: total length of the information elements + * @beacon_ies: the information elements from the last Beacon frame + * @len_beacon_ies: total length of the beacon_ies + * @proberesp_ies: the information elements from the last Probe Response frame + * @len_proberesp_ies: total length of the proberesp_ies * @signal: signal strength value (type depends on the wiphy's signal_type) * @free_priv: function pointer to free private data * @priv: private area for driver use, has at least wiphy->bss_priv_size bytes @@ -641,6 +647,10 @@ struct cfg80211_bss { u16 capability; u8 *information_elements; size_t len_information_elements; + u8 *beacon_ies; + size_t len_beacon_ies; + u8 *proberesp_ies; + size_t len_proberesp_ies; s32 signal; diff --git a/net/wireless/core.h b/net/wireless/core.h index 30ec95f05b5..2d6a6b9c0c4 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -111,7 +111,8 @@ struct cfg80211_internal_bss { unsigned long ts; struct kref ref; atomic_t hold; - bool ies_allocated; + bool beacon_ies_allocated; + bool proberesp_ies_allocated; /* must be last because of priv member */ struct cfg80211_bss pub; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index b804062e017..4af7991a9ec 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3163,6 +3163,10 @@ static int nl80211_send_bss(struct sk_buff *msg, u32 pid, u32 seq, int flags, NLA_PUT(msg, NL80211_BSS_INFORMATION_ELEMENTS, res->len_information_elements, res->information_elements); + if (res->beacon_ies && res->len_beacon_ies && + res->beacon_ies != res->information_elements) + NLA_PUT(msg, NL80211_BSS_BEACON_IES, + res->len_beacon_ies, res->beacon_ies); if (res->tsf) NLA_PUT_U64(msg, NL80211_BSS_TSF, res->tsf); if (res->beacon_interval) diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 0c2cbbebca9..06b0231ee5e 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -100,8 +100,10 @@ static void bss_release(struct kref *ref) if (bss->pub.free_priv) bss->pub.free_priv(&bss->pub); - if (bss->ies_allocated) - kfree(bss->pub.information_elements); + if (bss->beacon_ies_allocated) + kfree(bss->pub.beacon_ies); + if (bss->proberesp_ies_allocated) + kfree(bss->pub.proberesp_ies); BUG_ON(atomic_read(&bss->hold)); @@ -375,8 +377,7 @@ rb_find_bss(struct cfg80211_registered_device *dev, static struct cfg80211_internal_bss * cfg80211_bss_update(struct cfg80211_registered_device *dev, - struct cfg80211_internal_bss *res, - bool overwrite) + struct cfg80211_internal_bss *res) { struct cfg80211_internal_bss *found = NULL; const u8 *meshid, *meshcfg; @@ -418,28 +419,64 @@ cfg80211_bss_update(struct cfg80211_registered_device *dev, found->pub.capability = res->pub.capability; found->ts = res->ts; - /* overwrite IEs */ - if (overwrite) { + /* Update IEs */ + if (res->pub.proberesp_ies) { size_t used = dev->wiphy.bss_priv_size + sizeof(*res); - size_t ielen = res->pub.len_information_elements; + size_t ielen = res->pub.len_proberesp_ies; + + if (found->pub.proberesp_ies && + !found->proberesp_ies_allocated && + ksize(found) >= used + ielen) { + memcpy(found->pub.proberesp_ies, + res->pub.proberesp_ies, ielen); + found->pub.len_proberesp_ies = ielen; + } else { + u8 *ies = found->pub.proberesp_ies; + + if (found->proberesp_ies_allocated) + ies = krealloc(ies, ielen, GFP_ATOMIC); + else + ies = kmalloc(ielen, GFP_ATOMIC); + + if (ies) { + memcpy(ies, res->pub.proberesp_ies, + ielen); + found->proberesp_ies_allocated = true; + found->pub.proberesp_ies = ies; + found->pub.len_proberesp_ies = ielen; + } + } - if (!found->ies_allocated && ksize(found) >= used + ielen) { - memcpy(found->pub.information_elements, - res->pub.information_elements, ielen); - found->pub.len_information_elements = ielen; + /* Override possible earlier Beacon frame IEs */ + found->pub.information_elements = + found->pub.proberesp_ies; + found->pub.len_information_elements = + found->pub.len_proberesp_ies; + } + if (res->pub.beacon_ies) { + size_t used = dev->wiphy.bss_priv_size + sizeof(*res); + size_t ielen = res->pub.len_beacon_ies; + + if (found->pub.beacon_ies && + !found->beacon_ies_allocated && + ksize(found) >= used + ielen) { + memcpy(found->pub.beacon_ies, + res->pub.beacon_ies, ielen); + found->pub.len_beacon_ies = ielen; } else { - u8 *ies = found->pub.information_elements; + u8 *ies = found->pub.beacon_ies; - if (found->ies_allocated) + if (found->beacon_ies_allocated) ies = krealloc(ies, ielen, GFP_ATOMIC); else ies = kmalloc(ielen, GFP_ATOMIC); if (ies) { - memcpy(ies, res->pub.information_elements, ielen); - found->ies_allocated = true; - found->pub.information_elements = ies; - found->pub.len_information_elements = ielen; + memcpy(ies, res->pub.beacon_ies, + ielen); + found->beacon_ies_allocated = true; + found->pub.beacon_ies = ies; + found->pub.len_beacon_ies = ielen; } } } @@ -489,14 +526,26 @@ cfg80211_inform_bss(struct wiphy *wiphy, res->pub.tsf = timestamp; res->pub.beacon_interval = beacon_interval; res->pub.capability = capability; - /* point to after the private area */ - res->pub.information_elements = (u8 *)res + sizeof(*res) + privsz; - memcpy(res->pub.information_elements, ie, ielen); - res->pub.len_information_elements = ielen; + /* + * Since we do not know here whether the IEs are from a Beacon or Probe + * Response frame, we need to pick one of the options and only use it + * with the driver that does not provide the full Beacon/Probe Response + * frame. Use Beacon frame pointer to avoid indicating that this should + * override the information_elements pointer should we have received an + * earlier indication of Probe Response data. + * + * The initial buffer for the IEs is allocated with the BSS entry and + * is located after the private area. + */ + res->pub.beacon_ies = (u8 *)res + sizeof(*res) + privsz; + memcpy(res->pub.beacon_ies, ie, ielen); + res->pub.len_beacon_ies = ielen; + res->pub.information_elements = res->pub.beacon_ies; + res->pub.len_information_elements = res->pub.len_beacon_ies; kref_init(&res->ref); - res = cfg80211_bss_update(wiphy_to_dev(wiphy), res, 0); + res = cfg80211_bss_update(wiphy_to_dev(wiphy), res); if (!res) return NULL; @@ -517,7 +566,6 @@ cfg80211_inform_bss_frame(struct wiphy *wiphy, struct cfg80211_internal_bss *res; size_t ielen = len - offsetof(struct ieee80211_mgmt, u.probe_resp.variable); - bool overwrite; size_t privsz = wiphy->bss_priv_size; if (WARN_ON(wiphy->signal_type == NL80211_BSS_SIGNAL_UNSPEC && @@ -538,16 +586,28 @@ cfg80211_inform_bss_frame(struct wiphy *wiphy, res->pub.tsf = le64_to_cpu(mgmt->u.probe_resp.timestamp); res->pub.beacon_interval = le16_to_cpu(mgmt->u.probe_resp.beacon_int); res->pub.capability = le16_to_cpu(mgmt->u.probe_resp.capab_info); - /* point to after the private area */ - res->pub.information_elements = (u8 *)res + sizeof(*res) + privsz; - memcpy(res->pub.information_elements, mgmt->u.probe_resp.variable, ielen); - res->pub.len_information_elements = ielen; + /* + * The initial buffer for the IEs is allocated with the BSS entry and + * is located after the private area. + */ + if (ieee80211_is_probe_resp(mgmt->frame_control)) { + res->pub.proberesp_ies = (u8 *) res + sizeof(*res) + privsz; + memcpy(res->pub.proberesp_ies, mgmt->u.probe_resp.variable, + ielen); + res->pub.len_proberesp_ies = ielen; + res->pub.information_elements = res->pub.proberesp_ies; + res->pub.len_information_elements = res->pub.len_proberesp_ies; + } else { + res->pub.beacon_ies = (u8 *) res + sizeof(*res) + privsz; + memcpy(res->pub.beacon_ies, mgmt->u.beacon.variable, ielen); + res->pub.len_beacon_ies = ielen; + res->pub.information_elements = res->pub.beacon_ies; + res->pub.len_information_elements = res->pub.len_beacon_ies; + } kref_init(&res->ref); - overwrite = ieee80211_is_probe_resp(mgmt->frame_control); - - res = cfg80211_bss_update(wiphy_to_dev(wiphy), res, overwrite); + res = cfg80211_bss_update(wiphy_to_dev(wiphy), res); if (!res) return NULL; -- cgit v1.2.3-70-g09d2 From ab13315af97919fae0e014748105fdc2e30afb2d Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Tue, 12 Jan 2010 10:42:31 +0200 Subject: mac80211: add U-APSD client support Add Unscheduled Automatic Power-Save Delivery (U-APSD) client support. The idea is that the data frames from the client trigger AP to send the buffered frames with ACs which have U-APSD enabled. This decreases latency and makes it possible to save even more power. Driver needs to use IEEE80211_HW_UAPSD to enable the feature. The current implementation assumes that firmware takes care of the wakeup and hardware needing IEEE80211_HW_PS_NULLFUNC_STACK is not yet supported. Tested with wl1251 on a Nokia N900 and Cisco Aironet 1231G AP and running various test traffic with ping. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 18 ++++++++++++++++++ include/net/mac80211.h | 7 +++++++ net/mac80211/cfg.c | 7 +++++++ net/mac80211/ieee80211_i.h | 13 ++++++++++++- net/mac80211/main.c | 4 ++++ net/mac80211/mlme.c | 31 ++++++++++++++++++++++++++++--- net/mac80211/scan.c | 18 ++++++++++++++++++ net/mac80211/util.c | 2 ++ net/mac80211/work.c | 12 ++++++++++-- 9 files changed, 106 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 602c0692c3f..a8c6069a0d9 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -120,6 +120,24 @@ #define IEEE80211_QOS_CTL_TID_MASK 0x000F #define IEEE80211_QOS_CTL_TAG1D_MASK 0x0007 +/* U-APSD queue for WMM IEs sent by AP */ +#define IEEE80211_WMM_IE_AP_QOSINFO_UAPSD (1<<7) + +/* U-APSD queues for WMM IEs sent by STA */ +#define IEEE80211_WMM_IE_STA_QOSINFO_AC_VO (1<<0) +#define IEEE80211_WMM_IE_STA_QOSINFO_AC_VI (1<<1) +#define IEEE80211_WMM_IE_STA_QOSINFO_AC_BK (1<<2) +#define IEEE80211_WMM_IE_STA_QOSINFO_AC_BE (1<<3) +#define IEEE80211_WMM_IE_STA_QOSINFO_AC_MASK 0x0f + +/* U-APSD max SP length for WMM IEs sent by STA */ +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_ALL 0x00 +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_2 0x01 +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_4 0x02 +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_6 0x03 +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_MASK 0x03 +#define IEEE80211_WMM_IE_STA_QOSINFO_SP_SHIFT 5 + struct ieee80211_hdr { __le16 frame_control; __le16 duration_id; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index e1e73c6abef..f313a3cbabd 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -113,6 +113,7 @@ struct ieee80211_tx_queue_params { u16 cw_min; u16 cw_max; u8 aifs; + bool uapsd; }; /** @@ -929,6 +930,11 @@ enum ieee80211_tkip_key_type { * Hardware supports dynamic spatial multiplexing powersave, * ie. can turn off all but one chain and then wake the rest * up as required after, for example, rts/cts handshake. + * + * @IEEE80211_HW_SUPPORTS_UAPSD: + * Hardware supports Unscheduled Automatic Power Save Delivery + * (U-APSD) in managed mode. The mode is configured with + * conf_tx() operation. */ enum ieee80211_hw_flags { IEEE80211_HW_HAS_RATE_CONTROL = 1<<0, @@ -948,6 +954,7 @@ enum ieee80211_hw_flags { IEEE80211_HW_BEACON_FILTER = 1<<14, IEEE80211_HW_SUPPORTS_STATIC_SMPS = 1<<15, IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS = 1<<16, + IEEE80211_HW_SUPPORTS_UAPSD = 1<<17, }; /** diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index dc12e9466ff..8286df5822d 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1128,6 +1128,13 @@ static int ieee80211_set_txq_params(struct wiphy *wiphy, p.cw_max = params->cwmax; p.cw_min = params->cwmin; p.txop = params->txop; + + /* + * Setting tx queue params disables u-apsd because it's only + * called in master mode. + */ + p.uapsd = false; + if (drv_conf_tx(local, params->queue, &p)) { printk(KERN_DEBUG "%s: failed to set TX queue " "parameters for queue %d\n", diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 3e4ac3f3085..3468e378509 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -58,6 +58,15 @@ struct ieee80211_local; #define TU_TO_EXP_TIME(x) (jiffies + usecs_to_jiffies((x) * 1024)) +#define IEEE80211_DEFAULT_UAPSD_QUEUES \ + (IEEE80211_WMM_IE_STA_QOSINFO_AC_BK | \ + IEEE80211_WMM_IE_STA_QOSINFO_AC_BE | \ + IEEE80211_WMM_IE_STA_QOSINFO_AC_VI | \ + IEEE80211_WMM_IE_STA_QOSINFO_AC_VO) + +#define IEEE80211_DEFAULT_MAX_SP_LEN \ + IEEE80211_WMM_IE_STA_QOSINFO_SP_ALL + struct ieee80211_fragment_entry { unsigned long first_frag_time; unsigned int seq; @@ -78,6 +87,7 @@ struct ieee80211_bss { u8 dtim_period; bool wmm_used; + bool uapsd_supported; unsigned long last_probe_resp; @@ -285,7 +295,7 @@ struct ieee80211_work { u8 ssid[IEEE80211_MAX_SSID_LEN]; u8 ssid_len; u8 supp_rates_len; - bool wmm_used, use_11n; + bool wmm_used, use_11n, uapsd_used; } assoc; struct { u32 duration; @@ -306,6 +316,7 @@ enum ieee80211_sta_flags { IEEE80211_STA_DISABLE_11N = BIT(4), IEEE80211_STA_CSA_RECEIVED = BIT(5), IEEE80211_STA_MFP_ENABLED = BIT(6), + IEEE80211_STA_UAPSD_ENABLED = BIT(7), }; struct ieee80211_if_managed { diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 46882914399..0054bba08ce 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -491,6 +491,10 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) else if (local->hw.flags & IEEE80211_HW_SIGNAL_UNSPEC) local->hw.wiphy->signal_type = CFG80211_SIGNAL_TYPE_UNSPEC; + WARN((local->hw.flags & IEEE80211_HW_SUPPORTS_UAPSD) + && (local->hw.flags & IEEE80211_HW_PS_NULLFUNC_STACK), + "U-APSD not supported with HW_PS_NULLFUNC_STACK\n"); + /* * Calculate scan IE length -- we need this to alloc * memory and to subtract from the driver limit. It diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 86f025bc945..39c27d83a4f 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -569,7 +569,7 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, struct ieee80211_tx_queue_params params; size_t left; int count; - u8 *pos; + u8 *pos, uapsd_queues = 0; if (local->hw.queues < 4) return; @@ -579,6 +579,10 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, if (wmm_param_len < 8 || wmm_param[5] /* version */ != 1) return; + + if (ifmgd->flags & IEEE80211_STA_UAPSD_ENABLED) + uapsd_queues = IEEE80211_DEFAULT_UAPSD_QUEUES; + count = wmm_param[6] & 0x0f; if (count == ifmgd->wmm_last_param_set) return; @@ -593,6 +597,7 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, for (; left >= 4; left -= 4, pos += 4) { int aci = (pos[0] >> 5) & 0x03; int acm = (pos[0] >> 4) & 0x01; + bool uapsd = false; int queue; switch (aci) { @@ -600,22 +605,30 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, queue = 3; if (acm) local->wmm_acm |= BIT(1) | BIT(2); /* BK/- */ + if (uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BK) + uapsd = true; break; case 2: /* AC_VI */ queue = 1; if (acm) local->wmm_acm |= BIT(4) | BIT(5); /* CL/VI */ + if (uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VI) + uapsd = true; break; case 3: /* AC_VO */ queue = 0; if (acm) local->wmm_acm |= BIT(6) | BIT(7); /* VO/NC */ + if (uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_VO) + uapsd = true; break; case 0: /* AC_BE */ default: queue = 2; if (acm) local->wmm_acm |= BIT(0) | BIT(3); /* BE/EE */ + if (uapsd_queues & IEEE80211_WMM_IE_STA_QOSINFO_AC_BE) + uapsd = true; break; } @@ -623,11 +636,14 @@ static void ieee80211_sta_wmm_params(struct ieee80211_local *local, params.cw_max = ecw2cw((pos[1] & 0xf0) >> 4); params.cw_min = ecw2cw(pos[1] & 0x0f); params.txop = get_unaligned_le16(pos + 2); + params.uapsd = uapsd; + #ifdef CONFIG_MAC80211_VERBOSE_DEBUG printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d " - "cWmin=%d cWmax=%d txop=%d\n", + "cWmin=%d cWmax=%d txop=%d uapsd=%d\n", wiphy_name(local->hw.wiphy), queue, aci, acm, - params.aifs, params.cw_min, params.cw_max, params.txop); + params.aifs, params.cw_min, params.cw_max, params.txop, + params.uapsd); #endif if (drv_conf_tx(local, queue, ¶ms) && local->ops->conf_tx) printk(KERN_DEBUG "%s: failed to set TX queue " @@ -1906,6 +1922,15 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, wk->assoc.ht_information_ie = ieee80211_bss_get_ie(req->bss, WLAN_EID_HT_INFORMATION); + if (bss->wmm_used && bss->uapsd_supported && + (sdata->local->hw.flags & IEEE80211_HW_SUPPORTS_UAPSD)) { + wk->assoc.uapsd_used = true; + ifmgd->flags |= IEEE80211_STA_UAPSD_ENABLED; + } else { + wk->assoc.uapsd_used = false; + ifmgd->flags &= ~IEEE80211_STA_UAPSD_ENABLED; + } + ssid = ieee80211_bss_get_ie(req->bss, WLAN_EID_SSID); memcpy(wk->assoc.ssid, ssid + 2, ssid[1]); wk->assoc.ssid_len = ssid[1]; diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 30cb62bb45b..9afe2f9885d 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -54,6 +54,23 @@ void ieee80211_rx_bss_put(struct ieee80211_local *local, cfg80211_put_bss(container_of((void *)bss, struct cfg80211_bss, priv)); } +static bool is_uapsd_supported(struct ieee802_11_elems *elems) +{ + u8 qos_info; + + if (elems->wmm_info && elems->wmm_info_len == 7 + && elems->wmm_info[5] == 1) + qos_info = elems->wmm_info[6]; + else if (elems->wmm_param && elems->wmm_param_len == 24 + && elems->wmm_param[5] == 1) + qos_info = elems->wmm_param[6]; + else + /* no valid wmm information or parameter element found */ + return false; + + return qos_info & IEEE80211_WMM_IE_AP_QOSINFO_UAPSD; +} + struct ieee80211_bss * ieee80211_bss_info_update(struct ieee80211_local *local, struct ieee80211_rx_status *rx_status, @@ -117,6 +134,7 @@ ieee80211_bss_info_update(struct ieee80211_local *local, } bss->wmm_used = elems->wmm_param || elems->wmm_info; + bss->uapsd_supported = is_uapsd_supported(elems); if (!beacon) bss->last_probe_resp = jiffies; diff --git a/net/mac80211/util.c b/net/mac80211/util.c index a2ba6e29bd9..e278f97c830 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -792,6 +792,8 @@ void ieee80211_set_wmm_default(struct ieee80211_sub_if_data *sdata) break; } + qparam.uapsd = false; + drv_conf_tx(local, queue, &qparam); } } diff --git a/net/mac80211/work.c b/net/mac80211/work.c index 7c5d95b1bc0..a74fd6ee008 100644 --- a/net/mac80211/work.c +++ b/net/mac80211/work.c @@ -202,7 +202,7 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, struct ieee80211_local *local = sdata->local; struct sk_buff *skb; struct ieee80211_mgmt *mgmt; - u8 *pos; + u8 *pos, qos_info; const u8 *ies; size_t offset = 0, noffset; int i, len, count, rates_len, supp_rates_len; @@ -375,6 +375,14 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, } if (wk->assoc.wmm_used && local->hw.queues >= 4) { + if (wk->assoc.uapsd_used) { + qos_info = IEEE80211_DEFAULT_UAPSD_QUEUES; + qos_info |= (IEEE80211_DEFAULT_MAX_SP_LEN << + IEEE80211_WMM_IE_STA_QOSINFO_SP_SHIFT); + } else { + qos_info = 0; + } + pos = skb_put(skb, 9); *pos++ = WLAN_EID_VENDOR_SPECIFIC; *pos++ = 7; /* len */ @@ -384,7 +392,7 @@ static void ieee80211_send_assoc(struct ieee80211_sub_if_data *sdata, *pos++ = 2; /* WME */ *pos++ = 0; /* WME info */ *pos++ = 1; /* WME ver */ - *pos++ = 0; + *pos++ = qos_info; } /* add any remaining custom (i.e. vendor specific here) IEs */ -- cgit v1.2.3-70-g09d2 From 558a6669d7cb407fbb0b5aec184b5c3b9a893d30 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Tue, 12 Jan 2010 10:43:00 +0200 Subject: ieee80211: add struct ieee80211_hdr_qos The header can be used to create qos nullfunc frames, for example. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index a8c6069a0d9..842701906ae 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -157,6 +157,16 @@ struct ieee80211_hdr_3addr { __le16 seq_ctrl; } __attribute__ ((packed)); +struct ieee80211_qos_hdr { + __le16 frame_control; + __le16 duration_id; + u8 addr1[6]; + u8 addr2[6]; + u8 addr3[6]; + __le16 seq_ctrl; + __le16 qos_ctrl; +} __attribute__ ((packed)); + /** * ieee80211_has_tods - check if IEEE80211_FCTL_TODS is set * @fc: frame control bytes in little-endian byteorder -- cgit v1.2.3-70-g09d2 From cd8c20b650f49354722b8cc1f03320b004815a0a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 13 Jan 2010 16:02:14 +0100 Subject: netfilter: nfnetlink: netns support Make nfnl socket per-petns. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink.h | 8 ++--- include/net/net_namespace.h | 2 ++ net/netfilter/nf_conntrack_netlink.c | 13 ++++---- net/netfilter/nfnetlink.c | 65 +++++++++++++++++++++++------------- net/netfilter/nfnetlink_log.c | 3 +- net/netfilter/nfnetlink_queue.c | 2 +- 6 files changed, 58 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 49d321f3ccd..53923868c9b 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -73,11 +73,11 @@ struct nfnetlink_subsystem { extern int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n); extern int nfnetlink_subsys_unregister(const struct nfnetlink_subsystem *n); -extern int nfnetlink_has_listeners(unsigned int group); -extern int nfnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, +extern int nfnetlink_has_listeners(struct net *net, unsigned int group); +extern int nfnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned group, int echo, gfp_t flags); -extern void nfnetlink_set_err(u32 pid, u32 group, int error); -extern int nfnetlink_unicast(struct sk_buff *skb, u_int32_t pid, int flags); +extern void nfnetlink_set_err(struct net *net, u32 pid, u32 group, int error); +extern int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u_int32_t pid, int flags); extern void nfnl_lock(void); extern void nfnl_unlock(void); diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index f307e133d14..82b7be4db89 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -81,6 +81,8 @@ struct net { #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) struct netns_ct ct; #endif + struct sock *nfnl; + struct sock *nfnl_stash; #endif #ifdef CONFIG_XFRM struct netns_xfrm xfrm; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 59d8064eb52..d4c5d06677f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -482,7 +482,7 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) } else return 0; - if (!item->report && !nfnetlink_has_listeners(group)) + if (!item->report && !nfnetlink_has_listeners(&init_net, group)) return 0; skb = nlmsg_new(ctnetlink_nlmsg_size(ct), GFP_ATOMIC); @@ -559,7 +559,8 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) rcu_read_unlock(); nlmsg_end(skb, nlh); - err = nfnetlink_send(skb, item->pid, group, item->report, GFP_ATOMIC); + err = nfnetlink_send(skb, &init_net, item->pid, group, item->report, + GFP_ATOMIC); if (err == -ENOBUFS || err == -EAGAIN) return -ENOBUFS; @@ -571,7 +572,7 @@ nla_put_failure: nlmsg_failure: kfree_skb(skb); errout: - nfnetlink_set_err(0, group, -ENOBUFS); + nfnetlink_set_err(&init_net, 0, group, -ENOBUFS); return 0; } #endif /* CONFIG_NF_CONNTRACK_EVENTS */ @@ -1539,7 +1540,7 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item) return 0; if (!item->report && - !nfnetlink_has_listeners(NFNLGRP_CONNTRACK_EXP_NEW)) + !nfnetlink_has_listeners(&init_net, NFNLGRP_CONNTRACK_EXP_NEW)) return 0; skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); @@ -1562,7 +1563,7 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item) rcu_read_unlock(); nlmsg_end(skb, nlh); - nfnetlink_send(skb, item->pid, NFNLGRP_CONNTRACK_EXP_NEW, + nfnetlink_send(skb, &init_net, item->pid, NFNLGRP_CONNTRACK_EXP_NEW, item->report, GFP_ATOMIC); return 0; @@ -1572,7 +1573,7 @@ nla_put_failure: nlmsg_failure: kfree_skb(skb); errout: - nfnetlink_set_err(0, 0, -ENOBUFS); + nfnetlink_set_err(&init_net, 0, 0, -ENOBUFS); return 0; } #endif diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index eedc0c1ac7a..8eb0cc23ada 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -40,7 +40,6 @@ MODULE_ALIAS_NET_PF_PROTO(PF_NETLINK, NETLINK_NETFILTER); static char __initdata nfversion[] = "0.30"; -static struct sock *nfnl = NULL; static const struct nfnetlink_subsystem *subsys_table[NFNL_SUBSYS_COUNT]; static DEFINE_MUTEX(nfnl_mutex); @@ -101,34 +100,35 @@ nfnetlink_find_client(u_int16_t type, const struct nfnetlink_subsystem *ss) return &ss->cb[cb_id]; } -int nfnetlink_has_listeners(unsigned int group) +int nfnetlink_has_listeners(struct net *net, unsigned int group) { - return netlink_has_listeners(nfnl, group); + return netlink_has_listeners(net->nfnl, group); } EXPORT_SYMBOL_GPL(nfnetlink_has_listeners); -int nfnetlink_send(struct sk_buff *skb, u32 pid, +int nfnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, unsigned group, int echo, gfp_t flags) { - return nlmsg_notify(nfnl, skb, pid, group, echo, flags); + return nlmsg_notify(net->nfnl, skb, pid, group, echo, flags); } EXPORT_SYMBOL_GPL(nfnetlink_send); -void nfnetlink_set_err(u32 pid, u32 group, int error) +void nfnetlink_set_err(struct net *net, u32 pid, u32 group, int error) { - netlink_set_err(nfnl, pid, group, error); + netlink_set_err(net->nfnl, pid, group, error); } EXPORT_SYMBOL_GPL(nfnetlink_set_err); -int nfnetlink_unicast(struct sk_buff *skb, u_int32_t pid, int flags) +int nfnetlink_unicast(struct sk_buff *skb, struct net *net, u_int32_t pid, int flags) { - return netlink_unicast(nfnl, skb, pid, flags); + return netlink_unicast(net->nfnl, skb, pid, flags); } EXPORT_SYMBOL_GPL(nfnetlink_unicast); /* Process one complete nfnetlink message. */ static int nfnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) { + struct net *net = sock_net(skb->sk); const struct nfnl_callback *nc; const struct nfnetlink_subsystem *ss; int type, err; @@ -170,7 +170,7 @@ replay: if (err < 0) return err; - err = nc->call(nfnl, skb, nlh, (const struct nlattr **)cda); + err = nc->call(net->nfnl, skb, nlh, (const struct nlattr **)cda); if (err == -EAGAIN) goto replay; return err; @@ -184,26 +184,45 @@ static void nfnetlink_rcv(struct sk_buff *skb) nfnl_unlock(); } -static void __exit nfnetlink_exit(void) +static int __net_init nfnetlink_net_init(struct net *net) { - printk("Removing netfilter NETLINK layer.\n"); - netlink_kernel_release(nfnl); - return; + struct sock *nfnl; + + nfnl = netlink_kernel_create(net, NETLINK_NETFILTER, NFNLGRP_MAX, + nfnetlink_rcv, NULL, THIS_MODULE); + if (!nfnl) + return -ENOMEM; + net->nfnl_stash = nfnl; + rcu_assign_pointer(net->nfnl, nfnl); + return 0; } -static int __init nfnetlink_init(void) +static void __net_exit nfnetlink_net_exit_batch(struct list_head *net_exit_list) { - printk("Netfilter messages via NETLINK v%s.\n", nfversion); + struct net *net; - nfnl = netlink_kernel_create(&init_net, NETLINK_NETFILTER, NFNLGRP_MAX, - nfnetlink_rcv, NULL, THIS_MODULE); - if (!nfnl) { - printk(KERN_ERR "cannot initialize nfnetlink!\n"); - return -ENOMEM; - } + list_for_each_entry(net, net_exit_list, exit_list) + rcu_assign_pointer(net->nfnl, NULL); + synchronize_net(); + list_for_each_entry(net, net_exit_list, exit_list) + netlink_kernel_release(net->nfnl_stash); +} - return 0; +static struct pernet_operations nfnetlink_net_ops = { + .init = nfnetlink_net_init, + .exit_batch = nfnetlink_net_exit_batch, +}; + +static int __init nfnetlink_init(void) +{ + printk("Netfilter messages via NETLINK v%s.\n", nfversion); + return register_pernet_subsys(&nfnetlink_net_ops); } +static void __exit nfnetlink_exit(void) +{ + printk("Removing netfilter NETLINK layer.\n"); + unregister_pernet_subsys(&nfnetlink_net_ops); +} module_init(nfnetlink_init); module_exit(nfnetlink_exit); diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 9de0470d557..285e9029a9f 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -323,7 +323,8 @@ __nfulnl_send(struct nfulnl_instance *inst) NLMSG_DONE, sizeof(struct nfgenmsg)); - status = nfnetlink_unicast(inst->skb, inst->peer_pid, MSG_DONTWAIT); + status = nfnetlink_unicast(inst->skb, &init_net, inst->peer_pid, + MSG_DONTWAIT); inst->qlen = 0; inst->skb = NULL; diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 7e3fa410641..5c589b27d6e 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -420,7 +420,7 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum) } /* nfnetlink_unicast will either free the nskb or add it to a socket */ - err = nfnetlink_unicast(nskb, queue->peer_pid, MSG_DONTWAIT); + err = nfnetlink_unicast(nskb, &init_net, queue->peer_pid, MSG_DONTWAIT); if (err < 0) { queue->queue_user_dropped++; goto err_out_unlock; -- cgit v1.2.3-70-g09d2 From 508e14b4a4fb1a824a14f2c5b8d7df67b313f8e4 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 12 Jan 2010 14:27:30 +0000 Subject: netpoll: allow execution of multiple rx_hooks per interface Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/linux/netpoll.h | 11 +++- net/core/netpoll.c | 169 ++++++++++++++++++++++++++++++------------------ 2 files changed, 114 insertions(+), 66 deletions(-) (limited to 'include') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 2524267210d..a765ea89854 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -21,15 +21,20 @@ struct netpoll { __be32 local_ip, remote_ip; u16 local_port, remote_port; u8 remote_mac[ETH_ALEN]; + + struct list_head rx; /* rx_np list element */ }; struct netpoll_info { atomic_t refcnt; + int rx_flags; spinlock_t rx_lock; - struct netpoll *rx_np; /* netpoll that registered an rx_hook */ + struct list_head rx_np; /* netpolls that registered an rx_hook */ + struct sk_buff_head arp_tx; /* list of arp requests to reply to */ struct sk_buff_head txq; + struct delayed_work tx_work; }; @@ -51,7 +56,7 @@ static inline int netpoll_rx(struct sk_buff *skb) unsigned long flags; int ret = 0; - if (!npinfo || (!npinfo->rx_np && !npinfo->rx_flags)) + if (!npinfo || (list_empty(&npinfo->rx_np) && !npinfo->rx_flags)) return 0; spin_lock_irqsave(&npinfo->rx_lock, flags); @@ -67,7 +72,7 @@ static inline int netpoll_rx_on(struct sk_buff *skb) { struct netpoll_info *npinfo = skb->dev->npinfo; - return npinfo && (npinfo->rx_np || npinfo->rx_flags); + return npinfo && (!list_empty(&npinfo->rx_np) || npinfo->rx_flags); } static inline int netpoll_receive_skb(struct sk_buff *skb) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 0b4d0d35ef4..7aa69725376 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -407,11 +407,24 @@ static void arp_reply(struct sk_buff *skb) __be32 sip, tip; unsigned char *sha; struct sk_buff *send_skb; - struct netpoll *np = NULL; + struct netpoll *np, *tmp; + unsigned long flags; + int hits = 0; + + if (list_empty(&npinfo->rx_np)) + return; + + /* Before checking the packet, we do some early + inspection whether this is interesting at all */ + spin_lock_irqsave(&npinfo->rx_lock, flags); + list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) { + if (np->dev == skb->dev) + hits++; + } + spin_unlock_irqrestore(&npinfo->rx_lock, flags); - if (npinfo->rx_np && npinfo->rx_np->dev == skb->dev) - np = npinfo->rx_np; - if (!np) + /* No netpoll struct is using this dev */ + if (!hits) return; /* No arp on this interface */ @@ -437,77 +450,91 @@ static void arp_reply(struct sk_buff *skb) arp_ptr += skb->dev->addr_len; memcpy(&sip, arp_ptr, 4); arp_ptr += 4; - /* if we actually cared about dst hw addr, it would get copied here */ + /* If we actually cared about dst hw addr, + it would get copied here */ arp_ptr += skb->dev->addr_len; memcpy(&tip, arp_ptr, 4); /* Should we ignore arp? */ - if (tip != np->local_ip || - ipv4_is_loopback(tip) || ipv4_is_multicast(tip)) + if (ipv4_is_loopback(tip) || ipv4_is_multicast(tip)) return; size = arp_hdr_len(skb->dev); - send_skb = find_skb(np, size + LL_ALLOCATED_SPACE(np->dev), - LL_RESERVED_SPACE(np->dev)); - if (!send_skb) - return; - - skb_reset_network_header(send_skb); - arp = (struct arphdr *) skb_put(send_skb, size); - send_skb->dev = skb->dev; - send_skb->protocol = htons(ETH_P_ARP); + spin_lock_irqsave(&npinfo->rx_lock, flags); + list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) { + if (tip != np->local_ip) + continue; - /* Fill the device header for the ARP frame */ - if (dev_hard_header(send_skb, skb->dev, ptype, - sha, np->dev->dev_addr, - send_skb->len) < 0) { - kfree_skb(send_skb); - return; - } + send_skb = find_skb(np, size + LL_ALLOCATED_SPACE(np->dev), + LL_RESERVED_SPACE(np->dev)); + if (!send_skb) + continue; - /* - * Fill out the arp protocol part. - * - * we only support ethernet device type, - * which (according to RFC 1390) should always equal 1 (Ethernet). - */ + skb_reset_network_header(send_skb); + arp = (struct arphdr *) skb_put(send_skb, size); + send_skb->dev = skb->dev; + send_skb->protocol = htons(ETH_P_ARP); - arp->ar_hrd = htons(np->dev->type); - arp->ar_pro = htons(ETH_P_IP); - arp->ar_hln = np->dev->addr_len; - arp->ar_pln = 4; - arp->ar_op = htons(type); + /* Fill the device header for the ARP frame */ + if (dev_hard_header(send_skb, skb->dev, ptype, + sha, np->dev->dev_addr, + send_skb->len) < 0) { + kfree_skb(send_skb); + continue; + } - arp_ptr=(unsigned char *)(arp + 1); - memcpy(arp_ptr, np->dev->dev_addr, np->dev->addr_len); - arp_ptr += np->dev->addr_len; - memcpy(arp_ptr, &tip, 4); - arp_ptr += 4; - memcpy(arp_ptr, sha, np->dev->addr_len); - arp_ptr += np->dev->addr_len; - memcpy(arp_ptr, &sip, 4); + /* + * Fill out the arp protocol part. + * + * we only support ethernet device type, + * which (according to RFC 1390) should + * always equal 1 (Ethernet). + */ - netpoll_send_skb(np, send_skb); + arp->ar_hrd = htons(np->dev->type); + arp->ar_pro = htons(ETH_P_IP); + arp->ar_hln = np->dev->addr_len; + arp->ar_pln = 4; + arp->ar_op = htons(type); + + arp_ptr = (unsigned char *)(arp + 1); + memcpy(arp_ptr, np->dev->dev_addr, np->dev->addr_len); + arp_ptr += np->dev->addr_len; + memcpy(arp_ptr, &tip, 4); + arp_ptr += 4; + memcpy(arp_ptr, sha, np->dev->addr_len); + arp_ptr += np->dev->addr_len; + memcpy(arp_ptr, &sip, 4); + + netpoll_send_skb(np, send_skb); + + /* If there are several rx_hooks for the same address, + we're fine by sending a single reply */ + break; + } + spin_unlock_irqrestore(&npinfo->rx_lock, flags); } int __netpoll_rx(struct sk_buff *skb) { int proto, len, ulen; + int hits = 0; struct iphdr *iph; struct udphdr *uh; - struct netpoll_info *npi = skb->dev->npinfo; - struct netpoll *np = npi->rx_np; + struct netpoll_info *npinfo = skb->dev->npinfo; + struct netpoll *np, *tmp; - if (!np) + if (list_empty(&npinfo->rx_np)) goto out; + if (skb->dev->type != ARPHRD_ETHER) goto out; /* check if netpoll clients need ARP */ if (skb->protocol == htons(ETH_P_ARP) && atomic_read(&trapped)) { - skb_queue_tail(&npi->arp_tx, skb); + skb_queue_tail(&npinfo->arp_tx, skb); return 1; } @@ -551,16 +578,23 @@ int __netpoll_rx(struct sk_buff *skb) goto out; if (checksum_udp(skb, uh, ulen, iph->saddr, iph->daddr)) goto out; - if (np->local_ip && np->local_ip != iph->daddr) - goto out; - if (np->remote_ip && np->remote_ip != iph->saddr) - goto out; - if (np->local_port && np->local_port != ntohs(uh->dest)) - goto out; - np->rx_hook(np, ntohs(uh->source), - (char *)(uh+1), - ulen - sizeof(struct udphdr)); + list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) { + if (np->local_ip && np->local_ip != iph->daddr) + continue; + if (np->remote_ip && np->remote_ip != iph->saddr) + continue; + if (np->local_port && np->local_port != ntohs(uh->dest)) + continue; + + np->rx_hook(np, ntohs(uh->source), + (char *)(uh+1), + ulen - sizeof(struct udphdr)); + hits++; + } + + if (!hits) + goto out; kfree_skb(skb); return 1; @@ -684,6 +718,7 @@ int netpoll_setup(struct netpoll *np) struct net_device *ndev = NULL; struct in_device *in_dev; struct netpoll_info *npinfo; + struct netpoll *npe, *tmp; unsigned long flags; int err; @@ -704,7 +739,7 @@ int netpoll_setup(struct netpoll *np) } npinfo->rx_flags = 0; - npinfo->rx_np = NULL; + INIT_LIST_HEAD(&npinfo->rx_np); spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); @@ -785,7 +820,7 @@ int netpoll_setup(struct netpoll *np) if (np->rx_hook) { spin_lock_irqsave(&npinfo->rx_lock, flags); npinfo->rx_flags |= NETPOLL_RX_ENABLED; - npinfo->rx_np = np; + list_add_tail(&np->rx, &npinfo->rx_np); spin_unlock_irqrestore(&npinfo->rx_lock, flags); } @@ -801,9 +836,16 @@ int netpoll_setup(struct netpoll *np) return 0; release: - if (!ndev->npinfo) + if (!ndev->npinfo) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + list_for_each_entry_safe(npe, tmp, &npinfo->rx_np, rx) { + npe->dev = NULL; + } + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + kfree(npinfo); - np->dev = NULL; + } + dev_put(ndev); return err; } @@ -823,10 +865,11 @@ void netpoll_cleanup(struct netpoll *np) if (np->dev) { npinfo = np->dev->npinfo; if (npinfo) { - if (npinfo->rx_np == np) { + if (!list_empty(&npinfo->rx_np)) { spin_lock_irqsave(&npinfo->rx_lock, flags); - npinfo->rx_np = NULL; - npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; + list_del(&np->rx); + if (list_empty(&npinfo->rx_np)) + npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; spin_unlock_irqrestore(&npinfo->rx_lock, flags); } -- cgit v1.2.3-70-g09d2 From 9a58a80a701bdb2d220cdab4914218df5b48d781 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 14 Jan 2010 03:10:54 -0800 Subject: proc_fops: convert drivers/isdn/ to seq_file Convert code away from ->read_proc/->write_proc interfaces. Switch to proc_create()/proc_create_data() which make addition of proc entries reliable wrt NULL ->proc_fops, NULL ->data and so on. Problem with ->read_proc et al is described here commit 786d7e1612f0b0adb6046f19b906609e4fe8b1ba "Fix rmmod/read/write races in /proc entries" [akpm@linux-foundation.org: CONFIG_PROC_FS=n build fix] Signed-off-by: Alexey Dobriyan Signed-off-by: Tilman Schmidt Signed-off-by: Karsten Keil Signed-off-by: David S. Miller --- Documentation/isdn/INTERFACE.CAPI | 9 +- drivers/isdn/capi/capi.c | 99 ++++++---------- drivers/isdn/capi/capidrv.c | 55 +++------ drivers/isdn/capi/kcapi.c | 8 +- drivers/isdn/gigaset/capi.c | 75 ++++++------ drivers/isdn/hardware/avm/avmcard.h | 6 +- drivers/isdn/hardware/avm/b1.c | 54 +++++---- drivers/isdn/hardware/avm/b1dma.c | 71 ++++++------ drivers/isdn/hardware/avm/b1isa.c | 2 +- drivers/isdn/hardware/avm/b1pci.c | 4 +- drivers/isdn/hardware/avm/b1pcmcia.c | 2 +- drivers/isdn/hardware/avm/c4.c | 53 +++++---- drivers/isdn/hardware/avm/t1isa.c | 2 +- drivers/isdn/hardware/avm/t1pci.c | 2 +- drivers/isdn/hardware/eicon/capimain.c | 40 ++++--- drivers/isdn/hardware/eicon/diva_didd.c | 45 ++++---- drivers/isdn/hardware/eicon/divasi.c | 48 ++++---- drivers/isdn/hardware/eicon/divasproc.c | 198 ++++++++++++++------------------ drivers/isdn/hysdn/hycapi.c | 56 ++++----- include/linux/isdn/capilli.h | 3 +- net/bluetooth/cmtp/capi.c | 37 +++--- 21 files changed, 411 insertions(+), 458 deletions(-) (limited to 'include') diff --git a/Documentation/isdn/INTERFACE.CAPI b/Documentation/isdn/INTERFACE.CAPI index 5fe8de5cc72..f172091fb7c 100644 --- a/Documentation/isdn/INTERFACE.CAPI +++ b/Documentation/isdn/INTERFACE.CAPI @@ -149,10 +149,11 @@ char *(*procinfo)(struct capi_ctr *ctrlr) pointer to a callback function returning the entry for the device in the CAPI controller info table, /proc/capi/controller -read_proc_t *ctr_read_proc - pointer to the read_proc callback function for the device's proc file - system entry, /proc/capi/controllers/; will be called with a - pointer to the device's capi_ctr structure as the last (data) argument +const struct file_operations *proc_fops + pointers to callback functions for the device's proc file + system entry, /proc/capi/controllers/; pointer to the device's + capi_ctr structure is available from struct proc_dir_entry::data + which is available from struct inode. Note: Callback functions except send_message() are never called in interrupt context. diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 65bf91e16a4..79f9364aded 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -33,6 +33,7 @@ #endif /* CONFIG_ISDN_CAPI_MIDDLEWARE */ #include #include +#include #include #include #include @@ -1407,114 +1408,84 @@ static void capinc_tty_exit(void) * /proc/capi/capi20: * minor applid nrecvctlpkt nrecvdatapkt nsendctlpkt nsenddatapkt */ -static int proc_capidev_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int capi20_proc_show(struct seq_file *m, void *v) { struct capidev *cdev; struct list_head *l; - int len = 0; read_lock(&capidev_list_lock); list_for_each(l, &capidev_list) { cdev = list_entry(l, struct capidev, list); - len += sprintf(page+len, "0 %d %lu %lu %lu %lu\n", + seq_printf(m, "0 %d %lu %lu %lu %lu\n", cdev->ap.applid, cdev->ap.nrecvctlpkt, cdev->ap.nrecvdatapkt, cdev->ap.nsentctlpkt, cdev->ap.nsentdatapkt); - if (len <= off) { - off -= len; - len = 0; - } else { - if (len-off > count) - goto endloop; - } } - -endloop: read_unlock(&capidev_list_lock); - if (len < count) - *eof = 1; - if (len > count) len = count; - if (len < 0) len = 0; - return len; + return 0; } +static int capi20_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, capi20_proc_show, NULL); +} + +static const struct file_operations capi20_proc_fops = { + .owner = THIS_MODULE, + .open = capi20_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /* * /proc/capi/capi20ncci: * applid ncci */ -static int proc_capincci_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int capi20ncci_proc_show(struct seq_file *m, void *v) { struct capidev *cdev; struct capincci *np; struct list_head *l; - int len = 0; read_lock(&capidev_list_lock); list_for_each(l, &capidev_list) { cdev = list_entry(l, struct capidev, list); for (np=cdev->nccis; np; np = np->next) { - len += sprintf(page+len, "%d 0x%x\n", + seq_printf(m, "%d 0x%x\n", cdev->ap.applid, np->ncci); - if (len <= off) { - off -= len; - len = 0; - } else { - if (len-off > count) - goto endloop; - } } } -endloop: read_unlock(&capidev_list_lock); - *start = page+off; - if (len < count) - *eof = 1; - if (len>count) len = count; - if (len<0) len = 0; - return len; + return 0; } -static struct procfsentries { - char *name; - mode_t mode; - int (*read_proc)(char *page, char **start, off_t off, - int count, int *eof, void *data); - struct proc_dir_entry *procent; -} procfsentries[] = { - /* { "capi", S_IFDIR, 0 }, */ - { "capi/capi20", 0 , proc_capidev_read_proc }, - { "capi/capi20ncci", 0 , proc_capincci_read_proc }, +static int capi20ncci_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, capi20ncci_proc_show, NULL); +} + +static const struct file_operations capi20ncci_proc_fops = { + .owner = THIS_MODULE, + .open = capi20ncci_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, }; static void __init proc_init(void) { - int nelem = ARRAY_SIZE(procfsentries); - int i; - - for (i=0; i < nelem; i++) { - struct procfsentries *p = procfsentries + i; - p->procent = create_proc_entry(p->name, p->mode, NULL); - if (p->procent) p->procent->read_proc = p->read_proc; - } + proc_create("capi/capi20", 0, NULL, &capi20_proc_fops); + proc_create("capi/capi20ncci", 0, NULL, &capi20ncci_proc_fops); } static void __exit proc_exit(void) { - int nelem = ARRAY_SIZE(procfsentries); - int i; - - for (i=nelem-1; i >= 0; i--) { - struct procfsentries *p = procfsentries + i; - if (p->procent) { - remove_proc_entry(p->name, NULL); - p->procent = NULL; - } - } + remove_proc_entry("capi/capi20", NULL); + remove_proc_entry("capi/capi20ncci", NULL); } /* -------- init function and module interface ---------------------- */ diff --git a/drivers/isdn/capi/capidrv.c b/drivers/isdn/capi/capidrv.c index 66b7d7a8647..bb450152fb7 100644 --- a/drivers/isdn/capi/capidrv.c +++ b/drivers/isdn/capi/capidrv.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -2229,59 +2230,37 @@ static void lower_callback(unsigned int cmd, u32 contr, void *data) * /proc/capi/capidrv: * nrecvctlpkt nrecvdatapkt nsendctlpkt nsenddatapkt */ -static int proc_capidrv_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int capidrv_proc_show(struct seq_file *m, void *v) { - int len = 0; - - len += sprintf(page+len, "%lu %lu %lu %lu\n", + seq_printf(m, "%lu %lu %lu %lu\n", global.ap.nrecvctlpkt, global.ap.nrecvdatapkt, global.ap.nsentctlpkt, global.ap.nsentdatapkt); - if (off+count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len-off) ? count : len-off); + return 0; +} + +static int capidrv_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, capidrv_proc_show, NULL); } -static struct procfsentries { - char *name; - mode_t mode; - int (*read_proc)(char *page, char **start, off_t off, - int count, int *eof, void *data); - struct proc_dir_entry *procent; -} procfsentries[] = { - /* { "capi", S_IFDIR, 0 }, */ - { "capi/capidrv", 0 , proc_capidrv_read_proc }, +static const struct file_operations capidrv_proc_fops = { + .owner = THIS_MODULE, + .open = capidrv_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, }; static void __init proc_init(void) { - int nelem = ARRAY_SIZE(procfsentries); - int i; - - for (i=0; i < nelem; i++) { - struct procfsentries *p = procfsentries + i; - p->procent = create_proc_entry(p->name, p->mode, NULL); - if (p->procent) p->procent->read_proc = p->read_proc; - } + proc_create("capi/capidrv", 0, NULL, &capidrv_proc_fops); } static void __exit proc_exit(void) { - int nelem = ARRAY_SIZE(procfsentries); - int i; - - for (i=nelem-1; i >= 0; i--) { - struct procfsentries *p = procfsentries + i; - if (p->procent) { - remove_proc_entry(p->name, NULL); - p->procent = NULL; - } - } + remove_proc_entry("capi/capidrv", NULL); } static int __init capidrv_init(void) diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index dc506ab99ca..b0bacf377c1 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -490,13 +490,7 @@ attach_capi_ctr(struct capi_ctr *card) card->traceflag = showcapimsgs; sprintf(card->procfn, "capi/controllers/%d", card->cnr); - card->procent = create_proc_entry(card->procfn, 0, NULL); - if (card->procent) { - card->procent->read_proc = - (int (*)(char *,char **,off_t,int,int *,void *)) - card->ctr_read_proc; - card->procent->data = card; - } + card->procent = proc_create_data(card->procfn, 0, NULL, card->proc_fops, card); ncards++; printk(KERN_NOTICE "kcapi: Controller [%03d]: %s attached\n", diff --git a/drivers/isdn/gigaset/capi.c b/drivers/isdn/gigaset/capi.c index 3f5cd06af10..6f0ae32906b 100644 --- a/drivers/isdn/gigaset/capi.c +++ b/drivers/isdn/gigaset/capi.c @@ -13,6 +13,8 @@ #include "gigaset.h" #include +#include +#include #include #include #include @@ -2106,35 +2108,22 @@ static char *gigaset_procinfo(struct capi_ctr *ctr) return ctr->name; /* ToDo: more? */ } -/** - * gigaset_ctr_read_proc() - build controller proc file entry - * @page: buffer of PAGE_SIZE bytes for receiving the entry. - * @start: unused. - * @off: unused. - * @count: unused. - * @eof: unused. - * @ctr: controller descriptor structure. - * - * Return value: length of generated entry - */ -static int gigaset_ctr_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctr) +static int gigaset_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctr = m->private; struct cardstate *cs = ctr->driverdata; char *s; int i; - int len = 0; - len += sprintf(page+len, "%-16s %s\n", "name", ctr->name); - len += sprintf(page+len, "%-16s %s %s\n", "dev", + + seq_printf(m, "%-16s %s\n", "name", ctr->name); + seq_printf(m, "%-16s %s %s\n", "dev", dev_driver_string(cs->dev), dev_name(cs->dev)); - len += sprintf(page+len, "%-16s %d\n", "id", cs->myid); + seq_printf(m, "%-16s %d\n", "id", cs->myid); if (cs->gotfwver) - len += sprintf(page+len, "%-16s %d.%d.%d.%d\n", "firmware", + seq_printf(m, "%-16s %d.%d.%d.%d\n", "firmware", cs->fwver[0], cs->fwver[1], cs->fwver[2], cs->fwver[3]); - len += sprintf(page+len, "%-16s %d\n", "channels", - cs->channels); - len += sprintf(page+len, "%-16s %s\n", "onechannel", - cs->onechannel ? "yes" : "no"); + seq_printf(m, "%-16s %d\n", "channels", cs->channels); + seq_printf(m, "%-16s %s\n", "onechannel", cs->onechannel ? "yes" : "no"); switch (cs->mode) { case M_UNKNOWN: @@ -2152,7 +2141,7 @@ static int gigaset_ctr_read_proc(char *page, char **start, off_t off, default: s = "??"; } - len += sprintf(page+len, "%-16s %s\n", "mode", s); + seq_printf(m, "%-16s %s\n", "mode", s); switch (cs->mstate) { case MS_UNINITIALIZED: @@ -2176,25 +2165,21 @@ static int gigaset_ctr_read_proc(char *page, char **start, off_t off, default: s = "??"; } - len += sprintf(page+len, "%-16s %s\n", "mstate", s); + seq_printf(m, "%-16s %s\n", "mstate", s); - len += sprintf(page+len, "%-16s %s\n", "running", - cs->running ? "yes" : "no"); - len += sprintf(page+len, "%-16s %s\n", "connected", - cs->connected ? "yes" : "no"); - len += sprintf(page+len, "%-16s %s\n", "isdn_up", - cs->isdn_up ? "yes" : "no"); - len += sprintf(page+len, "%-16s %s\n", "cidmode", - cs->cidmode ? "yes" : "no"); + seq_printf(m, "%-16s %s\n", "running", cs->running ? "yes" : "no"); + seq_printf(m, "%-16s %s\n", "connected", cs->connected ? "yes" : "no"); + seq_printf(m, "%-16s %s\n", "isdn_up", cs->isdn_up ? "yes" : "no"); + seq_printf(m, "%-16s %s\n", "cidmode", cs->cidmode ? "yes" : "no"); for (i = 0; i < cs->channels; i++) { - len += sprintf(page+len, "[%d]%-13s %d\n", i, "corrupted", + seq_printf(m, "[%d]%-13s %d\n", i, "corrupted", cs->bcs[i].corrupted); - len += sprintf(page+len, "[%d]%-13s %d\n", i, "trans_down", + seq_printf(m, "[%d]%-13s %d\n", i, "trans_down", cs->bcs[i].trans_down); - len += sprintf(page+len, "[%d]%-13s %d\n", i, "trans_up", + seq_printf(m, "[%d]%-13s %d\n", i, "trans_up", cs->bcs[i].trans_up); - len += sprintf(page+len, "[%d]%-13s %d\n", i, "chstate", + seq_printf(m, "[%d]%-13s %d\n", i, "chstate", cs->bcs[i].chstate); switch (cs->bcs[i].proto2) { case L2_BITSYNC: @@ -2209,11 +2194,23 @@ static int gigaset_ctr_read_proc(char *page, char **start, off_t off, default: s = "??"; } - len += sprintf(page+len, "[%d]%-13s %s\n", i, "proto2", s); + seq_printf(m, "[%d]%-13s %s\n", i, "proto2", s); } - return len; + return 0; } +static int gigaset_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, gigaset_proc_show, PDE(inode)->data); +} + +static const struct file_operations gigaset_proc_fops = { + .owner = THIS_MODULE, + .open = gigaset_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; static struct capi_driver capi_driver_gigaset = { .name = "gigaset", @@ -2256,7 +2253,7 @@ int gigaset_isdn_register(struct cardstate *cs, const char *isdnid) iif->ctr.release_appl = gigaset_release_appl; iif->ctr.send_message = gigaset_send_message; iif->ctr.procinfo = gigaset_procinfo; - iif->ctr.ctr_read_proc = gigaset_ctr_read_proc; + iif->ctr.proc_fops = &gigaset_proc_fops; INIT_LIST_HEAD(&iif->appls); skb_queue_head_init(&iif->sendqueue); atomic_set(&iif->sendqlen, 0); diff --git a/drivers/isdn/hardware/avm/avmcard.h b/drivers/isdn/hardware/avm/avmcard.h index d964f07e4a5..a70e8854461 100644 --- a/drivers/isdn/hardware/avm/avmcard.h +++ b/drivers/isdn/hardware/avm/avmcard.h @@ -556,8 +556,7 @@ u16 b1_send_message(struct capi_ctr *ctrl, struct sk_buff *skb); void b1_parse_version(avmctrl_info *card); irqreturn_t b1_interrupt(int interrupt, void *devptr); -int b1ctl_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl); +extern const struct file_operations b1ctl_proc_fops; avmcard_dmainfo *avmcard_dma_alloc(char *name, struct pci_dev *, long rsize, long ssize); @@ -577,7 +576,6 @@ void b1dma_register_appl(struct capi_ctr *ctrl, capi_register_params *rp); void b1dma_release_appl(struct capi_ctr *ctrl, u16 appl); u16 b1dma_send_message(struct capi_ctr *ctrl, struct sk_buff *skb); -int b1dmactl_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl); +extern const struct file_operations b1dmactl_proc_fops; #endif /* _AVMCARD_H_ */ diff --git a/drivers/isdn/hardware/avm/b1.c b/drivers/isdn/hardware/avm/b1.c index a7c0083e78a..c38fa0f4c72 100644 --- a/drivers/isdn/hardware/avm/b1.c +++ b/drivers/isdn/hardware/avm/b1.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #include #include #include @@ -634,18 +636,17 @@ irqreturn_t b1_interrupt(int interrupt, void *devptr) } /* ------------------------------------------------------------- */ -int b1ctl_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl) +static int b1ctl_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; avmctrl_info *cinfo = (avmctrl_info *)(ctrl->driverdata); avmcard *card = cinfo->card; u8 flag; - int len = 0; char *s; - len += sprintf(page+len, "%-16s %s\n", "name", card->name); - len += sprintf(page+len, "%-16s 0x%x\n", "io", card->port); - len += sprintf(page+len, "%-16s %d\n", "irq", card->irq); + seq_printf(m, "%-16s %s\n", "name", card->name); + seq_printf(m, "%-16s 0x%x\n", "io", card->port); + seq_printf(m, "%-16s %d\n", "irq", card->irq); switch (card->cardtype) { case avm_b1isa: s = "B1 ISA"; break; case avm_b1pci: s = "B1 PCI"; break; @@ -658,20 +659,20 @@ int b1ctl_read_proc(char *page, char **start, off_t off, case avm_c2: s = "C2"; break; default: s = "???"; break; } - len += sprintf(page+len, "%-16s %s\n", "type", s); + seq_printf(m, "%-16s %s\n", "type", s); if (card->cardtype == avm_t1isa) - len += sprintf(page+len, "%-16s %d\n", "cardnr", card->cardnr); + seq_printf(m, "%-16s %d\n", "cardnr", card->cardnr); if ((s = cinfo->version[VER_DRIVER]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_driver", s); + seq_printf(m, "%-16s %s\n", "ver_driver", s); if ((s = cinfo->version[VER_CARDTYPE]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_cardtype", s); + seq_printf(m, "%-16s %s\n", "ver_cardtype", s); if ((s = cinfo->version[VER_SERIAL]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_serial", s); + seq_printf(m, "%-16s %s\n", "ver_serial", s); if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[3]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s%s%s%s\n", "protocol", (flag & 0x01) ? " DSS1" : "", (flag & 0x02) ? " CT1" : "", @@ -685,7 +686,7 @@ int b1ctl_read_proc(char *page, char **start, off_t off, if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[5]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s\n", "linetype", (flag & 0x01) ? " point to point" : "", (flag & 0x02) ? " point to multipoint" : "", @@ -693,16 +694,25 @@ int b1ctl_read_proc(char *page, char **start, off_t off, (flag & 0x04) ? " leased line with D-channel" : "" ); } - len += sprintf(page+len, "%-16s %s\n", "cardname", cinfo->cardname); - - if (off+count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len-off) ? count : len-off); + seq_printf(m, "%-16s %s\n", "cardname", cinfo->cardname); + + return 0; +} + +static int b1ctl_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, b1ctl_proc_show, PDE(inode)->data); } +const struct file_operations b1ctl_proc_fops = { + .owner = THIS_MODULE, + .open = b1ctl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +EXPORT_SYMBOL(b1ctl_proc_fops); + /* ------------------------------------------------------------- */ #ifdef CONFIG_PCI @@ -781,8 +791,6 @@ EXPORT_SYMBOL(b1_send_message); EXPORT_SYMBOL(b1_parse_version); EXPORT_SYMBOL(b1_interrupt); -EXPORT_SYMBOL(b1ctl_read_proc); - static int __init b1_init(void) { char *p; diff --git a/drivers/isdn/hardware/avm/b1dma.c b/drivers/isdn/hardware/avm/b1dma.c index 0e84aaae43f..124550d0dbf 100644 --- a/drivers/isdn/hardware/avm/b1dma.c +++ b/drivers/isdn/hardware/avm/b1dma.c @@ -11,6 +11,8 @@ #include #include +#include +#include #include #include #include @@ -855,21 +857,20 @@ u16 b1dma_send_message(struct capi_ctr *ctrl, struct sk_buff *skb) /* ------------------------------------------------------------- */ -int b1dmactl_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl) +static int b1dmactl_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; avmctrl_info *cinfo = (avmctrl_info *)(ctrl->driverdata); avmcard *card = cinfo->card; u8 flag; - int len = 0; char *s; u32 txoff, txlen, rxoff, rxlen, csr; unsigned long flags; - len += sprintf(page+len, "%-16s %s\n", "name", card->name); - len += sprintf(page+len, "%-16s 0x%x\n", "io", card->port); - len += sprintf(page+len, "%-16s %d\n", "irq", card->irq); - len += sprintf(page+len, "%-16s 0x%lx\n", "membase", card->membase); + seq_printf(m, "%-16s %s\n", "name", card->name); + seq_printf(m, "%-16s 0x%x\n", "io", card->port); + seq_printf(m, "%-16s %d\n", "irq", card->irq); + seq_printf(m, "%-16s 0x%lx\n", "membase", card->membase); switch (card->cardtype) { case avm_b1isa: s = "B1 ISA"; break; case avm_b1pci: s = "B1 PCI"; break; @@ -882,18 +883,18 @@ int b1dmactl_read_proc(char *page, char **start, off_t off, case avm_c2: s = "C2"; break; default: s = "???"; break; } - len += sprintf(page+len, "%-16s %s\n", "type", s); + seq_printf(m, "%-16s %s\n", "type", s); if ((s = cinfo->version[VER_DRIVER]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_driver", s); + seq_printf(m, "%-16s %s\n", "ver_driver", s); if ((s = cinfo->version[VER_CARDTYPE]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_cardtype", s); + seq_printf(m, "%-16s %s\n", "ver_cardtype", s); if ((s = cinfo->version[VER_SERIAL]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_serial", s); + seq_printf(m, "%-16s %s\n", "ver_serial", s); if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[3]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s%s%s%s\n", "protocol", (flag & 0x01) ? " DSS1" : "", (flag & 0x02) ? " CT1" : "", @@ -907,7 +908,7 @@ int b1dmactl_read_proc(char *page, char **start, off_t off, if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[5]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s\n", "linetype", (flag & 0x01) ? " point to point" : "", (flag & 0x02) ? " point to multipoint" : "", @@ -915,7 +916,7 @@ int b1dmactl_read_proc(char *page, char **start, off_t off, (flag & 0x04) ? " leased line with D-channel" : "" ); } - len += sprintf(page+len, "%-16s %s\n", "cardname", cinfo->cardname); + seq_printf(m, "%-16s %s\n", "cardname", cinfo->cardname); spin_lock_irqsave(&card->lock, flags); @@ -930,27 +931,30 @@ int b1dmactl_read_proc(char *page, char **start, off_t off, spin_unlock_irqrestore(&card->lock, flags); - len += sprintf(page+len, "%-16s 0x%lx\n", - "csr (cached)", (unsigned long)card->csr); - len += sprintf(page+len, "%-16s 0x%lx\n", - "csr", (unsigned long)csr); - len += sprintf(page+len, "%-16s %lu\n", - "txoff", (unsigned long)txoff); - len += sprintf(page+len, "%-16s %lu\n", - "txlen", (unsigned long)txlen); - len += sprintf(page+len, "%-16s %lu\n", - "rxoff", (unsigned long)rxoff); - len += sprintf(page+len, "%-16s %lu\n", - "rxlen", (unsigned long)rxlen); - - if (off+count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len-off) ? count : len-off); + seq_printf(m, "%-16s 0x%lx\n", "csr (cached)", (unsigned long)card->csr); + seq_printf(m, "%-16s 0x%lx\n", "csr", (unsigned long)csr); + seq_printf(m, "%-16s %lu\n", "txoff", (unsigned long)txoff); + seq_printf(m, "%-16s %lu\n", "txlen", (unsigned long)txlen); + seq_printf(m, "%-16s %lu\n", "rxoff", (unsigned long)rxoff); + seq_printf(m, "%-16s %lu\n", "rxlen", (unsigned long)rxlen); + + return 0; +} + +static int b1dmactl_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, b1dmactl_proc_show, PDE(inode)->data); } +const struct file_operations b1dmactl_proc_fops = { + .owner = THIS_MODULE, + .open = b1dmactl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +EXPORT_SYMBOL(b1dmactl_proc_fops); + /* ------------------------------------------------------------- */ EXPORT_SYMBOL(b1dma_reset); @@ -963,7 +967,6 @@ EXPORT_SYMBOL(b1dma_reset_ctr); EXPORT_SYMBOL(b1dma_register_appl); EXPORT_SYMBOL(b1dma_release_appl); EXPORT_SYMBOL(b1dma_send_message); -EXPORT_SYMBOL(b1dmactl_read_proc); static int __init b1dma_init(void) { diff --git a/drivers/isdn/hardware/avm/b1isa.c b/drivers/isdn/hardware/avm/b1isa.c index 6461a32bc83..ff5390546f9 100644 --- a/drivers/isdn/hardware/avm/b1isa.c +++ b/drivers/isdn/hardware/avm/b1isa.c @@ -121,7 +121,7 @@ static int b1isa_probe(struct pci_dev *pdev) cinfo->capi_ctrl.load_firmware = b1_load_firmware; cinfo->capi_ctrl.reset_ctr = b1_reset_ctr; cinfo->capi_ctrl.procinfo = b1isa_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1ctl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1ctl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/avm/b1pci.c b/drivers/isdn/hardware/avm/b1pci.c index 5b314a2c404..c97e4315079 100644 --- a/drivers/isdn/hardware/avm/b1pci.c +++ b/drivers/isdn/hardware/avm/b1pci.c @@ -112,7 +112,7 @@ static int b1pci_probe(struct capicardparams *p, struct pci_dev *pdev) cinfo->capi_ctrl.load_firmware = b1_load_firmware; cinfo->capi_ctrl.reset_ctr = b1_reset_ctr; cinfo->capi_ctrl.procinfo = b1pci_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1ctl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1ctl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); cinfo->capi_ctrl.owner = THIS_MODULE; @@ -251,7 +251,7 @@ static int b1pciv4_probe(struct capicardparams *p, struct pci_dev *pdev) cinfo->capi_ctrl.load_firmware = b1dma_load_firmware; cinfo->capi_ctrl.reset_ctr = b1dma_reset_ctr; cinfo->capi_ctrl.procinfo = b1pciv4_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1dmactl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1dmactl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/avm/b1pcmcia.c b/drivers/isdn/hardware/avm/b1pcmcia.c index 7740403b40e..d6391e0afee 100644 --- a/drivers/isdn/hardware/avm/b1pcmcia.c +++ b/drivers/isdn/hardware/avm/b1pcmcia.c @@ -108,7 +108,7 @@ static int b1pcmcia_add_card(unsigned int port, unsigned irq, cinfo->capi_ctrl.load_firmware = b1_load_firmware; cinfo->capi_ctrl.reset_ctr = b1_reset_ctr; cinfo->capi_ctrl.procinfo = b1pcmcia_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1ctl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1ctl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/avm/c4.c b/drivers/isdn/hardware/avm/c4.c index 6833301a45f..de6e6b31181 100644 --- a/drivers/isdn/hardware/avm/c4.c +++ b/drivers/isdn/hardware/avm/c4.c @@ -11,6 +11,8 @@ #include #include +#include +#include #include #include #include @@ -1062,19 +1064,18 @@ static char *c4_procinfo(struct capi_ctr *ctrl) return cinfo->infobuf; } -static int c4_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl) +static int c4_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; avmctrl_info *cinfo = (avmctrl_info *)(ctrl->driverdata); avmcard *card = cinfo->card; u8 flag; - int len = 0; char *s; - len += sprintf(page+len, "%-16s %s\n", "name", card->name); - len += sprintf(page+len, "%-16s 0x%x\n", "io", card->port); - len += sprintf(page+len, "%-16s %d\n", "irq", card->irq); - len += sprintf(page+len, "%-16s 0x%lx\n", "membase", card->membase); + seq_printf(m, "%-16s %s\n", "name", card->name); + seq_printf(m, "%-16s 0x%x\n", "io", card->port); + seq_printf(m, "%-16s %d\n", "irq", card->irq); + seq_printf(m, "%-16s 0x%lx\n", "membase", card->membase); switch (card->cardtype) { case avm_b1isa: s = "B1 ISA"; break; case avm_b1pci: s = "B1 PCI"; break; @@ -1087,18 +1088,18 @@ static int c4_read_proc(char *page, char **start, off_t off, case avm_c2: s = "C2"; break; default: s = "???"; break; } - len += sprintf(page+len, "%-16s %s\n", "type", s); + seq_printf(m, "%-16s %s\n", "type", s); if ((s = cinfo->version[VER_DRIVER]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_driver", s); + seq_printf(m, "%-16s %s\n", "ver_driver", s); if ((s = cinfo->version[VER_CARDTYPE]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_cardtype", s); + seq_printf(m, "%-16s %s\n", "ver_cardtype", s); if ((s = cinfo->version[VER_SERIAL]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_serial", s); + seq_printf(m, "%-16s %s\n", "ver_serial", s); if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[3]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s%s%s%s\n", "protocol", (flag & 0x01) ? " DSS1" : "", (flag & 0x02) ? " CT1" : "", @@ -1112,7 +1113,7 @@ static int c4_read_proc(char *page, char **start, off_t off, if (card->cardtype != avm_m1) { flag = ((u8 *)(ctrl->profile.manu))[5]; if (flag) - len += sprintf(page+len, "%-16s%s%s%s%s\n", + seq_printf(m, "%-16s%s%s%s%s\n", "linetype", (flag & 0x01) ? " point to point" : "", (flag & 0x02) ? " point to multipoint" : "", @@ -1120,16 +1121,24 @@ static int c4_read_proc(char *page, char **start, off_t off, (flag & 0x04) ? " leased line with D-channel" : "" ); } - len += sprintf(page+len, "%-16s %s\n", "cardname", cinfo->cardname); - - if (off+count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len-off) ? count : len-off); + seq_printf(m, "%-16s %s\n", "cardname", cinfo->cardname); + + return 0; } +static int c4_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, c4_proc_show, PDE(inode)->data); +} + +static const struct file_operations c4_proc_fops = { + .owner = THIS_MODULE, + .open = c4_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /* ------------------------------------------------------------- */ static int c4_add_card(struct capicardparams *p, struct pci_dev *dev, @@ -1201,7 +1210,7 @@ static int c4_add_card(struct capicardparams *p, struct pci_dev *dev, cinfo->capi_ctrl.load_firmware = c4_load_firmware; cinfo->capi_ctrl.reset_ctr = c4_reset_ctr; cinfo->capi_ctrl.procinfo = c4_procinfo; - cinfo->capi_ctrl.ctr_read_proc = c4_read_proc; + cinfo->capi_ctrl.proc_fops = &c4_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/avm/t1isa.c b/drivers/isdn/hardware/avm/t1isa.c index 1c53fd49adb..baeeb3c2a3e 100644 --- a/drivers/isdn/hardware/avm/t1isa.c +++ b/drivers/isdn/hardware/avm/t1isa.c @@ -429,7 +429,7 @@ static int t1isa_probe(struct pci_dev *pdev, int cardnr) cinfo->capi_ctrl.load_firmware = t1isa_load_firmware; cinfo->capi_ctrl.reset_ctr = t1isa_reset_ctr; cinfo->capi_ctrl.procinfo = t1isa_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1ctl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1ctl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/avm/t1pci.c b/drivers/isdn/hardware/avm/t1pci.c index e6d298d7514..5a3f8309801 100644 --- a/drivers/isdn/hardware/avm/t1pci.c +++ b/drivers/isdn/hardware/avm/t1pci.c @@ -119,7 +119,7 @@ static int t1pci_add_card(struct capicardparams *p, struct pci_dev *pdev) cinfo->capi_ctrl.load_firmware = b1dma_load_firmware; cinfo->capi_ctrl.reset_ctr = b1dma_reset_ctr; cinfo->capi_ctrl.procinfo = t1pci_procinfo; - cinfo->capi_ctrl.ctr_read_proc = b1dmactl_read_proc; + cinfo->capi_ctrl.proc_fops = &b1dmactl_proc_fops; strcpy(cinfo->capi_ctrl.name, card->name); retval = attach_capi_ctr(&cinfo->capi_ctrl); diff --git a/drivers/isdn/hardware/eicon/capimain.c b/drivers/isdn/hardware/eicon/capimain.c index 98fcdfc7ca5..0f073cd7376 100644 --- a/drivers/isdn/hardware/eicon/capimain.c +++ b/drivers/isdn/hardware/eicon/capimain.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include "os_capi.h" @@ -75,25 +76,32 @@ void diva_os_free_message_buffer(diva_os_message_buffer_s * dmb) /* * proc function for controller info */ -static int diva_ctl_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl) +static int diva_ctl_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; diva_card *card = (diva_card *) ctrl->driverdata; - int len = 0; - - len += sprintf(page + len, "%s\n", ctrl->name); - len += sprintf(page + len, "Serial No. : %s\n", ctrl->serial); - len += sprintf(page + len, "Id : %d\n", card->Id); - len += sprintf(page + len, "Channels : %d\n", card->d.channels); - - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); + + seq_printf(m, "%s\n", ctrl->name); + seq_printf(m, "Serial No. : %s\n", ctrl->serial); + seq_printf(m, "Id : %d\n", card->Id); + seq_printf(m, "Channels : %d\n", card->d.channels); + + return 0; +} + +static int diva_ctl_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, diva_ctl_proc_show, NULL); } +static const struct file_operations diva_ctl_proc_fops = { + .owner = THIS_MODULE, + .open = diva_ctl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /* * set additional os settings in capi_ctr struct */ @@ -102,7 +110,7 @@ void diva_os_set_controller_struct(struct capi_ctr *ctrl) ctrl->driver_name = DRIVERLNAME; ctrl->load_firmware = NULL; ctrl->reset_ctr = NULL; - ctrl->ctr_read_proc = diva_ctl_read_proc; + ctrl->proc_fops = &diva_ctl_proc_fops; ctrl->owner = THIS_MODULE; } diff --git a/drivers/isdn/hardware/eicon/diva_didd.c b/drivers/isdn/hardware/eicon/diva_didd.c index 993b14cf177..5d06a743782 100644 --- a/drivers/isdn/hardware/eicon/diva_didd.c +++ b/drivers/isdn/hardware/eicon/diva_didd.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include "platform.h" @@ -62,39 +63,41 @@ static char *getrev(const char *revision) return rev; } -static int -proc_read(char *page, char **start, off_t off, int count, int *eof, - void *data) +static int divadidd_proc_show(struct seq_file *m, void *v) { - int len = 0; char tmprev[32]; strcpy(tmprev, main_revision); - len += sprintf(page + len, "%s\n", DRIVERNAME); - len += sprintf(page + len, "name : %s\n", DRIVERLNAME); - len += sprintf(page + len, "release : %s\n", DRIVERRELEASE_DIDD); - len += sprintf(page + len, "build : %s(%s)\n", + seq_printf(m, "%s\n", DRIVERNAME); + seq_printf(m, "name : %s\n", DRIVERLNAME); + seq_printf(m, "release : %s\n", DRIVERRELEASE_DIDD); + seq_printf(m, "build : %s(%s)\n", diva_didd_common_code_build, DIVA_BUILD); - len += sprintf(page + len, "revision : %s\n", getrev(tmprev)); - - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); + seq_printf(m, "revision : %s\n", getrev(tmprev)); + + return 0; } +static int divadidd_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, divadidd_proc_show, NULL); +} + +static const struct file_operations divadidd_proc_fops = { + .owner = THIS_MODULE, + .open = divadidd_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + static int DIVA_INIT_FUNCTION create_proc(void) { proc_net_eicon = proc_mkdir("eicon", init_net.proc_net); if (proc_net_eicon) { - if ((proc_didd = - create_proc_entry(DRIVERLNAME, S_IFREG | S_IRUGO, - proc_net_eicon))) { - proc_didd->read_proc = proc_read; - } + proc_didd = proc_create(DRIVERLNAME, S_IRUGO, proc_net_eicon, + &divadidd_proc_fops); return (1); } return (0); diff --git a/drivers/isdn/hardware/eicon/divasi.c b/drivers/isdn/hardware/eicon/divasi.c index 69e71ebe784..f577719ab3f 100644 --- a/drivers/isdn/hardware/eicon/divasi.c +++ b/drivers/isdn/hardware/eicon/divasi.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -86,39 +87,40 @@ static void diva_um_timer_function(unsigned long data); extern struct proc_dir_entry *proc_net_eicon; static struct proc_dir_entry *um_idi_proc_entry = NULL; -static int -um_idi_proc_read(char *page, char **start, off_t off, int count, int *eof, - void *data) +static int um_idi_proc_show(struct seq_file *m, void *v) { - int len = 0; char tmprev[32]; - len += sprintf(page + len, "%s\n", DRIVERNAME); - len += sprintf(page + len, "name : %s\n", DRIVERLNAME); - len += sprintf(page + len, "release : %s\n", DRIVERRELEASE_IDI); + seq_printf(m, "%s\n", DRIVERNAME); + seq_printf(m, "name : %s\n", DRIVERLNAME); + seq_printf(m, "release : %s\n", DRIVERRELEASE_IDI); strcpy(tmprev, main_revision); - len += sprintf(page + len, "revision : %s\n", getrev(tmprev)); - len += sprintf(page + len, "build : %s\n", DIVA_BUILD); - len += sprintf(page + len, "major : %d\n", major); - - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); + seq_printf(m, "revision : %s\n", getrev(tmprev)); + seq_printf(m, "build : %s\n", DIVA_BUILD); + seq_printf(m, "major : %d\n", major); + + return 0; +} + +static int um_idi_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, um_idi_proc_show, NULL); } +static const struct file_operations um_idi_proc_fops = { + .owner = THIS_MODULE, + .open = um_idi_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + static int DIVA_INIT_FUNCTION create_um_idi_proc(void) { - um_idi_proc_entry = create_proc_entry(DRIVERLNAME, - S_IFREG | S_IRUGO | S_IWUSR, - proc_net_eicon); + um_idi_proc_entry = proc_create(DRIVERLNAME, S_IRUGO, proc_net_eicon, + &um_idi_proc_fops); if (!um_idi_proc_entry) return (0); - - um_idi_proc_entry->read_proc = um_idi_proc_read; - return (1); } diff --git a/drivers/isdn/hardware/eicon/divasproc.c b/drivers/isdn/hardware/eicon/divasproc.c index 040827288ec..46d44a94262 100644 --- a/drivers/isdn/hardware/eicon/divasproc.c +++ b/drivers/isdn/hardware/eicon/divasproc.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -141,14 +142,10 @@ void remove_divas_proc(void) } } -/* -** write group_optimization -*/ -static int -write_grp_opt(struct file *file, const char __user *buffer, unsigned long count, - void *data) +static ssize_t grp_opt_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *pos) { - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = PDE(file->f_path.dentry->d_inode)->data; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; if ((count == 1) || (count == 2)) { @@ -172,14 +169,10 @@ write_grp_opt(struct file *file, const char __user *buffer, unsigned long count, return (-EINVAL); } -/* -** write dynamic_l1_down -*/ -static int -write_d_l1_down(struct file *file, const char __user *buffer, unsigned long count, - void *data) +static ssize_t d_l1_down_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *pos) { - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = PDE(file->f_path.dentry->d_inode)->data; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; if ((count == 1) || (count == 2)) { @@ -203,63 +196,62 @@ write_d_l1_down(struct file *file, const char __user *buffer, unsigned long coun return (-EINVAL); } - -/* -** read dynamic_l1_down -*/ -static int -read_d_l1_down(char *page, char **start, off_t off, int count, int *eof, - void *data) +static int d_l1_down_proc_show(struct seq_file *m, void *v) { - int len = 0; - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = m->private; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; - len += sprintf(page + len, "%s\n", + seq_printf(m, "%s\n", (IoAdapter->capi_cfg. cfg_1 & DIVA_XDI_CAPI_CFG_1_DYNAMIC_L1_ON) ? "1" : "0"); + return 0; +} - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); +static int d_l1_down_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, d_l1_down_proc_show, PDE(inode)->data); } -/* -** read group_optimization -*/ -static int -read_grp_opt(char *page, char **start, off_t off, int count, int *eof, - void *data) +static const struct file_operations d_l1_down_proc_fops = { + .owner = THIS_MODULE, + .open = d_l1_down_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = d_l1_down_proc_write, +}; + +static int grp_opt_proc_show(struct seq_file *m, void *v) { - int len = 0; - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = m->private; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; - len += sprintf(page + len, "%s\n", + seq_printf(m, "%s\n", (IoAdapter->capi_cfg. cfg_1 & DIVA_XDI_CAPI_CFG_1_GROUP_POPTIMIZATION_ON) ? "1" : "0"); + return 0; +} - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); +static int grp_opt_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, grp_opt_proc_show, PDE(inode)->data); } -/* -** info write -*/ -static int -info_write(struct file *file, const char __user *buffer, unsigned long count, - void *data) +static const struct file_operations grp_opt_proc_fops = { + .owner = THIS_MODULE, + .open = grp_opt_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = grp_opt_proc_write, +}; + +static ssize_t info_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *pos) { - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = PDE(file->f_path.dentry->d_inode)->data; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; char c[4]; @@ -277,63 +269,46 @@ info_write(struct file *file, const char __user *buffer, unsigned long count, return (-EINVAL); } -/* -** info read -*/ -static int -info_read(char *page, char **start, off_t off, int count, int *eof, - void *data) +static int info_proc_show(struct seq_file *m, void *v) { int i = 0; - int len = 0; char *p; char tmpser[16]; - diva_os_xdi_adapter_t *a = (diva_os_xdi_adapter_t *) data; + diva_os_xdi_adapter_t *a = m->private; PISDN_ADAPTER IoAdapter = IoAdapters[a->controller - 1]; - len += - sprintf(page + len, "Name : %s\n", - IoAdapter->Properties.Name); - len += sprintf(page + len, "DSP state : %08x\n", a->dsp_mask); - len += sprintf(page + len, "Channels : %02d\n", - IoAdapter->Properties.Channels); - len += sprintf(page + len, "E. max/used : %03d/%03d\n", + seq_printf(m, "Name : %s\n", IoAdapter->Properties.Name); + seq_printf(m, "DSP state : %08x\n", a->dsp_mask); + seq_printf(m, "Channels : %02d\n", IoAdapter->Properties.Channels); + seq_printf(m, "E. max/used : %03d/%03d\n", IoAdapter->e_max, IoAdapter->e_count); diva_get_vserial_number(IoAdapter, tmpser); - len += sprintf(page + len, "Serial : %s\n", tmpser); - len += - sprintf(page + len, "IRQ : %d\n", - IoAdapter->irq_info.irq_nr); - len += sprintf(page + len, "CardIndex : %d\n", a->CardIndex); - len += sprintf(page + len, "CardOrdinal : %d\n", a->CardOrdinal); - len += sprintf(page + len, "Controller : %d\n", a->controller); - len += sprintf(page + len, "Bus-Type : %s\n", + seq_printf(m, "Serial : %s\n", tmpser); + seq_printf(m, "IRQ : %d\n", IoAdapter->irq_info.irq_nr); + seq_printf(m, "CardIndex : %d\n", a->CardIndex); + seq_printf(m, "CardOrdinal : %d\n", a->CardOrdinal); + seq_printf(m, "Controller : %d\n", a->controller); + seq_printf(m, "Bus-Type : %s\n", (a->Bus == DIVAS_XDI_ADAPTER_BUS_ISA) ? "ISA" : "PCI"); - len += sprintf(page + len, "Port-Name : %s\n", a->port_name); + seq_printf(m, "Port-Name : %s\n", a->port_name); if (a->Bus == DIVAS_XDI_ADAPTER_BUS_PCI) { - len += - sprintf(page + len, "PCI-bus : %d\n", - a->resources.pci.bus); - len += - sprintf(page + len, "PCI-func : %d\n", - a->resources.pci.func); + seq_printf(m, "PCI-bus : %d\n", a->resources.pci.bus); + seq_printf(m, "PCI-func : %d\n", a->resources.pci.func); for (i = 0; i < 8; i++) { if (a->resources.pci.bar[i]) { - len += - sprintf(page + len, + seq_printf(m, "Mem / I/O %d : 0x%x / mapped : 0x%lx", i, a->resources.pci.bar[i], (unsigned long) a->resources. pci.addr[i]); if (a->resources.pci.length[i]) { - len += - sprintf(page + len, + seq_printf(m, " / length : %d", a->resources.pci. length[i]); } - len += sprintf(page + len, "\n"); + seq_putc(m, '\n'); } } } @@ -353,16 +328,25 @@ info_read(char *page, char **start, off_t off, int count, int *eof, } else { p = "ready"; } - len += sprintf(page + len, "State : %s\n", p); + seq_printf(m, "State : %s\n", p); - if (off + count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len - off) ? count : len - off); + return 0; +} + +static int info_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, info_proc_show, PDE(inode)->data); } +static const struct file_operations info_proc_fops = { + .owner = THIS_MODULE, + .open = info_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = info_proc_write, +}; + /* ** adapter proc init/de-init */ @@ -380,28 +364,20 @@ int create_adapter_proc(diva_os_xdi_adapter_t * a) return (0); a->proc_adapter_dir = (void *) de; - if (!(pe = - create_proc_entry(info_proc_name, S_IFREG | S_IRUGO | S_IWUSR, de))) + pe = proc_create_data(info_proc_name, S_IRUGO | S_IWUSR, de, + &info_proc_fops, a); + if (!pe) return (0); a->proc_info = (void *) pe; - pe->write_proc = info_write; - pe->read_proc = info_read; - pe->data = a; - if ((pe = create_proc_entry(grp_opt_proc_name, - S_IFREG | S_IRUGO | S_IWUSR, de))) { + pe = proc_create_data(grp_opt_proc_name, S_IRUGO | S_IWUSR, de, + &grp_opt_proc_fops, a); + if (pe) a->proc_grp_opt = (void *) pe; - pe->write_proc = write_grp_opt; - pe->read_proc = read_grp_opt; - pe->data = a; - } - if ((pe = create_proc_entry(d_l1_down_proc_name, - S_IFREG | S_IRUGO | S_IWUSR, de))) { + pe = proc_create_data(d_l1_down_proc_name, S_IRUGO | S_IWUSR, de, + &d_l1_down_proc_fops, a); + if (pe) a->proc_d_l1_down = (void *) pe; - pe->write_proc = write_d_l1_down; - pe->read_proc = read_d_l1_down; - pe->data = a; - } DBG_TRC(("proc entry %s created", tmp)); diff --git a/drivers/isdn/hysdn/hycapi.c b/drivers/isdn/hysdn/hycapi.c index 4ffaa14b9fc..fe874afa4f8 100644 --- a/drivers/isdn/hysdn/hycapi.c +++ b/drivers/isdn/hysdn/hycapi.c @@ -11,6 +11,8 @@ */ #include +#include +#include #include #include #include @@ -432,26 +434,16 @@ static u16 hycapi_send_message(struct capi_ctr *ctrl, struct sk_buff *skb) return retval; } -/********************************************************************* -hycapi_read_proc - -Informations provided in the /proc/capi-entries. - -*********************************************************************/ - -static int hycapi_read_proc(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *ctrl) +static int hycapi_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; hycapictrl_info *cinfo = (hycapictrl_info *)(ctrl->driverdata); hysdn_card *card = cinfo->card; - int len = 0; char *s; -#ifdef HYCAPI_PRINTFNAMES - printk(KERN_NOTICE "hycapi_read_proc\n"); -#endif - len += sprintf(page+len, "%-16s %s\n", "name", cinfo->cardname); - len += sprintf(page+len, "%-16s 0x%x\n", "io", card->iobase); - len += sprintf(page+len, "%-16s %d\n", "irq", card->irq); + + seq_printf(m, "%-16s %s\n", "name", cinfo->cardname); + seq_printf(m, "%-16s 0x%x\n", "io", card->iobase); + seq_printf(m, "%-16s %d\n", "irq", card->irq); switch (card->brdtype) { case BD_PCCARD: s = "HYSDN Hycard"; break; @@ -461,24 +453,32 @@ static int hycapi_read_proc(char *page, char **start, off_t off, case BD_PLEXUS: s = "HYSDN Plexus30"; break; default: s = "???"; break; } - len += sprintf(page+len, "%-16s %s\n", "type", s); + seq_printf(m, "%-16s %s\n", "type", s); if ((s = cinfo->version[VER_DRIVER]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_driver", s); + seq_printf(m, "%-16s %s\n", "ver_driver", s); if ((s = cinfo->version[VER_CARDTYPE]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_cardtype", s); + seq_printf(m, "%-16s %s\n", "ver_cardtype", s); if ((s = cinfo->version[VER_SERIAL]) != NULL) - len += sprintf(page+len, "%-16s %s\n", "ver_serial", s); + seq_printf(m, "%-16s %s\n", "ver_serial", s); - len += sprintf(page+len, "%-16s %s\n", "cardname", cinfo->cardname); + seq_printf(m, "%-16s %s\n", "cardname", cinfo->cardname); - if (off+count >= len) - *eof = 1; - if (len < off) - return 0; - *start = page + off; - return ((count < len-off) ? count : len-off); + return 0; +} + +static int hycapi_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, hycapi_proc_show, PDE(inode)->data); } +static const struct file_operations hycapi_proc_fops = { + .owner = THIS_MODULE, + .open = hycapi_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /************************************************************** hycapi_load_firmware @@ -774,7 +774,7 @@ hycapi_capi_create(hysdn_card *card) ctrl->load_firmware = hycapi_load_firmware; ctrl->reset_ctr = hycapi_reset_ctr; ctrl->procinfo = hycapi_procinfo; - ctrl->ctr_read_proc = hycapi_read_proc; + ctrl->proc_fops = &hycapi_proc_fops; strcpy(ctrl->name, cinfo->cardname); ctrl->owner = THIS_MODULE; diff --git a/include/linux/isdn/capilli.h b/include/linux/isdn/capilli.h index 7acb87a4487..d3e5e9da0c8 100644 --- a/include/linux/isdn/capilli.h +++ b/include/linux/isdn/capilli.h @@ -50,8 +50,7 @@ struct capi_ctr { u16 (*send_message)(struct capi_ctr *, struct sk_buff *skb); char *(*procinfo)(struct capi_ctr *); - int (*ctr_read_proc)(char *page, char **start, off_t off, - int count, int *eof, struct capi_ctr *card); + const struct file_operations *proc_fops; /* filled in before calling ready callback */ u8 manu[CAPI_MANUFACTURER_LEN]; /* CAPI_GET_MANUFACTURER */ diff --git a/net/bluetooth/cmtp/capi.c b/net/bluetooth/cmtp/capi.c index 97f8d68d574..3487cfe74ae 100644 --- a/net/bluetooth/cmtp/capi.c +++ b/net/bluetooth/cmtp/capi.c @@ -21,7 +21,8 @@ */ #include - +#include +#include #include #include #include @@ -516,33 +517,37 @@ static char *cmtp_procinfo(struct capi_ctr *ctrl) return "CAPI Message Transport Protocol"; } -static int cmtp_ctr_read_proc(char *page, char **start, off_t off, int count, int *eof, struct capi_ctr *ctrl) +static int cmtp_proc_show(struct seq_file *m, void *v) { + struct capi_ctr *ctrl = m->private; struct cmtp_session *session = ctrl->driverdata; struct cmtp_application *app; struct list_head *p, *n; - int len = 0; - len += sprintf(page + len, "%s\n\n", cmtp_procinfo(ctrl)); - len += sprintf(page + len, "addr %s\n", session->name); - len += sprintf(page + len, "ctrl %d\n", session->num); + seq_printf(m, "%s\n\n", cmtp_procinfo(ctrl)); + seq_printf(m, "addr %s\n", session->name); + seq_printf(m, "ctrl %d\n", session->num); list_for_each_safe(p, n, &session->applications) { app = list_entry(p, struct cmtp_application, list); - len += sprintf(page + len, "appl %d -> %d\n", app->appl, app->mapping); + seq_printf(m, "appl %d -> %d\n", app->appl, app->mapping); } - if (off + count >= len) - *eof = 1; - - if (len < off) - return 0; - - *start = page + off; + return 0; +} - return ((count < len - off) ? count : len - off); +static int cmtp_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cmtp_proc_show, PDE(inode)->data); } +static const struct file_operations cmtp_proc_fops = { + .owner = THIS_MODULE, + .open = cmtp_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; int cmtp_attach_device(struct cmtp_session *session) { @@ -582,7 +587,7 @@ int cmtp_attach_device(struct cmtp_session *session) session->ctrl.send_message = cmtp_send_message; session->ctrl.procinfo = cmtp_procinfo; - session->ctrl.ctr_read_proc = cmtp_ctr_read_proc; + session->ctrl.proc_fops = &cmtp_proc_fops; if (attach_capi_ctr(&session->ctrl) < 0) { BT_ERR("Can't attach new controller"); -- cgit v1.2.3-70-g09d2 From 9d173fc5dfa8c1b4578b331ac7ff3ce8af27006e Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Thu, 14 Jan 2010 13:09:14 +0200 Subject: mac80211: fix mac80211.h documentation warnings There were some warnings about missing documentation and a missing reference. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- Documentation/DocBook/mac80211.tmpl | 2 +- include/net/mac80211.h | 7 ++++++- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/DocBook/mac80211.tmpl b/Documentation/DocBook/mac80211.tmpl index f3f37f141db..971d1c0c83e 100644 --- a/Documentation/DocBook/mac80211.tmpl +++ b/Documentation/DocBook/mac80211.tmpl @@ -144,7 +144,7 @@ usage should require reading the full document. this though and the recommendation to allow only a single interface in STA mode at first! -!Finclude/net/mac80211.h ieee80211_if_init_conf +!Finclude/net/mac80211.h ieee80211_vif diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f313a3cbabd..bbfa4750092 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -107,6 +107,7 @@ enum ieee80211_max_queues { * 2^n-1 in the range 1..32767] * @cw_max: maximum contention window [like @cw_min] * @txop: maximum burst time in units of 32 usecs, 0 meaning disabled + * @uapsd: is U-APSD mode enabled for the queue */ struct ieee80211_tx_queue_params { u16 txop; @@ -608,7 +609,11 @@ enum ieee80211_conf_changed { /** * enum ieee80211_smps_mode - spatial multiplexing power save mode * - * @ + * @IEEE80211_SMPS_AUTOMATIC: automatic + * @IEEE80211_SMPS_OFF: off + * @IEEE80211_SMPS_STATIC: static + * @IEEE80211_SMPS_DYNAMIC: dynamic + * @IEEE80211_SMPS_NUM_MODES: internal, don't use */ enum ieee80211_smps_mode { IEEE80211_SMPS_AUTOMATIC, -- cgit v1.2.3-70-g09d2 From c99445b14054e0c4ed4715df1dad1fc608cbab46 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Thu, 14 Jan 2010 13:09:21 +0200 Subject: mac80211: improve powersave documentation There has been some confusion how drivers should implement powersave support. Improve the documentation a bit to make it more clear what drivers need to do. Also mention about U-APSD. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/net/mac80211.h | 75 ++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 54 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index bbfa4750092..c90047de442 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -569,7 +569,13 @@ struct ieee80211_rx_status { * @IEEE80211_CONF_MONITOR: there's a monitor interface present -- use this * to determine for example whether to calculate timestamps for packets * or not, do not use instead of filter flags! - * @IEEE80211_CONF_PS: Enable 802.11 power save mode (managed mode only) + * @IEEE80211_CONF_PS: Enable 802.11 power save mode (managed mode only). + * This is the power save mode defined by IEEE 802.11-2007 section 11.2, + * meaning that the hardware still wakes up for beacons, is able to + * transmit frames and receive the possible acknowledgment frames. + * Not to be confused with hardware specific wakeup/sleep states, + * driver is responsible for that. See the section "Powersave support" + * for more. * @IEEE80211_CONF_IDLE: The device is running, but idle; if the flag is set * the driver should be prepared to handle configuration requests but * may turn the device off as much as possible. Typically, this flag will @@ -1138,18 +1144,24 @@ ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw, * * mac80211 has support for various powersave implementations. * - * First, it can support hardware that handles all powersaving by - * itself, such hardware should simply set the %IEEE80211_HW_SUPPORTS_PS - * hardware flag. In that case, it will be told about the desired - * powersave mode depending on the association status, and the driver - * must take care of sending nullfunc frames when necessary, i.e. when - * entering and leaving powersave mode. The driver is required to look at - * the AID in beacons and signal to the AP that it woke up when it finds - * traffic directed to it. This mode supports dynamic PS by simply - * enabling/disabling PS. - * - * Additionally, such hardware may set the %IEEE80211_HW_SUPPORTS_DYNAMIC_PS - * flag to indicate that it can support dynamic PS mode itself (see below). + * First, it can support hardware that handles all powersaving by itself, + * such hardware should simply set the %IEEE80211_HW_SUPPORTS_PS hardware + * flag. In that case, it will be told about the desired powersave mode + * with the %IEEE80211_CONF_PS flag depending on the association status. + * The hardware must take care of sending nullfunc frames when necessary, + * i.e. when entering and leaving powersave mode. The hardware is required + * to look at the AID in beacons and signal to the AP that it woke up when + * it finds traffic directed to it. + * + * %IEEE80211_CONF_PS flag enabled means that the powersave mode defined in + * IEEE 802.11-2007 section 11.2 is enabled. This is not to be confused + * with hardware wakeup and sleep states. Driver is responsible for waking + * up the hardware before issueing commands to the hardware and putting it + * back to sleep at approriate times. + * + * When PS is enabled, hardware needs to wakeup for beacons and receive the + * buffered multicast/broadcast frames after the beacon. Also it must be + * possible to send frames and receive the acknowledment frame. * * Other hardware designs cannot send nullfunc frames by themselves and also * need software support for parsing the TIM bitmap. This is also supported @@ -1157,14 +1169,35 @@ ieee80211_get_alt_retry_rate(const struct ieee80211_hw *hw, * %IEEE80211_HW_PS_NULLFUNC_STACK flags. The hardware is of course still * required to pass up beacons. The hardware is still required to handle * waking up for multicast traffic; if it cannot the driver must handle that - * as best as it can, mac80211 is too slow. - * - * Dynamic powersave mode is an extension to normal powersave mode in which - * the hardware stays awake for a user-specified period of time after sending - * a frame so that reply frames need not be buffered and therefore delayed - * to the next wakeup. This can either be supported by hardware, in which case - * the driver needs to look at the @dynamic_ps_timeout hardware configuration - * value, or by the stack if all nullfunc handling is in the stack. + * as best as it can, mac80211 is too slow to do that. + * + * Dynamic powersave is an extension to normal powersave in which the + * hardware stays awake for a user-specified period of time after sending a + * frame so that reply frames need not be buffered and therefore delayed to + * the next wakeup. It's compromise of getting good enough latency when + * there's data traffic and still saving significantly power in idle + * periods. + * + * Dynamic powersave is supported by simply mac80211 enabling and disabling + * PS based on traffic. Driver needs to only set %IEEE80211_HW_SUPPORTS_PS + * flag and mac80211 will handle everything automatically. Additionally, + * hardware having support for the dynamic PS feature may set the + * %IEEE80211_HW_SUPPORTS_DYNAMIC_PS flag to indicate that it can support + * dynamic PS mode itself. The driver needs to look at the + * @dynamic_ps_timeout hardware configuration value and use it that value + * whenever %IEEE80211_CONF_PS is set. In this case mac80211 will disable + * dynamic PS feature in stack and will just keep %IEEE80211_CONF_PS + * enabled whenever user has enabled powersave. + * + * Driver informs U-APSD client support by enabling + * %IEEE80211_HW_SUPPORTS_UAPSD flag. The mode is configured through the + * uapsd paramater in conf_tx() operation. Hardware needs to send the QoS + * Nullfunc frames and stay awake until the service period has ended. To + * utilize U-APSD, dynamic powersave is disabled for voip AC and all frames + * from that AC are transmitted with powersave enabled. + * + * Note: U-APSD client mode is not yet supported with + * %IEEE80211_HW_PS_NULLFUNC_STACK. */ /** -- cgit v1.2.3-70-g09d2 From ad72c347e56bf3a0231b9d686e17764157d2961c Mon Sep 17 00:00:00 2001 From: Christian Pellegrin Date: Thu, 14 Jan 2010 07:08:34 +0000 Subject: can: Proper ctrlmode handling for CAN devices This patch adds error checking of ctrlmode values for CAN devices. As an example all availabe bits are implemented in the mcp251x driver. Signed-off-by: Christian Pellegrin Acked-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/at91_can.c | 1 + drivers/net/can/bfin_can.c | 1 + drivers/net/can/dev.c | 2 ++ drivers/net/can/mcp251x.c | 11 ++++++++++- drivers/net/can/mscan/mscan.c | 1 + drivers/net/can/sja1000/sja1000.c | 1 + drivers/net/can/ti_hecc.c | 1 + drivers/net/can/usb/ems_usb.c | 1 + include/linux/can/dev.h | 1 + 9 files changed, 19 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c index f7287497ba6..a2f29a38798 100644 --- a/drivers/net/can/at91_can.c +++ b/drivers/net/can/at91_can.c @@ -1073,6 +1073,7 @@ static int __init at91_can_probe(struct platform_device *pdev) priv->can.bittiming_const = &at91_bittiming_const; priv->can.do_set_bittiming = at91_set_bittiming; priv->can.do_set_mode = at91_set_mode; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; priv->reg_base = addr; priv->dev = dev; priv->clk = clk; diff --git a/drivers/net/can/bfin_can.c b/drivers/net/can/bfin_can.c index 7e1926e79e9..bf7f9ba2d90 100644 --- a/drivers/net/can/bfin_can.c +++ b/drivers/net/can/bfin_can.c @@ -603,6 +603,7 @@ struct net_device *alloc_bfin_candev(void) priv->can.bittiming_const = &bfin_can_bittiming_const; priv->can.do_set_bittiming = bfin_can_set_bittiming; priv->can.do_set_mode = bfin_can_set_mode; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; return dev; } diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index c1bb29f0322..f08f1202ff0 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -592,6 +592,8 @@ static int can_changelink(struct net_device *dev, if (dev->flags & IFF_UP) return -EBUSY; cm = nla_data(data[IFLA_CAN_CTRLMODE]); + if (cm->flags & ~priv->ctrlmode_supported) + return -EOPNOTSUPP; priv->ctrlmode &= ~cm->mask; priv->ctrlmode |= cm->flags; } diff --git a/drivers/net/can/mcp251x.c b/drivers/net/can/mcp251x.c index afa2fa45fed..bbe186b5a0e 100644 --- a/drivers/net/can/mcp251x.c +++ b/drivers/net/can/mcp251x.c @@ -539,9 +539,14 @@ static void mcp251x_set_normal_mode(struct spi_device *spi) if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK) { /* Put device into loopback mode */ mcp251x_write_reg(spi, CANCTRL, CANCTRL_REQOP_LOOPBACK); + } else if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) { + /* Put device into listen-only mode */ + mcp251x_write_reg(spi, CANCTRL, CANCTRL_REQOP_LISTEN_ONLY); } else { /* Put device into normal mode */ - mcp251x_write_reg(spi, CANCTRL, CANCTRL_REQOP_NORMAL); + mcp251x_write_reg(spi, CANCTRL, CANCTRL_REQOP_NORMAL | + (priv->can.ctrlmode & CAN_CTRLMODE_ONE_SHOT ? + CANCTRL_OSM : 0)); /* Wait for the device to enter normal mode */ timeout = jiffies + HZ; @@ -948,6 +953,10 @@ static int __devinit mcp251x_can_probe(struct spi_device *spi) priv->can.bittiming_const = &mcp251x_bittiming_const; priv->can.do_set_mode = mcp251x_do_set_mode; priv->can.clock.freq = pdata->oscillator_frequency / 2; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES | + CAN_CTRLMODE_LOOPBACK | CAN_CTRLMODE_LISTENONLY; + if (pdata->model == CAN_MCP251X_MCP2515) + priv->can.ctrlmode_supported |= CAN_CTRLMODE_ONE_SHOT; priv->net = net; dev_set_drvdata(&spi->dev, priv); diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 40827c128b6..6b7dd578d41 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -686,6 +686,7 @@ struct net_device *alloc_mscandev(void) priv->can.bittiming_const = &mscan_bittiming_const; priv->can.do_set_bittiming = mscan_do_set_bittiming; priv->can.do_set_mode = mscan_do_set_mode; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; for (i = 0; i < TX_QUEUE_SIZE; i++) { priv->tx_queue[i].id = i; diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index 345304d779b..ace103a4483 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -567,6 +567,7 @@ struct net_device *alloc_sja1000dev(int sizeof_priv) priv->can.bittiming_const = &sja1000_bittiming_const; priv->can.do_set_bittiming = sja1000_set_bittiming; priv->can.do_set_mode = sja1000_set_mode; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; if (sizeof_priv) priv->priv = (void *)priv + sizeof(struct sja1000_priv); diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index 7d370e32a7a..8332e242b0b 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -909,6 +909,7 @@ static int ti_hecc_probe(struct platform_device *pdev) priv->can.bittiming_const = &ti_hecc_bittiming_const; priv->can.do_set_mode = ti_hecc_do_set_mode; priv->can.do_get_state = ti_hecc_get_state; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; ndev->irq = irq->start; ndev->flags |= IFF_ECHO; diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index ddb17e25665..bfab283ba9b 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -1022,6 +1022,7 @@ static int ems_usb_probe(struct usb_interface *intf, dev->can.bittiming_const = &ems_usb_bittiming_const; dev->can.do_set_bittiming = ems_usb_set_bittiming; dev->can.do_set_mode = ems_usb_set_mode; + dev->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; netdev->flags |= IFF_ECHO; /* we support local echo */ diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 7e7c98a3e90..c8c660a79f9 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -38,6 +38,7 @@ struct can_priv { enum can_state state; u32 ctrlmode; + u32 ctrlmode_supported; int restart_ms; struct timer_list restart_timer; -- cgit v1.2.3-70-g09d2 From 05c2828c72c4eabf62376adfe27bd24797621f62 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 14 Jan 2010 06:17:09 +0000 Subject: tun: export underlying socket Tun device looks similar to a packet socket in that both pass complete frames from/to userspace. This patch fills in enough fields in the socket underlying tun driver to support sendmsg/recvmsg operations, and message flags MSG_TRUNC and MSG_DONTWAIT, and exports access to this socket to modules. Regular read/write behaviour is unchanged. This way, code using raw sockets to inject packets into a physical device, can support injecting packets into host network stack almost without modification. First user of this interface will be vhost virtualization accelerator. Signed-off-by: Michael S. Tsirkin Acked-by: Herbert Xu Acked-by: David S. Miller Signed-off-by: David S. Miller --- drivers/net/tun.c | 101 +++++++++++++++++++++++++++++++++++++++---------- include/linux/if_tun.h | 14 +++++++ 2 files changed, 96 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 2834a01bae2..5adb3d15055 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -144,6 +144,7 @@ static int tun_attach(struct tun_struct *tun, struct file *file) err = 0; tfile->tun = tun; tun->tfile = tfile; + tun->socket.file = file; dev_hold(tun->dev); sock_hold(tun->socket.sk); atomic_inc(&tfile->count); @@ -158,6 +159,7 @@ static void __tun_detach(struct tun_struct *tun) /* Detach from net device */ netif_tx_lock_bh(tun->dev); tun->tfile = NULL; + tun->socket.file = NULL; netif_tx_unlock_bh(tun->dev); /* Drop read queue */ @@ -387,7 +389,8 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev) /* Notify and wake up reader process */ if (tun->flags & TUN_FASYNC) kill_fasync(&tun->fasync, SIGIO, POLL_IN); - wake_up_interruptible(&tun->socket.wait); + wake_up_interruptible_poll(&tun->socket.wait, POLLIN | + POLLRDNORM | POLLRDBAND); return NETDEV_TX_OK; drop: @@ -743,7 +746,7 @@ static __inline__ ssize_t tun_put_user(struct tun_struct *tun, len = min_t(int, skb->len, len); skb_copy_datagram_const_iovec(skb, 0, iv, total, len); - total += len; + total += skb->len; tun->dev->stats.tx_packets++; tun->dev->stats.tx_bytes += len; @@ -751,34 +754,23 @@ static __inline__ ssize_t tun_put_user(struct tun_struct *tun, return total; } -static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv, - unsigned long count, loff_t pos) +static ssize_t tun_do_read(struct tun_struct *tun, + struct kiocb *iocb, const struct iovec *iv, + ssize_t len, int noblock) { - struct file *file = iocb->ki_filp; - struct tun_file *tfile = file->private_data; - struct tun_struct *tun = __tun_get(tfile); DECLARE_WAITQUEUE(wait, current); struct sk_buff *skb; - ssize_t len, ret = 0; - - if (!tun) - return -EBADFD; + ssize_t ret = 0; DBG(KERN_INFO "%s: tun_chr_read\n", tun->dev->name); - len = iov_length(iv, count); - if (len < 0) { - ret = -EINVAL; - goto out; - } - add_wait_queue(&tun->socket.wait, &wait); while (len) { current->state = TASK_INTERRUPTIBLE; /* Read frames from the queue */ if (!(skb=skb_dequeue(&tun->socket.sk->sk_receive_queue))) { - if (file->f_flags & O_NONBLOCK) { + if (noblock) { ret = -EAGAIN; break; } @@ -805,6 +797,27 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv, current->state = TASK_RUNNING; remove_wait_queue(&tun->socket.wait, &wait); + return ret; +} + +static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv, + unsigned long count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct tun_file *tfile = file->private_data; + struct tun_struct *tun = __tun_get(tfile); + ssize_t len, ret; + + if (!tun) + return -EBADFD; + len = iov_length(iv, count); + if (len < 0) { + ret = -EINVAL; + goto out; + } + + ret = tun_do_read(tun, iocb, iv, len, file->f_flags & O_NONBLOCK); + ret = min_t(ssize_t, ret, len); out: tun_put(tun); return ret; @@ -847,7 +860,8 @@ static void tun_sock_write_space(struct sock *sk) return; if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) - wake_up_interruptible_sync(sk->sk_sleep); + wake_up_interruptible_sync_poll(sk->sk_sleep, POLLOUT | + POLLWRNORM | POLLWRBAND); tun = tun_sk(sk)->tun; kill_fasync(&tun->fasync, SIGIO, POLL_OUT); @@ -858,6 +872,37 @@ static void tun_sock_destruct(struct sock *sk) free_netdev(tun_sk(sk)->tun->dev); } +static int tun_sendmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct tun_struct *tun = container_of(sock, struct tun_struct, socket); + return tun_get_user(tun, m->msg_iov, total_len, + m->msg_flags & MSG_DONTWAIT); +} + +static int tun_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len, + int flags) +{ + struct tun_struct *tun = container_of(sock, struct tun_struct, socket); + int ret; + if (flags & ~(MSG_DONTWAIT|MSG_TRUNC)) + return -EINVAL; + ret = tun_do_read(tun, iocb, m->msg_iov, total_len, + flags & MSG_DONTWAIT); + if (ret > total_len) { + m->msg_flags |= MSG_TRUNC; + ret = flags & MSG_TRUNC ? ret : total_len; + } + return ret; +} + +/* Ops structure to mimic raw sockets with tun */ +static const struct proto_ops tun_socket_ops = { + .sendmsg = tun_sendmsg, + .recvmsg = tun_recvmsg, +}; + static struct proto tun_proto = { .name = "tun", .owner = THIS_MODULE, @@ -986,6 +1031,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) goto err_free_dev; init_waitqueue_head(&tun->socket.wait); + tun->socket.ops = &tun_socket_ops; sock_init_data(&tun->socket, sk); sk->sk_write_space = tun_sock_write_space; sk->sk_sndbuf = INT_MAX; @@ -1525,6 +1571,23 @@ static void tun_cleanup(void) rtnl_link_unregister(&tun_link_ops); } +/* Get an underlying socket object from tun file. Returns error unless file is + * attached to a device. The returned object works like a packet socket, it + * can be used for sock_sendmsg/sock_recvmsg. The caller is responsible for + * holding a reference to the file for as long as the socket is in use. */ +struct socket *tun_get_socket(struct file *file) +{ + struct tun_struct *tun; + if (file->f_op != &tun_fops) + return ERR_PTR(-EINVAL); + tun = tun_get(file); + if (!tun) + return ERR_PTR(-EBADFD); + tun_put(tun); + return &tun->socket; +} +EXPORT_SYMBOL_GPL(tun_get_socket); + module_init(tun_init); module_exit(tun_cleanup); MODULE_DESCRIPTION(DRV_DESCRIPTION); diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h index 3f5fd523b49..404abe00162 100644 --- a/include/linux/if_tun.h +++ b/include/linux/if_tun.h @@ -86,4 +86,18 @@ struct tun_filter { __u8 addr[0][ETH_ALEN]; }; +#ifdef __KERNEL__ +#if defined(CONFIG_TUN) || defined(CONFIG_TUN_MODULE) +struct socket *tun_get_socket(struct file *); +#else +#include +#include +struct file; +struct socket; +static inline struct socket *tun_get_socket(struct file *f) +{ + return ERR_PTR(-EINVAL); +} +#endif /* CONFIG_TUN */ +#endif /* __KERNEL__ */ #endif /* __IF_TUN_H */ -- cgit v1.2.3-70-g09d2 From 3a4d5c94e959359ece6d6b55045c3f046677f55c Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 14 Jan 2010 06:17:27 +0000 Subject: vhost_net: a kernel-level virtio server What it is: vhost net is a character device that can be used to reduce the number of system calls involved in virtio networking. Existing virtio net code is used in the guest without modification. There's similarity with vringfd, with some differences and reduced scope - uses eventfd for signalling - structures can be moved around in memory at any time (good for migration, bug work-arounds in userspace) - write logging is supported (good for migration) - support memory table and not just an offset (needed for kvm) common virtio related code has been put in a separate file vhost.c and can be made into a separate module if/when more backends appear. I used Rusty's lguest.c as the source for developing this part : this supplied me with witty comments I wouldn't be able to write myself. What it is not: vhost net is not a bus, and not a generic new system call. No assumptions are made on how guest performs hypercalls. Userspace hypervisors are supported as well as kvm. How it works: Basically, we connect virtio frontend (configured by userspace) to a backend. The backend could be a network device, or a tap device. Backend is also configured by userspace, including vlan/mac etc. Status: This works for me, and I haven't see any crashes. Compared to userspace, people reported improved latency (as I save up to 4 system calls per packet), as well as better bandwidth and CPU utilization. Features that I plan to look at in the future: - mergeable buffers - zero copy - scalability tuning: figure out the best threading model to use Note on RCU usage (this is also documented in vhost.h, near private_pointer which is the value protected by this variant of RCU): what is happening is that the rcu_dereference() is being used in a workqueue item. The role of rcu_read_lock() is taken on by the start of execution of the workqueue item, of rcu_read_unlock() by the end of execution of the workqueue item, and of synchronize_rcu() by flush_workqueue()/flush_work(). In the future we might need to apply some gcc attribute or sparse annotation to the function passed to INIT_WORK(). Paul's ack below is for this RCU usage. (Includes fixes by Alan Cox , David L Stevens , Chris Wright ) Acked-by: Rusty Russell Acked-by: Arnd Bergmann Acked-by: "Paul E. McKenney" Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- MAINTAINERS | 9 + arch/ia64/kvm/Kconfig | 1 + arch/powerpc/kvm/Kconfig | 1 + arch/s390/kvm/Kconfig | 1 + arch/x86/kvm/Kconfig | 1 + drivers/Makefile | 1 + drivers/vhost/Kconfig | 11 + drivers/vhost/Makefile | 2 + drivers/vhost/net.c | 661 ++++++++++++++++++++++++++ drivers/vhost/vhost.c | 1098 ++++++++++++++++++++++++++++++++++++++++++++ drivers/vhost/vhost.h | 161 +++++++ include/linux/Kbuild | 1 + include/linux/miscdevice.h | 1 + include/linux/vhost.h | 130 ++++++ 14 files changed, 2079 insertions(+) create mode 100644 drivers/vhost/Kconfig create mode 100644 drivers/vhost/Makefile create mode 100644 drivers/vhost/net.c create mode 100644 drivers/vhost/vhost.c create mode 100644 drivers/vhost/vhost.h create mode 100644 include/linux/vhost.h (limited to 'include') diff --git a/MAINTAINERS b/MAINTAINERS index 745643b8c34..337dffbe9a4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5803,6 +5803,15 @@ S: Maintained F: Documentation/filesystems/vfat.txt F: fs/fat/ +VIRTIO HOST (VHOST) +M: "Michael S. Tsirkin" +L: kvm@vger.kernel.org +L: virtualization@lists.osdl.org +L: netdev@vger.kernel.org +S: Maintained +F: drivers/vhost/ +F: include/linux/vhost.h + VIA RHINE NETWORK DRIVER M: Roger Luethi S: Maintained diff --git a/arch/ia64/kvm/Kconfig b/arch/ia64/kvm/Kconfig index ef3e7be29ca..01c75797119 100644 --- a/arch/ia64/kvm/Kconfig +++ b/arch/ia64/kvm/Kconfig @@ -47,6 +47,7 @@ config KVM_INTEL Provides support for KVM on Itanium 2 processors equipped with the VT extensions. +source drivers/vhost/Kconfig source drivers/virtio/Kconfig endif # VIRTUALIZATION diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig index 07703f72330..e28841fbfb8 100644 --- a/arch/powerpc/kvm/Kconfig +++ b/arch/powerpc/kvm/Kconfig @@ -75,6 +75,7 @@ config KVM_E500 If unsure, say N. +source drivers/vhost/Kconfig source drivers/virtio/Kconfig endif # VIRTUALIZATION diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig index 6ee55ae84ce..a7251580891 100644 --- a/arch/s390/kvm/Kconfig +++ b/arch/s390/kvm/Kconfig @@ -35,6 +35,7 @@ config KVM # OK, it's a little counter-intuitive to do this, but it puts it neatly under # the virtualization menu. +source drivers/vhost/Kconfig source drivers/virtio/Kconfig endif # VIRTUALIZATION diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index 4cd49833246..3c4d0109ad2 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -65,6 +65,7 @@ config KVM_AMD # OK, it's a little counter-intuitive to do this, but it puts it neatly under # the virtualization menu. +source drivers/vhost/Kconfig source drivers/lguest/Kconfig source drivers/virtio/Kconfig diff --git a/drivers/Makefile b/drivers/Makefile index 6ee53c7a57a..81e36596b1e 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -106,6 +106,7 @@ obj-$(CONFIG_HID) += hid/ obj-$(CONFIG_PPC_PS3) += ps3/ obj-$(CONFIG_OF) += of/ obj-$(CONFIG_SSB) += ssb/ +obj-$(CONFIG_VHOST_NET) += vhost/ obj-$(CONFIG_VIRTIO) += virtio/ obj-$(CONFIG_VLYNQ) += vlynq/ obj-$(CONFIG_STAGING) += staging/ diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig new file mode 100644 index 00000000000..9f409f447ae --- /dev/null +++ b/drivers/vhost/Kconfig @@ -0,0 +1,11 @@ +config VHOST_NET + tristate "Host kernel accelerator for virtio net (EXPERIMENTAL)" + depends on NET && EVENTFD && EXPERIMENTAL + ---help--- + This kernel module can be loaded in host kernel to accelerate + guest networking with virtio_net. Not to be confused with virtio_net + module itself which needs to be loaded in guest kernel. + + To compile this driver as a module, choose M here: the module will + be called vhost_net. + diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile new file mode 100644 index 00000000000..72dd02050bb --- /dev/null +++ b/drivers/vhost/Makefile @@ -0,0 +1,2 @@ +obj-$(CONFIG_VHOST_NET) += vhost_net.o +vhost_net-y := vhost.o net.o diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c new file mode 100644 index 00000000000..4c8928319e1 --- /dev/null +++ b/drivers/vhost/net.c @@ -0,0 +1,661 @@ +/* Copyright (C) 2009 Red Hat, Inc. + * Author: Michael S. Tsirkin + * + * This work is licensed under the terms of the GNU GPL, version 2. + * + * virtio-net server in host kernel. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#include + +#include "vhost.h" + +/* Max number of bytes transferred before requeueing the job. + * Using this limit prevents one virtqueue from starving others. */ +#define VHOST_NET_WEIGHT 0x80000 + +enum { + VHOST_NET_VQ_RX = 0, + VHOST_NET_VQ_TX = 1, + VHOST_NET_VQ_MAX = 2, +}; + +enum vhost_net_poll_state { + VHOST_NET_POLL_DISABLED = 0, + VHOST_NET_POLL_STARTED = 1, + VHOST_NET_POLL_STOPPED = 2, +}; + +struct vhost_net { + struct vhost_dev dev; + struct vhost_virtqueue vqs[VHOST_NET_VQ_MAX]; + struct vhost_poll poll[VHOST_NET_VQ_MAX]; + /* Tells us whether we are polling a socket for TX. + * We only do this when socket buffer fills up. + * Protected by tx vq lock. */ + enum vhost_net_poll_state tx_poll_state; +}; + +/* Pop first len bytes from iovec. Return number of segments used. */ +static int move_iovec_hdr(struct iovec *from, struct iovec *to, + size_t len, int iov_count) +{ + int seg = 0; + size_t size; + while (len && seg < iov_count) { + size = min(from->iov_len, len); + to->iov_base = from->iov_base; + to->iov_len = size; + from->iov_len -= size; + from->iov_base += size; + len -= size; + ++from; + ++to; + ++seg; + } + return seg; +} + +/* Caller must have TX VQ lock */ +static void tx_poll_stop(struct vhost_net *net) +{ + if (likely(net->tx_poll_state != VHOST_NET_POLL_STARTED)) + return; + vhost_poll_stop(net->poll + VHOST_NET_VQ_TX); + net->tx_poll_state = VHOST_NET_POLL_STOPPED; +} + +/* Caller must have TX VQ lock */ +static void tx_poll_start(struct vhost_net *net, struct socket *sock) +{ + if (unlikely(net->tx_poll_state != VHOST_NET_POLL_STOPPED)) + return; + vhost_poll_start(net->poll + VHOST_NET_VQ_TX, sock->file); + net->tx_poll_state = VHOST_NET_POLL_STARTED; +} + +/* Expects to be always run from workqueue - which acts as + * read-size critical section for our kind of RCU. */ +static void handle_tx(struct vhost_net *net) +{ + struct vhost_virtqueue *vq = &net->dev.vqs[VHOST_NET_VQ_TX]; + unsigned head, out, in, s; + struct msghdr msg = { + .msg_name = NULL, + .msg_namelen = 0, + .msg_control = NULL, + .msg_controllen = 0, + .msg_iov = vq->iov, + .msg_flags = MSG_DONTWAIT, + }; + size_t len, total_len = 0; + int err, wmem; + size_t hdr_size; + struct socket *sock = rcu_dereference(vq->private_data); + if (!sock) + return; + + wmem = atomic_read(&sock->sk->sk_wmem_alloc); + if (wmem >= sock->sk->sk_sndbuf) + return; + + use_mm(net->dev.mm); + mutex_lock(&vq->mutex); + vhost_disable_notify(vq); + + if (wmem < sock->sk->sk_sndbuf * 2) + tx_poll_stop(net); + hdr_size = vq->hdr_size; + + for (;;) { + head = vhost_get_vq_desc(&net->dev, vq, vq->iov, + ARRAY_SIZE(vq->iov), + &out, &in, + NULL, NULL); + /* Nothing new? Wait for eventfd to tell us they refilled. */ + if (head == vq->num) { + wmem = atomic_read(&sock->sk->sk_wmem_alloc); + if (wmem >= sock->sk->sk_sndbuf * 3 / 4) { + tx_poll_start(net, sock); + set_bit(SOCK_ASYNC_NOSPACE, &sock->flags); + break; + } + if (unlikely(vhost_enable_notify(vq))) { + vhost_disable_notify(vq); + continue; + } + break; + } + if (in) { + vq_err(vq, "Unexpected descriptor format for TX: " + "out %d, int %d\n", out, in); + break; + } + /* Skip header. TODO: support TSO. */ + s = move_iovec_hdr(vq->iov, vq->hdr, hdr_size, out); + msg.msg_iovlen = out; + len = iov_length(vq->iov, out); + /* Sanity check */ + if (!len) { + vq_err(vq, "Unexpected header len for TX: " + "%zd expected %zd\n", + iov_length(vq->hdr, s), hdr_size); + break; + } + /* TODO: Check specific error and bomb out unless ENOBUFS? */ + err = sock->ops->sendmsg(NULL, sock, &msg, len); + if (unlikely(err < 0)) { + vhost_discard_vq_desc(vq); + tx_poll_start(net, sock); + break; + } + if (err != len) + pr_err("Truncated TX packet: " + " len %d != %zd\n", err, len); + vhost_add_used_and_signal(&net->dev, vq, head, 0); + total_len += len; + if (unlikely(total_len >= VHOST_NET_WEIGHT)) { + vhost_poll_queue(&vq->poll); + break; + } + } + + mutex_unlock(&vq->mutex); + unuse_mm(net->dev.mm); +} + +/* Expects to be always run from workqueue - which acts as + * read-size critical section for our kind of RCU. */ +static void handle_rx(struct vhost_net *net) +{ + struct vhost_virtqueue *vq = &net->dev.vqs[VHOST_NET_VQ_RX]; + unsigned head, out, in, log, s; + struct vhost_log *vq_log; + struct msghdr msg = { + .msg_name = NULL, + .msg_namelen = 0, + .msg_control = NULL, /* FIXME: get and handle RX aux data. */ + .msg_controllen = 0, + .msg_iov = vq->iov, + .msg_flags = MSG_DONTWAIT, + }; + + struct virtio_net_hdr hdr = { + .flags = 0, + .gso_type = VIRTIO_NET_HDR_GSO_NONE + }; + + size_t len, total_len = 0; + int err; + size_t hdr_size; + struct socket *sock = rcu_dereference(vq->private_data); + if (!sock || skb_queue_empty(&sock->sk->sk_receive_queue)) + return; + + use_mm(net->dev.mm); + mutex_lock(&vq->mutex); + vhost_disable_notify(vq); + hdr_size = vq->hdr_size; + + vq_log = unlikely(vhost_has_feature(&net->dev, VHOST_F_LOG_ALL)) ? + vq->log : NULL; + + for (;;) { + head = vhost_get_vq_desc(&net->dev, vq, vq->iov, + ARRAY_SIZE(vq->iov), + &out, &in, + vq_log, &log); + /* OK, now we need to know about added descriptors. */ + if (head == vq->num) { + if (unlikely(vhost_enable_notify(vq))) { + /* They have slipped one in as we were + * doing that: check again. */ + vhost_disable_notify(vq); + continue; + } + /* Nothing new? Wait for eventfd to tell us + * they refilled. */ + break; + } + /* We don't need to be notified again. */ + if (out) { + vq_err(vq, "Unexpected descriptor format for RX: " + "out %d, int %d\n", + out, in); + break; + } + /* Skip header. TODO: support TSO/mergeable rx buffers. */ + s = move_iovec_hdr(vq->iov, vq->hdr, hdr_size, in); + msg.msg_iovlen = in; + len = iov_length(vq->iov, in); + /* Sanity check */ + if (!len) { + vq_err(vq, "Unexpected header len for RX: " + "%zd expected %zd\n", + iov_length(vq->hdr, s), hdr_size); + break; + } + err = sock->ops->recvmsg(NULL, sock, &msg, + len, MSG_DONTWAIT | MSG_TRUNC); + /* TODO: Check specific error and bomb out unless EAGAIN? */ + if (err < 0) { + vhost_discard_vq_desc(vq); + break; + } + /* TODO: Should check and handle checksum. */ + if (err > len) { + pr_err("Discarded truncated rx packet: " + " len %d > %zd\n", err, len); + vhost_discard_vq_desc(vq); + continue; + } + len = err; + err = memcpy_toiovec(vq->hdr, (unsigned char *)&hdr, hdr_size); + if (err) { + vq_err(vq, "Unable to write vnet_hdr at addr %p: %d\n", + vq->iov->iov_base, err); + break; + } + len += hdr_size; + vhost_add_used_and_signal(&net->dev, vq, head, len); + if (unlikely(vq_log)) + vhost_log_write(vq, vq_log, log, len); + total_len += len; + if (unlikely(total_len >= VHOST_NET_WEIGHT)) { + vhost_poll_queue(&vq->poll); + break; + } + } + + mutex_unlock(&vq->mutex); + unuse_mm(net->dev.mm); +} + +static void handle_tx_kick(struct work_struct *work) +{ + struct vhost_virtqueue *vq; + struct vhost_net *net; + vq = container_of(work, struct vhost_virtqueue, poll.work); + net = container_of(vq->dev, struct vhost_net, dev); + handle_tx(net); +} + +static void handle_rx_kick(struct work_struct *work) +{ + struct vhost_virtqueue *vq; + struct vhost_net *net; + vq = container_of(work, struct vhost_virtqueue, poll.work); + net = container_of(vq->dev, struct vhost_net, dev); + handle_rx(net); +} + +static void handle_tx_net(struct work_struct *work) +{ + struct vhost_net *net; + net = container_of(work, struct vhost_net, poll[VHOST_NET_VQ_TX].work); + handle_tx(net); +} + +static void handle_rx_net(struct work_struct *work) +{ + struct vhost_net *net; + net = container_of(work, struct vhost_net, poll[VHOST_NET_VQ_RX].work); + handle_rx(net); +} + +static int vhost_net_open(struct inode *inode, struct file *f) +{ + struct vhost_net *n = kmalloc(sizeof *n, GFP_KERNEL); + int r; + if (!n) + return -ENOMEM; + n->vqs[VHOST_NET_VQ_TX].handle_kick = handle_tx_kick; + n->vqs[VHOST_NET_VQ_RX].handle_kick = handle_rx_kick; + r = vhost_dev_init(&n->dev, n->vqs, VHOST_NET_VQ_MAX); + if (r < 0) { + kfree(n); + return r; + } + + vhost_poll_init(n->poll + VHOST_NET_VQ_TX, handle_tx_net, POLLOUT); + vhost_poll_init(n->poll + VHOST_NET_VQ_RX, handle_rx_net, POLLIN); + n->tx_poll_state = VHOST_NET_POLL_DISABLED; + + f->private_data = n; + + return 0; +} + +static void vhost_net_disable_vq(struct vhost_net *n, + struct vhost_virtqueue *vq) +{ + if (!vq->private_data) + return; + if (vq == n->vqs + VHOST_NET_VQ_TX) { + tx_poll_stop(n); + n->tx_poll_state = VHOST_NET_POLL_DISABLED; + } else + vhost_poll_stop(n->poll + VHOST_NET_VQ_RX); +} + +static void vhost_net_enable_vq(struct vhost_net *n, + struct vhost_virtqueue *vq) +{ + struct socket *sock = vq->private_data; + if (!sock) + return; + if (vq == n->vqs + VHOST_NET_VQ_TX) { + n->tx_poll_state = VHOST_NET_POLL_STOPPED; + tx_poll_start(n, sock); + } else + vhost_poll_start(n->poll + VHOST_NET_VQ_RX, sock->file); +} + +static struct socket *vhost_net_stop_vq(struct vhost_net *n, + struct vhost_virtqueue *vq) +{ + struct socket *sock; + + mutex_lock(&vq->mutex); + sock = vq->private_data; + vhost_net_disable_vq(n, vq); + rcu_assign_pointer(vq->private_data, NULL); + mutex_unlock(&vq->mutex); + return sock; +} + +static void vhost_net_stop(struct vhost_net *n, struct socket **tx_sock, + struct socket **rx_sock) +{ + *tx_sock = vhost_net_stop_vq(n, n->vqs + VHOST_NET_VQ_TX); + *rx_sock = vhost_net_stop_vq(n, n->vqs + VHOST_NET_VQ_RX); +} + +static void vhost_net_flush_vq(struct vhost_net *n, int index) +{ + vhost_poll_flush(n->poll + index); + vhost_poll_flush(&n->dev.vqs[index].poll); +} + +static void vhost_net_flush(struct vhost_net *n) +{ + vhost_net_flush_vq(n, VHOST_NET_VQ_TX); + vhost_net_flush_vq(n, VHOST_NET_VQ_RX); +} + +static int vhost_net_release(struct inode *inode, struct file *f) +{ + struct vhost_net *n = f->private_data; + struct socket *tx_sock; + struct socket *rx_sock; + + vhost_net_stop(n, &tx_sock, &rx_sock); + vhost_net_flush(n); + vhost_dev_cleanup(&n->dev); + if (tx_sock) + fput(tx_sock->file); + if (rx_sock) + fput(rx_sock->file); + /* We do an extra flush before freeing memory, + * since jobs can re-queue themselves. */ + vhost_net_flush(n); + kfree(n); + return 0; +} + +static struct socket *get_raw_socket(int fd) +{ + struct { + struct sockaddr_ll sa; + char buf[MAX_ADDR_LEN]; + } uaddr; + int uaddr_len = sizeof uaddr, r; + struct socket *sock = sockfd_lookup(fd, &r); + if (!sock) + return ERR_PTR(-ENOTSOCK); + + /* Parameter checking */ + if (sock->sk->sk_type != SOCK_RAW) { + r = -ESOCKTNOSUPPORT; + goto err; + } + + r = sock->ops->getname(sock, (struct sockaddr *)&uaddr.sa, + &uaddr_len, 0); + if (r) + goto err; + + if (uaddr.sa.sll_family != AF_PACKET) { + r = -EPFNOSUPPORT; + goto err; + } + return sock; +err: + fput(sock->file); + return ERR_PTR(r); +} + +static struct socket *get_tun_socket(int fd) +{ + struct file *file = fget(fd); + struct socket *sock; + if (!file) + return ERR_PTR(-EBADF); + sock = tun_get_socket(file); + if (IS_ERR(sock)) + fput(file); + return sock; +} + +static struct socket *get_socket(int fd) +{ + struct socket *sock; + /* special case to disable backend */ + if (fd == -1) + return NULL; + sock = get_raw_socket(fd); + if (!IS_ERR(sock)) + return sock; + sock = get_tun_socket(fd); + if (!IS_ERR(sock)) + return sock; + return ERR_PTR(-ENOTSOCK); +} + +static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd) +{ + struct socket *sock, *oldsock; + struct vhost_virtqueue *vq; + int r; + + mutex_lock(&n->dev.mutex); + r = vhost_dev_check_owner(&n->dev); + if (r) + goto err; + + if (index >= VHOST_NET_VQ_MAX) { + r = -ENOBUFS; + goto err; + } + vq = n->vqs + index; + mutex_lock(&vq->mutex); + + /* Verify that ring has been setup correctly. */ + if (!vhost_vq_access_ok(vq)) { + r = -EFAULT; + goto err; + } + sock = get_socket(fd); + if (IS_ERR(sock)) { + r = PTR_ERR(sock); + goto err; + } + + /* start polling new socket */ + oldsock = vq->private_data; + if (sock == oldsock) + goto done; + + vhost_net_disable_vq(n, vq); + rcu_assign_pointer(vq->private_data, sock); + vhost_net_enable_vq(n, vq); + mutex_unlock(&vq->mutex); +done: + if (oldsock) { + vhost_net_flush_vq(n, index); + fput(oldsock->file); + } +err: + mutex_unlock(&n->dev.mutex); + return r; +} + +static long vhost_net_reset_owner(struct vhost_net *n) +{ + struct socket *tx_sock = NULL; + struct socket *rx_sock = NULL; + long err; + mutex_lock(&n->dev.mutex); + err = vhost_dev_check_owner(&n->dev); + if (err) + goto done; + vhost_net_stop(n, &tx_sock, &rx_sock); + vhost_net_flush(n); + err = vhost_dev_reset_owner(&n->dev); +done: + mutex_unlock(&n->dev.mutex); + if (tx_sock) + fput(tx_sock->file); + if (rx_sock) + fput(rx_sock->file); + return err; +} + +static int vhost_net_set_features(struct vhost_net *n, u64 features) +{ + size_t hdr_size = features & (1 << VHOST_NET_F_VIRTIO_NET_HDR) ? + sizeof(struct virtio_net_hdr) : 0; + int i; + mutex_lock(&n->dev.mutex); + if ((features & (1 << VHOST_F_LOG_ALL)) && + !vhost_log_access_ok(&n->dev)) { + mutex_unlock(&n->dev.mutex); + return -EFAULT; + } + n->dev.acked_features = features; + smp_wmb(); + for (i = 0; i < VHOST_NET_VQ_MAX; ++i) { + mutex_lock(&n->vqs[i].mutex); + n->vqs[i].hdr_size = hdr_size; + mutex_unlock(&n->vqs[i].mutex); + } + vhost_net_flush(n); + mutex_unlock(&n->dev.mutex); + return 0; +} + +static long vhost_net_ioctl(struct file *f, unsigned int ioctl, + unsigned long arg) +{ + struct vhost_net *n = f->private_data; + void __user *argp = (void __user *)arg; + u64 __user *featurep = argp; + struct vhost_vring_file backend; + u64 features; + int r; + switch (ioctl) { + case VHOST_NET_SET_BACKEND: + r = copy_from_user(&backend, argp, sizeof backend); + if (r < 0) + return r; + return vhost_net_set_backend(n, backend.index, backend.fd); + case VHOST_GET_FEATURES: + features = VHOST_FEATURES; + return copy_to_user(featurep, &features, sizeof features); + case VHOST_SET_FEATURES: + r = copy_from_user(&features, featurep, sizeof features); + if (r < 0) + return r; + if (features & ~VHOST_FEATURES) + return -EOPNOTSUPP; + return vhost_net_set_features(n, features); + case VHOST_RESET_OWNER: + return vhost_net_reset_owner(n); + default: + mutex_lock(&n->dev.mutex); + r = vhost_dev_ioctl(&n->dev, ioctl, arg); + vhost_net_flush(n); + mutex_unlock(&n->dev.mutex); + return r; + } +} + +#ifdef CONFIG_COMPAT +static long vhost_net_compat_ioctl(struct file *f, unsigned int ioctl, + unsigned long arg) +{ + return vhost_net_ioctl(f, ioctl, (unsigned long)compat_ptr(arg)); +} +#endif + +const static struct file_operations vhost_net_fops = { + .owner = THIS_MODULE, + .release = vhost_net_release, + .unlocked_ioctl = vhost_net_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = vhost_net_compat_ioctl, +#endif + .open = vhost_net_open, +}; + +static struct miscdevice vhost_net_misc = { + VHOST_NET_MINOR, + "vhost-net", + &vhost_net_fops, +}; + +int vhost_net_init(void) +{ + int r = vhost_init(); + if (r) + goto err_init; + r = misc_register(&vhost_net_misc); + if (r) + goto err_reg; + return 0; +err_reg: + vhost_cleanup(); +err_init: + return r; + +} +module_init(vhost_net_init); + +void vhost_net_exit(void) +{ + misc_deregister(&vhost_net_misc); + vhost_cleanup(); +} +module_exit(vhost_net_exit); + +MODULE_VERSION("0.0.1"); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Michael S. Tsirkin"); +MODULE_DESCRIPTION("Host kernel accelerator for virtio net"); diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c new file mode 100644 index 00000000000..c8c25dbc585 --- /dev/null +++ b/drivers/vhost/vhost.c @@ -0,0 +1,1098 @@ +/* Copyright (C) 2009 Red Hat, Inc. + * Copyright (C) 2006 Rusty Russell IBM Corporation + * + * Author: Michael S. Tsirkin + * + * Inspiration, some code, and most witty comments come from + * Documentation/lguest/lguest.c, by Rusty Russell + * + * This work is licensed under the terms of the GNU GPL, version 2. + * + * Generic code for virtio server in host kernel. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include + +#include "vhost.h" + +enum { + VHOST_MEMORY_MAX_NREGIONS = 64, + VHOST_MEMORY_F_LOG = 0x1, +}; + +static struct workqueue_struct *vhost_workqueue; + +static void vhost_poll_func(struct file *file, wait_queue_head_t *wqh, + poll_table *pt) +{ + struct vhost_poll *poll; + poll = container_of(pt, struct vhost_poll, table); + + poll->wqh = wqh; + add_wait_queue(wqh, &poll->wait); +} + +static int vhost_poll_wakeup(wait_queue_t *wait, unsigned mode, int sync, + void *key) +{ + struct vhost_poll *poll; + poll = container_of(wait, struct vhost_poll, wait); + if (!((unsigned long)key & poll->mask)) + return 0; + + queue_work(vhost_workqueue, &poll->work); + return 0; +} + +/* Init poll structure */ +void vhost_poll_init(struct vhost_poll *poll, work_func_t func, + unsigned long mask) +{ + INIT_WORK(&poll->work, func); + init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup); + init_poll_funcptr(&poll->table, vhost_poll_func); + poll->mask = mask; +} + +/* Start polling a file. We add ourselves to file's wait queue. The caller must + * keep a reference to a file until after vhost_poll_stop is called. */ +void vhost_poll_start(struct vhost_poll *poll, struct file *file) +{ + unsigned long mask; + mask = file->f_op->poll(file, &poll->table); + if (mask) + vhost_poll_wakeup(&poll->wait, 0, 0, (void *)mask); +} + +/* Stop polling a file. After this function returns, it becomes safe to drop the + * file reference. You must also flush afterwards. */ +void vhost_poll_stop(struct vhost_poll *poll) +{ + remove_wait_queue(poll->wqh, &poll->wait); +} + +/* Flush any work that has been scheduled. When calling this, don't hold any + * locks that are also used by the callback. */ +void vhost_poll_flush(struct vhost_poll *poll) +{ + flush_work(&poll->work); +} + +void vhost_poll_queue(struct vhost_poll *poll) +{ + queue_work(vhost_workqueue, &poll->work); +} + +static void vhost_vq_reset(struct vhost_dev *dev, + struct vhost_virtqueue *vq) +{ + vq->num = 1; + vq->desc = NULL; + vq->avail = NULL; + vq->used = NULL; + vq->last_avail_idx = 0; + vq->avail_idx = 0; + vq->last_used_idx = 0; + vq->used_flags = 0; + vq->used_flags = 0; + vq->log_used = false; + vq->log_addr = -1ull; + vq->hdr_size = 0; + vq->private_data = NULL; + vq->log_base = NULL; + vq->error_ctx = NULL; + vq->error = NULL; + vq->kick = NULL; + vq->call_ctx = NULL; + vq->call = NULL; +} + +long vhost_dev_init(struct vhost_dev *dev, + struct vhost_virtqueue *vqs, int nvqs) +{ + int i; + dev->vqs = vqs; + dev->nvqs = nvqs; + mutex_init(&dev->mutex); + dev->log_ctx = NULL; + dev->log_file = NULL; + dev->memory = NULL; + dev->mm = NULL; + + for (i = 0; i < dev->nvqs; ++i) { + dev->vqs[i].dev = dev; + mutex_init(&dev->vqs[i].mutex); + vhost_vq_reset(dev, dev->vqs + i); + if (dev->vqs[i].handle_kick) + vhost_poll_init(&dev->vqs[i].poll, + dev->vqs[i].handle_kick, + POLLIN); + } + return 0; +} + +/* Caller should have device mutex */ +long vhost_dev_check_owner(struct vhost_dev *dev) +{ + /* Are you the owner? If not, I don't think you mean to do that */ + return dev->mm == current->mm ? 0 : -EPERM; +} + +/* Caller should have device mutex */ +static long vhost_dev_set_owner(struct vhost_dev *dev) +{ + /* Is there an owner already? */ + if (dev->mm) + return -EBUSY; + /* No owner, become one */ + dev->mm = get_task_mm(current); + return 0; +} + +/* Caller should have device mutex */ +long vhost_dev_reset_owner(struct vhost_dev *dev) +{ + struct vhost_memory *memory; + + /* Restore memory to default empty mapping. */ + memory = kmalloc(offsetof(struct vhost_memory, regions), GFP_KERNEL); + if (!memory) + return -ENOMEM; + + vhost_dev_cleanup(dev); + + memory->nregions = 0; + dev->memory = memory; + return 0; +} + +/* Caller should have device mutex */ +void vhost_dev_cleanup(struct vhost_dev *dev) +{ + int i; + for (i = 0; i < dev->nvqs; ++i) { + if (dev->vqs[i].kick && dev->vqs[i].handle_kick) { + vhost_poll_stop(&dev->vqs[i].poll); + vhost_poll_flush(&dev->vqs[i].poll); + } + if (dev->vqs[i].error_ctx) + eventfd_ctx_put(dev->vqs[i].error_ctx); + if (dev->vqs[i].error) + fput(dev->vqs[i].error); + if (dev->vqs[i].kick) + fput(dev->vqs[i].kick); + if (dev->vqs[i].call_ctx) + eventfd_ctx_put(dev->vqs[i].call_ctx); + if (dev->vqs[i].call) + fput(dev->vqs[i].call); + vhost_vq_reset(dev, dev->vqs + i); + } + if (dev->log_ctx) + eventfd_ctx_put(dev->log_ctx); + dev->log_ctx = NULL; + if (dev->log_file) + fput(dev->log_file); + dev->log_file = NULL; + /* No one will access memory at this point */ + kfree(dev->memory); + dev->memory = NULL; + if (dev->mm) + mmput(dev->mm); + dev->mm = NULL; +} + +static int log_access_ok(void __user *log_base, u64 addr, unsigned long sz) +{ + u64 a = addr / VHOST_PAGE_SIZE / 8; + /* Make sure 64 bit math will not overflow. */ + if (a > ULONG_MAX - (unsigned long)log_base || + a + (unsigned long)log_base > ULONG_MAX) + return -EFAULT; + + return access_ok(VERIFY_WRITE, log_base + a, + (sz + VHOST_PAGE_SIZE * 8 - 1) / VHOST_PAGE_SIZE / 8); +} + +/* Caller should have vq mutex and device mutex. */ +static int vq_memory_access_ok(void __user *log_base, struct vhost_memory *mem, + int log_all) +{ + int i; + for (i = 0; i < mem->nregions; ++i) { + struct vhost_memory_region *m = mem->regions + i; + unsigned long a = m->userspace_addr; + if (m->memory_size > ULONG_MAX) + return 0; + else if (!access_ok(VERIFY_WRITE, (void __user *)a, + m->memory_size)) + return 0; + else if (log_all && !log_access_ok(log_base, + m->guest_phys_addr, + m->memory_size)) + return 0; + } + return 1; +} + +/* Can we switch to this memory table? */ +/* Caller should have device mutex but not vq mutex */ +static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem, + int log_all) +{ + int i; + for (i = 0; i < d->nvqs; ++i) { + int ok; + mutex_lock(&d->vqs[i].mutex); + /* If ring is inactive, will check when it's enabled. */ + if (d->vqs[i].private_data) + ok = vq_memory_access_ok(d->vqs[i].log_base, mem, + log_all); + else + ok = 1; + mutex_unlock(&d->vqs[i].mutex); + if (!ok) + return 0; + } + return 1; +} + +static int vq_access_ok(unsigned int num, + struct vring_desc __user *desc, + struct vring_avail __user *avail, + struct vring_used __user *used) +{ + return access_ok(VERIFY_READ, desc, num * sizeof *desc) && + access_ok(VERIFY_READ, avail, + sizeof *avail + num * sizeof *avail->ring) && + access_ok(VERIFY_WRITE, used, + sizeof *used + num * sizeof *used->ring); +} + +/* Can we log writes? */ +/* Caller should have device mutex but not vq mutex */ +int vhost_log_access_ok(struct vhost_dev *dev) +{ + return memory_access_ok(dev, dev->memory, 1); +} + +/* Verify access for write logging. */ +/* Caller should have vq mutex and device mutex */ +static int vq_log_access_ok(struct vhost_virtqueue *vq, void __user *log_base) +{ + return vq_memory_access_ok(log_base, vq->dev->memory, + vhost_has_feature(vq->dev, VHOST_F_LOG_ALL)) && + (!vq->log_used || log_access_ok(log_base, vq->log_addr, + sizeof *vq->used + + vq->num * sizeof *vq->used->ring)); +} + +/* Can we start vq? */ +/* Caller should have vq mutex and device mutex */ +int vhost_vq_access_ok(struct vhost_virtqueue *vq) +{ + return vq_access_ok(vq->num, vq->desc, vq->avail, vq->used) && + vq_log_access_ok(vq, vq->log_base); +} + +static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) +{ + struct vhost_memory mem, *newmem, *oldmem; + unsigned long size = offsetof(struct vhost_memory, regions); + long r; + r = copy_from_user(&mem, m, size); + if (r) + return r; + if (mem.padding) + return -EOPNOTSUPP; + if (mem.nregions > VHOST_MEMORY_MAX_NREGIONS) + return -E2BIG; + newmem = kmalloc(size + mem.nregions * sizeof *m->regions, GFP_KERNEL); + if (!newmem) + return -ENOMEM; + + memcpy(newmem, &mem, size); + r = copy_from_user(newmem->regions, m->regions, + mem.nregions * sizeof *m->regions); + if (r) { + kfree(newmem); + return r; + } + + if (!memory_access_ok(d, newmem, vhost_has_feature(d, VHOST_F_LOG_ALL))) + return -EFAULT; + oldmem = d->memory; + rcu_assign_pointer(d->memory, newmem); + synchronize_rcu(); + kfree(oldmem); + return 0; +} + +static int init_used(struct vhost_virtqueue *vq, + struct vring_used __user *used) +{ + int r = put_user(vq->used_flags, &used->flags); + if (r) + return r; + return get_user(vq->last_used_idx, &used->idx); +} + +static long vhost_set_vring(struct vhost_dev *d, int ioctl, void __user *argp) +{ + struct file *eventfp, *filep = NULL, + *pollstart = NULL, *pollstop = NULL; + struct eventfd_ctx *ctx = NULL; + u32 __user *idxp = argp; + struct vhost_virtqueue *vq; + struct vhost_vring_state s; + struct vhost_vring_file f; + struct vhost_vring_addr a; + u32 idx; + long r; + + r = get_user(idx, idxp); + if (r < 0) + return r; + if (idx > d->nvqs) + return -ENOBUFS; + + vq = d->vqs + idx; + + mutex_lock(&vq->mutex); + + switch (ioctl) { + case VHOST_SET_VRING_NUM: + /* Resizing ring with an active backend? + * You don't want to do that. */ + if (vq->private_data) { + r = -EBUSY; + break; + } + r = copy_from_user(&s, argp, sizeof s); + if (r < 0) + break; + if (!s.num || s.num > 0xffff || (s.num & (s.num - 1))) { + r = -EINVAL; + break; + } + vq->num = s.num; + break; + case VHOST_SET_VRING_BASE: + /* Moving base with an active backend? + * You don't want to do that. */ + if (vq->private_data) { + r = -EBUSY; + break; + } + r = copy_from_user(&s, argp, sizeof s); + if (r < 0) + break; + if (s.num > 0xffff) { + r = -EINVAL; + break; + } + vq->last_avail_idx = s.num; + /* Forget the cached index value. */ + vq->avail_idx = vq->last_avail_idx; + break; + case VHOST_GET_VRING_BASE: + s.index = idx; + s.num = vq->last_avail_idx; + r = copy_to_user(argp, &s, sizeof s); + break; + case VHOST_SET_VRING_ADDR: + r = copy_from_user(&a, argp, sizeof a); + if (r < 0) + break; + if (a.flags & ~(0x1 << VHOST_VRING_F_LOG)) { + r = -EOPNOTSUPP; + break; + } + /* For 32bit, verify that the top 32bits of the user + data are set to zero. */ + if ((u64)(unsigned long)a.desc_user_addr != a.desc_user_addr || + (u64)(unsigned long)a.used_user_addr != a.used_user_addr || + (u64)(unsigned long)a.avail_user_addr != a.avail_user_addr) { + r = -EFAULT; + break; + } + if ((a.avail_user_addr & (sizeof *vq->avail->ring - 1)) || + (a.used_user_addr & (sizeof *vq->used->ring - 1)) || + (a.log_guest_addr & (sizeof *vq->used->ring - 1))) { + r = -EINVAL; + break; + } + + /* We only verify access here if backend is configured. + * If it is not, we don't as size might not have been setup. + * We will verify when backend is configured. */ + if (vq->private_data) { + if (!vq_access_ok(vq->num, + (void __user *)(unsigned long)a.desc_user_addr, + (void __user *)(unsigned long)a.avail_user_addr, + (void __user *)(unsigned long)a.used_user_addr)) { + r = -EINVAL; + break; + } + + /* Also validate log access for used ring if enabled. */ + if ((a.flags & (0x1 << VHOST_VRING_F_LOG)) && + !log_access_ok(vq->log_base, a.log_guest_addr, + sizeof *vq->used + + vq->num * sizeof *vq->used->ring)) { + r = -EINVAL; + break; + } + } + + r = init_used(vq, (struct vring_used __user *)(unsigned long) + a.used_user_addr); + if (r) + break; + vq->log_used = !!(a.flags & (0x1 << VHOST_VRING_F_LOG)); + vq->desc = (void __user *)(unsigned long)a.desc_user_addr; + vq->avail = (void __user *)(unsigned long)a.avail_user_addr; + vq->log_addr = a.log_guest_addr; + vq->used = (void __user *)(unsigned long)a.used_user_addr; + break; + case VHOST_SET_VRING_KICK: + r = copy_from_user(&f, argp, sizeof f); + if (r < 0) + break; + eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd); + if (IS_ERR(eventfp)) + return PTR_ERR(eventfp); + if (eventfp != vq->kick) { + pollstop = filep = vq->kick; + pollstart = vq->kick = eventfp; + } else + filep = eventfp; + break; + case VHOST_SET_VRING_CALL: + r = copy_from_user(&f, argp, sizeof f); + if (r < 0) + break; + eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd); + if (IS_ERR(eventfp)) + return PTR_ERR(eventfp); + if (eventfp != vq->call) { + filep = vq->call; + ctx = vq->call_ctx; + vq->call = eventfp; + vq->call_ctx = eventfp ? + eventfd_ctx_fileget(eventfp) : NULL; + } else + filep = eventfp; + break; + case VHOST_SET_VRING_ERR: + r = copy_from_user(&f, argp, sizeof f); + if (r < 0) + break; + eventfp = f.fd == -1 ? NULL : eventfd_fget(f.fd); + if (IS_ERR(eventfp)) + return PTR_ERR(eventfp); + if (eventfp != vq->error) { + filep = vq->error; + vq->error = eventfp; + ctx = vq->error_ctx; + vq->error_ctx = eventfp ? + eventfd_ctx_fileget(eventfp) : NULL; + } else + filep = eventfp; + break; + default: + r = -ENOIOCTLCMD; + } + + if (pollstop && vq->handle_kick) + vhost_poll_stop(&vq->poll); + + if (ctx) + eventfd_ctx_put(ctx); + if (filep) + fput(filep); + + if (pollstart && vq->handle_kick) + vhost_poll_start(&vq->poll, vq->kick); + + mutex_unlock(&vq->mutex); + + if (pollstop && vq->handle_kick) + vhost_poll_flush(&vq->poll); + return r; +} + +/* Caller must have device mutex */ +long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, unsigned long arg) +{ + void __user *argp = (void __user *)arg; + struct file *eventfp, *filep = NULL; + struct eventfd_ctx *ctx = NULL; + u64 p; + long r; + int i, fd; + + /* If you are not the owner, you can become one */ + if (ioctl == VHOST_SET_OWNER) { + r = vhost_dev_set_owner(d); + goto done; + } + + /* You must be the owner to do anything else */ + r = vhost_dev_check_owner(d); + if (r) + goto done; + + switch (ioctl) { + case VHOST_SET_MEM_TABLE: + r = vhost_set_memory(d, argp); + break; + case VHOST_SET_LOG_BASE: + r = copy_from_user(&p, argp, sizeof p); + if (r < 0) + break; + if ((u64)(unsigned long)p != p) { + r = -EFAULT; + break; + } + for (i = 0; i < d->nvqs; ++i) { + struct vhost_virtqueue *vq; + void __user *base = (void __user *)(unsigned long)p; + vq = d->vqs + i; + mutex_lock(&vq->mutex); + /* If ring is inactive, will check when it's enabled. */ + if (vq->private_data && !vq_log_access_ok(vq, base)) + r = -EFAULT; + else + vq->log_base = base; + mutex_unlock(&vq->mutex); + } + break; + case VHOST_SET_LOG_FD: + r = get_user(fd, (int __user *)argp); + if (r < 0) + break; + eventfp = fd == -1 ? NULL : eventfd_fget(fd); + if (IS_ERR(eventfp)) { + r = PTR_ERR(eventfp); + break; + } + if (eventfp != d->log_file) { + filep = d->log_file; + ctx = d->log_ctx; + d->log_ctx = eventfp ? + eventfd_ctx_fileget(eventfp) : NULL; + } else + filep = eventfp; + for (i = 0; i < d->nvqs; ++i) { + mutex_lock(&d->vqs[i].mutex); + d->vqs[i].log_ctx = d->log_ctx; + mutex_unlock(&d->vqs[i].mutex); + } + if (ctx) + eventfd_ctx_put(ctx); + if (filep) + fput(filep); + break; + default: + r = vhost_set_vring(d, ioctl, argp); + break; + } +done: + return r; +} + +static const struct vhost_memory_region *find_region(struct vhost_memory *mem, + __u64 addr, __u32 len) +{ + struct vhost_memory_region *reg; + int i; + /* linear search is not brilliant, but we really have on the order of 6 + * regions in practice */ + for (i = 0; i < mem->nregions; ++i) { + reg = mem->regions + i; + if (reg->guest_phys_addr <= addr && + reg->guest_phys_addr + reg->memory_size - 1 >= addr) + return reg; + } + return NULL; +} + +/* TODO: This is really inefficient. We need something like get_user() + * (instruction directly accesses the data, with an exception table entry + * returning -EFAULT). See Documentation/x86/exception-tables.txt. + */ +static int set_bit_to_user(int nr, void __user *addr) +{ + unsigned long log = (unsigned long)addr; + struct page *page; + void *base; + int bit = nr + (log % PAGE_SIZE) * 8; + int r; + r = get_user_pages_fast(log, 1, 1, &page); + if (r) + return r; + base = kmap_atomic(page, KM_USER0); + set_bit(bit, base); + kunmap_atomic(base, KM_USER0); + set_page_dirty_lock(page); + put_page(page); + return 0; +} + +static int log_write(void __user *log_base, + u64 write_address, u64 write_length) +{ + int r; + if (!write_length) + return 0; + write_address /= VHOST_PAGE_SIZE; + for (;;) { + u64 base = (u64)(unsigned long)log_base; + u64 log = base + write_address / 8; + int bit = write_address % 8; + if ((u64)(unsigned long)log != log) + return -EFAULT; + r = set_bit_to_user(bit, (void __user *)(unsigned long)log); + if (r < 0) + return r; + if (write_length <= VHOST_PAGE_SIZE) + break; + write_length -= VHOST_PAGE_SIZE; + write_address += VHOST_PAGE_SIZE; + } + return r; +} + +int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log, + unsigned int log_num, u64 len) +{ + int i, r; + + /* Make sure data written is seen before log. */ + wmb(); + for (i = 0; i < log_num; ++i) { + u64 l = min(log[i].len, len); + r = log_write(vq->log_base, log[i].addr, l); + if (r < 0) + return r; + len -= l; + if (!len) + return 0; + } + if (vq->log_ctx) + eventfd_signal(vq->log_ctx, 1); + /* Length written exceeds what we have stored. This is a bug. */ + BUG(); + return 0; +} + +int translate_desc(struct vhost_dev *dev, u64 addr, u32 len, + struct iovec iov[], int iov_size) +{ + const struct vhost_memory_region *reg; + struct vhost_memory *mem; + struct iovec *_iov; + u64 s = 0; + int ret = 0; + + rcu_read_lock(); + + mem = rcu_dereference(dev->memory); + while ((u64)len > s) { + u64 size; + if (ret >= iov_size) { + ret = -ENOBUFS; + break; + } + reg = find_region(mem, addr, len); + if (!reg) { + ret = -EFAULT; + break; + } + _iov = iov + ret; + size = reg->memory_size - addr + reg->guest_phys_addr; + _iov->iov_len = min((u64)len, size); + _iov->iov_base = (void *)(unsigned long) + (reg->userspace_addr + addr - reg->guest_phys_addr); + s += size; + addr += size; + ++ret; + } + + rcu_read_unlock(); + return ret; +} + +/* Each buffer in the virtqueues is actually a chain of descriptors. This + * function returns the next descriptor in the chain, + * or -1U if we're at the end. */ +static unsigned next_desc(struct vring_desc *desc) +{ + unsigned int next; + + /* If this descriptor says it doesn't chain, we're done. */ + if (!(desc->flags & VRING_DESC_F_NEXT)) + return -1U; + + /* Check they're not leading us off end of descriptors. */ + next = desc->next; + /* Make sure compiler knows to grab that: we don't want it changing! */ + /* We will use the result as an index in an array, so most + * architectures only need a compiler barrier here. */ + read_barrier_depends(); + + return next; +} + +static unsigned get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq, + struct iovec iov[], unsigned int iov_size, + unsigned int *out_num, unsigned int *in_num, + struct vhost_log *log, unsigned int *log_num, + struct vring_desc *indirect) +{ + struct vring_desc desc; + unsigned int i = 0, count, found = 0; + int ret; + + /* Sanity check */ + if (indirect->len % sizeof desc) { + vq_err(vq, "Invalid length in indirect descriptor: " + "len 0x%llx not multiple of 0x%zx\n", + (unsigned long long)indirect->len, + sizeof desc); + return -EINVAL; + } + + ret = translate_desc(dev, indirect->addr, indirect->len, vq->indirect, + ARRAY_SIZE(vq->indirect)); + if (ret < 0) { + vq_err(vq, "Translation failure %d in indirect.\n", ret); + return ret; + } + + /* We will use the result as an address to read from, so most + * architectures only need a compiler barrier here. */ + read_barrier_depends(); + + count = indirect->len / sizeof desc; + /* Buffers are chained via a 16 bit next field, so + * we can have at most 2^16 of these. */ + if (count > USHORT_MAX + 1) { + vq_err(vq, "Indirect buffer length too big: %d\n", + indirect->len); + return -E2BIG; + } + + do { + unsigned iov_count = *in_num + *out_num; + if (++found > count) { + vq_err(vq, "Loop detected: last one at %u " + "indirect size %u\n", + i, count); + return -EINVAL; + } + if (memcpy_fromiovec((unsigned char *)&desc, vq->indirect, + sizeof desc)) { + vq_err(vq, "Failed indirect descriptor: idx %d, %zx\n", + i, (size_t)indirect->addr + i * sizeof desc); + return -EINVAL; + } + if (desc.flags & VRING_DESC_F_INDIRECT) { + vq_err(vq, "Nested indirect descriptor: idx %d, %zx\n", + i, (size_t)indirect->addr + i * sizeof desc); + return -EINVAL; + } + + ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count, + iov_size - iov_count); + if (ret < 0) { + vq_err(vq, "Translation failure %d indirect idx %d\n", + ret, i); + return ret; + } + /* If this is an input descriptor, increment that count. */ + if (desc.flags & VRING_DESC_F_WRITE) { + *in_num += ret; + if (unlikely(log)) { + log[*log_num].addr = desc.addr; + log[*log_num].len = desc.len; + ++*log_num; + } + } else { + /* If it's an output descriptor, they're all supposed + * to come before any input descriptors. */ + if (*in_num) { + vq_err(vq, "Indirect descriptor " + "has out after in: idx %d\n", i); + return -EINVAL; + } + *out_num += ret; + } + } while ((i = next_desc(&desc)) != -1); + return 0; +} + +/* This looks in the virtqueue and for the first available buffer, and converts + * it to an iovec for convenient access. Since descriptors consist of some + * number of output then some number of input descriptors, it's actually two + * iovecs, but we pack them into one and note how many of each there were. + * + * This function returns the descriptor number found, or vq->num (which + * is never a valid descriptor number) if none was found. */ +unsigned vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq, + struct iovec iov[], unsigned int iov_size, + unsigned int *out_num, unsigned int *in_num, + struct vhost_log *log, unsigned int *log_num) +{ + struct vring_desc desc; + unsigned int i, head, found = 0; + u16 last_avail_idx; + int ret; + + /* Check it isn't doing very strange things with descriptor numbers. */ + last_avail_idx = vq->last_avail_idx; + if (get_user(vq->avail_idx, &vq->avail->idx)) { + vq_err(vq, "Failed to access avail idx at %p\n", + &vq->avail->idx); + return vq->num; + } + + if ((u16)(vq->avail_idx - last_avail_idx) > vq->num) { + vq_err(vq, "Guest moved used index from %u to %u", + last_avail_idx, vq->avail_idx); + return vq->num; + } + + /* If there's nothing new since last we looked, return invalid. */ + if (vq->avail_idx == last_avail_idx) + return vq->num; + + /* Only get avail ring entries after they have been exposed by guest. */ + rmb(); + + /* Grab the next descriptor number they're advertising, and increment + * the index we've seen. */ + if (get_user(head, &vq->avail->ring[last_avail_idx % vq->num])) { + vq_err(vq, "Failed to read head: idx %d address %p\n", + last_avail_idx, + &vq->avail->ring[last_avail_idx % vq->num]); + return vq->num; + } + + /* If their number is silly, that's an error. */ + if (head >= vq->num) { + vq_err(vq, "Guest says index %u > %u is available", + head, vq->num); + return vq->num; + } + + /* When we start there are none of either input nor output. */ + *out_num = *in_num = 0; + if (unlikely(log)) + *log_num = 0; + + i = head; + do { + unsigned iov_count = *in_num + *out_num; + if (i >= vq->num) { + vq_err(vq, "Desc index is %u > %u, head = %u", + i, vq->num, head); + return vq->num; + } + if (++found > vq->num) { + vq_err(vq, "Loop detected: last one at %u " + "vq size %u head %u\n", + i, vq->num, head); + return vq->num; + } + ret = copy_from_user(&desc, vq->desc + i, sizeof desc); + if (ret) { + vq_err(vq, "Failed to get descriptor: idx %d addr %p\n", + i, vq->desc + i); + return vq->num; + } + if (desc.flags & VRING_DESC_F_INDIRECT) { + ret = get_indirect(dev, vq, iov, iov_size, + out_num, in_num, + log, log_num, &desc); + if (ret < 0) { + vq_err(vq, "Failure detected " + "in indirect descriptor at idx %d\n", i); + return vq->num; + } + continue; + } + + ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count, + iov_size - iov_count); + if (ret < 0) { + vq_err(vq, "Translation failure %d descriptor idx %d\n", + ret, i); + return vq->num; + } + if (desc.flags & VRING_DESC_F_WRITE) { + /* If this is an input descriptor, + * increment that count. */ + *in_num += ret; + if (unlikely(log)) { + log[*log_num].addr = desc.addr; + log[*log_num].len = desc.len; + ++*log_num; + } + } else { + /* If it's an output descriptor, they're all supposed + * to come before any input descriptors. */ + if (*in_num) { + vq_err(vq, "Descriptor has out after in: " + "idx %d\n", i); + return vq->num; + } + *out_num += ret; + } + } while ((i = next_desc(&desc)) != -1); + + /* On success, increment avail index. */ + vq->last_avail_idx++; + return head; +} + +/* Reverse the effect of vhost_get_vq_desc. Useful for error handling. */ +void vhost_discard_vq_desc(struct vhost_virtqueue *vq) +{ + vq->last_avail_idx--; +} + +/* After we've used one of their buffers, we tell them about it. We'll then + * want to notify the guest, using eventfd. */ +int vhost_add_used(struct vhost_virtqueue *vq, unsigned int head, int len) +{ + struct vring_used_elem *used; + + /* The virtqueue contains a ring of used buffers. Get a pointer to the + * next entry in that used ring. */ + used = &vq->used->ring[vq->last_used_idx % vq->num]; + if (put_user(head, &used->id)) { + vq_err(vq, "Failed to write used id"); + return -EFAULT; + } + if (put_user(len, &used->len)) { + vq_err(vq, "Failed to write used len"); + return -EFAULT; + } + /* Make sure buffer is written before we update index. */ + wmb(); + if (put_user(vq->last_used_idx + 1, &vq->used->idx)) { + vq_err(vq, "Failed to increment used idx"); + return -EFAULT; + } + if (unlikely(vq->log_used)) { + /* Make sure data is seen before log. */ + wmb(); + log_write(vq->log_base, vq->log_addr + sizeof *vq->used->ring * + (vq->last_used_idx % vq->num), + sizeof *vq->used->ring); + log_write(vq->log_base, vq->log_addr, sizeof *vq->used->ring); + if (vq->log_ctx) + eventfd_signal(vq->log_ctx, 1); + } + vq->last_used_idx++; + return 0; +} + +/* This actually signals the guest, using eventfd. */ +void vhost_signal(struct vhost_dev *dev, struct vhost_virtqueue *vq) +{ + __u16 flags = 0; + if (get_user(flags, &vq->avail->flags)) { + vq_err(vq, "Failed to get flags"); + return; + } + + /* If they don't want an interrupt, don't signal, unless empty. */ + if ((flags & VRING_AVAIL_F_NO_INTERRUPT) && + (vq->avail_idx != vq->last_avail_idx || + !vhost_has_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY))) + return; + + /* Signal the Guest tell them we used something up. */ + if (vq->call_ctx) + eventfd_signal(vq->call_ctx, 1); +} + +/* And here's the combo meal deal. Supersize me! */ +void vhost_add_used_and_signal(struct vhost_dev *dev, + struct vhost_virtqueue *vq, + unsigned int head, int len) +{ + vhost_add_used(vq, head, len); + vhost_signal(dev, vq); +} + +/* OK, now we need to know about added descriptors. */ +bool vhost_enable_notify(struct vhost_virtqueue *vq) +{ + u16 avail_idx; + int r; + if (!(vq->used_flags & VRING_USED_F_NO_NOTIFY)) + return false; + vq->used_flags &= ~VRING_USED_F_NO_NOTIFY; + r = put_user(vq->used_flags, &vq->used->flags); + if (r) { + vq_err(vq, "Failed to enable notification at %p: %d\n", + &vq->used->flags, r); + return false; + } + /* They could have slipped one in as we were doing that: make + * sure it's written, then check again. */ + mb(); + r = get_user(avail_idx, &vq->avail->idx); + if (r) { + vq_err(vq, "Failed to check avail idx at %p: %d\n", + &vq->avail->idx, r); + return false; + } + + return avail_idx != vq->last_avail_idx; +} + +/* We don't need to be notified again. */ +void vhost_disable_notify(struct vhost_virtqueue *vq) +{ + int r; + if (vq->used_flags & VRING_USED_F_NO_NOTIFY) + return; + vq->used_flags |= VRING_USED_F_NO_NOTIFY; + r = put_user(vq->used_flags, &vq->used->flags); + if (r) + vq_err(vq, "Failed to enable notification at %p: %d\n", + &vq->used->flags, r); +} + +int vhost_init(void) +{ + vhost_workqueue = create_singlethread_workqueue("vhost"); + if (!vhost_workqueue) + return -ENOMEM; + return 0; +} + +void vhost_cleanup(void) +{ + destroy_workqueue(vhost_workqueue); +} diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h new file mode 100644 index 00000000000..44591ba9b07 --- /dev/null +++ b/drivers/vhost/vhost.h @@ -0,0 +1,161 @@ +#ifndef _VHOST_H +#define _VHOST_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct vhost_device; + +enum { + /* Enough place for all fragments, head, and virtio net header. */ + VHOST_NET_MAX_SG = MAX_SKB_FRAGS + 2, +}; + +/* Poll a file (eventfd or socket) */ +/* Note: there's nothing vhost specific about this structure. */ +struct vhost_poll { + poll_table table; + wait_queue_head_t *wqh; + wait_queue_t wait; + /* struct which will handle all actual work. */ + struct work_struct work; + unsigned long mask; +}; + +void vhost_poll_init(struct vhost_poll *poll, work_func_t func, + unsigned long mask); +void vhost_poll_start(struct vhost_poll *poll, struct file *file); +void vhost_poll_stop(struct vhost_poll *poll); +void vhost_poll_flush(struct vhost_poll *poll); +void vhost_poll_queue(struct vhost_poll *poll); + +struct vhost_log { + u64 addr; + u64 len; +}; + +/* The virtqueue structure describes a queue attached to a device. */ +struct vhost_virtqueue { + struct vhost_dev *dev; + + /* The actual ring of buffers. */ + struct mutex mutex; + unsigned int num; + struct vring_desc __user *desc; + struct vring_avail __user *avail; + struct vring_used __user *used; + struct file *kick; + struct file *call; + struct file *error; + struct eventfd_ctx *call_ctx; + struct eventfd_ctx *error_ctx; + struct eventfd_ctx *log_ctx; + + struct vhost_poll poll; + + /* The routine to call when the Guest pings us, or timeout. */ + work_func_t handle_kick; + + /* Last available index we saw. */ + u16 last_avail_idx; + + /* Caches available index value from user. */ + u16 avail_idx; + + /* Last index we used. */ + u16 last_used_idx; + + /* Used flags */ + u16 used_flags; + + /* Log writes to used structure. */ + bool log_used; + u64 log_addr; + + struct iovec indirect[VHOST_NET_MAX_SG]; + struct iovec iov[VHOST_NET_MAX_SG]; + struct iovec hdr[VHOST_NET_MAX_SG]; + size_t hdr_size; + /* We use a kind of RCU to access private pointer. + * All readers access it from workqueue, which makes it possible to + * flush the workqueue instead of synchronize_rcu. Therefore readers do + * not need to call rcu_read_lock/rcu_read_unlock: the beginning of + * work item execution acts instead of rcu_read_lock() and the end of + * work item execution acts instead of rcu_read_lock(). + * Writers use virtqueue mutex. */ + void *private_data; + /* Log write descriptors */ + void __user *log_base; + struct vhost_log log[VHOST_NET_MAX_SG]; +}; + +struct vhost_dev { + /* Readers use RCU to access memory table pointer + * log base pointer and features. + * Writers use mutex below.*/ + struct vhost_memory *memory; + struct mm_struct *mm; + struct mutex mutex; + unsigned acked_features; + struct vhost_virtqueue *vqs; + int nvqs; + struct file *log_file; + struct eventfd_ctx *log_ctx; +}; + +long vhost_dev_init(struct vhost_dev *, struct vhost_virtqueue *vqs, int nvqs); +long vhost_dev_check_owner(struct vhost_dev *); +long vhost_dev_reset_owner(struct vhost_dev *); +void vhost_dev_cleanup(struct vhost_dev *); +long vhost_dev_ioctl(struct vhost_dev *, unsigned int ioctl, unsigned long arg); +int vhost_vq_access_ok(struct vhost_virtqueue *vq); +int vhost_log_access_ok(struct vhost_dev *); + +unsigned vhost_get_vq_desc(struct vhost_dev *, struct vhost_virtqueue *, + struct iovec iov[], unsigned int iov_count, + unsigned int *out_num, unsigned int *in_num, + struct vhost_log *log, unsigned int *log_num); +void vhost_discard_vq_desc(struct vhost_virtqueue *); + +int vhost_add_used(struct vhost_virtqueue *, unsigned int head, int len); +void vhost_signal(struct vhost_dev *, struct vhost_virtqueue *); +void vhost_add_used_and_signal(struct vhost_dev *, struct vhost_virtqueue *, + unsigned int head, int len); +void vhost_disable_notify(struct vhost_virtqueue *); +bool vhost_enable_notify(struct vhost_virtqueue *); + +int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log, + unsigned int log_num, u64 len); + +int vhost_init(void); +void vhost_cleanup(void); + +#define vq_err(vq, fmt, ...) do { \ + pr_debug(pr_fmt(fmt), ##__VA_ARGS__); \ + if ((vq)->error_ctx) \ + eventfd_signal((vq)->error_ctx, 1);\ + } while (0) + +enum { + VHOST_FEATURES = (1 << VIRTIO_F_NOTIFY_ON_EMPTY) | + (1 << VIRTIO_RING_F_INDIRECT_DESC) | + (1 << VHOST_F_LOG_ALL) | + (1 << VHOST_NET_F_VIRTIO_NET_HDR), +}; + +static inline int vhost_has_feature(struct vhost_dev *dev, int bit) +{ + unsigned acked_features = rcu_dereference(dev->acked_features); + return acked_features & (1 << bit); +} + +#endif diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 756f831cbdd..d93080748a9 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -362,6 +362,7 @@ unifdef-y += uio.h unifdef-y += unistd.h unifdef-y += usbdevice_fs.h unifdef-y += utsname.h +unifdef-y += vhost.h unifdef-y += videodev2.h unifdef-y += videodev.h unifdef-y += virtio_config.h diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index adaf3c15e44..8b5f7cc0fba 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -30,6 +30,7 @@ #define HPET_MINOR 228 #define FUSE_MINOR 229 #define KVM_MINOR 232 +#define VHOST_NET_MINOR 233 #define MISC_DYNAMIC_MINOR 255 struct device; diff --git a/include/linux/vhost.h b/include/linux/vhost.h new file mode 100644 index 00000000000..e847f1e3075 --- /dev/null +++ b/include/linux/vhost.h @@ -0,0 +1,130 @@ +#ifndef _LINUX_VHOST_H +#define _LINUX_VHOST_H +/* Userspace interface for in-kernel virtio accelerators. */ + +/* vhost is used to reduce the number of system calls involved in virtio. + * + * Existing virtio net code is used in the guest without modification. + * + * This header includes interface used by userspace hypervisor for + * device configuration. + */ + +#include +#include +#include +#include +#include + +struct vhost_vring_state { + unsigned int index; + unsigned int num; +}; + +struct vhost_vring_file { + unsigned int index; + int fd; /* Pass -1 to unbind from file. */ + +}; + +struct vhost_vring_addr { + unsigned int index; + /* Option flags. */ + unsigned int flags; + /* Flag values: */ + /* Whether log address is valid. If set enables logging. */ +#define VHOST_VRING_F_LOG 0 + + /* Start of array of descriptors (virtually contiguous) */ + __u64 desc_user_addr; + /* Used structure address. Must be 32 bit aligned */ + __u64 used_user_addr; + /* Available structure address. Must be 16 bit aligned */ + __u64 avail_user_addr; + /* Logging support. */ + /* Log writes to used structure, at offset calculated from specified + * address. Address must be 32 bit aligned. */ + __u64 log_guest_addr; +}; + +struct vhost_memory_region { + __u64 guest_phys_addr; + __u64 memory_size; /* bytes */ + __u64 userspace_addr; + __u64 flags_padding; /* No flags are currently specified. */ +}; + +/* All region addresses and sizes must be 4K aligned. */ +#define VHOST_PAGE_SIZE 0x1000 + +struct vhost_memory { + __u32 nregions; + __u32 padding; + struct vhost_memory_region regions[0]; +}; + +/* ioctls */ + +#define VHOST_VIRTIO 0xAF + +/* Features bitmask for forward compatibility. Transport bits are used for + * vhost specific features. */ +#define VHOST_GET_FEATURES _IOR(VHOST_VIRTIO, 0x00, __u64) +#define VHOST_SET_FEATURES _IOW(VHOST_VIRTIO, 0x00, __u64) + +/* Set current process as the (exclusive) owner of this file descriptor. This + * must be called before any other vhost command. Further calls to + * VHOST_OWNER_SET fail until VHOST_OWNER_RESET is called. */ +#define VHOST_SET_OWNER _IO(VHOST_VIRTIO, 0x01) +/* Give up ownership, and reset the device to default values. + * Allows subsequent call to VHOST_OWNER_SET to succeed. */ +#define VHOST_RESET_OWNER _IO(VHOST_VIRTIO, 0x02) + +/* Set up/modify memory layout */ +#define VHOST_SET_MEM_TABLE _IOW(VHOST_VIRTIO, 0x03, struct vhost_memory) + +/* Write logging setup. */ +/* Memory writes can optionally be logged by setting bit at an offset + * (calculated from the physical address) from specified log base. + * The bit is set using an atomic 32 bit operation. */ +/* Set base address for logging. */ +#define VHOST_SET_LOG_BASE _IOW(VHOST_VIRTIO, 0x04, __u64) +/* Specify an eventfd file descriptor to signal on log write. */ +#define VHOST_SET_LOG_FD _IOW(VHOST_VIRTIO, 0x07, int) + +/* Ring setup. */ +/* Set number of descriptors in ring. This parameter can not + * be modified while ring is running (bound to a device). */ +#define VHOST_SET_VRING_NUM _IOW(VHOST_VIRTIO, 0x10, struct vhost_vring_state) +/* Set addresses for the ring. */ +#define VHOST_SET_VRING_ADDR _IOW(VHOST_VIRTIO, 0x11, struct vhost_vring_addr) +/* Base value where queue looks for available descriptors */ +#define VHOST_SET_VRING_BASE _IOW(VHOST_VIRTIO, 0x12, struct vhost_vring_state) +/* Get accessor: reads index, writes value in num */ +#define VHOST_GET_VRING_BASE _IOWR(VHOST_VIRTIO, 0x12, struct vhost_vring_state) + +/* The following ioctls use eventfd file descriptors to signal and poll + * for events. */ + +/* Set eventfd to poll for added buffers */ +#define VHOST_SET_VRING_KICK _IOW(VHOST_VIRTIO, 0x20, struct vhost_vring_file) +/* Set eventfd to signal when buffers have beed used */ +#define VHOST_SET_VRING_CALL _IOW(VHOST_VIRTIO, 0x21, struct vhost_vring_file) +/* Set eventfd to signal an error */ +#define VHOST_SET_VRING_ERR _IOW(VHOST_VIRTIO, 0x22, struct vhost_vring_file) + +/* VHOST_NET specific defines */ + +/* Attach virtio net ring to a raw socket, or tap device. + * The socket must be already bound to an ethernet device, this device will be + * used for transmit. Pass fd -1 to unbind from the socket and the transmit + * device. This can be used to stop the ring (e.g. for migration). */ +#define VHOST_NET_SET_BACKEND _IOW(VHOST_VIRTIO, 0x30, struct vhost_vring_file) + +/* Feature bits */ +/* Log all write descriptors. Can be changed while device is active. */ +#define VHOST_F_LOG_ALL 26 +/* vhost-net should add virtio_net_hdr for RX, and strip for TX packets. */ +#define VHOST_NET_F_VIRTIO_NET_HDR 27 + +#endif -- cgit v1.2.3-70-g09d2 From 72659ecce68588b74f6c46862c2b4cec137d7a5a Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Sun, 17 Jan 2010 19:09:39 -0800 Subject: tcp: account SYN-ACK timeouts & retransmissions Currently we don't increment SYN-ACK timeouts & retransmissions although we do increment the same stats for SYN. We seem to have lost the SYN-ACK accounting with the introduction of tcp_syn_recv_timer (commit 2248761e in the netdev-vger-cvs tree). This patch fixes this issue. In the process we also rename the v4/v6 syn/ack retransmit functions for clarity. We also add a new request_socket operations (syn_ack_timeout) so we can keep code in inet_connection_sock.c protocol agnostic. Signed-off-by: Octavian Purdila Signed-off-by: David S. Miller --- include/net/request_sock.h | 2 ++ include/net/tcp.h | 2 ++ net/ipv4/inet_connection_sock.c | 2 ++ net/ipv4/tcp_ipv4.c | 18 ++++++++++-------- net/ipv4/tcp_timer.c | 6 ++++++ net/ipv6/tcp_ipv6.c | 12 ++++++++++-- 6 files changed, 32 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index c9b50ebd9ce..99e6e19b57c 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -45,6 +45,8 @@ struct request_sock_ops { void (*send_reset)(struct sock *sk, struct sk_buff *skb); void (*destructor)(struct request_sock *req); + void (*syn_ack_timeout)(struct sock *sk, + struct request_sock *req); }; /* struct request_sock - mini sock to represent a connection request diff --git a/include/net/tcp.h b/include/net/tcp.h index 788c99f9859..87d164b9bd8 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -400,6 +400,8 @@ extern int compat_tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, unsigned int optlen); extern void tcp_set_keepalive(struct sock *sk, int val); +extern void tcp_syn_ack_timeout(struct sock *sk, + struct request_sock *req); extern int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len, int nonblock, diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index ee16475f8fc..8da6429269d 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -529,6 +529,8 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, syn_ack_recalc(req, thresh, max_retries, queue->rskq_defer_accept, &expire, &resend); + if (req->rsk_ops->syn_ack_timeout) + req->rsk_ops->syn_ack_timeout(parent, req); if (!expire && (!resend || !req->rsk_ops->rtx_syn_ack(parent, req, NULL) || diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 382f667238e..356f544c4c1 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -742,9 +742,9 @@ static void tcp_v4_reqsk_send_ack(struct sock *sk, struct sk_buff *skb, * This still operates on a request_sock only, not on a big * socket. */ -static int __tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, - struct request_sock *req, - struct request_values *rvp) +static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, + struct request_sock *req, + struct request_values *rvp) { const struct inet_request_sock *ireq = inet_rsk(req); int err = -1; @@ -775,10 +775,11 @@ static int __tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, return err; } -static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, +static int tcp_v4_rtx_synack(struct sock *sk, struct request_sock *req, struct request_values *rvp) { - return __tcp_v4_send_synack(sk, NULL, req, rvp); + TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS); + return tcp_v4_send_synack(sk, NULL, req, rvp); } /* @@ -1192,10 +1193,11 @@ static int tcp_v4_inbound_md5_hash(struct sock *sk, struct sk_buff *skb) struct request_sock_ops tcp_request_sock_ops __read_mostly = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), - .rtx_syn_ack = tcp_v4_send_synack, + .rtx_syn_ack = tcp_v4_rtx_synack, .send_ack = tcp_v4_reqsk_send_ack, .destructor = tcp_v4_reqsk_destructor, .send_reset = tcp_v4_send_reset, + .syn_ack_timeout = tcp_syn_ack_timeout, }; #ifdef CONFIG_TCP_MD5SIG @@ -1373,8 +1375,8 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) } tcp_rsk(req)->snt_isn = isn; - if (__tcp_v4_send_synack(sk, dst, req, - (struct request_values *)&tmp_ext) || + if (tcp_v4_send_synack(sk, dst, req, + (struct request_values *)&tmp_ext) || want_cookie) goto drop_and_free; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 8816a20c259..de7d1bf9114 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -474,6 +474,12 @@ static void tcp_synack_timer(struct sock *sk) TCP_TIMEOUT_INIT, TCP_RTO_MAX); } +void tcp_syn_ack_timeout(struct sock *sk, struct request_sock *req) +{ + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPTIMEOUTS); +} +EXPORT_SYMBOL(tcp_syn_ack_timeout); + void tcp_set_keepalive(struct sock *sk, int val) { if ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 1c832bf198b..82f2dea0e39 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -520,6 +520,13 @@ done: return err; } +static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req, + struct request_values *rvp) +{ + TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS); + return tcp_v6_send_synack(sk, req, rvp); +} + static inline void syn_flood_warning(struct sk_buff *skb) { #ifdef CONFIG_SYN_COOKIES @@ -890,10 +897,11 @@ static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb) struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), - .rtx_syn_ack = tcp_v6_send_synack, + .rtx_syn_ack = tcp_v6_rtx_synack, .send_ack = tcp_v6_reqsk_send_ack, .destructor = tcp_v6_reqsk_destructor, - .send_reset = tcp_v6_send_reset + .send_reset = tcp_v6_send_reset, + .syn_ack_timeout = tcp_syn_ack_timeout, }; #ifdef CONFIG_TCP_MD5SIG -- cgit v1.2.3-70-g09d2 From e9d3897cc2205eec8b7afcc022e4730914b4f48c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 18 Jan 2010 08:08:37 +0100 Subject: netfilter: netns: #ifdef ->iptable_security, ->ip6table_security 'security' tables depend on SECURITY, so ifdef them. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/net/netns/ipv4.h | 2 ++ include/net/netns/ipv6.h | 2 ++ 2 files changed, 4 insertions(+) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 2eb3814d625..8098f6b8319 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -37,7 +37,9 @@ struct netns_ipv4 { struct xt_table *iptable_mangle; struct xt_table *iptable_raw; struct xt_table *arptable_filter; +#ifdef CONFIG_SECURITY struct xt_table *iptable_security; +#endif struct xt_table *nat_table; struct hlist_head *nat_bysource; int nat_vmalloced; diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index dfeb2d7c425..1f11ebc2215 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -36,7 +36,9 @@ struct netns_ipv6 { struct xt_table *ip6table_filter; struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; +#ifdef CONFIG_SECURITY struct xt_table *ip6table_security; +#endif #endif struct rt6_info *ip6_null_entry; struct rt6_statistics *rt6_stats; -- cgit v1.2.3-70-g09d2 From a83d8e8d099fc373a5ca7112ad08c553bb2c180f Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 18 Jan 2010 08:21:13 +0100 Subject: netfilter: xtables: add struct xt_mtchk_param::net Some complex match modules (like xt_hashlimit/xt_recent) want netns information at constructor and destructor time. We propably can play games at match destruction time, because netns can be passed in object, but I think it's cleaner to explicitly pass netns. Add ->net, make sure it's set from ebtables/iptables/ip6tables code. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 1 + net/bridge/netfilter/ebtables.c | 14 +++++++++----- net/ipv4/netfilter/ip_tables.c | 24 ++++++++++++++---------- net/ipv6/netfilter/ip6_tables.c | 14 ++++++++------ 4 files changed, 32 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 378f27ae777..88261b9829a 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -205,6 +205,7 @@ struct xt_match_param { * @hook_mask: via which hooks the new rule is reachable */ struct xt_mtchk_param { + struct net *net; const char *table; const void *entryinfo; const struct xt_match *match; diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index bd1c65425d4..c77bab98669 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -619,7 +619,9 @@ ebt_cleanup_entry(struct ebt_entry *e, unsigned int *cnt) } static inline int -ebt_check_entry(struct ebt_entry *e, struct ebt_table_info *newinfo, +ebt_check_entry(struct ebt_entry *e, + struct net *net, + struct ebt_table_info *newinfo, const char *name, unsigned int *cnt, struct ebt_cl_stack *cl_s, unsigned int udc_cnt) { @@ -671,6 +673,7 @@ ebt_check_entry(struct ebt_entry *e, struct ebt_table_info *newinfo, } i = 0; + mtpar.net = net; mtpar.table = tgpar.table = name; mtpar.entryinfo = tgpar.entryinfo = e; mtpar.hook_mask = tgpar.hook_mask = hookmask; @@ -808,7 +811,8 @@ letscontinue: } /* do the parsing of the table/chains/entries/matches/watchers/targets, heh */ -static int translate_table(char *name, struct ebt_table_info *newinfo) +static int translate_table(struct net *net, char *name, + struct ebt_table_info *newinfo) { unsigned int i, j, k, udc_cnt; int ret; @@ -917,7 +921,7 @@ static int translate_table(char *name, struct ebt_table_info *newinfo) /* used to know what we need to clean up if something goes wrong */ i = 0; ret = EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, - ebt_check_entry, newinfo, name, &i, cl_s, udc_cnt); + ebt_check_entry, net, newinfo, name, &i, cl_s, udc_cnt); if (ret != 0) { EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, ebt_cleanup_entry, &i); @@ -1017,7 +1021,7 @@ static int do_replace(struct net *net, void __user *user, unsigned int len) if (ret != 0) goto free_counterstmp; - ret = translate_table(tmp.name, newinfo); + ret = translate_table(net, tmp.name, newinfo); if (ret != 0) goto free_counterstmp; @@ -1154,7 +1158,7 @@ ebt_register_table(struct net *net, const struct ebt_table *input_table) newinfo->hook_entry[i] = p + ((char *)repl->hook_entry[i] - repl->entries); } - ret = translate_table(repl->name, newinfo); + ret = translate_table(net, repl->name, newinfo); if (ret != 0) { BUGPRINT("Translate_table failed\n"); goto free_chainstack; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 572330a552e..a069d72d948 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -661,8 +661,8 @@ static int check_target(struct ipt_entry *e, const char *name) } static int -find_check_entry(struct ipt_entry *e, const char *name, unsigned int size, - unsigned int *i) +find_check_entry(struct ipt_entry *e, struct net *net, const char *name, + unsigned int size, unsigned int *i) { struct ipt_entry_target *t; struct xt_target *target; @@ -675,6 +675,7 @@ find_check_entry(struct ipt_entry *e, const char *name, unsigned int size, return ret; j = 0; + mtpar.net = net; mtpar.table = name; mtpar.entryinfo = &e->ip; mtpar.hook_mask = e->comefrom; @@ -798,7 +799,8 @@ cleanup_entry(struct ipt_entry *e, unsigned int *i) /* Checks and translates the user-supplied table segment (held in newinfo) */ static int -translate_table(const char *name, +translate_table(struct net *net, + const char *name, unsigned int valid_hooks, struct xt_table_info *newinfo, void *entry0, @@ -860,7 +862,7 @@ translate_table(const char *name, /* Finally, each sanity check must pass */ i = 0; ret = IPT_ENTRY_ITERATE(entry0, newinfo->size, - find_check_entry, name, size, &i); + find_check_entry, net, name, size, &i); if (ret != 0) { IPT_ENTRY_ITERATE(entry0, newinfo->size, @@ -1303,7 +1305,7 @@ do_replace(struct net *net, void __user *user, unsigned int len) goto free_newinfo; } - ret = translate_table(tmp.name, tmp.valid_hooks, + ret = translate_table(net, tmp.name, tmp.valid_hooks, newinfo, loc_cpu_entry, tmp.size, tmp.num_entries, tmp.hook_entry, tmp.underflow); if (ret != 0) @@ -1655,7 +1657,7 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, } static int -compat_check_entry(struct ipt_entry *e, const char *name, +compat_check_entry(struct ipt_entry *e, struct net *net, const char *name, unsigned int *i) { struct xt_mtchk_param mtpar; @@ -1663,6 +1665,7 @@ compat_check_entry(struct ipt_entry *e, const char *name, int ret; j = 0; + mtpar.net = net; mtpar.table = name; mtpar.entryinfo = &e->ip; mtpar.hook_mask = e->comefrom; @@ -1684,7 +1687,8 @@ compat_check_entry(struct ipt_entry *e, const char *name, } static int -translate_compat_table(const char *name, +translate_compat_table(struct net *net, + const char *name, unsigned int valid_hooks, struct xt_table_info **pinfo, void **pentry0, @@ -1773,7 +1777,7 @@ translate_compat_table(const char *name, i = 0; ret = IPT_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry, - name, &i); + net, name, &i); if (ret) { j -= i; COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, @@ -1833,7 +1837,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) goto free_newinfo; } - ret = translate_compat_table(tmp.name, tmp.valid_hooks, + ret = translate_compat_table(net, tmp.name, tmp.valid_hooks, &newinfo, &loc_cpu_entry, tmp.size, tmp.num_entries, tmp.hook_entry, tmp.underflow); @@ -2086,7 +2090,7 @@ struct xt_table *ipt_register_table(struct net *net, loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; memcpy(loc_cpu_entry, repl->entries, repl->size); - ret = translate_table(table->name, table->valid_hooks, + ret = translate_table(net, table->name, table->valid_hooks, newinfo, loc_cpu_entry, repl->size, repl->num_entries, repl->hook_entry, diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 480d7f8c980..a825940a92e 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -693,8 +693,8 @@ static int check_target(struct ip6t_entry *e, const char *name) } static int -find_check_entry(struct ip6t_entry *e, const char *name, unsigned int size, - unsigned int *i) +find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, + unsigned int size, unsigned int *i) { struct ip6t_entry_target *t; struct xt_target *target; @@ -707,6 +707,7 @@ find_check_entry(struct ip6t_entry *e, const char *name, unsigned int size, return ret; j = 0; + mtpar.net = net; mtpar.table = name; mtpar.entryinfo = &e->ipv6; mtpar.hook_mask = e->comefrom; @@ -830,7 +831,8 @@ cleanup_entry(struct ip6t_entry *e, unsigned int *i) /* Checks and translates the user-supplied table segment (held in newinfo) */ static int -translate_table(const char *name, +translate_table(struct net *net, + const char *name, unsigned int valid_hooks, struct xt_table_info *newinfo, void *entry0, @@ -892,7 +894,7 @@ translate_table(const char *name, /* Finally, each sanity check must pass */ i = 0; ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size, - find_check_entry, name, size, &i); + find_check_entry, net, name, size, &i); if (ret != 0) { IP6T_ENTRY_ITERATE(entry0, newinfo->size, @@ -1336,7 +1338,7 @@ do_replace(struct net *net, void __user *user, unsigned int len) goto free_newinfo; } - ret = translate_table(tmp.name, tmp.valid_hooks, + ret = translate_table(net, tmp.name, tmp.valid_hooks, newinfo, loc_cpu_entry, tmp.size, tmp.num_entries, tmp.hook_entry, tmp.underflow); if (ret != 0) @@ -2121,7 +2123,7 @@ struct xt_table *ip6t_register_table(struct net *net, loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; memcpy(loc_cpu_entry, repl->entries, repl->size); - ret = translate_table(table->name, table->valid_hooks, + ret = translate_table(net, table->name, table->valid_hooks, newinfo, loc_cpu_entry, repl->size, repl->num_entries, repl->hook_entry, -- cgit v1.2.3-70-g09d2 From f54e9367f8499a9bf6b2afbc0dce63e1d53c525a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 18 Jan 2010 08:25:47 +0100 Subject: netfilter: xtables: add struct xt_mtdtor_param::net Add ->net to match destructor list like ->net in constructor list. Make sure it's set in ebtables/iptables/ip6tables, this requires to propagate netns up to *_unregister_table(). Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 1 + include/linux/netfilter_bridge/ebtables.h | 2 +- include/linux/netfilter_ipv4/ip_tables.h | 2 +- include/linux/netfilter_ipv6/ip6_tables.h | 2 +- net/bridge/netfilter/ebtable_broute.c | 2 +- net/bridge/netfilter/ebtable_filter.c | 2 +- net/bridge/netfilter/ebtable_nat.c | 2 +- net/bridge/netfilter/ebtables.c | 19 ++++++++-------- net/ipv4/netfilter/ip_tables.c | 25 +++++++++++---------- net/ipv4/netfilter/iptable_filter.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 2 +- net/ipv4/netfilter/iptable_raw.c | 2 +- net/ipv4/netfilter/iptable_security.c | 2 +- net/ipv4/netfilter/nf_nat_rule.c | 2 +- net/ipv6/netfilter/ip6_tables.c | 37 +++++++++++++++++-------------- net/ipv6/netfilter/ip6table_filter.c | 2 +- net/ipv6/netfilter/ip6table_mangle.c | 2 +- net/ipv6/netfilter/ip6table_raw.c | 2 +- net/ipv6/netfilter/ip6table_security.c | 2 +- 19 files changed, 59 insertions(+), 53 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 88261b9829a..3caf5e15110 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -216,6 +216,7 @@ struct xt_mtchk_param { /* Match destructor parameters */ struct xt_mtdtor_param { + struct net *net; const struct xt_match *match; void *matchinfo; u_int8_t family; diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index 3cc40c131cc..1c6f0c5f530 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -289,7 +289,7 @@ struct ebt_table { ~(__alignof__(struct ebt_replace)-1)) extern struct ebt_table *ebt_register_table(struct net *net, const struct ebt_table *table); -extern void ebt_unregister_table(struct ebt_table *table); +extern void ebt_unregister_table(struct net *net, struct ebt_table *table); extern unsigned int ebt_do_table(unsigned int hook, struct sk_buff *skb, const struct net_device *in, const struct net_device *out, struct ebt_table *table); diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 27b3f580730..8d1f273d350 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -242,7 +242,7 @@ extern void ipt_init(void) __init; extern struct xt_table *ipt_register_table(struct net *net, const struct xt_table *table, const struct ipt_replace *repl); -extern void ipt_unregister_table(struct xt_table *table); +extern void ipt_unregister_table(struct net *net, struct xt_table *table); /* Standard entry. */ struct ipt_standard { diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index b31050d20ae..d2952d2fa65 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -300,7 +300,7 @@ extern void ip6t_init(void) __init; extern struct xt_table *ip6t_register_table(struct net *net, const struct xt_table *table, const struct ip6t_replace *repl); -extern void ip6t_unregister_table(struct xt_table *table); +extern void ip6t_unregister_table(struct net *net, struct xt_table *table); extern unsigned int ip6t_do_table(struct sk_buff *skb, unsigned int hook, const struct net_device *in, diff --git a/net/bridge/netfilter/ebtable_broute.c b/net/bridge/netfilter/ebtable_broute.c index d32ab13e728..ae3f106c390 100644 --- a/net/bridge/netfilter/ebtable_broute.c +++ b/net/bridge/netfilter/ebtable_broute.c @@ -71,7 +71,7 @@ static int __net_init broute_net_init(struct net *net) static void __net_exit broute_net_exit(struct net *net) { - ebt_unregister_table(net->xt.broute_table); + ebt_unregister_table(net, net->xt.broute_table); } static struct pernet_operations broute_net_ops = { diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index 60b1a6ca718..42e6bd09457 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -107,7 +107,7 @@ static int __net_init frame_filter_net_init(struct net *net) static void __net_exit frame_filter_net_exit(struct net *net) { - ebt_unregister_table(net->xt.frame_filter); + ebt_unregister_table(net, net->xt.frame_filter); } static struct pernet_operations frame_filter_net_ops = { diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 4a98804203b..6dc2f878ae0 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -107,7 +107,7 @@ static int __net_init frame_nat_net_init(struct net *net) static void __net_exit frame_nat_net_exit(struct net *net) { - ebt_unregister_table(net->xt.frame_nat); + ebt_unregister_table(net, net->xt.frame_nat); } static struct pernet_operations frame_nat_net_ops = { diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index c77bab98669..1aa0e4c1f52 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -561,13 +561,14 @@ ebt_get_udc_positions(struct ebt_entry *e, struct ebt_table_info *newinfo, } static inline int -ebt_cleanup_match(struct ebt_entry_match *m, unsigned int *i) +ebt_cleanup_match(struct ebt_entry_match *m, struct net *net, unsigned int *i) { struct xt_mtdtor_param par; if (i && (*i)-- == 0) return 1; + par.net = net; par.match = m->u.match; par.matchinfo = m->data; par.family = NFPROTO_BRIDGE; @@ -595,7 +596,7 @@ ebt_cleanup_watcher(struct ebt_entry_watcher *w, unsigned int *i) } static inline int -ebt_cleanup_entry(struct ebt_entry *e, unsigned int *cnt) +ebt_cleanup_entry(struct ebt_entry *e, struct net *net, unsigned int *cnt) { struct xt_tgdtor_param par; struct ebt_entry_target *t; @@ -606,7 +607,7 @@ ebt_cleanup_entry(struct ebt_entry *e, unsigned int *cnt) if (cnt && (*cnt)-- == 0) return 1; EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, NULL); - EBT_MATCH_ITERATE(e, ebt_cleanup_match, NULL); + EBT_MATCH_ITERATE(e, ebt_cleanup_match, net, NULL); t = (struct ebt_entry_target *)(((char *)e) + e->target_offset); par.target = t->u.target; @@ -731,7 +732,7 @@ ebt_check_entry(struct ebt_entry *e, cleanup_watchers: EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, &j); cleanup_matches: - EBT_MATCH_ITERATE(e, ebt_cleanup_match, &i); + EBT_MATCH_ITERATE(e, ebt_cleanup_match, net, &i); return ret; } @@ -924,7 +925,7 @@ static int translate_table(struct net *net, char *name, ebt_check_entry, net, newinfo, name, &i, cl_s, udc_cnt); if (ret != 0) { EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, - ebt_cleanup_entry, &i); + ebt_cleanup_entry, net, &i); } vfree(cl_s); return ret; @@ -1074,7 +1075,7 @@ static int do_replace(struct net *net, void __user *user, unsigned int len) /* decrease module count and free resources */ EBT_ENTRY_ITERATE(table->entries, table->entries_size, - ebt_cleanup_entry, NULL); + ebt_cleanup_entry, net, NULL); vfree(table->entries); if (table->chainstack) { @@ -1091,7 +1092,7 @@ free_unlock: mutex_unlock(&ebt_mutex); free_iterate: EBT_ENTRY_ITERATE(newinfo->entries, newinfo->entries_size, - ebt_cleanup_entry, NULL); + ebt_cleanup_entry, net, NULL); free_counterstmp: vfree(counterstmp); /* can be initialized in translate_table() */ @@ -1208,7 +1209,7 @@ out: return ERR_PTR(ret); } -void ebt_unregister_table(struct ebt_table *table) +void ebt_unregister_table(struct net *net, struct ebt_table *table) { int i; @@ -1220,7 +1221,7 @@ void ebt_unregister_table(struct ebt_table *table) list_del(&table->list); mutex_unlock(&ebt_mutex); EBT_ENTRY_ITERATE(table->private->entries, table->private->entries_size, - ebt_cleanup_entry, NULL); + ebt_cleanup_entry, net, NULL); if (table->private->nentries) module_put(table->me); vfree(table->private->entries); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index a069d72d948..cfaba0e2e6f 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -553,13 +553,14 @@ mark_source_chains(struct xt_table_info *newinfo, } static int -cleanup_match(struct ipt_entry_match *m, unsigned int *i) +cleanup_match(struct ipt_entry_match *m, struct net *net, unsigned int *i) { struct xt_mtdtor_param par; if (i && (*i)-- == 0) return 1; + par.net = net; par.match = m->u.kernel.match; par.matchinfo = m->data; par.family = NFPROTO_IPV4; @@ -705,7 +706,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, err: module_put(t->u.kernel.target->me); cleanup_matches: - IPT_MATCH_ITERATE(e, cleanup_match, &j); + IPT_MATCH_ITERATE(e, cleanup_match, net, &j); return ret; } @@ -775,7 +776,7 @@ check_entry_size_and_hooks(struct ipt_entry *e, } static int -cleanup_entry(struct ipt_entry *e, unsigned int *i) +cleanup_entry(struct ipt_entry *e, struct net *net, unsigned int *i) { struct xt_tgdtor_param par; struct ipt_entry_target *t; @@ -784,7 +785,7 @@ cleanup_entry(struct ipt_entry *e, unsigned int *i) return 1; /* Cleanup all matches */ - IPT_MATCH_ITERATE(e, cleanup_match, NULL); + IPT_MATCH_ITERATE(e, cleanup_match, net, NULL); t = ipt_get_target(e); par.target = t->u.kernel.target; @@ -866,7 +867,7 @@ translate_table(struct net *net, if (ret != 0) { IPT_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); + cleanup_entry, net, &i); return ret; } @@ -1260,7 +1261,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; IPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry, - NULL); + net, NULL); xt_free_table_info(oldinfo); if (copy_to_user(counters_ptr, counters, sizeof(struct xt_counters) * num_counters) != 0) @@ -1320,7 +1321,7 @@ do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1682,7 +1683,7 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name, return 0; cleanup_matches: - IPT_MATCH_ITERATE(e, cleanup_match, &j); + IPT_MATCH_ITERATE(e, cleanup_match, net, &j); return ret; } @@ -1782,7 +1783,7 @@ translate_compat_table(struct net *net, j -= i; COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, compat_release_entry, &j); - IPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i); + IPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, net, &i); xt_free_table_info(newinfo); return ret; } @@ -1853,7 +1854,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); free_newinfo: xt_free_table_info(newinfo); return ret; @@ -2112,7 +2113,7 @@ out: return ERR_PTR(ret); } -void ipt_unregister_table(struct xt_table *table) +void ipt_unregister_table(struct net *net, struct xt_table *table) { struct xt_table_info *private; void *loc_cpu_entry; @@ -2122,7 +2123,7 @@ void ipt_unregister_table(struct xt_table *table) /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; - IPT_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL); + IPT_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, net, NULL); if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index df566cbd68e..dee90eb8aa4 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -138,7 +138,7 @@ static int __net_init iptable_filter_net_init(struct net *net) static void __net_exit iptable_filter_net_exit(struct net *net) { - ipt_unregister_table(net->ipv4.iptable_filter); + ipt_unregister_table(net, net->ipv4.iptable_filter); } static struct pernet_operations iptable_filter_net_ops = { diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index fae78c3076c..e07bf242343 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -208,7 +208,7 @@ static int __net_init iptable_mangle_net_init(struct net *net) static void __net_exit iptable_mangle_net_exit(struct net *net) { - ipt_unregister_table(net->ipv4.iptable_mangle); + ipt_unregister_table(net, net->ipv4.iptable_mangle); } static struct pernet_operations iptable_mangle_net_ops = { diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 993edc23be0..40f2b9f611a 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -100,7 +100,7 @@ static int __net_init iptable_raw_net_init(struct net *net) static void __net_exit iptable_raw_net_exit(struct net *net) { - ipt_unregister_table(net->ipv4.iptable_raw); + ipt_unregister_table(net, net->ipv4.iptable_raw); } static struct pernet_operations iptable_raw_net_ops = { diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index 3bd3d6388da..7ce2366e430 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -138,7 +138,7 @@ static int __net_init iptable_security_net_init(struct net *net) static void __net_exit iptable_security_net_exit(struct net *net) { - ipt_unregister_table(net->ipv4.iptable_security); + ipt_unregister_table(net, net->ipv4.iptable_security); } static struct pernet_operations iptable_security_net_ops = { diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c index 9e81e0dfb4e..85da34fdc75 100644 --- a/net/ipv4/netfilter/nf_nat_rule.c +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -195,7 +195,7 @@ static int __net_init nf_nat_rule_net_init(struct net *net) static void __net_exit nf_nat_rule_net_exit(struct net *net) { - ipt_unregister_table(net->ipv4.nat_table); + ipt_unregister_table(net, net->ipv4.nat_table); } static struct pernet_operations nf_nat_rule_net_ops = { diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index a825940a92e..9f1d45f2ba8 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -585,13 +585,14 @@ mark_source_chains(struct xt_table_info *newinfo, } static int -cleanup_match(struct ip6t_entry_match *m, unsigned int *i) +cleanup_match(struct ip6t_entry_match *m, struct net *net, unsigned int *i) { struct xt_mtdtor_param par; if (i && (*i)-- == 0) return 1; + par.net = net; par.match = m->u.kernel.match; par.matchinfo = m->data; par.family = NFPROTO_IPV6; @@ -737,7 +738,7 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, err: module_put(t->u.kernel.target->me); cleanup_matches: - IP6T_MATCH_ITERATE(e, cleanup_match, &j); + IP6T_MATCH_ITERATE(e, cleanup_match, net, &j); return ret; } @@ -807,7 +808,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e, } static int -cleanup_entry(struct ip6t_entry *e, unsigned int *i) +cleanup_entry(struct ip6t_entry *e, struct net *net, unsigned int *i) { struct xt_tgdtor_param par; struct ip6t_entry_target *t; @@ -816,7 +817,7 @@ cleanup_entry(struct ip6t_entry *e, unsigned int *i) return 1; /* Cleanup all matches */ - IP6T_MATCH_ITERATE(e, cleanup_match, NULL); + IP6T_MATCH_ITERATE(e, cleanup_match, net, NULL); t = ip6t_get_target(e); par.target = t->u.kernel.target; @@ -898,7 +899,7 @@ translate_table(struct net *net, if (ret != 0) { IP6T_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); + cleanup_entry, net, &i); return ret; } @@ -1293,7 +1294,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry, - NULL); + net, NULL); xt_free_table_info(oldinfo); if (copy_to_user(counters_ptr, counters, sizeof(struct xt_counters) * num_counters) != 0) @@ -1353,7 +1354,7 @@ do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1692,14 +1693,15 @@ compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr, return ret; } -static int compat_check_entry(struct ip6t_entry *e, const char *name, - unsigned int *i) +static int compat_check_entry(struct ip6t_entry *e, struct net *net, + const char *name, unsigned int *i) { unsigned int j; int ret; struct xt_mtchk_param mtpar; j = 0; + mtpar.net = net; mtpar.table = name; mtpar.entryinfo = &e->ipv6; mtpar.hook_mask = e->comefrom; @@ -1716,12 +1718,13 @@ static int compat_check_entry(struct ip6t_entry *e, const char *name, return 0; cleanup_matches: - IP6T_MATCH_ITERATE(e, cleanup_match, &j); + IP6T_MATCH_ITERATE(e, cleanup_match, net, &j); return ret; } static int -translate_compat_table(const char *name, +translate_compat_table(struct net *net, + const char *name, unsigned int valid_hooks, struct xt_table_info **pinfo, void **pentry0, @@ -1810,12 +1813,12 @@ translate_compat_table(const char *name, i = 0; ret = IP6T_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry, - name, &i); + net, name, &i); if (ret) { j -= i; COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, compat_release_entry, &j); - IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i); + IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, net, &i); xt_free_table_info(newinfo); return ret; } @@ -1870,7 +1873,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) goto free_newinfo; } - ret = translate_compat_table(tmp.name, tmp.valid_hooks, + ret = translate_compat_table(net, tmp.name, tmp.valid_hooks, &newinfo, &loc_cpu_entry, tmp.size, tmp.num_entries, tmp.hook_entry, tmp.underflow); @@ -1886,7 +1889,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); free_newinfo: xt_free_table_info(newinfo); return ret; @@ -2144,7 +2147,7 @@ out: return ERR_PTR(ret); } -void ip6t_unregister_table(struct xt_table *table) +void ip6t_unregister_table(struct net *net, struct xt_table *table) { struct xt_table_info *private; void *loc_cpu_entry; @@ -2154,7 +2157,7 @@ void ip6t_unregister_table(struct xt_table *table) /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; - IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL); + IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, net, NULL); if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index ad378efd0eb..33ddfe53e18 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -131,7 +131,7 @@ static int __net_init ip6table_filter_net_init(struct net *net) static void __net_exit ip6table_filter_net_exit(struct net *net) { - ip6t_unregister_table(net->ipv6.ip6table_filter); + ip6t_unregister_table(net, net->ipv6.ip6table_filter); } static struct pernet_operations ip6table_filter_net_ops = { diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index a929c19d30e..9bc483f000e 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -182,7 +182,7 @@ static int __net_init ip6table_mangle_net_init(struct net *net) static void __net_exit ip6table_mangle_net_exit(struct net *net) { - ip6t_unregister_table(net->ipv6.ip6table_mangle); + ip6t_unregister_table(net, net->ipv6.ip6table_mangle); } static struct pernet_operations ip6table_mangle_net_ops = { diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index ed1a1180f3b..4c90b552e43 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -94,7 +94,7 @@ static int __net_init ip6table_raw_net_init(struct net *net) static void __net_exit ip6table_raw_net_exit(struct net *net) { - ip6t_unregister_table(net->ipv6.ip6table_raw); + ip6t_unregister_table(net, net->ipv6.ip6table_raw); } static struct pernet_operations ip6table_raw_net_ops = { diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index 41b444c6093..baa8d4ef3b0 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -134,7 +134,7 @@ static int __net_init ip6table_security_net_init(struct net *net) static void __net_exit ip6table_security_net_exit(struct net *net) { - ip6t_unregister_table(net->ipv6.ip6table_security); + ip6t_unregister_table(net, net->ipv6.ip6table_security); } static struct pernet_operations ip6table_security_net_ops = { -- cgit v1.2.3-70-g09d2 From c6fcf6bcfc3cfc1c00cc7fd9610cfa2b1a18041f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 17 Jan 2010 01:47:59 +0100 Subject: mac80211: re-enable re-transmission of filtered frames In an earlier commit, mac80211: disable software retry for now Pavel Roskin reported a problem that seems to be due to software retry of already transmitted frames. It turns out that we've never done that correctly, but due to some recent changes it now crashes in the TX code. I've added a comment in the patch that explains the problem better and also points to possible solutions -- which I can't implement right now. I disabled software retry of failed/filtered frames because it was broken. With the work of the previous patches, it now becomes fairly easy to re-enable it by adding a flag indicating that the frame shouldn't be modified, but still running it through the transmit handlers to populate the control information. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 4 ++++ net/mac80211/status.c | 32 +++++--------------------------- net/mac80211/tx.c | 7 ++++++- 3 files changed, 15 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index c90047de442..f03f97b627f 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -271,6 +271,9 @@ struct ieee80211_bss_conf { * transmit function after the current frame, this can be used * by drivers to kick the DMA queue only if unset or when the * queue gets full. + * @IEEE80211_TX_INTFL_RETRANSMISSION: This frame is being retransmitted + * after TX status because the destination was asleep, it must not + * be modified again (no seqno assignment, crypto, etc.) */ enum mac80211_tx_control_flags { IEEE80211_TX_CTL_REQ_TX_STATUS = BIT(0), @@ -291,6 +294,7 @@ enum mac80211_tx_control_flags { IEEE80211_TX_INTFL_DONT_ENCRYPT = BIT(16), IEEE80211_TX_CTL_PSPOLL_RESPONSE = BIT(17), IEEE80211_TX_CTL_MORE_FRAMES = BIT(18), + IEEE80211_TX_INTFL_RETRANSMISSION = BIT(19), }; /** diff --git a/net/mac80211/status.c b/net/mac80211/status.c index 9e171b17827..800b6777e0e 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -44,38 +44,17 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, { struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - /* - * XXX: This is temporary! - * - * The problem here is that when we get here, the driver will - * quite likely have pretty much overwritten info->control by - * using info->driver_data or info->rate_driver_data. Thus, - * when passing out the frame to the driver again, we would be - * passing completely bogus data since the driver would then - * expect a properly filled info->control. In mac80211 itself - * the same problem occurs, since we need info->control.vif - * internally. - * - * To fix this, we should send the frame through TX processing - * again. However, it's not that simple, since the frame will - * have been software-encrypted (if applicable) already, and - * encrypting it again doesn't do much good. So to properly do - * that, we not only have to skip the actual 'raw' encryption - * (key selection etc. still has to be done!) but also the - * sequence number assignment since that impacts the crypto - * encapsulation, of course. - * - * Hence, for now, fix the bug by just dropping the frame. - */ - goto drop; - /* * This skb 'survived' a round-trip through the driver, and * hopefully the driver didn't mangle it too badly. However, * we can definitely not rely on the the control information - * being correct. Clear it so we don't get junk there. + * being correct. Clear it so we don't get junk there, and + * indicate that it needs new processing, but must not be + * modified/encrypted again. */ memset(&info->control, 0, sizeof(info->control)); + info->flags |= IEEE80211_TX_INTFL_NEED_TXPROCESSING | + IEEE80211_TX_INTFL_RETRANSMISSION; sta->tx_filtered_count++; @@ -130,7 +109,6 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, return; } - drop: #ifdef CONFIG_MAC80211_VERBOSE_DEBUG if (net_ratelimit()) printk(KERN_DEBUG "%s: dropped TX filtered frame, " diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index e3d8ff533ee..da557b0d011 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1285,6 +1285,7 @@ static int __ieee80211_tx(struct ieee80211_local *local, static int invoke_tx_handlers(struct ieee80211_tx_data *tx) { struct sk_buff *skb = tx->skb; + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); ieee80211_tx_result res = TX_DROP; #define CALL_TXH(txh) \ @@ -1299,9 +1300,13 @@ static int invoke_tx_handlers(struct ieee80211_tx_data *tx) CALL_TXH(ieee80211_tx_h_ps_buf); CALL_TXH(ieee80211_tx_h_select_key); CALL_TXH(ieee80211_tx_h_sta); - CALL_TXH(ieee80211_tx_h_michael_mic_add); if (!(tx->local->hw.flags & IEEE80211_HW_HAS_RATE_CONTROL)) CALL_TXH(ieee80211_tx_h_rate_ctrl); + + if (unlikely(info->flags & IEEE80211_TX_INTFL_RETRANSMISSION)) + goto txh_done; + + CALL_TXH(ieee80211_tx_h_michael_mic_add); CALL_TXH(ieee80211_tx_h_sequence); CALL_TXH(ieee80211_tx_h_fragment); /* handlers after fragment must be aware of tx info fragmentation! */ -- cgit v1.2.3-70-g09d2 From 11380a4b2d86fae9a6bce75c9373668cc323fe57 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 19 Jan 2010 13:46:10 -0800 Subject: net: Unexport napi_gro_flush(). Nothing outside of net/core/dev.c uses it. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 - net/core/dev.c | 3 +-- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a3fccc85b1a..468a11dea58 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1527,7 +1527,6 @@ extern int netif_rx(struct sk_buff *skb); extern int netif_rx_ni(struct sk_buff *skb); #define HAVE_NETIF_RECEIVE_SKB 1 extern int netif_receive_skb(struct sk_buff *skb); -extern void napi_gro_flush(struct napi_struct *napi); extern gro_result_t dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb); extern gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb); diff --git a/net/core/dev.c b/net/core/dev.c index a008f6987a9..5747b9edc1b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2582,7 +2582,7 @@ out: return netif_receive_skb(skb); } -void napi_gro_flush(struct napi_struct *napi) +static void napi_gro_flush(struct napi_struct *napi) { struct sk_buff *skb, *next; @@ -2595,7 +2595,6 @@ void napi_gro_flush(struct napi_struct *napi) napi->gro_count = 0; napi->gro_list = NULL; } -EXPORT_SYMBOL(napi_gro_flush); enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb) { -- cgit v1.2.3-70-g09d2 From 7c070aa947d1a4105742378579c267f6e7fd08a1 Mon Sep 17 00:00:00 2001 From: Shan Wei Date: Wed, 20 Jan 2010 10:42:41 +0100 Subject: IPv6: reassembly: replace magic number with macro definitions Use macro to define high/low thresh value, refer to IPV6_FRAG_TIMEOUT. Signed-off-by: Shan Wei Acked-by: David S. Miller Signed-off-by: Patrick McHardy --- include/net/ipv6.h | 2 ++ net/ipv6/netfilter/nf_conntrack_reasm.c | 4 ++-- net/ipv6/reassembly.c | 4 ++-- 3 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index ccab5946c83..299bbf5adfb 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -246,6 +246,8 @@ extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb); int ip6_frag_nqueues(struct net *net); int ip6_frag_mem(struct net *net); +#define IPV6_FRAG_HIGH_THRESH 262144 /* == 256*1024 */ +#define IPV6_FRAG_LOW_THRESH 196608 /* == 192*1024 */ #define IPV6_FRAG_TIMEOUT (60*HZ) /* 60 seconds */ extern int __ipv6_addr_type(const struct in6_addr *addr); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 1030ce1e6c7..744ea49de35 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -666,8 +666,8 @@ int nf_ct_frag6_init(void) nf_frags.frag_expire = nf_ct_frag6_expire; nf_frags.secret_interval = 10 * 60 * HZ; nf_init_frags.timeout = IPV6_FRAG_TIMEOUT; - nf_init_frags.high_thresh = 256 * 1024; - nf_init_frags.low_thresh = 192 * 1024; + nf_init_frags.high_thresh = IPV6_FRAG_HIGH_THRESH; + nf_init_frags.low_thresh = IPV6_FRAG_LOW_THRESH; inet_frags_init_net(&nf_init_frags); inet_frags_init(&nf_frags); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 2cddea3bd6b..15bb122e1ce 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -747,8 +747,8 @@ static inline void ip6_frags_sysctl_unregister(void) static int ipv6_frags_init_net(struct net *net) { - net->ipv6.frags.high_thresh = 256 * 1024; - net->ipv6.frags.low_thresh = 192 * 1024; + net->ipv6.frags.high_thresh = IPV6_FRAG_HIGH_THRESH; + net->ipv6.frags.low_thresh = IPV6_FRAG_LOW_THRESH; net->ipv6.frags.timeout = IPV6_FRAG_TIMEOUT; inet_frags_init_net(&net->ipv6.frags); -- cgit v1.2.3-70-g09d2 From b3fbdcf49f940d0703c356441e0daf045e64e076 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 21 Jan 2010 11:40:47 +0100 Subject: mac80211: pass vif and station to update_tkip_key When a TKIP key is updated, we should pass the station pointer instead of just the address, since drivers can use that to store their own data. We also need to pass the virtual interface pointer. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/b43/main.c | 11 ++++++++--- drivers/net/wireless/iwlwifi/iwl-agn.c | 10 +++++++--- include/net/mac80211.h | 6 ++++-- net/mac80211/driver-ops.h | 14 ++++++++++---- net/mac80211/driver-trace.h | 15 +++++++++------ net/mac80211/tkip.c | 12 +++++------- 6 files changed, 43 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c index c238468bca7..c699e46534d 100644 --- a/drivers/net/wireless/b43/main.c +++ b/drivers/net/wireless/b43/main.c @@ -844,8 +844,10 @@ static void rx_tkip_phase1_write(struct b43_wldev *dev, u8 index, u32 iv32, } static void b43_op_update_tkip_key(struct ieee80211_hw *hw, - struct ieee80211_key_conf *keyconf, const u8 *addr, - u32 iv32, u16 *phase1key) + struct ieee80211_vif *vif, + struct ieee80211_key_conf *keyconf, + struct ieee80211_sta *sta, + u32 iv32, u16 *phase1key) { struct b43_wl *wl = hw_to_b43_wl(hw); struct b43_wldev *dev; @@ -863,7 +865,10 @@ static void b43_op_update_tkip_key(struct ieee80211_hw *hw, keymac_write(dev, index, NULL); /* First zero out mac to avoid race */ rx_tkip_phase1_write(dev, index, iv32, phase1key); - keymac_write(dev, index, addr); + /* only pairwise TKIP keys are supported right now */ + if (WARN_ON(!sta)) + goto out_unlock; + keymac_write(dev, index, sta->addr); out_unlock: mutex_unlock(&wl->mutex); diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c index 8db86239bd6..62b6939df52 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn.c @@ -2839,14 +2839,18 @@ void iwl_config_ap(struct iwl_priv *priv) } static void iwl_mac_update_tkip_key(struct ieee80211_hw *hw, - struct ieee80211_key_conf *keyconf, const u8 *addr, - u32 iv32, u16 *phase1key) + struct ieee80211_vif *vif, + struct ieee80211_key_conf *keyconf, + struct ieee80211_sta *sta, + u32 iv32, u16 *phase1key) { struct iwl_priv *priv = hw->priv; IWL_DEBUG_MAC80211(priv, "enter\n"); - iwl_update_tkip_key(priv, keyconf, addr, iv32, phase1key); + iwl_update_tkip_key(priv, keyconf, + sta ? sta->addr : iwl_bcast_addr, + iv32, phase1key); IWL_DEBUG_MAC80211(priv, "leave\n"); } diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f03f97b627f..f56d6f47953 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1614,8 +1614,10 @@ struct ieee80211_ops { struct ieee80211_vif *vif, struct ieee80211_sta *sta, struct ieee80211_key_conf *key); void (*update_tkip_key)(struct ieee80211_hw *hw, - struct ieee80211_key_conf *conf, const u8 *address, - u32 iv32, u16 *phase1key); + struct ieee80211_vif *vif, + struct ieee80211_key_conf *conf, + struct ieee80211_sta *sta, + u32 iv32, u16 *phase1key); int (*hw_scan)(struct ieee80211_hw *hw, struct cfg80211_scan_request *req); void (*sw_scan_start)(struct ieee80211_hw *hw); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index de91d39e027..40c6e9a8986 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -137,16 +137,22 @@ static inline int drv_set_key(struct ieee80211_local *local, } static inline void drv_update_tkip_key(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, struct ieee80211_key_conf *conf, - const u8 *address, u32 iv32, + struct sta_info *sta, u32 iv32, u16 *phase1key) { + struct ieee80211_sta *ista = NULL; + might_sleep(); + if (sta) + ista = &sta->sta; + if (local->ops->update_tkip_key) - local->ops->update_tkip_key(&local->hw, conf, address, - iv32, phase1key); - trace_drv_update_tkip_key(local, conf, address, iv32); + local->ops->update_tkip_key(&local->hw, &sdata->vif, conf, + ista, iv32, phase1key); + trace_drv_update_tkip_key(local, sdata, conf, ista, iv32); } static inline int drv_hw_scan(struct ieee80211_local *local, diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index 0ea258123b8..fefa6e6b01b 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -331,26 +331,29 @@ TRACE_EVENT(drv_set_key, TRACE_EVENT(drv_update_tkip_key, TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, struct ieee80211_key_conf *conf, - const u8 *address, u32 iv32), + struct ieee80211_sta *sta, u32 iv32), - TP_ARGS(local, conf, address, iv32), + TP_ARGS(local, sdata, conf, sta, iv32), TP_STRUCT__entry( LOCAL_ENTRY - __array(u8, addr, 6) + VIF_ENTRY + STA_ENTRY __field(u32, iv32) ), TP_fast_assign( LOCAL_ASSIGN; - memcpy(__entry->addr, address, 6); + VIF_ASSIGN; + STA_ASSIGN; __entry->iv32 = iv32; ), TP_printk( - LOCAL_PR_FMT " addr:%pM iv32:%#x", - LOCAL_PR_ARG, __entry->addr, __entry->iv32 + LOCAL_PR_FMT VIF_PR_FMT STA_PR_FMT " iv32:%#x", + LOCAL_PR_ARG,VIF_PR_ARG,STA_PR_ARG, __entry->iv32 ) ); diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c index 14fe49332c0..7ef491e9d66 100644 --- a/net/mac80211/tkip.c +++ b/net/mac80211/tkip.c @@ -304,14 +304,12 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm, if (key->local->ops->update_tkip_key && key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE && key->u.tkip.rx[queue].state != TKIP_STATE_PHASE1_HW_UPLOADED) { - static const u8 bcast[ETH_ALEN] = - {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; - const u8 *sta_addr = key->sta->sta.addr; + struct ieee80211_sub_if_data *sdata = key->sdata; - if (is_multicast_ether_addr(ra)) - sta_addr = bcast; - - drv_update_tkip_key(key->local, &key->conf, sta_addr, + if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + sdata = container_of(key->sdata->bss, + struct ieee80211_sub_if_data, u.ap); + drv_update_tkip_key(key->local, sdata, &key->conf, key->sta, iv32, key->u.tkip.rx[queue].p1k); key->u.tkip.rx[queue].state = TKIP_STATE_PHASE1_HW_UPLOADED; } -- cgit v1.2.3-70-g09d2 From ef15aac6073b27fd4f70007784d2d52ed394bf43 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Jan 2010 12:02:33 +0100 Subject: cfg80211: export multiple MAC addresses in sysfs If a device has multiple MAC addresses, userspace will need to know about that. Similarly, if it allows the MAC addresses to vary by a bitmask. If a driver exports multiple addresses, it is assumed that it will be able to deal with that many different addresses, which need not necessarily match the ones programmed into the device; if a mask is set then the device should deal addresses within that mask based on an arbitrary "base address". To test it all and show how it is used, add support to hwsim even though it can't actually deal with addresses different from the default. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/mac80211_hwsim.c | 8 +++++++- include/net/cfg80211.h | 22 +++++++++++++++++++++- net/wireless/core.c | 12 ++++++++++++ net/wireless/sysfs.c | 20 ++++++++++++++++++++ 4 files changed, 60 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 84df3fcf37b..0dbda8dfbd9 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -281,6 +281,8 @@ struct mac80211_hwsim_data { struct ieee80211_channel channels_5ghz[ARRAY_SIZE(hwsim_channels_5ghz)]; struct ieee80211_rate rates[ARRAY_SIZE(hwsim_rates)]; + struct mac_address addresses[2]; + struct ieee80211_channel *channel; unsigned long beacon_int; /* in jiffies unit */ unsigned int rx_filter; @@ -1154,7 +1156,11 @@ static int __init init_mac80211_hwsim(void) SET_IEEE80211_DEV(hw, data->dev); addr[3] = i >> 8; addr[4] = i; - SET_IEEE80211_PERM_ADDR(hw, addr); + memcpy(data->addresses[0].addr, addr, ETH_ALEN); + memcpy(data->addresses[1].addr, addr, ETH_ALEN); + data->addresses[1].addr[0] |= 0x40; + hw->wiphy->n_addresses = 2; + hw->wiphy->addresses = data->addresses; hw->channel_change_time = 1; hw->queues = 4; diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 2af52704e67..c5d16f299d6 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1195,6 +1195,10 @@ enum wiphy_flags { WIPHY_FLAG_4ADDR_STATION = BIT(6), }; +struct mac_address { + u8 addr[ETH_ALEN]; +}; + /** * struct wiphy - wireless hardware description * @idx: the wiphy index assigned to this item @@ -1213,12 +1217,28 @@ enum wiphy_flags { * -1 = fragmentation disabled, only odd values >= 256 used * @rts_threshold: RTS threshold (dot11RTSThreshold); -1 = RTS/CTS disabled * @net: the network namespace this wiphy currently lives in + * @perm_addr: permanent MAC address of this device + * @addr_mask: If the device supports multiple MAC addresses by masking, + * set this to a mask with variable bits set to 1, e.g. if the last + * four bits are variable then set it to 00:...:00:0f. The actual + * variable bits shall be determined by the interfaces added, with + * interfaces not matching the mask being rejected to be brought up. + * @n_addresses: number of addresses in @addresses. + * @addresses: If the device has more than one address, set this pointer + * to a list of addresses (6 bytes each). The first one will be used + * by default for perm_addr. In this case, the mask should be set to + * all-zeroes. In this case it is assumed that the device can handle + * the same number of arbitrary MAC addresses. */ struct wiphy { /* assign these fields before you register the wiphy */ - /* permanent MAC address */ + /* permanent MAC address(es) */ u8 perm_addr[ETH_ALEN]; + u8 addr_mask[ETH_ALEN]; + + u16 n_addresses; + struct mac_address *addresses; /* Supported interface modes, OR together BIT(NL80211_IFTYPE_...) */ u16 interface_modes; diff --git a/net/wireless/core.c b/net/wireless/core.c index d07f57c906d..71b6b3a9cf1 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -413,6 +413,18 @@ int wiphy_register(struct wiphy *wiphy) int i; u16 ifmodes = wiphy->interface_modes; + if (WARN_ON(wiphy->addresses && !wiphy->n_addresses)) + return -EINVAL; + + if (WARN_ON(wiphy->addresses && + !is_zero_ether_addr(wiphy->perm_addr) && + memcmp(wiphy->perm_addr, wiphy->addresses[0].addr, + ETH_ALEN))) + return -EINVAL; + + if (wiphy->addresses) + memcpy(wiphy->perm_addr, wiphy->addresses[0].addr, ETH_ALEN); + /* sanity check ifmodes */ WARN_ON(!ifmodes); ifmodes &= ((1 << __NL80211_IFTYPE_AFTER_LAST) - 1) & ~1; diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c index efe3c5c92b2..9f2cef3e0ca 100644 --- a/net/wireless/sysfs.c +++ b/net/wireless/sysfs.c @@ -33,10 +33,30 @@ static ssize_t name ## _show(struct device *dev, \ SHOW_FMT(index, "%d", wiphy_idx); SHOW_FMT(macaddress, "%pM", wiphy.perm_addr); +SHOW_FMT(address_mask, "%pM", wiphy.addr_mask); + +static ssize_t addresses_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct wiphy *wiphy = &dev_to_rdev(dev)->wiphy; + char *start = buf; + int i; + + if (!wiphy->addresses) + return sprintf(buf, "%pM\n", wiphy->perm_addr); + + for (i = 0; i < wiphy->n_addresses; i++) + buf += sprintf(buf, "%pM\n", &wiphy->addresses[i].addr); + + return buf - start; +} static struct device_attribute ieee80211_dev_attrs[] = { __ATTR_RO(index), __ATTR_RO(macaddress), + __ATTR_RO(address_mask), + __ATTR_RO(addresses), {} }; -- cgit v1.2.3-70-g09d2 From d0dd2de0d055f0ffb1e2ecdc21380de9d12a85e2 Mon Sep 17 00:00:00 2001 From: Andriy Tkachuk Date: Wed, 20 Jan 2010 13:55:06 +0200 Subject: mac80211: Account HT Control field in Data frame hdrlen according to 802.11n-2009 ieee80211_hdrlen() should account account new HT Control field in 802.11 data frame header introduced by IEEE 802.11n standard. According to 802.11n-2009 HT Control field is present in data frames when both of following are met: 1. It is QoS data frame. 2. Order bit is set in Frame Control field. The change might be totally compatible with legacy non-11n aware frames, because 802.11-2007 standard states that "all QoS STAs set this subfield to 0". Signed-off-by: Andriy V. Tkachuk Acked-by : Benoit Papillault Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 2 ++ net/wireless/util.c | 5 ++++- 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 842701906ae..19984958ab7 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -138,6 +138,8 @@ #define IEEE80211_WMM_IE_STA_QOSINFO_SP_MASK 0x03 #define IEEE80211_WMM_IE_STA_QOSINFO_SP_SHIFT 5 +#define IEEE80211_HT_CTL_LEN 4 + struct ieee80211_hdr { __le16 frame_control; __le16 duration_id; diff --git a/net/wireless/util.c b/net/wireless/util.c index 23557c1d0a9..be2ab8c59e3 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -227,8 +227,11 @@ unsigned int ieee80211_hdrlen(__le16 fc) if (ieee80211_is_data(fc)) { if (ieee80211_has_a4(fc)) hdrlen = 30; - if (ieee80211_is_data_qos(fc)) + if (ieee80211_is_data_qos(fc)) { hdrlen += IEEE80211_QOS_CTL_LEN; + if (ieee80211_has_order(fc)) + hdrlen += IEEE80211_HT_CTL_LEN; + } goto out; } -- cgit v1.2.3-70-g09d2 From a271623f871dda970319ca15dfad3a8c8c36249f Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 22 Jan 2010 10:13:10 +0000 Subject: netdev: remove certain HAVE_ macros After netdev_ops compat code HAVE_* macros aren't needed, in fact they _will_ result in compile breakage for out of tree drivers. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/netdevice.h | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 468a11dea58..b5fb51d0b8b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -621,30 +621,21 @@ struct net_device_ops { struct net_device *dev); u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb); -#define HAVE_CHANGE_RX_FLAGS void (*ndo_change_rx_flags)(struct net_device *dev, int flags); -#define HAVE_SET_RX_MODE void (*ndo_set_rx_mode)(struct net_device *dev); -#define HAVE_MULTICAST void (*ndo_set_multicast_list)(struct net_device *dev); -#define HAVE_SET_MAC_ADDR int (*ndo_set_mac_address)(struct net_device *dev, void *addr); -#define HAVE_VALIDATE_ADDR int (*ndo_validate_addr)(struct net_device *dev); -#define HAVE_PRIVATE_IOCTL int (*ndo_do_ioctl)(struct net_device *dev, struct ifreq *ifr, int cmd); -#define HAVE_SET_CONFIG int (*ndo_set_config)(struct net_device *dev, struct ifmap *map); -#define HAVE_CHANGE_MTU int (*ndo_change_mtu)(struct net_device *dev, int new_mtu); int (*ndo_neigh_setup)(struct net_device *dev, struct neigh_parms *); -#define HAVE_TX_TIMEOUT void (*ndo_tx_timeout) (struct net_device *dev); struct net_device_stats* (*ndo_get_stats)(struct net_device *dev); @@ -656,7 +647,6 @@ struct net_device_ops { void (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid); #ifdef CONFIG_NET_POLL_CONTROLLER -#define HAVE_NETDEV_POLL void (*ndo_poll_controller)(struct net_device *dev); #endif #if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE) -- cgit v1.2.3-70-g09d2 From 5833929cc2ad2b3064b4fac8c44e293972d240d8 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 22 Jan 2010 10:17:26 +0000 Subject: net: constify MIB name tables Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/snmp.h | 2 +- net/ipv4/proc.c | 4 ++-- net/ipv6/proc.c | 12 ++++++------ net/sctp/proc.c | 2 +- net/xfrm/xfrm_proc.c | 2 +- 5 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/snmp.h b/include/net/snmp.h index f0d756f2ac9..da02ee027d6 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -32,7 +32,7 @@ * - name of entries. */ struct snmp_mib { - char *name; + const char *name; int entry; }; diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index f25542c48b7..1b09a6dde7c 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -127,8 +127,8 @@ static const struct snmp_mib snmp4_ipextstats_list[] = { SNMP_MIB_SENTINEL }; -static struct { - char *name; +static const struct { + const char *name; int index; } icmpmibmap[] = { { "DestUnreachs", ICMP_DEST_UNREACH }, diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 02f20016b4c..bfe2598dd56 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -59,7 +59,7 @@ static const struct file_operations sockstat6_seq_fops = { .release = single_release_net, }; -static struct snmp_mib snmp6_ipstats_list[] = { +static const struct snmp_mib snmp6_ipstats_list[] = { /* ipv6 mib according to RFC 2465 */ SNMP_MIB_ITEM("Ip6InReceives", IPSTATS_MIB_INPKTS), SNMP_MIB_ITEM("Ip6InHdrErrors", IPSTATS_MIB_INHDRERRORS), @@ -92,7 +92,7 @@ static struct snmp_mib snmp6_ipstats_list[] = { SNMP_MIB_SENTINEL }; -static struct snmp_mib snmp6_icmp6_list[] = { +static const struct snmp_mib snmp6_icmp6_list[] = { /* icmpv6 mib according to RFC 2466 */ SNMP_MIB_ITEM("Icmp6InMsgs", ICMP6_MIB_INMSGS), SNMP_MIB_ITEM("Icmp6InErrors", ICMP6_MIB_INERRORS), @@ -120,7 +120,7 @@ static const char *const icmp6type2name[256] = { }; -static struct snmp_mib snmp6_udp6_list[] = { +static const struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_ITEM("Udp6InDatagrams", UDP_MIB_INDATAGRAMS), SNMP_MIB_ITEM("Udp6NoPorts", UDP_MIB_NOPORTS), SNMP_MIB_ITEM("Udp6InErrors", UDP_MIB_INERRORS), @@ -128,7 +128,7 @@ static struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_SENTINEL }; -static struct snmp_mib snmp6_udplite6_list[] = { +static const struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS), SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS), SNMP_MIB_ITEM("UdpLite6InErrors", UDP_MIB_INERRORS), @@ -170,8 +170,8 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) return; } -static inline void -snmp6_seq_show_item(struct seq_file *seq, void **mib, struct snmp_mib *itemlist) +static void snmp6_seq_show_item(struct seq_file *seq, void **mib, + const struct snmp_mib *itemlist) { int i; for (i=0; itemlist[i].name; i++) diff --git a/net/sctp/proc.c b/net/sctp/proc.c index d093cbfeaac..a5ac6e0a8d9 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -40,7 +40,7 @@ #include #include /* for snmp_fold_field */ -static struct snmp_mib sctp_snmp_list[] = { +static const struct snmp_mib sctp_snmp_list[] = { SNMP_MIB_ITEM("SctpCurrEstab", SCTP_MIB_CURRESTAB), SNMP_MIB_ITEM("SctpActiveEstabs", SCTP_MIB_ACTIVEESTABS), SNMP_MIB_ITEM("SctpPassiveEstabs", SCTP_MIB_PASSIVEESTABS), diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index fef8db553e8..c083a4e4e79 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -15,7 +15,7 @@ #include #include -static struct snmp_mib xfrm_mib_list[] = { +static const struct snmp_mib xfrm_mib_list[] = { SNMP_MIB_ITEM("XfrmInError", LINUX_MIB_XFRMINERROR), SNMP_MIB_ITEM("XfrmInBufferError", LINUX_MIB_XFRMINBUFFERERROR), SNMP_MIB_ITEM("XfrmInHdrError", LINUX_MIB_XFRMINHDRERROR), -- cgit v1.2.3-70-g09d2 From e754834e65220b2b674c55c3b6dfb2fb1a2804d0 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 22 Jan 2010 10:18:25 +0000 Subject: icmp: move icmp_err_convert[] to .rodata Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/icmp.h | 2 +- net/ipv4/icmp.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/icmp.h b/include/net/icmp.h index dfa72d4e890..15b3dfe9fce 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -28,7 +28,7 @@ struct icmp_err { unsigned fatal:1; }; -extern struct icmp_err icmp_err_convert[]; +extern const struct icmp_err icmp_err_convert[]; #define ICMP_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmp_statistics, field) #define ICMP_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.icmp_statistics, field) #define ICMPMSGOUT_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmpmsg_statistics, field+256) diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index fe11f60ce41..4b4c2bcd15d 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -114,7 +114,7 @@ struct icmp_bxm { /* An array of errno for error messages from dest unreach. */ /* RFC 1122: 3.2.2.1 States that NET_UNREACH, HOST_UNREACH and SR_FAILED MUST be considered 'transient errs'. */ -struct icmp_err icmp_err_convert[] = { +const struct icmp_err icmp_err_convert[] = { { .errno = ENETUNREACH, /* ICMP_NET_UNREACH */ .fatal = 0, -- cgit v1.2.3-70-g09d2 From 32e7bfc41110bc8f29ec0f293c3bcee6645fef34 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 25 Jan 2010 13:36:10 -0800 Subject: net: use helpers to access uc list V2 This patch introduces three macros to work with uc list from net drivers. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 5 ++--- drivers/net/e1000/e1000_main.c | 4 ++-- drivers/net/igb/igb_main.c | 7 ++++--- drivers/net/ixgbe/ixgbe_common.c | 7 +++---- drivers/net/ixgbe/ixgbe_common.h | 2 +- drivers/net/ixgbe/ixgbe_main.c | 2 +- drivers/net/ixgbe/ixgbe_type.h | 4 ++-- drivers/net/mv643xx_eth.c | 3 +-- drivers/net/niu.c | 4 ++-- drivers/net/stmmac/dwmac1000_core.c | 10 +++++----- drivers/net/virtio_net.c | 12 +++++++----- drivers/s390/net/qeth_l2_main.c | 2 +- include/linux/netdevice.h | 5 +++++ net/core/dev.c | 4 ++-- 14 files changed, 38 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index d83512d3e02..a7b6b12c1c0 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -48,7 +48,6 @@ #include #include #include -#include #if defined(CONFIG_CNIC) || defined(CONFIG_CNIC_MODULE) #define BCM_CNIC 1 @@ -3579,14 +3578,14 @@ bnx2_set_rx_mode(struct net_device *dev) sort_mode |= BNX2_RPM_SORT_USER0_MC_HSH_EN; } - if (dev->uc.count > BNX2_MAX_UNICAST_ADDRESSES) { + if (netdev_uc_count(dev) > BNX2_MAX_UNICAST_ADDRESSES) { rx_mode |= BNX2_EMAC_RX_MODE_PROMISCUOUS; sort_mode |= BNX2_RPM_SORT_USER0_PROM_EN | BNX2_RPM_SORT_USER0_PROM_VLAN; } else if (!(dev->flags & IFF_PROMISC)) { /* Add all entries into to the match filter list */ i = 0; - list_for_each_entry(ha, &dev->uc.list, list) { + netdev_for_each_uc_addr(ha, dev) { bnx2_set_mac_addr(bp, ha->addr, i + BNX2_START_UNICAST_ADDRESS_INDEX); sort_mode |= (1 << diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 87f575ca427..2ce88c5f75c 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2139,7 +2139,7 @@ static void e1000_set_rx_mode(struct net_device *netdev) rctl |= E1000_RCTL_VFE; } - if (netdev->uc.count > rar_entries - 1) { + if (netdev_uc_count(netdev) > rar_entries - 1) { rctl |= E1000_RCTL_UPE; } else if (!(netdev->flags & IFF_PROMISC)) { rctl &= ~E1000_RCTL_UPE; @@ -2162,7 +2162,7 @@ static void e1000_set_rx_mode(struct net_device *netdev) */ i = 1; if (use_uc) - list_for_each_entry(ha, &netdev->uc.list, list) { + netdev_for_each_uc_addr(ha, netdev) { if (i == rar_entries) break; e1000_rar_set(hw, ha->addr, i++); diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index d9679493c63..01cc29483e2 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -2905,12 +2905,13 @@ static int igb_write_uc_addr_list(struct net_device *netdev) int count = 0; /* return ENOMEM indicating insufficient memory for addresses */ - if (netdev->uc.count > rar_entries) + if (netdev_uc_count(netdev) > rar_entries) return -ENOMEM; - if (netdev->uc.count && rar_entries) { + if (!netdev_uc_empty(netdev) && rar_entries) { struct netdev_hw_addr *ha; - list_for_each_entry(ha, &netdev->uc.list, list) { + + netdev_for_each_uc_addr(ha, netdev) { if (!rar_entries) break; igb_rar_set_qsel(adapter, ha->addr, diff --git a/drivers/net/ixgbe/ixgbe_common.c b/drivers/net/ixgbe/ixgbe_common.c index 276c2aaa800..eb49020903c 100644 --- a/drivers/net/ixgbe/ixgbe_common.c +++ b/drivers/net/ixgbe/ixgbe_common.c @@ -28,7 +28,6 @@ #include #include #include -#include #include #include "ixgbe.h" @@ -1347,7 +1346,7 @@ static void ixgbe_add_uc_addr(struct ixgbe_hw *hw, u8 *addr, u32 vmdq) /** * ixgbe_update_uc_addr_list_generic - Updates MAC list of secondary addresses * @hw: pointer to hardware structure - * @uc_list: the list of new addresses + * @netdev: pointer to net device structure * * The given list replaces any existing list. Clears the secondary addrs from * receive address registers. Uses unused receive address registers for the @@ -1357,7 +1356,7 @@ static void ixgbe_add_uc_addr(struct ixgbe_hw *hw, u8 *addr, u32 vmdq) * manually putting the device into promiscuous mode. **/ s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, - struct list_head *uc_list) + struct net_device *netdev) { u32 i; u32 old_promisc_setting = hw->addr_ctrl.overflow_promisc; @@ -1381,7 +1380,7 @@ s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, } /* Add the new addresses */ - list_for_each_entry(ha, uc_list, list) { + netdev_for_each_uc_addr(ha, netdev) { hw_dbg(hw, " Adding the secondary addresses:\n"); ixgbe_add_uc_addr(hw, ha->addr, 0); } diff --git a/drivers/net/ixgbe/ixgbe_common.h b/drivers/net/ixgbe/ixgbe_common.h index dfff0ffaa50..13606d4809c 100644 --- a/drivers/net/ixgbe/ixgbe_common.h +++ b/drivers/net/ixgbe/ixgbe_common.h @@ -60,7 +60,7 @@ s32 ixgbe_update_mc_addr_list_generic(struct ixgbe_hw *hw, u8 *mc_addr_list, u32 mc_addr_count, ixgbe_mc_addr_itr func); s32 ixgbe_update_uc_addr_list_generic(struct ixgbe_hw *hw, - struct list_head *uc_list); + struct net_device *netdev); s32 ixgbe_enable_mc_generic(struct ixgbe_hw *hw); s32 ixgbe_disable_mc_generic(struct ixgbe_hw *hw); s32 ixgbe_enable_rx_dma_generic(struct ixgbe_hw *hw, u32 regval); diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index ee41d331a35..439645d2aee 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -2568,7 +2568,7 @@ void ixgbe_set_rx_mode(struct net_device *netdev) IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl); /* reprogram secondary unicast list */ - hw->mac.ops.update_uc_addr_list(hw, &netdev->uc.list); + hw->mac.ops.update_uc_addr_list(hw, netdev); /* reprogram multicast list */ addr_count = netdev->mc_count; diff --git a/drivers/net/ixgbe/ixgbe_type.h b/drivers/net/ixgbe/ixgbe_type.h index b4caa7011a2..0db67c19b2c 100644 --- a/drivers/net/ixgbe/ixgbe_type.h +++ b/drivers/net/ixgbe/ixgbe_type.h @@ -30,7 +30,7 @@ #include #include -#include +#include /* Vendor ID */ #define IXGBE_INTEL_VENDOR_ID 0x8086 @@ -2405,7 +2405,7 @@ struct ixgbe_mac_operations { s32 (*set_vmdq)(struct ixgbe_hw *, u32, u32); s32 (*clear_vmdq)(struct ixgbe_hw *, u32, u32); s32 (*init_rx_addrs)(struct ixgbe_hw *); - s32 (*update_uc_addr_list)(struct ixgbe_hw *, struct list_head *); + s32 (*update_uc_addr_list)(struct ixgbe_hw *, struct net_device *); s32 (*update_mc_addr_list)(struct ixgbe_hw *, u8 *, u32, ixgbe_mc_addr_itr); s32 (*enable_mc)(struct ixgbe_hw *); diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index af67af55efe..e24072a9a97 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -55,7 +55,6 @@ #include #include #include -#include static char mv643xx_eth_driver_name[] = "mv643xx_eth"; static char mv643xx_eth_driver_version[] = "1.4"; @@ -1697,7 +1696,7 @@ static u32 uc_addr_filter_mask(struct net_device *dev) return 0; nibbles = 1 << (dev->dev_addr[5] & 0x0f); - list_for_each_entry(ha, &dev->uc.list, list) { + netdev_for_each_uc_addr(ha, dev) { if (memcmp(dev->dev_addr, ha->addr, 5)) return 0; if ((dev->dev_addr[5] ^ ha->addr[5]) & 0xf0) diff --git a/drivers/net/niu.c b/drivers/net/niu.c index 0e260cfbff7..af9a8647c7e 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -6372,7 +6372,7 @@ static void niu_set_rx_mode(struct net_device *dev) if ((dev->flags & IFF_ALLMULTI) || (dev->mc_count > 0)) np->flags |= NIU_FLAGS_MCAST; - alt_cnt = dev->uc.count; + alt_cnt = netdev_uc_count(dev); if (alt_cnt > niu_num_alt_addr(np)) { alt_cnt = 0; np->flags |= NIU_FLAGS_PROMISC; @@ -6381,7 +6381,7 @@ static void niu_set_rx_mode(struct net_device *dev) if (alt_cnt) { int index = 0; - list_for_each_entry(ha, &dev->uc.list, list) { + netdev_for_each_uc_addr(ha, dev) { err = niu_set_alt_mac(np, index, ha->addr); if (err) printk(KERN_WARNING PFX "%s: Error %d " diff --git a/drivers/net/stmmac/dwmac1000_core.c b/drivers/net/stmmac/dwmac1000_core.c index 928eac05b91..d812e9cdb3d 100644 --- a/drivers/net/stmmac/dwmac1000_core.c +++ b/drivers/net/stmmac/dwmac1000_core.c @@ -83,7 +83,7 @@ static void dwmac1000_set_filter(struct net_device *dev) unsigned int value = 0; DBG(KERN_INFO "%s: # mcasts %d, # unicast %d\n", - __func__, dev->mc_count, dev->uc.count); + __func__, dev->mc_count, netdev_uc_count(dev)); if (dev->flags & IFF_PROMISC) value = GMAC_FRAME_FILTER_PR; @@ -117,7 +117,7 @@ static void dwmac1000_set_filter(struct net_device *dev) } /* Handle multiple unicast addresses (perfect filtering)*/ - if (dev->uc.count > GMAC_MAX_UNICAST_ADDRESSES) + if (netdev_uc_count(dev) > GMAC_MAX_UNICAST_ADDRESSES) /* Switch to promiscuous mode is more than 16 addrs are required */ value |= GMAC_FRAME_FILTER_PR; @@ -125,9 +125,9 @@ static void dwmac1000_set_filter(struct net_device *dev) int reg = 1; struct netdev_hw_addr *ha; - list_for_each_entry(ha, &dev->uc.list, list) { - dwmac1000_set_umac_addr(ioaddr, ha->addr, reg); - reg++; + netdev_for_each_uc_addr(ha, dev) { + dwmac1000_set_umac_addr(ioaddr, ha->addr, reg); + reg++; } } diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index c708ecc3cb2..088332a943f 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -675,6 +675,7 @@ static void virtnet_set_rx_mode(struct net_device *dev) struct virtio_net_ctrl_mac *mac_data; struct dev_addr_list *addr; struct netdev_hw_addr *ha; + int uc_count; void *buf; int i; @@ -701,8 +702,9 @@ static void virtnet_set_rx_mode(struct net_device *dev) dev_warn(&dev->dev, "Failed to %sable allmulti mode.\n", allmulti ? "en" : "dis"); + uc_count = netdev_uc_count(dev); /* MAC filter - use one buffer for both lists */ - mac_data = buf = kzalloc(((dev->uc.count + dev->mc_count) * ETH_ALEN) + + mac_data = buf = kzalloc(((uc_count + dev->mc_count) * ETH_ALEN) + (2 * sizeof(mac_data->entries)), GFP_ATOMIC); if (!buf) { dev_warn(&dev->dev, "No memory for MAC address buffer\n"); @@ -712,16 +714,16 @@ static void virtnet_set_rx_mode(struct net_device *dev) sg_init_table(sg, 2); /* Store the unicast list and count in the front of the buffer */ - mac_data->entries = dev->uc.count; + mac_data->entries = uc_count; i = 0; - list_for_each_entry(ha, &dev->uc.list, list) + netdev_for_each_uc_addr(ha, dev) memcpy(&mac_data->macs[i++][0], ha->addr, ETH_ALEN); sg_set_buf(&sg[0], mac_data, - sizeof(mac_data->entries) + (dev->uc.count * ETH_ALEN)); + sizeof(mac_data->entries) + (uc_count * ETH_ALEN)); /* multicast list and count fill the end */ - mac_data = (void *)&mac_data->macs[dev->uc.count][0]; + mac_data = (void *)&mac_data->macs[uc_count][0]; mac_data->entries = dev->mc_count; addr = dev->mc_list; diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index c3258b0dd64..51fde6f2e0b 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -622,7 +622,7 @@ static void qeth_l2_set_multicast_list(struct net_device *dev) for (dm = dev->mc_list; dm; dm = dm->next) qeth_l2_add_mc(card, dm->da_addr, 0); - list_for_each_entry(ha, &dev->uc.list, list) + netdev_for_each_uc_addr(ha, dev) qeth_l2_add_mc(card, ha->addr, 1); spin_unlock_bh(&card->mclock); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b5fb51d0b8b..93a32a5ca74 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -263,6 +263,11 @@ struct netdev_hw_addr_list { int count; }; +#define netdev_uc_count(dev) ((dev)->uc.count) +#define netdev_uc_empty(dev) ((dev)->uc.count == 0) +#define netdev_for_each_uc_addr(ha, dev) \ + list_for_each_entry(ha, &dev->uc.list, list) + struct hh_cache { struct hh_cache *hh_next; /* Next entry */ atomic_t hh_refcnt; /* number of users */ diff --git a/net/core/dev.c b/net/core/dev.c index 4fad9db417b..2cba5c521e5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3665,10 +3665,10 @@ void __dev_set_rx_mode(struct net_device *dev) /* Unicast addresses changes may only happen under the rtnl, * therefore calling __dev_set_promiscuity here is safe. */ - if (dev->uc.count > 0 && !dev->uc_promisc) { + if (!netdev_uc_empty(dev) && !dev->uc_promisc) { __dev_set_promiscuity(dev, 1); dev->uc_promisc = 1; - } else if (dev->uc.count == 0 && dev->uc_promisc) { + } else if (netdev_uc_empty(dev) && dev->uc_promisc) { __dev_set_promiscuity(dev, -1); dev->uc_promisc = 0; } -- cgit v1.2.3-70-g09d2 From eb807fb23878bc319e029ed8ce3d835d239723a5 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Sun, 24 Jan 2010 14:55:12 +0200 Subject: mac80211: fix update_tkip_key() documentation about the context Johannes noticed that I had incorrectly documented the context of update_tkip_key() driver operation. It must be atomic because all RX code is run inside rcu critical section. Reported-by: Johannes Berg Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/net/mac80211.h | 2 +- net/mac80211/driver-ops.h | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f56d6f47953..f64402f6312 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1489,7 +1489,7 @@ enum ieee80211_ampdu_mlme_action { * @update_tkip_key: See the section "Hardware crypto acceleration" * This callback will be called in the context of Rx. Called for drivers * which set IEEE80211_KEY_FLAG_TKIP_REQ_RX_P1_KEY. - * The callback can sleep. + * The callback must be atomic. * * @hw_scan: Ask the hardware to service the scan request, no need to start * the scan state machine in stack. The scan must honour the channel diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 40c6e9a8986..6c31f38ac7f 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -144,8 +144,6 @@ static inline void drv_update_tkip_key(struct ieee80211_local *local, { struct ieee80211_sta *ista = NULL; - might_sleep(); - if (sta) ista = &sta->sta; -- cgit v1.2.3-70-g09d2 From c21dbf9214bce129f92e1af05552553ff0e318ed Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 26 Jan 2010 14:15:46 +0100 Subject: cfg80211: export cfg80211_find_ie This new function (previously a static function called just "find_ie" can be used to find a specific IE in a buffer of IEs. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/cfg80211.h | 16 ++++++++++++++++ net/wireless/scan.c | 38 ++++++++++++++++++++------------------ 2 files changed, 36 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index c5d16f299d6..a3f0a7ed31a 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1658,6 +1658,22 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, */ unsigned int cfg80211_classify8021d(struct sk_buff *skb); +/** + * cfg80211_find_ie - find information element in data + * + * @eid: element ID + * @ies: data consisting of IEs + * @len: length of data + * + * This function will return %NULL if the element ID could + * not be found or if the element is invalid (claims to be + * longer than the given data), or a pointer to the first byte + * of the requested element, that is the byte containing the + * element ID. There are no checks on the element length + * other than having to fit into the given data. + */ +const u8 *cfg80211_find_ie(u8 eid, const u8 *ies, int len); + /* * Regulatory helper functions for wiphys */ diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 06b0231ee5e..978cac3414b 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -143,9 +143,9 @@ void cfg80211_bss_expire(struct cfg80211_registered_device *dev) dev->bss_generation++; } -static u8 *find_ie(u8 num, u8 *ies, int len) +const u8 *cfg80211_find_ie(u8 eid, const u8 *ies, int len) { - while (len > 2 && ies[0] != num) { + while (len > 2 && ies[0] != eid) { len -= ies[1] + 2; ies += ies[1] + 2; } @@ -155,11 +155,12 @@ static u8 *find_ie(u8 num, u8 *ies, int len) return NULL; return ies; } +EXPORT_SYMBOL(cfg80211_find_ie); static int cmp_ies(u8 num, u8 *ies1, size_t len1, u8 *ies2, size_t len2) { - const u8 *ie1 = find_ie(num, ies1, len1); - const u8 *ie2 = find_ie(num, ies2, len2); + const u8 *ie1 = cfg80211_find_ie(num, ies1, len1); + const u8 *ie2 = cfg80211_find_ie(num, ies2, len2); int r; if (!ie1 && !ie2) @@ -185,9 +186,9 @@ static bool is_bss(struct cfg80211_bss *a, if (!ssid) return true; - ssidie = find_ie(WLAN_EID_SSID, - a->information_elements, - a->len_information_elements); + ssidie = cfg80211_find_ie(WLAN_EID_SSID, + a->information_elements, + a->len_information_elements); if (!ssidie) return false; if (ssidie[1] != ssid_len) @@ -204,9 +205,9 @@ static bool is_mesh(struct cfg80211_bss *a, if (!is_zero_ether_addr(a->bssid)) return false; - ie = find_ie(WLAN_EID_MESH_ID, - a->information_elements, - a->len_information_elements); + ie = cfg80211_find_ie(WLAN_EID_MESH_ID, + a->information_elements, + a->len_information_elements); if (!ie) return false; if (ie[1] != meshidlen) @@ -214,9 +215,9 @@ static bool is_mesh(struct cfg80211_bss *a, if (memcmp(ie + 2, meshid, meshidlen)) return false; - ie = find_ie(WLAN_EID_MESH_CONFIG, - a->information_elements, - a->len_information_elements); + ie = cfg80211_find_ie(WLAN_EID_MESH_CONFIG, + a->information_elements, + a->len_information_elements); if (!ie) return false; if (ie[1] != sizeof(struct ieee80211_meshconf_ie)) @@ -395,11 +396,12 @@ cfg80211_bss_update(struct cfg80211_registered_device *dev, if (is_zero_ether_addr(res->pub.bssid)) { /* must be mesh, verify */ - meshid = find_ie(WLAN_EID_MESH_ID, res->pub.information_elements, - res->pub.len_information_elements); - meshcfg = find_ie(WLAN_EID_MESH_CONFIG, - res->pub.information_elements, - res->pub.len_information_elements); + meshid = cfg80211_find_ie(WLAN_EID_MESH_ID, + res->pub.information_elements, + res->pub.len_information_elements); + meshcfg = cfg80211_find_ie(WLAN_EID_MESH_CONFIG, + res->pub.information_elements, + res->pub.len_information_elements); if (!meshid || !meshcfg || meshcfg[1] != sizeof(struct ieee80211_meshconf_ie)) { /* bogus mesh */ -- cgit v1.2.3-70-g09d2 From 56007a028c51cbf800a6c969d6f6431d23443b99 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 26 Jan 2010 14:19:52 +0100 Subject: mac80211: wait for beacon before enabling powersave Because DTIM information is required for powersave but is only conveyed in beacons, wait for a beacon before enabling powersave, and change the way the information is conveyed to the driver accordingly. mwl8k doesn't currently seem to implement PS but requires the DTIM period in a different way; after talking to Lennert we agreed to just have mwl8k do the parsing itself in the finalize_join work. Signed-off-by: Johannes Berg Acked-by: Lennert Buytenhek Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-power.c | 2 +- drivers/net/wireless/mwl8k.c | 14 +++++++++----- drivers/net/wireless/wl12xx/wl1251.h | 3 --- drivers/net/wireless/wl12xx/wl1251_main.c | 25 +++++++------------------ include/net/mac80211.h | 7 ++++++- net/mac80211/mlme.c | 27 ++++++++++++++++++++++++--- net/mac80211/scan.c | 4 ---- 7 files changed, 47 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/iwlwifi/iwl-power.c b/drivers/net/wireless/iwlwifi/iwl-power.c index 8599444bef0..9e3ca064145 100644 --- a/drivers/net/wireless/iwlwifi/iwl-power.c +++ b/drivers/net/wireless/iwlwifi/iwl-power.c @@ -319,7 +319,7 @@ int iwl_power_update_mode(struct iwl_priv *priv, bool force) priv->chain_noise_data.state == IWL_CHAIN_NOISE_ALIVE; if (priv->vif) - dtimper = priv->vif->bss_conf.dtim_period; + dtimper = priv->hw->conf.ps_dtim_period; else dtimper = 1; diff --git a/drivers/net/wireless/mwl8k.c b/drivers/net/wireless/mwl8k.c index 68546ca0ba3..f0f08f3919c 100644 --- a/drivers/net/wireless/mwl8k.c +++ b/drivers/net/wireless/mwl8k.c @@ -3881,12 +3881,16 @@ static void mwl8k_finalize_join_worker(struct work_struct *work) struct mwl8k_priv *priv = container_of(work, struct mwl8k_priv, finalize_join_worker); struct sk_buff *skb = priv->beacon_skb; - struct mwl8k_vif *mwl8k_vif; + struct ieee80211_mgmt *mgmt = (void *)skb->data; + int len = skb->len - offsetof(struct ieee80211_mgmt, u.beacon.variable); + const u8 *tim = cfg80211_find_ie(WLAN_EID_TIM, + mgmt->u.beacon.variable, len); + int dtim_period = 1; + + if (tim && tim[1] >= 2) + dtim_period = tim[3]; - mwl8k_vif = mwl8k_first_vif(priv); - if (mwl8k_vif != NULL) - mwl8k_cmd_finalize_join(priv->hw, skb->data, skb->len, - mwl8k_vif->vif->bss_conf.dtim_period); + mwl8k_cmd_finalize_join(priv->hw, skb->data, skb->len, dtim_period); dev_kfree_skb(skb); priv->beacon_skb = NULL; diff --git a/drivers/net/wireless/wl12xx/wl1251.h b/drivers/net/wireless/wl12xx/wl1251.h index 6301578d156..37c61c19cae 100644 --- a/drivers/net/wireless/wl12xx/wl1251.h +++ b/drivers/net/wireless/wl12xx/wl1251.h @@ -341,9 +341,6 @@ struct wl1251 { /* Are we currently scanning */ bool scanning; - /* Our association ID */ - u16 aid; - /* Default key (for WEP) */ u32 default_key; diff --git a/drivers/net/wireless/wl12xx/wl1251_main.c b/drivers/net/wireless/wl12xx/wl1251_main.c index 595f0f94d16..a717dde4822 100644 --- a/drivers/net/wireless/wl12xx/wl1251_main.c +++ b/drivers/net/wireless/wl12xx/wl1251_main.c @@ -617,10 +617,13 @@ static int wl1251_op_config(struct ieee80211_hw *hw, u32 changed) wl->psm_requested = true; + wl->dtim_period = conf->ps_dtim_period; + + ret = wl1251_acx_wr_tbtt_and_dtim(wl, wl->beacon_int, + wl->dtim_period); + /* - * We enter PSM only if we're already associated. - * If we're not, we'll enter it when joining an SSID, - * through the bss_info_changed() hook. + * mac80211 enables PSM only if we're already associated. */ ret = wl1251_ps_set_mode(wl, STATION_POWER_SAVE_MODE); if (ret < 0) @@ -943,7 +946,6 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_bss_conf *bss_conf, u32 changed) { - enum wl1251_cmd_ps_mode mode; struct wl1251 *wl = hw->priv; struct sk_buff *beacon, *skb; int ret; @@ -984,11 +986,6 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw, if (changed & BSS_CHANGED_ASSOC) { if (bss_conf->assoc) { wl->beacon_int = bss_conf->beacon_int; - wl->dtim_period = bss_conf->dtim_period; - - ret = wl1251_acx_wr_tbtt_and_dtim(wl, wl->beacon_int, - wl->dtim_period); - wl->aid = bss_conf->aid; skb = ieee80211_pspoll_get(wl->hw, wl->vif); if (!skb) @@ -1001,17 +998,9 @@ static void wl1251_op_bss_info_changed(struct ieee80211_hw *hw, if (ret < 0) goto out_sleep; - ret = wl1251_acx_aid(wl, wl->aid); + ret = wl1251_acx_aid(wl, bss_conf->aid); if (ret < 0) goto out_sleep; - - /* If we want to go in PSM but we're not there yet */ - if (wl->psm_requested && !wl->psm) { - mode = STATION_POWER_SAVE_MODE; - ret = wl1251_ps_set_mode(wl, mode); - if (ret < 0) - goto out_sleep; - } } else { /* use defaults when not associated */ wl->beacon_int = WL1251_DEFAULT_BEACON_INT; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f64402f6312..1e9c93024cf 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -186,7 +186,8 @@ enum ieee80211_bss_change { * @use_short_slot: use short slot time (only relevant for ERP); * if the hardware cannot handle this it must set the * IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE hardware flag - * @dtim_period: num of beacons before the next DTIM, for PSM + * @dtim_period: num of beacons before the next DTIM, for beaconing, + * not valid in station mode (cf. hw conf ps_dtim_period) * @timestamp: beacon timestamp * @beacon_int: beacon interval * @assoc_capability: capabilities taken from assoc resp @@ -648,6 +649,9 @@ enum ieee80211_smps_mode { * value will be only achievable between DTIM frames, the hardware * needs to check for the multicast traffic bit in DTIM beacons. * This variable is valid only when the CONF_PS flag is set. + * @ps_dtim_period: The DTIM period of the AP we're connected to, for use + * in power saving. Power saving will not be enabled until a beacon + * has been received and the DTIM period is known. * @dynamic_ps_timeout: The dynamic powersave timeout (in ms), see the * powersave documentation below. This variable is valid only when * the CONF_PS flag is set. @@ -674,6 +678,7 @@ struct ieee80211_conf { int max_sleep_period; u16 listen_interval; + u8 ps_dtim_period; u8 long_frame_max_tx_count, short_frame_max_tx_count; diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 1e1d16c55ee..86c6ad1b058 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -484,6 +484,7 @@ void ieee80211_recalc_ps(struct ieee80211_local *local, s32 latency) if (count == 1 && found->u.mgd.powersave && found->u.mgd.associated && + found->u.mgd.associated->beacon_ies && !(found->u.mgd.flags & (IEEE80211_STA_BEACON_POLL | IEEE80211_STA_CONNECTION_POLL))) { s32 beaconint_us; @@ -497,14 +498,22 @@ void ieee80211_recalc_ps(struct ieee80211_local *local, s32 latency) if (beaconint_us > latency) { local->ps_sdata = NULL; } else { - u8 dtimper = found->vif.bss_conf.dtim_period; + struct ieee80211_bss *bss; int maxslp = 1; + u8 dtimper; - if (dtimper > 1) + bss = (void *)found->u.mgd.associated->priv; + dtimper = bss->dtim_period; + + /* If the TIM IE is invalid, pretend the value is 1 */ + if (!dtimper) + dtimper = 1; + else if (dtimper > 1) maxslp = min_t(int, dtimper, latency / beaconint_us); local->hw.conf.max_sleep_period = maxslp; + local->hw.conf.ps_dtim_period = dtimper; local->ps_sdata = found; } } else { @@ -702,7 +711,6 @@ static void ieee80211_set_associated(struct ieee80211_sub_if_data *sdata, /* set timing information */ sdata->vif.bss_conf.beacon_int = cbss->beacon_interval; sdata->vif.bss_conf.timestamp = cbss->tsf; - sdata->vif.bss_conf.dtim_period = bss->dtim_period; bss_info_changed |= BSS_CHANGED_BEACON_INT; bss_info_changed |= ieee80211_handle_bss_capability(sdata, @@ -1168,6 +1176,13 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, int freq; struct ieee80211_bss *bss; struct ieee80211_channel *channel; + bool need_ps = false; + + if (sdata->u.mgd.associated) { + bss = (void *)sdata->u.mgd.associated->priv; + /* not previously set so we may need to recalc */ + need_ps = !bss->dtim_period; + } if (elems->ds_params && elems->ds_params_len == 1) freq = ieee80211_channel_to_frequency(elems->ds_params[0]); @@ -1187,6 +1202,12 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, if (!sdata->u.mgd.associated) return; + if (need_ps) { + mutex_lock(&local->iflist_mtx); + ieee80211_recalc_ps(local, -1); + mutex_unlock(&local->iflist_mtx); + } + if (elems->ch_switch_elem && (elems->ch_switch_elem_len == 3) && (memcmp(mgmt->bssid, sdata->u.mgd.associated->bssid, ETH_ALEN) == 0)) { diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 9afe2f9885d..bc061f62967 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -111,10 +111,6 @@ ieee80211_bss_info_update(struct ieee80211_local *local, bss->dtim_period = tim_ie->dtim_period; } - /* set default value for buggy AP/no TIM element */ - if (bss->dtim_period == 0) - bss->dtim_period = 1; - bss->supp_rates_len = 0; if (elems->supp_rates) { clen = IEEE80211_MAX_SUPP_RATES - bss->supp_rates_len; -- cgit v1.2.3-70-g09d2 From a1664773907a2b69e2a3019598dcbeffa6bc724b Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 25 Jan 2010 10:37:54 +0000 Subject: netns xfrm: xfrm6_tunnel in netns I'm not sure about rcu stuff near kmem cache destruction: * checks for non-empty hashes look bogus, they're done _before_ rcu_berrier() * unregistering netns ops is done before kmem_cache destoy (as it should), and unregistering involves rcu barriers by itself So it looks nothing should be done. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/xfrm.h | 6 +-- net/ipv6/ipcomp6.c | 4 +- net/ipv6/xfrm6_tunnel.c | 140 ++++++++++++++++++++++++++++-------------------- 3 files changed, 88 insertions(+), 62 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 60c27706e7b..fcee547ca7e 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1408,9 +1408,9 @@ extern int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto); extern int xfrm6_tunnel_register(struct xfrm6_tunnel *handler, unsigned short family); extern int xfrm6_tunnel_deregister(struct xfrm6_tunnel *handler, unsigned short family); -extern __be32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr); -extern void xfrm6_tunnel_free_spi(xfrm_address_t *saddr); -extern __be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr); +extern __be32 xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr); +extern void xfrm6_tunnel_free_spi(struct net *net, xfrm_address_t *saddr); +extern __be32 xfrm6_tunnel_spi_lookup(struct net *net, xfrm_address_t *saddr); extern int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb); extern int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb); extern int xfrm6_output(struct sk_buff *skb); diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 2f2a5ca2c87..1d1faf757c9 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -81,7 +81,7 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x) goto out; t->id.proto = IPPROTO_IPV6; - t->id.spi = xfrm6_tunnel_alloc_spi((xfrm_address_t *)&x->props.saddr); + t->id.spi = xfrm6_tunnel_alloc_spi(&init_net, (xfrm_address_t *)&x->props.saddr); if (!t->id.spi) goto error; @@ -112,7 +112,7 @@ static int ipcomp6_tunnel_attach(struct xfrm_state *x) struct xfrm_state *t = NULL; __be32 spi; - spi = xfrm6_tunnel_spi_lookup((xfrm_address_t *)&x->props.saddr); + spi = xfrm6_tunnel_spi_lookup(&init_net, (xfrm_address_t *)&x->props.saddr); if (spi) t = xfrm_state_lookup(&init_net, (xfrm_address_t *)&x->id.daddr, spi, IPPROTO_IPV6, AF_INET6); diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index 23fb1002124..d6f9aeec69f 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -30,6 +30,25 @@ #include #include #include +#include + +#define XFRM6_TUNNEL_SPI_BYADDR_HSIZE 256 +#define XFRM6_TUNNEL_SPI_BYSPI_HSIZE 256 + +#define XFRM6_TUNNEL_SPI_MIN 1 +#define XFRM6_TUNNEL_SPI_MAX 0xffffffff + +struct xfrm6_tunnel_net { + struct hlist_head spi_byaddr[XFRM6_TUNNEL_SPI_BYADDR_HSIZE]; + struct hlist_head spi_byspi[XFRM6_TUNNEL_SPI_BYSPI_HSIZE]; + u32 spi; +}; + +static int xfrm6_tunnel_net_id __read_mostly; +static inline struct xfrm6_tunnel_net *xfrm6_tunnel_pernet(struct net *net) +{ + return net_generic(net, xfrm6_tunnel_net_id); +} /* * xfrm_tunnel_spi things are for allocating unique id ("spi") @@ -46,19 +65,8 @@ struct xfrm6_tunnel_spi { static DEFINE_SPINLOCK(xfrm6_tunnel_spi_lock); -static u32 xfrm6_tunnel_spi; - -#define XFRM6_TUNNEL_SPI_MIN 1 -#define XFRM6_TUNNEL_SPI_MAX 0xffffffff - static struct kmem_cache *xfrm6_tunnel_spi_kmem __read_mostly; -#define XFRM6_TUNNEL_SPI_BYADDR_HSIZE 256 -#define XFRM6_TUNNEL_SPI_BYSPI_HSIZE 256 - -static struct hlist_head xfrm6_tunnel_spi_byaddr[XFRM6_TUNNEL_SPI_BYADDR_HSIZE]; -static struct hlist_head xfrm6_tunnel_spi_byspi[XFRM6_TUNNEL_SPI_BYSPI_HSIZE]; - static inline unsigned xfrm6_tunnel_spi_hash_byaddr(xfrm_address_t *addr) { unsigned h; @@ -77,49 +85,30 @@ static inline unsigned xfrm6_tunnel_spi_hash_byspi(u32 spi) } -static int xfrm6_tunnel_spi_init(void) +static int __init xfrm6_tunnel_spi_init(void) { - int i; - - xfrm6_tunnel_spi = 0; xfrm6_tunnel_spi_kmem = kmem_cache_create("xfrm6_tunnel_spi", sizeof(struct xfrm6_tunnel_spi), 0, SLAB_HWCACHE_ALIGN, NULL); if (!xfrm6_tunnel_spi_kmem) return -ENOMEM; - - for (i = 0; i < XFRM6_TUNNEL_SPI_BYADDR_HSIZE; i++) - INIT_HLIST_HEAD(&xfrm6_tunnel_spi_byaddr[i]); - for (i = 0; i < XFRM6_TUNNEL_SPI_BYSPI_HSIZE; i++) - INIT_HLIST_HEAD(&xfrm6_tunnel_spi_byspi[i]); return 0; } static void xfrm6_tunnel_spi_fini(void) { - int i; - - for (i = 0; i < XFRM6_TUNNEL_SPI_BYADDR_HSIZE; i++) { - if (!hlist_empty(&xfrm6_tunnel_spi_byaddr[i])) - return; - } - for (i = 0; i < XFRM6_TUNNEL_SPI_BYSPI_HSIZE; i++) { - if (!hlist_empty(&xfrm6_tunnel_spi_byspi[i])) - return; - } - rcu_barrier(); kmem_cache_destroy(xfrm6_tunnel_spi_kmem); - xfrm6_tunnel_spi_kmem = NULL; } -static struct xfrm6_tunnel_spi *__xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) +static struct xfrm6_tunnel_spi *__xfrm6_tunnel_spi_lookup(struct net *net, xfrm_address_t *saddr) { + struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); struct xfrm6_tunnel_spi *x6spi; struct hlist_node *pos; hlist_for_each_entry_rcu(x6spi, pos, - &xfrm6_tunnel_spi_byaddr[xfrm6_tunnel_spi_hash_byaddr(saddr)], + &xfrm6_tn->spi_byaddr[xfrm6_tunnel_spi_hash_byaddr(saddr)], list_byaddr) { if (memcmp(&x6spi->addr, saddr, sizeof(x6spi->addr)) == 0) return x6spi; @@ -128,13 +117,13 @@ static struct xfrm6_tunnel_spi *__xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) return NULL; } -__be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) +__be32 xfrm6_tunnel_spi_lookup(struct net *net, xfrm_address_t *saddr) { struct xfrm6_tunnel_spi *x6spi; u32 spi; rcu_read_lock_bh(); - x6spi = __xfrm6_tunnel_spi_lookup(saddr); + x6spi = __xfrm6_tunnel_spi_lookup(net, saddr); spi = x6spi ? x6spi->spi : 0; rcu_read_unlock_bh(); return htonl(spi); @@ -142,14 +131,15 @@ __be32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) EXPORT_SYMBOL(xfrm6_tunnel_spi_lookup); -static int __xfrm6_tunnel_spi_check(u32 spi) +static int __xfrm6_tunnel_spi_check(struct net *net, u32 spi) { + struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); struct xfrm6_tunnel_spi *x6spi; int index = xfrm6_tunnel_spi_hash_byspi(spi); struct hlist_node *pos; hlist_for_each_entry(x6spi, pos, - &xfrm6_tunnel_spi_byspi[index], + &xfrm6_tn->spi_byspi[index], list_byspi) { if (x6spi->spi == spi) return -1; @@ -157,32 +147,33 @@ static int __xfrm6_tunnel_spi_check(u32 spi) return index; } -static u32 __xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr) +static u32 __xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr) { + struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); u32 spi; struct xfrm6_tunnel_spi *x6spi; int index; - if (xfrm6_tunnel_spi < XFRM6_TUNNEL_SPI_MIN || - xfrm6_tunnel_spi >= XFRM6_TUNNEL_SPI_MAX) - xfrm6_tunnel_spi = XFRM6_TUNNEL_SPI_MIN; + if (xfrm6_tn->spi < XFRM6_TUNNEL_SPI_MIN || + xfrm6_tn->spi >= XFRM6_TUNNEL_SPI_MAX) + xfrm6_tn->spi = XFRM6_TUNNEL_SPI_MIN; else - xfrm6_tunnel_spi++; + xfrm6_tn->spi++; - for (spi = xfrm6_tunnel_spi; spi <= XFRM6_TUNNEL_SPI_MAX; spi++) { - index = __xfrm6_tunnel_spi_check(spi); + for (spi = xfrm6_tn->spi; spi <= XFRM6_TUNNEL_SPI_MAX; spi++) { + index = __xfrm6_tunnel_spi_check(net, spi); if (index >= 0) goto alloc_spi; } - for (spi = XFRM6_TUNNEL_SPI_MIN; spi < xfrm6_tunnel_spi; spi++) { - index = __xfrm6_tunnel_spi_check(spi); + for (spi = XFRM6_TUNNEL_SPI_MIN; spi < xfrm6_tn->spi; spi++) { + index = __xfrm6_tunnel_spi_check(net, spi); if (index >= 0) goto alloc_spi; } spi = 0; goto out; alloc_spi: - xfrm6_tunnel_spi = spi; + xfrm6_tn->spi = spi; x6spi = kmem_cache_alloc(xfrm6_tunnel_spi_kmem, GFP_ATOMIC); if (!x6spi) goto out; @@ -192,26 +183,26 @@ alloc_spi: x6spi->spi = spi; atomic_set(&x6spi->refcnt, 1); - hlist_add_head_rcu(&x6spi->list_byspi, &xfrm6_tunnel_spi_byspi[index]); + hlist_add_head_rcu(&x6spi->list_byspi, &xfrm6_tn->spi_byspi[index]); index = xfrm6_tunnel_spi_hash_byaddr(saddr); - hlist_add_head_rcu(&x6spi->list_byaddr, &xfrm6_tunnel_spi_byaddr[index]); + hlist_add_head_rcu(&x6spi->list_byaddr, &xfrm6_tn->spi_byaddr[index]); out: return spi; } -__be32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr) +__be32 xfrm6_tunnel_alloc_spi(struct net *net, xfrm_address_t *saddr) { struct xfrm6_tunnel_spi *x6spi; u32 spi; spin_lock_bh(&xfrm6_tunnel_spi_lock); - x6spi = __xfrm6_tunnel_spi_lookup(saddr); + x6spi = __xfrm6_tunnel_spi_lookup(net, saddr); if (x6spi) { atomic_inc(&x6spi->refcnt); spi = x6spi->spi; } else - spi = __xfrm6_tunnel_alloc_spi(saddr); + spi = __xfrm6_tunnel_alloc_spi(net, saddr); spin_unlock_bh(&xfrm6_tunnel_spi_lock); return htonl(spi); @@ -225,15 +216,16 @@ static void x6spi_destroy_rcu(struct rcu_head *head) container_of(head, struct xfrm6_tunnel_spi, rcu_head)); } -void xfrm6_tunnel_free_spi(xfrm_address_t *saddr) +void xfrm6_tunnel_free_spi(struct net *net, xfrm_address_t *saddr) { + struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); struct xfrm6_tunnel_spi *x6spi; struct hlist_node *pos, *n; spin_lock_bh(&xfrm6_tunnel_spi_lock); hlist_for_each_entry_safe(x6spi, pos, n, - &xfrm6_tunnel_spi_byaddr[xfrm6_tunnel_spi_hash_byaddr(saddr)], + &xfrm6_tn->spi_byaddr[xfrm6_tunnel_spi_hash_byaddr(saddr)], list_byaddr) { if (memcmp(&x6spi->addr, saddr, sizeof(x6spi->addr)) == 0) { @@ -263,10 +255,11 @@ static int xfrm6_tunnel_input(struct xfrm_state *x, struct sk_buff *skb) static int xfrm6_tunnel_rcv(struct sk_buff *skb) { + struct net *net = dev_net(skb->dev); struct ipv6hdr *iph = ipv6_hdr(skb); __be32 spi; - spi = xfrm6_tunnel_spi_lookup((xfrm_address_t *)&iph->saddr); + spi = xfrm6_tunnel_spi_lookup(net, (xfrm_address_t *)&iph->saddr); return xfrm6_rcv_spi(skb, IPPROTO_IPV6, spi) > 0 ? : 0; } @@ -326,7 +319,9 @@ static int xfrm6_tunnel_init_state(struct xfrm_state *x) static void xfrm6_tunnel_destroy(struct xfrm_state *x) { - xfrm6_tunnel_free_spi((xfrm_address_t *)&x->props.saddr); + struct net *net = xs_net(x); + + xfrm6_tunnel_free_spi(net, (xfrm_address_t *)&x->props.saddr); } static const struct xfrm_type xfrm6_tunnel_type = { @@ -351,6 +346,31 @@ static struct xfrm6_tunnel xfrm46_tunnel_handler = { .priority = 2, }; +static int __net_init xfrm6_tunnel_net_init(struct net *net) +{ + struct xfrm6_tunnel_net *xfrm6_tn = xfrm6_tunnel_pernet(net); + unsigned int i; + + for (i = 0; i < XFRM6_TUNNEL_SPI_BYADDR_HSIZE; i++) + INIT_HLIST_HEAD(&xfrm6_tn->spi_byaddr[i]); + for (i = 0; i < XFRM6_TUNNEL_SPI_BYSPI_HSIZE; i++) + INIT_HLIST_HEAD(&xfrm6_tn->spi_byspi[i]); + xfrm6_tn->spi = 0; + + return 0; +} + +static void __net_exit xfrm6_tunnel_net_exit(struct net *net) +{ +} + +static struct pernet_operations xfrm6_tunnel_net_ops = { + .init = xfrm6_tunnel_net_init, + .exit = xfrm6_tunnel_net_exit, + .id = &xfrm6_tunnel_net_id, + .size = sizeof(struct xfrm6_tunnel_net), +}; + static int __init xfrm6_tunnel_init(void) { int rv; @@ -367,8 +387,13 @@ static int __init xfrm6_tunnel_init(void) rv = xfrm6_tunnel_spi_init(); if (rv < 0) goto dereg46; + rv = register_pernet_subsys(&xfrm6_tunnel_net_ops); + if (rv < 0) + goto deregspi; return 0; +deregspi: + xfrm6_tunnel_spi_fini(); dereg46: xfrm6_tunnel_deregister(&xfrm46_tunnel_handler, AF_INET); dereg6: @@ -381,6 +406,7 @@ err: static void __exit xfrm6_tunnel_fini(void) { + unregister_pernet_subsys(&xfrm6_tunnel_net_ops); xfrm6_tunnel_spi_fini(); xfrm6_tunnel_deregister(&xfrm46_tunnel_handler, AF_INET); xfrm6_tunnel_deregister(&xfrm6_tunnel_handler, AF_INET6); -- cgit v1.2.3-70-g09d2 From 57dbb2d83d100ea601c54fe129bfde0678db5dee Mon Sep 17 00:00:00 2001 From: Hagen Paul Pfeifer Date: Sun, 24 Jan 2010 12:30:59 +0000 Subject: sched: add head drop fifo queue This adds an additional queuing strategy, called pfifo_head_drop, to remove the oldest skb in the case of an overflow within the queue - the head element - instead of the last skb (tail). To remove the oldest skb in congested situations is useful for sensor network environments where newer packets reflect the superior information. Reviewed-by: Florian Westphal Acked-by: Patrick McHardy Signed-off-by: Hagen Paul Pfeifer Signed-off-by: David S. Miller --- include/net/pkt_sched.h | 1 + include/net/sch_generic.h | 19 +++++++++++++++++++ net/sched/sch_api.c | 1 + net/sched/sch_fifo.c | 34 ++++++++++++++++++++++++++++++++++ 4 files changed, 55 insertions(+) (limited to 'include') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 2d567265363..b6cdc33b39c 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -71,6 +71,7 @@ extern void qdisc_watchdog_cancel(struct qdisc_watchdog *wd); extern struct Qdisc_ops pfifo_qdisc_ops; extern struct Qdisc_ops bfifo_qdisc_ops; +extern struct Qdisc_ops pfifo_head_drop_qdisc_ops; extern int fifo_set_limit(struct Qdisc *q, unsigned int limit); extern struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops, diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index dad558bc06f..67dc08eaaa4 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -427,6 +427,25 @@ static inline struct sk_buff *qdisc_dequeue_head(struct Qdisc *sch) return __qdisc_dequeue_head(sch, &sch->q); } +static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch, + struct sk_buff_head *list) +{ + struct sk_buff *skb = __qdisc_dequeue_head(sch, list); + + if (likely(skb != NULL)) { + unsigned int len = qdisc_pkt_len(skb); + kfree_skb(skb); + return len; + } + + return 0; +} + +static inline unsigned int qdisc_queue_drop_head(struct Qdisc *sch) +{ + return __qdisc_queue_drop_head(sch, &sch->q); +} + static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch, struct sk_buff_head *list) { diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 75fd1c672c6..6cd491013b5 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1707,6 +1707,7 @@ static int __init pktsched_init(void) { register_qdisc(&pfifo_qdisc_ops); register_qdisc(&bfifo_qdisc_ops); + register_qdisc(&pfifo_head_drop_qdisc_ops); register_qdisc(&mq_qdisc_ops); proc_net_fops_create(&init_net, "psched", 0, &psched_fops); diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 69188e8358b..4b0a6cc44c7 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -43,6 +43,26 @@ static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc* sch) return qdisc_reshape_fail(skb, sch); } +static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc* sch) +{ + struct sk_buff *skb_head; + struct fifo_sched_data *q = qdisc_priv(sch); + + if (likely(skb_queue_len(&sch->q) < q->limit)) + return qdisc_enqueue_tail(skb, sch); + + /* queue full, remove one skb to fulfill the limit */ + skb_head = qdisc_dequeue_head(sch); + sch->bstats.bytes -= qdisc_pkt_len(skb_head); + sch->bstats.packets--; + sch->qstats.drops++; + kfree_skb(skb_head); + + qdisc_enqueue_tail(skb, sch); + + return NET_XMIT_CN; +} + static int fifo_init(struct Qdisc *sch, struct nlattr *opt) { struct fifo_sched_data *q = qdisc_priv(sch); @@ -108,6 +128,20 @@ struct Qdisc_ops bfifo_qdisc_ops __read_mostly = { }; EXPORT_SYMBOL(bfifo_qdisc_ops); +struct Qdisc_ops pfifo_head_drop_qdisc_ops __read_mostly = { + .id = "pfifo_head_drop", + .priv_size = sizeof(struct fifo_sched_data), + .enqueue = pfifo_tail_enqueue, + .dequeue = qdisc_dequeue_head, + .peek = qdisc_peek_head, + .drop = qdisc_queue_drop_head, + .init = fifo_init, + .reset = qdisc_reset_queue, + .change = fifo_init, + .dump = fifo_dump, + .owner = THIS_MODULE, +}; + /* Pass size change message down to embedded FIFO */ int fifo_set_limit(struct Qdisc *q, unsigned int limit) { -- cgit v1.2.3-70-g09d2 From 09d989d179d0c679043556dda77c51b41a2dae7e Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Fri, 29 Jan 2010 19:58:57 -0500 Subject: cfg80211: add regulatory hint disconnect support This adds a new regulatory hint to be used when we know all devices have been disconnected and idle. This can happen when we suspend, for instance. When we disconnect we can no longer assume the same regulatory rules learned from a country IE or beacon hints are applicable so restore regulatory settings to an initial state. Since driver hints are cached on the wiphy that called the hint, those hints are not reproduced onto cfg80211 as the wiphy will respect its own wiphy->regd regardless. Signed-off-by: Luis R. Rodriguez Signed-off-by: John W. Linville --- include/net/regulatory.h | 1 + net/wireless/reg.c | 157 ++++++++++++++++++++++++++++++++++++++++++++++- net/wireless/reg.h | 18 ++++++ net/wireless/sme.c | 40 ++++++++++++ 4 files changed, 213 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/regulatory.h b/include/net/regulatory.h index 47995b81c5d..f873ee37f7e 100644 --- a/include/net/regulatory.h +++ b/include/net/regulatory.h @@ -39,6 +39,7 @@ enum environment_cap { * 00 - World regulatory domain * 99 - built by driver but a specific alpha2 cannot be determined * 98 - result of an intersection between two regulatory domains + * 97 - regulatory domain has not yet been configured * @intersect: indicates whether the wireless core should intersect * the requested regulatory domain with the presently set regulatory * domain. diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 5dcda28b6f0..ed89c59bb43 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -134,6 +134,7 @@ static const struct ieee80211_regdomain *cfg80211_world_regdom = &world_regdom; static char *ieee80211_regdom = "00"; +static char user_alpha2[2]; module_param(ieee80211_regdom, charp, 0444); MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain code"); @@ -252,6 +253,27 @@ static bool regdom_changes(const char *alpha2) return true; } +/* + * The NL80211_REGDOM_SET_BY_USER regdom alpha2 is cached, this lets + * you know if a valid regulatory hint with NL80211_REGDOM_SET_BY_USER + * has ever been issued. + */ +static bool is_user_regdom_saved(void) +{ + if (user_alpha2[0] == '9' && user_alpha2[1] == '7') + return false; + + /* This would indicate a mistake on the design */ + if (WARN((!is_world_regdom(user_alpha2) && + !is_an_alpha2(user_alpha2)), + "Unexpected user alpha2: %c%c\n", + user_alpha2[0], + user_alpha2[1])) + return false; + + return true; +} + /** * country_ie_integrity_changes - tells us if the country IE has changed * @checksum: checksum of country IE of fields we are interested in @@ -1646,7 +1668,7 @@ static int ignore_request(struct wiphy *wiphy, switch (pending_request->initiator) { case NL80211_REGDOM_SET_BY_CORE: - return -EINVAL; + return 0; case NL80211_REGDOM_SET_BY_COUNTRY_IE: last_wiphy = wiphy_idx_to_wiphy(last_request->wiphy_idx); @@ -1785,6 +1807,11 @@ new_request: pending_request = NULL; + if (last_request->initiator == NL80211_REGDOM_SET_BY_USER) { + user_alpha2[0] = last_request->alpha2[0]; + user_alpha2[1] = last_request->alpha2[1]; + } + /* When r == REG_INTERSECT we do need to call CRDA */ if (r < 0) { /* @@ -1904,12 +1931,16 @@ static void queue_regulatory_request(struct regulatory_request *request) schedule_work(®_work); } -/* Core regulatory hint -- happens once during cfg80211_init() */ +/* + * Core regulatory hint -- happens during cfg80211_init() + * and when we restore regulatory settings. + */ static int regulatory_hint_core(const char *alpha2) { struct regulatory_request *request; - BUG_ON(last_request); + kfree(last_request); + last_request = NULL; request = kzalloc(sizeof(struct regulatory_request), GFP_KERNEL); @@ -2107,6 +2138,123 @@ out: mutex_unlock(®_mutex); } +static void restore_alpha2(char *alpha2, bool reset_user) +{ + /* indicates there is no alpha2 to consider for restoration */ + alpha2[0] = '9'; + alpha2[1] = '7'; + + /* The user setting has precedence over the module parameter */ + if (is_user_regdom_saved()) { + /* Unless we're asked to ignore it and reset it */ + if (reset_user) { + REG_DBG_PRINT("cfg80211: Restoring regulatory settings " + "including user preference\n"); + user_alpha2[0] = '9'; + user_alpha2[1] = '7'; + + /* + * If we're ignoring user settings, we still need to + * check the module parameter to ensure we put things + * back as they were for a full restore. + */ + if (!is_world_regdom(ieee80211_regdom)) { + REG_DBG_PRINT("cfg80211: Keeping preference on " + "module parameter ieee80211_regdom: %c%c\n", + ieee80211_regdom[0], + ieee80211_regdom[1]); + alpha2[0] = ieee80211_regdom[0]; + alpha2[1] = ieee80211_regdom[1]; + } + } else { + REG_DBG_PRINT("cfg80211: Restoring regulatory settings " + "while preserving user preference for: %c%c\n", + user_alpha2[0], + user_alpha2[1]); + alpha2[0] = user_alpha2[0]; + alpha2[1] = user_alpha2[1]; + } + } else if (!is_world_regdom(ieee80211_regdom)) { + REG_DBG_PRINT("cfg80211: Keeping preference on " + "module parameter ieee80211_regdom: %c%c\n", + ieee80211_regdom[0], + ieee80211_regdom[1]); + alpha2[0] = ieee80211_regdom[0]; + alpha2[1] = ieee80211_regdom[1]; + } else + REG_DBG_PRINT("cfg80211: Restoring regulatory settings\n"); +} + +/* + * Restoring regulatory settings involves ingoring any + * possibly stale country IE information and user regulatory + * settings if so desired, this includes any beacon hints + * learned as we could have traveled outside to another country + * after disconnection. To restore regulatory settings we do + * exactly what we did at bootup: + * + * - send a core regulatory hint + * - send a user regulatory hint if applicable + * + * Device drivers that send a regulatory hint for a specific country + * keep their own regulatory domain on wiphy->regd so that does does + * not need to be remembered. + */ +static void restore_regulatory_settings(bool reset_user) +{ + char alpha2[2]; + struct reg_beacon *reg_beacon, *btmp; + + mutex_lock(&cfg80211_mutex); + mutex_lock(®_mutex); + + reset_regdomains(); + restore_alpha2(alpha2, reset_user); + + /* Clear beacon hints */ + spin_lock_bh(®_pending_beacons_lock); + if (!list_empty(®_pending_beacons)) { + list_for_each_entry_safe(reg_beacon, btmp, + ®_pending_beacons, list) { + list_del(®_beacon->list); + kfree(reg_beacon); + } + } + spin_unlock_bh(®_pending_beacons_lock); + + if (!list_empty(®_beacon_list)) { + list_for_each_entry_safe(reg_beacon, btmp, + ®_beacon_list, list) { + list_del(®_beacon->list); + kfree(reg_beacon); + } + } + + /* First restore to the basic regulatory settings */ + cfg80211_regdomain = cfg80211_world_regdom; + + mutex_unlock(®_mutex); + mutex_unlock(&cfg80211_mutex); + + regulatory_hint_core(cfg80211_regdomain->alpha2); + + /* + * This restores the ieee80211_regdom module parameter + * preference or the last user requested regulatory + * settings, user regulatory settings takes precedence. + */ + if (is_an_alpha2(alpha2)) + regulatory_hint_user(user_alpha2); +} + + +void regulatory_hint_disconnect(void) +{ + REG_DBG_PRINT("cfg80211: All devices are disconnected, going to " + "restore regulatory settings\n"); + restore_regulatory_settings(false); +} + static bool freq_is_chan_12_13_14(u16 freq) { if (freq == ieee80211_channel_to_frequency(12) || @@ -2496,6 +2644,9 @@ int regulatory_init(void) cfg80211_regdomain = cfg80211_world_regdom; + user_alpha2[0] = '9'; + user_alpha2[1] = '7'; + /* We always try to get an update for the static regdomain */ err = regulatory_hint_core(cfg80211_regdomain->alpha2); if (err) { diff --git a/net/wireless/reg.h b/net/wireless/reg.h index 3018508226a..b26224a9f3b 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -63,4 +63,22 @@ void regulatory_hint_11d(struct wiphy *wiphy, u8 *country_ie, u8 country_ie_len); +/** + * regulatory_hint_disconnect - informs all devices have been disconneted + * + * Regulotory rules can be enhanced further upon scanning and upon + * connection to an AP. These rules become stale if we disconnect + * and go to another country, whether or not we suspend and resume. + * If we suspend, go to another country and resume we'll automatically + * get disconnected shortly after resuming and things will be reset as well. + * This routine is a helper to restore regulatory settings to how they were + * prior to our first connect attempt. This includes ignoring country IE and + * beacon regulatory hints. The ieee80211_regdom module parameter will always + * be respected but if a user had set the regulatory domain that will take + * precedence. + * + * Must be called from process context. + */ +void regulatory_hint_disconnect(void); + #endif /* __NET_WIRELESS_REG_H */ diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 745c37e7992..17fde0da1b0 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -34,6 +34,44 @@ struct cfg80211_conn { bool auto_auth, prev_bssid_valid; }; +bool cfg80211_is_all_idle(void) +{ + struct cfg80211_registered_device *rdev; + struct wireless_dev *wdev; + bool is_all_idle = true; + + mutex_lock(&cfg80211_mutex); + + /* + * All devices must be idle as otherwise if you are actively + * scanning some new beacon hints could be learned and would + * count as new regulatory hints. + */ + list_for_each_entry(rdev, &cfg80211_rdev_list, list) { + cfg80211_lock_rdev(rdev); + list_for_each_entry(wdev, &rdev->netdev_list, list) { + wdev_lock(wdev); + if (wdev->sme_state != CFG80211_SME_IDLE) + is_all_idle = false; + wdev_unlock(wdev); + } + cfg80211_unlock_rdev(rdev); + } + + mutex_unlock(&cfg80211_mutex); + + return is_all_idle; +} + +static void disconnect_work(struct work_struct *work) +{ + if (!cfg80211_is_all_idle()) + return; + + regulatory_hint_disconnect(); +} + +static DECLARE_WORK(cfg80211_disconnect_work, disconnect_work); static int cfg80211_conn_scan(struct wireless_dev *wdev) { @@ -658,6 +696,8 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, wireless_send_event(dev, SIOCGIWAP, &wrqu, NULL); wdev->wext.connect.ssid_len = 0; #endif + + schedule_work(&cfg80211_disconnect_work); } void cfg80211_disconnected(struct net_device *dev, u16 reason, -- cgit v1.2.3-70-g09d2 From 17ad353b8d9843731258b5d23556667b764939e9 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sun, 31 Jan 2010 21:56:25 +0100 Subject: mac80211: fix monitor mode tx radiotap header handling When an injected frame gets buffered for a powersave STA or filtered and retransmitted, mac80211 attempts to parse the radiotap header again, which doesn't work because it's gone at that point. This patch adds a new flag for checking the availability of a radiotap header, so that it only attempts to parse it once, reusing the tx info on the next call to ieee80211_tx(). This fixes severe issues with rekeying in AP mode. Signed-off-by: Felix Fietkau Cc: stable@kernel.org Signed-off-by: John W. Linville --- include/net/mac80211.h | 3 +++ net/mac80211/tx.c | 6 ++++-- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 1e9c93024cf..74ccf30fdf8 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -275,6 +275,8 @@ struct ieee80211_bss_conf { * @IEEE80211_TX_INTFL_RETRANSMISSION: This frame is being retransmitted * after TX status because the destination was asleep, it must not * be modified again (no seqno assignment, crypto, etc.) + * @IEEE80211_TX_INTFL_HAS_RADIOTAP: This frame was injected and still + * has a radiotap header at skb->data. */ enum mac80211_tx_control_flags { IEEE80211_TX_CTL_REQ_TX_STATUS = BIT(0), @@ -296,6 +298,7 @@ enum mac80211_tx_control_flags { IEEE80211_TX_CTL_PSPOLL_RESPONSE = BIT(17), IEEE80211_TX_CTL_MORE_FRAMES = BIT(18), IEEE80211_TX_INTFL_RETRANSMISSION = BIT(19), + IEEE80211_TX_INTFL_HAS_RADIOTAP = BIT(20), }; /** diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 14c70452c24..e7b1cdc7651 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1108,7 +1108,7 @@ ieee80211_tx_prepare(struct ieee80211_sub_if_data *sdata, tx->flags |= IEEE80211_TX_FRAGMENTED; /* process and remove the injection radiotap header */ - if (unlikely(info->flags & IEEE80211_TX_CTL_INJECTED)) { + if (unlikely(info->flags & IEEE80211_TX_INTFL_HAS_RADIOTAP)) { if (!__ieee80211_parse_tx_radiotap(tx, skb)) return TX_DROP; @@ -1117,6 +1117,7 @@ ieee80211_tx_prepare(struct ieee80211_sub_if_data *sdata, * the radiotap header that was present and pre-filled * 'tx' with tx control information. */ + info->flags &= ~IEEE80211_TX_INTFL_HAS_RADIOTAP; } /* @@ -1499,7 +1500,8 @@ static void ieee80211_xmit(struct ieee80211_sub_if_data *sdata, int hdrlen; u16 len_rthdr; - info->flags |= IEEE80211_TX_CTL_INJECTED; + info->flags |= IEEE80211_TX_CTL_INJECTED | + IEEE80211_TX_INTFL_HAS_RADIOTAP; len_rthdr = ieee80211_get_radiotap_len(skb->data); hdr = (struct ieee80211_hdr *)(skb->data + len_rthdr); -- cgit v1.2.3-70-g09d2 From c30f540b63047437ffa894b5353216410c480d1a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 2 Feb 2010 15:03:24 +0100 Subject: netfilter: xtables: CONFIG_COMPAT redux Ifdef out struct nf_sockopt_ops::compat_set struct nf_sockopt_ops::compat_get struct xt_match::compat_from_user struct xt_match::compat_to_user struct xt_match::compatsize to make structures smaller on COMPAT=n kernels. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 9 ++++++--- include/linux/netfilter/x_tables.h | 12 ++++++++---- 2 files changed, 14 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 48c54960773..78f33d22368 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -114,15 +114,17 @@ struct nf_sockopt_ops { int set_optmin; int set_optmax; int (*set)(struct sock *sk, int optval, void __user *user, unsigned int len); +#ifdef CONFIG_COMPAT int (*compat_set)(struct sock *sk, int optval, void __user *user, unsigned int len); - +#endif int get_optmin; int get_optmax; int (*get)(struct sock *sk, int optval, void __user *user, int *len); +#ifdef CONFIG_COMPAT int (*compat_get)(struct sock *sk, int optval, void __user *user, int *len); - +#endif /* Use the module struct to lock set/get code in place */ struct module *owner; }; @@ -222,11 +224,12 @@ int nf_setsockopt(struct sock *sk, u_int8_t pf, int optval, char __user *opt, unsigned int len); int nf_getsockopt(struct sock *sk, u_int8_t pf, int optval, char __user *opt, int *len); - +#ifdef CONFIG_COMPAT int compat_nf_setsockopt(struct sock *sk, u_int8_t pf, int optval, char __user *opt, unsigned int len); int compat_nf_getsockopt(struct sock *sk, u_int8_t pf, int optval, char __user *opt, int *len); +#endif /* Call this before modifying an existing packet: ensures it is modifiable and linear to the point you care about (writable_len). diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 3caf5e15110..026eb78ee83 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -283,11 +283,11 @@ struct xt_match { /* Called when entry of this type deleted. */ void (*destroy)(const struct xt_mtdtor_param *); - +#ifdef CONFIG_COMPAT /* Called when userspace align differs from kernel space one */ void (*compat_from_user)(void *dst, void *src); int (*compat_to_user)(void __user *dst, void *src); - +#endif /* Set this to THIS_MODULE if you are a module, otherwise NULL */ struct module *me; @@ -296,7 +296,9 @@ struct xt_match { const char *table; unsigned int matchsize; +#ifdef CONFIG_COMPAT unsigned int compatsize; +#endif unsigned int hooks; unsigned short proto; @@ -323,17 +325,19 @@ struct xt_target { /* Called when entry of this type deleted. */ void (*destroy)(const struct xt_tgdtor_param *); - +#ifdef CONFIG_COMPAT /* Called when userspace align differs from kernel space one */ void (*compat_from_user)(void *dst, void *src); int (*compat_to_user)(void __user *dst, void *src); - +#endif /* Set this to THIS_MODULE if you are a module, otherwise NULL */ struct module *me; const char *table; unsigned int targetsize; +#ifdef CONFIG_COMPAT unsigned int compatsize; +#endif unsigned int hooks; unsigned short proto; -- cgit v1.2.3-70-g09d2 From c85bb41e93184bf5494dde6d8fe5a81b564c84c8 Mon Sep 17 00:00:00 2001 From: Flavio Leitner Date: Tue, 2 Feb 2010 07:32:29 -0800 Subject: igmp: fix ip_mc_sf_allow race [v5] Almost all igmp functions accessing inet->mc_list are protected by rtnl_lock(), but there is one exception which is ip_mc_sf_allow(), so there is a chance of either ip_mc_drop_socket or ip_mc_leave_group remove an entry while ip_mc_sf_allow is running causing a crash. Signed-off-by: Flavio Leitner Signed-off-by: David S. Miller --- include/linux/igmp.h | 2 ++ net/ipv4/igmp.c | 83 +++++++++++++++++++++++++++++++++++++++------------- 2 files changed, 64 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 724c27e5d17..93fc2449af1 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -153,6 +153,7 @@ extern int sysctl_igmp_max_msf; struct ip_sf_socklist { unsigned int sl_max; unsigned int sl_count; + struct rcu_head rcu; __be32 sl_addr[0]; }; @@ -170,6 +171,7 @@ struct ip_mc_socklist { struct ip_mreqn multi; unsigned int sfmode; /* MCAST_{INCLUDE,EXCLUDE} */ struct ip_sf_socklist *sflist; + struct rcu_head rcu; }; struct ip_sf_list { diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 8f5468393f0..d2836399874 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1799,7 +1799,7 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr) iml->next = inet->mc_list; iml->sflist = NULL; iml->sfmode = MCAST_EXCLUDE; - inet->mc_list = iml; + rcu_assign_pointer(inet->mc_list, iml); ip_mc_inc_group(in_dev, addr); err = 0; done: @@ -1807,24 +1807,46 @@ done: return err; } +static void ip_sf_socklist_reclaim(struct rcu_head *rp) +{ + struct ip_sf_socklist *psf; + + psf = container_of(rp, struct ip_sf_socklist, rcu); + /* sk_omem_alloc should have been decreased by the caller*/ + kfree(psf); +} + static int ip_mc_leave_src(struct sock *sk, struct ip_mc_socklist *iml, struct in_device *in_dev) { + struct ip_sf_socklist *psf = iml->sflist; int err; - if (iml->sflist == NULL) { + if (psf == NULL) { /* any-source empty exclude case */ return ip_mc_del_src(in_dev, &iml->multi.imr_multiaddr.s_addr, iml->sfmode, 0, NULL, 0); } err = ip_mc_del_src(in_dev, &iml->multi.imr_multiaddr.s_addr, - iml->sfmode, iml->sflist->sl_count, - iml->sflist->sl_addr, 0); - sock_kfree_s(sk, iml->sflist, IP_SFLSIZE(iml->sflist->sl_max)); - iml->sflist = NULL; + iml->sfmode, psf->sl_count, psf->sl_addr, 0); + rcu_assign_pointer(iml->sflist, NULL); + /* decrease mem now to avoid the memleak warning */ + atomic_sub(IP_SFLSIZE(psf->sl_max), &sk->sk_omem_alloc); + call_rcu(&psf->rcu, ip_sf_socklist_reclaim); return err; } + +static void ip_mc_socklist_reclaim(struct rcu_head *rp) +{ + struct ip_mc_socklist *iml; + + iml = container_of(rp, struct ip_mc_socklist, rcu); + /* sk_omem_alloc should have been decreased by the caller*/ + kfree(iml); +} + + /* * Ask a socket to leave a group. */ @@ -1854,12 +1876,14 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr) (void) ip_mc_leave_src(sk, iml, in_dev); - *imlp = iml->next; + rcu_assign_pointer(*imlp, iml->next); if (in_dev) ip_mc_dec_group(in_dev, group); rtnl_unlock(); - sock_kfree_s(sk, iml, sizeof(*iml)); + /* decrease mem now to avoid the memleak warning */ + atomic_sub(sizeof(*iml), &sk->sk_omem_alloc); + call_rcu(&iml->rcu, ip_mc_socklist_reclaim); return 0; } if (!in_dev) @@ -1974,9 +1998,12 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct if (psl) { for (i=0; isl_count; i++) newpsl->sl_addr[i] = psl->sl_addr[i]; - sock_kfree_s(sk, psl, IP_SFLSIZE(psl->sl_max)); + /* decrease mem now to avoid the memleak warning */ + atomic_sub(IP_SFLSIZE(psl->sl_max), &sk->sk_omem_alloc); + call_rcu(&psl->rcu, ip_sf_socklist_reclaim); } - pmc->sflist = psl = newpsl; + rcu_assign_pointer(pmc->sflist, newpsl); + psl = newpsl; } rv = 1; /* > 0 for insert logic below if sl_count is 0 */ for (i=0; isl_count; i++) { @@ -2072,11 +2099,13 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) if (psl) { (void) ip_mc_del_src(in_dev, &msf->imsf_multiaddr, pmc->sfmode, psl->sl_count, psl->sl_addr, 0); - sock_kfree_s(sk, psl, IP_SFLSIZE(psl->sl_max)); + /* decrease mem now to avoid the memleak warning */ + atomic_sub(IP_SFLSIZE(psl->sl_max), &sk->sk_omem_alloc); + call_rcu(&psl->rcu, ip_sf_socklist_reclaim); } else (void) ip_mc_del_src(in_dev, &msf->imsf_multiaddr, pmc->sfmode, 0, NULL, 0); - pmc->sflist = newpsl; + rcu_assign_pointer(pmc->sflist, newpsl); pmc->sfmode = msf->imsf_fmode; err = 0; done: @@ -2209,30 +2238,40 @@ int ip_mc_sf_allow(struct sock *sk, __be32 loc_addr, __be32 rmt_addr, int dif) struct ip_mc_socklist *pmc; struct ip_sf_socklist *psl; int i; + int ret; + ret = 1; if (!ipv4_is_multicast(loc_addr)) - return 1; + goto out; - for (pmc=inet->mc_list; pmc; pmc=pmc->next) { + rcu_read_lock(); + for (pmc=rcu_dereference(inet->mc_list); pmc; pmc=rcu_dereference(pmc->next)) { if (pmc->multi.imr_multiaddr.s_addr == loc_addr && pmc->multi.imr_ifindex == dif) break; } + ret = inet->mc_all; if (!pmc) - return inet->mc_all; + goto unlock; psl = pmc->sflist; + ret = (pmc->sfmode == MCAST_EXCLUDE); if (!psl) - return pmc->sfmode == MCAST_EXCLUDE; + goto unlock; for (i=0; isl_count; i++) { if (psl->sl_addr[i] == rmt_addr) break; } + ret = 0; if (pmc->sfmode == MCAST_INCLUDE && i >= psl->sl_count) - return 0; + goto unlock; if (pmc->sfmode == MCAST_EXCLUDE && i < psl->sl_count) - return 0; - return 1; + goto unlock; + ret = 1; +unlock: + rcu_read_unlock(); +out: + return ret; } /* @@ -2251,7 +2290,7 @@ void ip_mc_drop_socket(struct sock *sk) rtnl_lock(); while ((iml = inet->mc_list) != NULL) { struct in_device *in_dev; - inet->mc_list = iml->next; + rcu_assign_pointer(inet->mc_list, iml->next); in_dev = inetdev_by_index(net, iml->multi.imr_ifindex); (void) ip_mc_leave_src(sk, iml, in_dev); @@ -2259,7 +2298,9 @@ void ip_mc_drop_socket(struct sock *sk) ip_mc_dec_group(in_dev, iml->multi.imr_multiaddr.s_addr); in_dev_put(in_dev); } - sock_kfree_s(sk, iml, sizeof(*iml)); + /* decrease mem now to avoid the memleak warning */ + atomic_sub(sizeof(*iml), &sk->sk_omem_alloc); + call_rcu(&iml->rcu, ip_mc_socklist_reclaim); } rtnl_unlock(); } -- cgit v1.2.3-70-g09d2 From f9bfbebf34eab707b065116cdc9699d25ba4252a Mon Sep 17 00:00:00 2001 From: Shirley Ma Date: Fri, 29 Jan 2010 03:19:05 +0000 Subject: virtio: Add ability to detach unused buffers from vrings There's currently no way for a virtio driver to ask for unused buffers, so it has to keep a list itself to reclaim them at shutdown. This is redundant, since virtio_ring stores that information. So add a new hook to do this. Signed-off-by: Shirley Ma Signed-off-by: Amit Shah Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- drivers/virtio/virtio_ring.c | 25 +++++++++++++++++++++++++ include/linux/virtio.h | 4 ++++ 2 files changed, 29 insertions(+) (limited to 'include') diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index fbd2ecde93e..71929ee00d6 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -334,6 +334,30 @@ static bool vring_enable_cb(struct virtqueue *_vq) return true; } +static void *vring_detach_unused_buf(struct virtqueue *_vq) +{ + struct vring_virtqueue *vq = to_vvq(_vq); + unsigned int i; + void *buf; + + START_USE(vq); + + for (i = 0; i < vq->vring.num; i++) { + if (!vq->data[i]) + continue; + /* detach_buf clears data, so grab it now. */ + buf = vq->data[i]; + detach_buf(vq, i); + END_USE(vq); + return buf; + } + /* That should have freed everything. */ + BUG_ON(vq->num_free != vq->vring.num); + + END_USE(vq); + return NULL; +} + irqreturn_t vring_interrupt(int irq, void *_vq) { struct vring_virtqueue *vq = to_vvq(_vq); @@ -360,6 +384,7 @@ static struct virtqueue_ops vring_vq_ops = { .kick = vring_kick, .disable_cb = vring_disable_cb, .enable_cb = vring_enable_cb, + .detach_unused_buf = vring_detach_unused_buf, }; struct virtqueue *vring_new_virtqueue(unsigned int num, diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 057a2e01075..f508c651e53 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -51,6 +51,9 @@ struct virtqueue { * This re-enables callbacks; it returns "false" if there are pending * buffers in the queue, to detect a possible race between the driver * checking for more work, and enabling callbacks. + * @detach_unused_buf: detach first unused buffer + * vq: the struct virtqueue we're talking about. + * Returns NULL or the "data" token handed to add_buf * * Locking rules are straightforward: the driver is responsible for * locking. No two operations may be invoked simultaneously, with the exception @@ -71,6 +74,7 @@ struct virtqueue_ops { void (*disable_cb)(struct virtqueue *vq); bool (*enable_cb)(struct virtqueue *vq); + void *(*detach_unused_buf)(struct virtqueue *vq); }; /** -- cgit v1.2.3-70-g09d2 From 794e68716bab578ae8f8912dc934496d7c7abc90 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 13:41:29 +0100 Subject: netfilter: ctnetlink: only assign helpers for matching protocols Make sure not to assign a helper for a different network or transport layer protocol to a connection. Additionally change expectation deletion by helper to compare the name directly - there might be multiple helper registrations using the same name, currently one of them is chosen in an unpredictable manner and only those expectations are removed. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_helper.h | 2 +- net/netfilter/nf_conntrack_helper.c | 8 +++++--- net/netfilter/nf_conntrack_netlink.c | 23 +++++++++++------------ 3 files changed, 17 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index d015de92e03..86be7c4816d 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -40,7 +40,7 @@ struct nf_conntrack_helper { }; extern struct nf_conntrack_helper * -__nf_conntrack_helper_find_byname(const char *name); +__nf_conntrack_helper_find(const char *name, u16 l3num, u8 protonum); extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 65c2a7bc3af..c0e461f466a 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -65,7 +65,7 @@ __nf_ct_helper_find(const struct nf_conntrack_tuple *tuple) } struct nf_conntrack_helper * -__nf_conntrack_helper_find_byname(const char *name) +__nf_conntrack_helper_find(const char *name, u16 l3num, u8 protonum) { struct nf_conntrack_helper *h; struct hlist_node *n; @@ -73,13 +73,15 @@ __nf_conntrack_helper_find_byname(const char *name) for (i = 0; i < nf_ct_helper_hsize; i++) { hlist_for_each_entry_rcu(h, n, &nf_ct_helper_hash[i], hnode) { - if (!strcmp(h->name, name)) + if (!strcmp(h->name, name) && + h->tuple.src.l3num == l3num && + h->tuple.dst.protonum == protonum) return h; } } return NULL; } -EXPORT_SYMBOL_GPL(__nf_conntrack_helper_find_byname); +EXPORT_SYMBOL_GPL(__nf_conntrack_helper_find); struct nf_conn_help *nf_ct_helper_ext_add(struct nf_conn *ct, gfp_t gfp) { diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 79478dfba27..16f86d61e5d 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1001,7 +1001,8 @@ ctnetlink_change_helper(struct nf_conn *ct, const struct nlattr * const cda[]) return 0; } - helper = __nf_conntrack_helper_find_byname(helpname); + helper = __nf_conntrack_helper_find(helpname, nf_ct_l3num(ct), + nf_ct_protonum(ct)); if (helper == NULL) { #ifdef CONFIG_MODULES spin_unlock_bh(&nf_conntrack_lock); @@ -1012,7 +1013,8 @@ ctnetlink_change_helper(struct nf_conn *ct, const struct nlattr * const cda[]) } spin_lock_bh(&nf_conntrack_lock); - helper = __nf_conntrack_helper_find_byname(helpname); + helper = __nf_conntrack_helper_find(helpname, nf_ct_l3num(ct), + nf_ct_protonum(ct)); if (helper) return -EAGAIN; #endif @@ -1211,7 +1213,8 @@ ctnetlink_create_conntrack(struct net *net, if (err < 0) goto err2; - helper = __nf_conntrack_helper_find_byname(helpname); + helper = __nf_conntrack_helper_find(helpname, nf_ct_l3num(ct), + nf_ct_protonum(ct)); if (helper == NULL) { rcu_read_unlock(); #ifdef CONFIG_MODULES @@ -1221,7 +1224,9 @@ ctnetlink_create_conntrack(struct net *net, } rcu_read_lock(); - helper = __nf_conntrack_helper_find_byname(helpname); + helper = __nf_conntrack_helper_find(helpname, + nf_ct_l3num(ct), + nf_ct_protonum(ct)); if (helper) { err = -EAGAIN; goto err2; @@ -1714,7 +1719,6 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, struct net *net = sock_net(ctnl); struct nf_conntrack_expect *exp; struct nf_conntrack_tuple tuple; - struct nf_conntrack_helper *h; struct nfgenmsg *nfmsg = nlmsg_data(nlh); struct hlist_node *n, *next; u_int8_t u3 = nfmsg->nfgen_family; @@ -1751,18 +1755,13 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, /* delete all expectations for this helper */ spin_lock_bh(&nf_conntrack_lock); - h = __nf_conntrack_helper_find_byname(name); - if (!h) { - spin_unlock_bh(&nf_conntrack_lock); - return -EOPNOTSUPP; - } for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, &net->ct.expect_hash[i], hnode) { m_help = nfct_help(exp->master); - if (m_help->helper == h - && del_timer(&exp->timeout)) { + if (!strcmp(m_help->helper->name, name) && + del_timer(&exp->timeout)) { nf_ct_unlink_expect(exp); nf_ct_expect_put(exp); } -- cgit v1.2.3-70-g09d2 From add67461240c1dadc7c8d97e66f8f92b556ca523 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 13:45:12 +0100 Subject: netfilter: add struct net * to target parameters Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 2 ++ net/bridge/netfilter/ebtables.c | 10 ++++++---- net/ipv4/netfilter/ip_tables.c | 8 +++++--- net/ipv6/netfilter/ip6_tables.c | 8 +++++--- 4 files changed, 18 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 026eb78ee83..365fabe1b16 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -249,6 +249,7 @@ struct xt_target_param { * Other fields see above. */ struct xt_tgchk_param { + struct net *net; const char *table; const void *entryinfo; const struct xt_target *target; @@ -259,6 +260,7 @@ struct xt_tgchk_param { /* Target destructor parameters */ struct xt_tgdtor_param { + struct net *net; const struct xt_target *target; void *targinfo; u_int8_t family; diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 1aa0e4c1f52..12beb580aa2 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -579,13 +579,14 @@ ebt_cleanup_match(struct ebt_entry_match *m, struct net *net, unsigned int *i) } static inline int -ebt_cleanup_watcher(struct ebt_entry_watcher *w, unsigned int *i) +ebt_cleanup_watcher(struct ebt_entry_watcher *w, struct net *net, unsigned int *i) { struct xt_tgdtor_param par; if (i && (*i)-- == 0) return 1; + par.net = net; par.target = w->u.watcher; par.targinfo = w->data; par.family = NFPROTO_BRIDGE; @@ -606,10 +607,11 @@ ebt_cleanup_entry(struct ebt_entry *e, struct net *net, unsigned int *cnt) /* we're done */ if (cnt && (*cnt)-- == 0) return 1; - EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, NULL); + EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, net, NULL); EBT_MATCH_ITERATE(e, ebt_cleanup_match, net, NULL); t = (struct ebt_entry_target *)(((char *)e) + e->target_offset); + par.net = net; par.target = t->u.target; par.targinfo = t->data; par.family = NFPROTO_BRIDGE; @@ -674,7 +676,7 @@ ebt_check_entry(struct ebt_entry *e, } i = 0; - mtpar.net = net; + mtpar.net = tgpar.net = net; mtpar.table = tgpar.table = name; mtpar.entryinfo = tgpar.entryinfo = e; mtpar.hook_mask = tgpar.hook_mask = hookmask; @@ -730,7 +732,7 @@ ebt_check_entry(struct ebt_entry *e, (*cnt)++; return 0; cleanup_watchers: - EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, &j); + EBT_WATCHER_ITERATE(e, ebt_cleanup_watcher, net, &j); cleanup_matches: EBT_MATCH_ITERATE(e, ebt_cleanup_match, net, &i); return ret; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index cfaba0e2e6f..7fde8f6950d 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -638,10 +638,11 @@ err: return ret; } -static int check_target(struct ipt_entry *e, const char *name) +static int check_target(struct ipt_entry *e, struct net *net, const char *name) { struct ipt_entry_target *t = ipt_get_target(e); struct xt_tgchk_param par = { + .net = net, .table = name, .entryinfo = e, .target = t->u.kernel.target, @@ -697,7 +698,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, } t->u.kernel.target = target; - ret = check_target(e, name); + ret = check_target(e, net, name); if (ret) goto err; @@ -788,6 +789,7 @@ cleanup_entry(struct ipt_entry *e, struct net *net, unsigned int *i) IPT_MATCH_ITERATE(e, cleanup_match, net, NULL); t = ipt_get_target(e); + par.net = net; par.target = t->u.kernel.target; par.targinfo = t->data; par.family = NFPROTO_IPV4; @@ -1675,7 +1677,7 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name, if (ret) goto cleanup_matches; - ret = check_target(e, name); + ret = check_target(e, net, name); if (ret) goto cleanup_matches; diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 9f1d45f2ba8..0376ed6d559 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -669,10 +669,11 @@ err: return ret; } -static int check_target(struct ip6t_entry *e, const char *name) +static int check_target(struct ip6t_entry *e, struct net *net, const char *name) { struct ip6t_entry_target *t = ip6t_get_target(e); struct xt_tgchk_param par = { + .net = net, .table = name, .entryinfo = e, .target = t->u.kernel.target, @@ -729,7 +730,7 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, } t->u.kernel.target = target; - ret = check_target(e, name); + ret = check_target(e, net, name); if (ret) goto err; @@ -820,6 +821,7 @@ cleanup_entry(struct ip6t_entry *e, struct net *net, unsigned int *i) IP6T_MATCH_ITERATE(e, cleanup_match, net, NULL); t = ip6t_get_target(e); + par.net = net; par.target = t->u.kernel.target; par.targinfo = t->data; par.family = NFPROTO_IPV6; @@ -1710,7 +1712,7 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, if (ret) goto cleanup_matches; - ret = check_target(e, name); + ret = check_target(e, net, name); if (ret) goto cleanup_matches; -- cgit v1.2.3-70-g09d2 From 858b31330054a9ad259feceea0ad1ce5385c47f0 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 13:48:53 +0100 Subject: netfilter: nf_conntrack: split up IPCT_STATUS event Split up the IPCT_STATUS event into an IPCT_REPLY event, which is generated when the IPS_SEEN_REPLY bit is set, and an IPCT_ASSURED event, which is generated when the IPS_ASSURED bit is set. In combination with a following patch to support selective event delivery, this can be used for "sparse" conntrack replication: start replicating the conntrack entry after it reached the ASSURED state and that way it's SYN-flood resistant. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_ecache.h | 21 +++++++++++---------- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 6 ++++-- net/netfilter/nf_conntrack_proto_gre.c | 2 +- net/netfilter/nf_conntrack_proto_sctp.c | 2 +- net/netfilter/nf_conntrack_proto_tcp.c | 2 +- net/netfilter/nf_conntrack_proto_udp.c | 2 +- net/netfilter/nf_conntrack_proto_udplite.c | 2 +- 8 files changed, 21 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index 475facc3051..5e05fb883ab 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -14,19 +14,20 @@ /* Connection tracking event types */ enum ip_conntrack_events { - IPCT_NEW = 0, /* new conntrack */ - IPCT_RELATED = 1, /* related conntrack */ - IPCT_DESTROY = 2, /* destroyed conntrack */ - IPCT_STATUS = 3, /* status has changed */ - IPCT_PROTOINFO = 4, /* protocol information has changed */ - IPCT_HELPER = 5, /* new helper has been set */ - IPCT_MARK = 6, /* new mark has been set */ - IPCT_NATSEQADJ = 7, /* NAT is doing sequence adjustment */ - IPCT_SECMARK = 8, /* new security mark has been set */ + IPCT_NEW, /* new conntrack */ + IPCT_RELATED, /* related conntrack */ + IPCT_DESTROY, /* destroyed conntrack */ + IPCT_REPLY, /* connection has seen two-way traffic */ + IPCT_ASSURED, /* connection status has changed to assured */ + IPCT_PROTOINFO, /* protocol information has changed */ + IPCT_HELPER, /* new helper has been set */ + IPCT_MARK, /* new mark has been set */ + IPCT_NATSEQADJ, /* NAT is doing sequence adjustment */ + IPCT_SECMARK, /* new security mark has been set */ }; enum ip_conntrack_expect_events { - IPEXP_NEW = 0, /* new expectation */ + IPEXP_NEW, /* new expectation */ }; struct nf_conntrack_ecache { diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 0e98c3282d4..091ff770eb7 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -825,7 +825,7 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, } if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_REPLY, ct); return ret; } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 16f86d61e5d..ff594eb138c 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1371,7 +1371,8 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, else events = IPCT_NEW; - nf_conntrack_eventmask_report((1 << IPCT_STATUS) | + nf_conntrack_eventmask_report((1 << IPCT_REPLY) | + (1 << IPCT_ASSURED) | (1 << IPCT_HELPER) | (1 << IPCT_PROTOINFO) | (1 << IPCT_NATSEQADJ) | @@ -1396,7 +1397,8 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err == 0) { nf_conntrack_get(&ct->ct_general); spin_unlock_bh(&nf_conntrack_lock); - nf_conntrack_eventmask_report((1 << IPCT_STATUS) | + nf_conntrack_eventmask_report((1 << IPCT_REPLY) | + (1 << IPCT_ASSURED) | (1 << IPCT_HELPER) | (1 << IPCT_PROTOINFO) | (1 << IPCT_NATSEQADJ) | diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index c99cfba64dd..d899b1a6994 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -241,7 +241,7 @@ static int gre_packet(struct nf_conn *ct, ct->proto.gre.stream_timeout); /* Also, more likely to be important, and not a probe. */ set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_ASSURED, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, ct->proto.gre.timeout); diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index f9d930f8027..b68ff15ed97 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -377,7 +377,7 @@ static int sctp_packet(struct nf_conn *ct, new_state == SCTP_CONNTRACK_ESTABLISHED) { pr_debug("Setting assured bit\n"); set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_ASSURED, ct); } return NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 3c96437b45a..ad118053971 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1045,7 +1045,7 @@ static int tcp_packet(struct nf_conn *ct, after SYN_RECV or a valid answer for a picked up connection. */ set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_ASSURED, ct); } nf_ct_refresh_acct(ct, ctinfo, skb, timeout); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index 5c5518bedb4..8d38f9a4bed 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -77,7 +77,7 @@ static int udp_packet(struct nf_conn *ct, nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udp_timeout_stream); /* Also, more likely to be important, and not a probe */ if (!test_and_set_bit(IPS_ASSURED_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_ASSURED, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udp_timeout); diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c index 458655bb210..0b1bc9ba667 100644 --- a/net/netfilter/nf_conntrack_proto_udplite.c +++ b/net/netfilter/nf_conntrack_proto_udplite.c @@ -75,7 +75,7 @@ static int udplite_packet(struct nf_conn *ct, nf_ct_udplite_timeout_stream); /* Also, more likely to be important, and not a probe */ if (!test_and_set_bit(IPS_ASSURED_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, ct); + nf_conntrack_event_cache(IPCT_ASSURED, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udplite_timeout); -- cgit v1.2.3-70-g09d2 From 0cebe4b4163b6373c9d24c1a192939777bc27e55 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 13:51:51 +0100 Subject: netfilter: ctnetlink: support selective event delivery Add two masks for conntrack end expectation events to struct nf_conntrack_ecache and use them to filter events. Their default value is "all events" when the event sysctl is on and "no events" when it is off. A following patch will add specific initializations. Expectation events depend on the ecache struct of their master conntrack. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_common.h | 18 ++++++++ include/net/netfilter/nf_conntrack_ecache.h | 59 +++++++++++++-------------- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 2 +- 4 files changed, 48 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h index a374787ed9b..ebfed90733f 100644 --- a/include/linux/netfilter/nf_conntrack_common.h +++ b/include/linux/netfilter/nf_conntrack_common.h @@ -74,6 +74,24 @@ enum ip_conntrack_status { IPS_FIXED_TIMEOUT = (1 << IPS_FIXED_TIMEOUT_BIT), }; +/* Connection tracking event types */ +enum ip_conntrack_events { + IPCT_NEW, /* new conntrack */ + IPCT_RELATED, /* related conntrack */ + IPCT_DESTROY, /* destroyed conntrack */ + IPCT_REPLY, /* connection has seen two-way traffic */ + IPCT_ASSURED, /* connection status has changed to assured */ + IPCT_PROTOINFO, /* protocol information has changed */ + IPCT_HELPER, /* new helper has been set */ + IPCT_MARK, /* new mark has been set */ + IPCT_NATSEQADJ, /* NAT is doing sequence adjustment */ + IPCT_SECMARK, /* new security mark has been set */ +}; + +enum ip_conntrack_expect_events { + IPEXP_NEW, /* new expectation */ +}; + #ifdef __KERNEL__ struct ip_conntrack_stat { unsigned int searched; diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index 5e05fb883ab..96ba5f7dcab 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -12,28 +12,12 @@ #include #include -/* Connection tracking event types */ -enum ip_conntrack_events { - IPCT_NEW, /* new conntrack */ - IPCT_RELATED, /* related conntrack */ - IPCT_DESTROY, /* destroyed conntrack */ - IPCT_REPLY, /* connection has seen two-way traffic */ - IPCT_ASSURED, /* connection status has changed to assured */ - IPCT_PROTOINFO, /* protocol information has changed */ - IPCT_HELPER, /* new helper has been set */ - IPCT_MARK, /* new mark has been set */ - IPCT_NATSEQADJ, /* NAT is doing sequence adjustment */ - IPCT_SECMARK, /* new security mark has been set */ -}; - -enum ip_conntrack_expect_events { - IPEXP_NEW, /* new expectation */ -}; - struct nf_conntrack_ecache { - unsigned long cache; /* bitops want long */ - unsigned long missed; /* missed events */ - u32 pid; /* netlink pid of destroyer */ + unsigned long cache; /* bitops want long */ + unsigned long missed; /* missed events */ + u16 ctmask; /* bitmask of ct events to be delivered */ + u16 expmask; /* bitmask of expect events to be delivered */ + u32 pid; /* netlink pid of destroyer */ }; static inline struct nf_conntrack_ecache * @@ -43,14 +27,24 @@ nf_ct_ecache_find(const struct nf_conn *ct) } static inline struct nf_conntrack_ecache * -nf_ct_ecache_ext_add(struct nf_conn *ct, gfp_t gfp) +nf_ct_ecache_ext_add(struct nf_conn *ct, u16 ctmask, u16 expmask, gfp_t gfp) { struct net *net = nf_ct_net(ct); + struct nf_conntrack_ecache *e; - if (!net->ct.sysctl_events) + if (!ctmask && !expmask && net->ct.sysctl_events) { + ctmask = ~0; + expmask = ~0; + } + if (!ctmask && !expmask) return NULL; - return nf_ct_ext_add(ct, NF_CT_EXT_ECACHE, gfp); + e = nf_ct_ext_add(ct, NF_CT_EXT_ECACHE, gfp); + if (e) { + e->ctmask = ctmask; + e->expmask = expmask; + } + return e; }; #ifdef CONFIG_NF_CONNTRACK_EVENTS @@ -83,6 +77,9 @@ nf_conntrack_event_cache(enum ip_conntrack_events event, struct nf_conn *ct) if (e == NULL) return; + if (!(e->ctmask & (1 << event))) + return; + set_bit(event, &e->cache); } @@ -93,7 +90,6 @@ nf_conntrack_eventmask_report(unsigned int eventmask, int report) { int ret = 0; - struct net *net = nf_ct_net(ct); struct nf_ct_event_notifier *notify; struct nf_conntrack_ecache *e; @@ -102,9 +98,6 @@ nf_conntrack_eventmask_report(unsigned int eventmask, if (notify == NULL) goto out_unlock; - if (!net->ct.sysctl_events) - goto out_unlock; - e = nf_ct_ecache_find(ct); if (e == NULL) goto out_unlock; @@ -118,6 +111,9 @@ nf_conntrack_eventmask_report(unsigned int eventmask, /* This is a resent of a destroy event? If so, skip missed */ unsigned long missed = e->pid ? 0 : e->missed; + if (!((eventmask | missed) & e->ctmask)) + goto out_unlock; + ret = notify->fcn(eventmask | missed, &item); if (unlikely(ret < 0 || missed)) { spin_lock_bh(&ct->lock); @@ -173,18 +169,19 @@ nf_ct_expect_event_report(enum ip_conntrack_expect_events event, u32 pid, int report) { - struct net *net = nf_ct_exp_net(exp); struct nf_exp_event_notifier *notify; + struct nf_conntrack_ecache *e; rcu_read_lock(); notify = rcu_dereference(nf_expect_event_cb); if (notify == NULL) goto out_unlock; - if (!net->ct.sysctl_events) + e = nf_ct_ecache_find(exp->master); + if (e == NULL) goto out_unlock; - { + if (e->expmask & (1 << event)) { struct nf_exp_event item = { .exp = exp, .pid = pid, diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 091ff770eb7..53b8da6ad6b 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -648,7 +648,7 @@ init_conntrack(struct net *net, } nf_ct_acct_ext_add(ct, GFP_ATOMIC); - nf_ct_ecache_ext_add(ct, GFP_ATOMIC); + nf_ct_ecache_ext_add(ct, 0, 0, GFP_ATOMIC); spin_lock_bh(&nf_conntrack_lock); exp = nf_ct_find_expectation(net, tuple); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index ff594eb138c..f5c0b09e12f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1281,7 +1281,7 @@ ctnetlink_create_conntrack(struct net *net, } nf_ct_acct_ext_add(ct, GFP_ATOMIC); - nf_ct_ecache_ext_add(ct, GFP_ATOMIC); + nf_ct_ecache_ext_add(ct, 0, 0, GFP_ATOMIC); #if defined(CONFIG_NF_CONNTRACK_MARK) if (cda[CTA_MARK]) -- cgit v1.2.3-70-g09d2 From b2a15a604d379af323645e330638e2cfcc696aff Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 14:13:03 +0100 Subject: netfilter: nf_conntrack: support conntrack templates Support initializing selected parameters of new conntrack entries from a "conntrack template", which is a specially marked conntrack entry attached to the skb. Currently the helper and the event delivery masks can be initialized this way. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_common.h | 4 +++ include/net/netfilter/nf_conntrack.h | 5 +++ include/net/netfilter/nf_conntrack_helper.h | 3 +- net/ipv4/netfilter/nf_defrag_ipv4.c | 2 +- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 2 +- net/netfilter/nf_conntrack_core.c | 50 +++++++++++++++++--------- net/netfilter/nf_conntrack_helper.c | 17 ++++++--- net/netfilter/nf_conntrack_netlink.c | 2 +- 8 files changed, 61 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h index ebfed90733f..c608677dda6 100644 --- a/include/linux/netfilter/nf_conntrack_common.h +++ b/include/linux/netfilter/nf_conntrack_common.h @@ -72,6 +72,10 @@ enum ip_conntrack_status { /* Connection has fixed timeout. */ IPS_FIXED_TIMEOUT_BIT = 10, IPS_FIXED_TIMEOUT = (1 << IPS_FIXED_TIMEOUT_BIT), + + /* Conntrack is a template */ + IPS_TEMPLATE_BIT = 11, + IPS_TEMPLATE = (1 << IPS_TEMPLATE_BIT), }; /* Connection tracking event types */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index a0904adfb8f..5043d61c99a 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -272,6 +272,11 @@ nf_conntrack_alloc(struct net *net, const struct nf_conntrack_tuple *repl, gfp_t gfp); +static inline int nf_ct_is_template(const struct nf_conn *ct) +{ + return test_bit(IPS_TEMPLATE_BIT, &ct->status); +} + /* It's confirmed if it is, or has been in the hash table. */ static inline int nf_ct_is_confirmed(struct nf_conn *ct) { diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 86be7c4816d..e17aaa3e19f 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -47,7 +47,8 @@ extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); extern struct nf_conn_help *nf_ct_helper_ext_add(struct nf_conn *ct, gfp_t gfp); -extern int __nf_ct_try_assign_helper(struct nf_conn *ct, gfp_t flags); +extern int __nf_ct_try_assign_helper(struct nf_conn *ct, struct nf_conn *tmpl, + gfp_t flags); extern void nf_ct_helper_destroy(struct nf_conn *ct); diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index 331ead3ebd1..77627fa8056 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -59,7 +59,7 @@ static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, #if !defined(CONFIG_NF_NAT) && !defined(CONFIG_NF_NAT_MODULE) /* Previously seen (loopback)? Ignore. Do this before fragment check. */ - if (skb->nfct) + if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct)) return NF_ACCEPT; #endif #endif diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index 0956ebabbff..55ce22e5de4 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -212,7 +212,7 @@ static unsigned int ipv6_defrag(unsigned int hooknum, struct sk_buff *reasm; /* Previously seen (loopback)? */ - if (skb->nfct) + if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct)) return NF_ACCEPT; reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(hooknum, skb)); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 53b8da6ad6b..471e2a79d26 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -618,7 +618,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_free); /* Allocate a new conntrack: we return -ENOMEM if classification failed due to stress. Otherwise it really is unclassifiable. */ static struct nf_conntrack_tuple_hash * -init_conntrack(struct net *net, +init_conntrack(struct net *net, struct nf_conn *tmpl, const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, struct nf_conntrack_l4proto *l4proto, @@ -628,6 +628,7 @@ init_conntrack(struct net *net, struct nf_conn *ct; struct nf_conn_help *help; struct nf_conntrack_tuple repl_tuple; + struct nf_conntrack_ecache *ecache; struct nf_conntrack_expect *exp; if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { @@ -648,7 +649,11 @@ init_conntrack(struct net *net, } nf_ct_acct_ext_add(ct, GFP_ATOMIC); - nf_ct_ecache_ext_add(ct, 0, 0, GFP_ATOMIC); + + ecache = tmpl ? nf_ct_ecache_find(tmpl) : NULL; + nf_ct_ecache_ext_add(ct, ecache ? ecache->ctmask : 0, + ecache ? ecache->expmask : 0, + GFP_ATOMIC); spin_lock_bh(&nf_conntrack_lock); exp = nf_ct_find_expectation(net, tuple); @@ -673,7 +678,7 @@ init_conntrack(struct net *net, nf_conntrack_get(&ct->master->ct_general); NF_CT_STAT_INC(net, expect_new); } else { - __nf_ct_try_assign_helper(ct, GFP_ATOMIC); + __nf_ct_try_assign_helper(ct, tmpl, GFP_ATOMIC); NF_CT_STAT_INC(net, new); } @@ -694,7 +699,7 @@ init_conntrack(struct net *net, /* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ static inline struct nf_conn * -resolve_normal_ct(struct net *net, +resolve_normal_ct(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, u_int16_t l3num, @@ -718,7 +723,8 @@ resolve_normal_ct(struct net *net, /* look for tuple match */ h = nf_conntrack_find_get(net, &tuple); if (!h) { - h = init_conntrack(net, &tuple, l3proto, l4proto, skb, dataoff); + h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto, + skb, dataoff); if (!h) return NULL; if (IS_ERR(h)) @@ -755,7 +761,7 @@ unsigned int nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, struct sk_buff *skb) { - struct nf_conn *ct; + struct nf_conn *ct, *tmpl = NULL; enum ip_conntrack_info ctinfo; struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -764,10 +770,14 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, int set_reply = 0; int ret; - /* Previously seen (loopback or untracked)? Ignore. */ if (skb->nfct) { - NF_CT_STAT_INC_ATOMIC(net, ignore); - return NF_ACCEPT; + /* Previously seen (loopback or untracked)? Ignore. */ + tmpl = (struct nf_conn *)skb->nfct; + if (!nf_ct_is_template(tmpl)) { + NF_CT_STAT_INC_ATOMIC(net, ignore); + return NF_ACCEPT; + } + skb->nfct = NULL; } /* rcu_read_lock()ed by nf_hook_slow */ @@ -778,7 +788,8 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, pr_debug("not prepared to track yet or error occured\n"); NF_CT_STAT_INC_ATOMIC(net, error); NF_CT_STAT_INC_ATOMIC(net, invalid); - return -ret; + ret = -ret; + goto out; } l4proto = __nf_ct_l4proto_find(pf, protonum); @@ -791,22 +802,25 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, if (ret <= 0) { NF_CT_STAT_INC_ATOMIC(net, error); NF_CT_STAT_INC_ATOMIC(net, invalid); - return -ret; + ret = -ret; + goto out; } } - ct = resolve_normal_ct(net, skb, dataoff, pf, protonum, + ct = resolve_normal_ct(net, tmpl, skb, dataoff, pf, protonum, l3proto, l4proto, &set_reply, &ctinfo); if (!ct) { /* Not valid part of a connection */ NF_CT_STAT_INC_ATOMIC(net, invalid); - return NF_ACCEPT; + ret = NF_ACCEPT; + goto out; } if (IS_ERR(ct)) { /* Too stressed to deal. */ NF_CT_STAT_INC_ATOMIC(net, drop); - return NF_DROP; + ret = NF_DROP; + goto out; } NF_CT_ASSERT(skb->nfct); @@ -821,11 +835,15 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, NF_CT_STAT_INC_ATOMIC(net, invalid); if (ret == -NF_DROP) NF_CT_STAT_INC_ATOMIC(net, drop); - return -ret; + ret = -ret; + goto out; } if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) nf_conntrack_event_cache(IPCT_REPLY, ct); +out: + if (tmpl) + nf_ct_put(tmpl); return ret; } @@ -864,7 +882,7 @@ void nf_conntrack_alter_reply(struct nf_conn *ct, return; rcu_read_lock(); - __nf_ct_try_assign_helper(ct, GFP_ATOMIC); + __nf_ct_try_assign_helper(ct, NULL, GFP_ATOMIC); rcu_read_unlock(); } EXPORT_SYMBOL_GPL(nf_conntrack_alter_reply); diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index c0e461f466a..8144b0da551 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -96,13 +96,22 @@ struct nf_conn_help *nf_ct_helper_ext_add(struct nf_conn *ct, gfp_t gfp) } EXPORT_SYMBOL_GPL(nf_ct_helper_ext_add); -int __nf_ct_try_assign_helper(struct nf_conn *ct, gfp_t flags) +int __nf_ct_try_assign_helper(struct nf_conn *ct, struct nf_conn *tmpl, + gfp_t flags) { + struct nf_conntrack_helper *helper = NULL; + struct nf_conn_help *help; int ret = 0; - struct nf_conntrack_helper *helper; - struct nf_conn_help *help = nfct_help(ct); - helper = __nf_ct_helper_find(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + if (tmpl != NULL) { + help = nfct_help(tmpl); + if (help != NULL) + helper = help->helper; + } + + help = nfct_help(ct); + if (helper == NULL) + helper = __nf_ct_helper_find(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); if (helper == NULL) { if (help) rcu_assign_pointer(help->helper, NULL); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index f5c0b09e12f..09044f9f4b2 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1249,7 +1249,7 @@ ctnetlink_create_conntrack(struct net *net, } } else { /* try an implicit helper assignation */ - err = __nf_ct_try_assign_helper(ct, GFP_ATOMIC); + err = __nf_ct_try_assign_helper(ct, NULL, GFP_ATOMIC); if (err < 0) goto err2; } -- cgit v1.2.3-70-g09d2 From 84f3bb9ae9db90f7fb15d98b55279a58ab1b2363 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 Feb 2010 17:17:06 +0100 Subject: netfilter: xtables: add CT target Add a new target for the raw table, which can be used to specify conntrack parameters for specific connections, f.i. the conntrack helper. The target attaches a "template" connection tracking entry to the skb, which is used by the conntrack core when initializing a new conntrack. Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_CT.h | 17 +++ include/net/netfilter/nf_conntrack_helper.h | 3 + net/netfilter/Kconfig | 12 +++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_helper.c | 19 ++++ net/netfilter/xt_CT.c | 158 ++++++++++++++++++++++++++++ 7 files changed, 211 insertions(+) create mode 100644 include/linux/netfilter/xt_CT.h create mode 100644 net/netfilter/xt_CT.c (limited to 'include') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index 2aea50399c0..a5a63e41b8a 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -6,6 +6,7 @@ header-y += nfnetlink_queue.h header-y += xt_CLASSIFY.h header-y += xt_CONNMARK.h header-y += xt_CONNSECMARK.h +header-y += xt_CT.h header-y += xt_DSCP.h header-y += xt_LED.h header-y += xt_MARK.h diff --git a/include/linux/netfilter/xt_CT.h b/include/linux/netfilter/xt_CT.h new file mode 100644 index 00000000000..7fd0effe131 --- /dev/null +++ b/include/linux/netfilter/xt_CT.h @@ -0,0 +1,17 @@ +#ifndef _XT_CT_H +#define _XT_CT_H + +#define XT_CT_NOTRACK 0x1 + +struct xt_ct_target_info { + u_int16_t flags; + u_int16_t __unused; + u_int32_t ct_events; + u_int32_t exp_events; + char helper[16]; + + /* Used internally by the kernel */ + struct nf_conn *ct __attribute__((aligned(8))); +}; + +#endif /* _XT_CT_H */ diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index e17aaa3e19f..32c305dbdab 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -42,6 +42,9 @@ struct nf_conntrack_helper { extern struct nf_conntrack_helper * __nf_conntrack_helper_find(const char *name, u16 l3num, u8 protonum); +extern struct nf_conntrack_helper * +nf_conntrack_helper_try_module_get(const char *name, u16 l3num, u8 protonum); + extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 634d14affc8..4469d45261f 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -341,6 +341,18 @@ config NETFILTER_XT_TARGET_CONNSECMARK To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_TARGET_CT + tristate '"CT" target support' + depends on NF_CONNTRACK + depends on IP_NF_RAW || IP6_NF_RAW + depends on NETFILTER_ADVANCED + help + This options adds a `CT' target, which allows to specify initial + connection tracking parameters like events to be delivered and + the helper to be used. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_DSCP tristate '"DSCP" and "TOS" target support' depends on IP_NF_MANGLE || IP6_NF_MANGLE diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 49f62ee4e9f..f873644f02f 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -44,6 +44,7 @@ obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o obj-$(CONFIG_NETFILTER_XT_TARGET_CLASSIFY) += xt_CLASSIFY.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o +obj-$(CONFIG_NETFILTER_XT_TARGET_CT) += xt_CT.o obj-$(CONFIG_NETFILTER_XT_TARGET_DSCP) += xt_DSCP.o obj-$(CONFIG_NETFILTER_XT_TARGET_HL) += xt_HL.o obj-$(CONFIG_NETFILTER_XT_TARGET_LED) += xt_LED.o diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 8144b0da551..a74a5769877 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -83,6 +83,25 @@ __nf_conntrack_helper_find(const char *name, u16 l3num, u8 protonum) } EXPORT_SYMBOL_GPL(__nf_conntrack_helper_find); +struct nf_conntrack_helper * +nf_conntrack_helper_try_module_get(const char *name, u16 l3num, u8 protonum) +{ + struct nf_conntrack_helper *h; + + h = __nf_conntrack_helper_find(name, l3num, protonum); +#ifdef CONFIG_MODULES + if (h == NULL) { + if (request_module("nfct-helper-%s", name) == 0) + h = __nf_conntrack_helper_find(name, l3num, protonum); + } +#endif + if (h != NULL && !try_module_get(h->me)) + h = NULL; + + return h; +} +EXPORT_SYMBOL_GPL(nf_conntrack_helper_try_module_get); + struct nf_conn_help *nf_ct_helper_ext_add(struct nf_conn *ct, gfp_t gfp) { struct nf_conn_help *help; diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c new file mode 100644 index 00000000000..8183a054256 --- /dev/null +++ b/net/netfilter/xt_CT.c @@ -0,0 +1,158 @@ +/* + * Copyright (c) 2010 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static unsigned int xt_ct_target(struct sk_buff *skb, + const struct xt_target_param *par) +{ + const struct xt_ct_target_info *info = par->targinfo; + struct nf_conn *ct = info->ct; + + /* Previously seen (loopback)? Ignore. */ + if (skb->nfct != NULL) + return XT_CONTINUE; + + atomic_inc(&ct->ct_general.use); + skb->nfct = &ct->ct_general; + skb->nfctinfo = IP_CT_NEW; + + return XT_CONTINUE; +} + +static u8 xt_ct_find_proto(const struct xt_tgchk_param *par) +{ + if (par->family == AF_INET) { + const struct ipt_entry *e = par->entryinfo; + + if (e->ip.invflags & IPT_INV_PROTO) + return 0; + return e->ip.proto; + } else if (par->family == AF_INET6) { + const struct ip6t_entry *e = par->entryinfo; + + if (e->ipv6.invflags & IP6T_INV_PROTO) + return 0; + return e->ipv6.proto; + } else + return 0; +} + +static bool xt_ct_tg_check(const struct xt_tgchk_param *par) +{ + struct xt_ct_target_info *info = par->targinfo; + struct nf_conntrack_tuple t; + struct nf_conn_help *help; + struct nf_conn *ct; + u8 proto; + + if (info->flags & ~XT_CT_NOTRACK) + return false; + + if (info->flags & XT_CT_NOTRACK) { + ct = &nf_conntrack_untracked; + atomic_inc(&ct->ct_general.use); + goto out; + } + + if (nf_ct_l3proto_try_module_get(par->family) < 0) + goto err1; + + memset(&t, 0, sizeof(t)); + ct = nf_conntrack_alloc(par->net, &t, &t, GFP_KERNEL); + if (IS_ERR(ct)) + goto err2; + + if ((info->ct_events || info->exp_events) && + !nf_ct_ecache_ext_add(ct, info->ct_events, info->exp_events, + GFP_KERNEL)) + goto err3; + + if (info->helper[0]) { + proto = xt_ct_find_proto(par); + if (!proto) + goto err3; + + help = nf_ct_helper_ext_add(ct, GFP_KERNEL); + if (help == NULL) + goto err3; + + help->helper = nf_conntrack_helper_try_module_get(info->helper, + par->family, + proto); + if (help->helper == NULL) + goto err3; + } + + __set_bit(IPS_TEMPLATE_BIT, &ct->status); + __set_bit(IPS_CONFIRMED_BIT, &ct->status); +out: + info->ct = ct; + return true; + +err3: + nf_conntrack_free(ct); +err2: + nf_ct_l3proto_module_put(par->family); +err1: + return false; +} + +static void xt_ct_tg_destroy(const struct xt_tgdtor_param *par) +{ + struct xt_ct_target_info *info = par->targinfo; + struct nf_conn *ct = info->ct; + struct nf_conn_help *help; + + if (ct != &nf_conntrack_untracked) { + help = nfct_help(ct); + if (help) + module_put(help->helper->me); + + nf_ct_l3proto_module_put(par->family); + } + nf_ct_put(info->ct); +} + +static struct xt_target xt_ct_tg __read_mostly = { + .name = "CT", + .family = NFPROTO_UNSPEC, + .targetsize = XT_ALIGN(sizeof(struct xt_ct_target_info)), + .checkentry = xt_ct_tg_check, + .destroy = xt_ct_tg_destroy, + .target = xt_ct_target, + .table = "raw", + .me = THIS_MODULE, +}; + +static int __init xt_ct_tg_init(void) +{ + return xt_register_target(&xt_ct_tg); +} + +static void __exit xt_ct_tg_exit(void) +{ + xt_unregister_target(&xt_ct_tg); +} + +module_init(xt_ct_tg_init); +module_exit(xt_ct_tg_exit); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Xtables: connection tracking target"); +MODULE_ALIAS("ipt_CT"); +MODULE_ALIAS("ip6t_CT"); -- cgit v1.2.3-70-g09d2 From 8a83a00b0735190384a348156837918271034144 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sat, 30 Jan 2010 12:23:03 +0000 Subject: net: maintain namespace isolation between vlan and real device In the vlan and macvlan drivers, the start_xmit function forwards data to the dev_queue_xmit function for another device, which may potentially belong to a different namespace. To make sure that classification stays within a single namespace, this resets the potentially critical fields. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 2 +- include/linux/netdevice.h | 9 +++++++++ net/8021q/vlan_dev.c | 2 +- net/core/dev.c | 35 +++++++++++++++++++++++++++++++---- 4 files changed, 42 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index fa0dc514dba..d32e0bdfc5e 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -269,7 +269,7 @@ static int macvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev) } xmit_world: - skb->dev = vlan->lowerdev; + skb_set_dev(skb, vlan->lowerdev); return dev_queue_xmit(skb); } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 93a32a5ca74..622ba5aa93c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1004,6 +1004,15 @@ static inline bool netdev_uses_dsa_tags(struct net_device *dev) return 0; } +#ifndef CONFIG_NET_NS +static inline void skb_set_dev(struct sk_buff *skb, struct net_device *dev) +{ + skb->dev = dev; +} +#else /* CONFIG_NET_NS */ +void skb_set_dev(struct sk_buff *skb, struct net_device *dev); +#endif + static inline bool netdev_uses_trailer_tags(struct net_device *dev) { #ifdef CONFIG_NET_DSA_TAG_TRAILER diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index a9e1f178561..9e83272fc5b 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -322,7 +322,7 @@ static netdev_tx_t vlan_dev_hard_start_xmit(struct sk_buff *skb, } - skb->dev = vlan_dev_info(dev)->real_dev; + skb_set_dev(skb, vlan_dev_info(dev)->real_dev); len = skb->len; ret = dev_queue_xmit(skb); diff --git a/net/core/dev.c b/net/core/dev.c index 2cba5c521e5..94c1eeed25e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1448,13 +1448,10 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) if (skb->len > (dev->mtu + dev->hard_header_len)) return NET_RX_DROP; - skb_dst_drop(skb); + skb_set_dev(skb, dev); skb->tstamp.tv64 = 0; skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, dev); - skb->mark = 0; - secpath_reset(skb); - nf_reset(skb); return netif_rx(skb); } EXPORT_SYMBOL_GPL(dev_forward_skb); @@ -1614,6 +1611,36 @@ static bool dev_can_checksum(struct net_device *dev, struct sk_buff *skb) return false; } +/** + * skb_dev_set -- assign a new device to a buffer + * @skb: buffer for the new device + * @dev: network device + * + * If an skb is owned by a device already, we have to reset + * all data private to the namespace a device belongs to + * before assigning it a new device. + */ +#ifdef CONFIG_NET_NS +void skb_set_dev(struct sk_buff *skb, struct net_device *dev) +{ + skb_dst_drop(skb); + if (skb->dev && !net_eq(dev_net(skb->dev), dev_net(dev))) { + secpath_reset(skb); + nf_reset(skb); + skb_init_secmark(skb); + skb->mark = 0; + skb->priority = 0; + skb->nf_trace = 0; + skb->ipvs_property = 0; +#ifdef CONFIG_NET_SCHED + skb->tc_index = 0; +#endif + } + skb->dev = dev; +} +EXPORT_SYMBOL(skb_set_dev); +#endif /* CONFIG_NET_NS */ + /* * Invalidate hardware checksum when packet is to be mangled, and * complete checksum manually on outgoing path. -- cgit v1.2.3-70-g09d2 From fc0663d6b5e6d8e9b57f872a644c0aafd82361b7 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sat, 30 Jan 2010 12:23:40 +0000 Subject: macvlan: allow multiple driver backends This makes it possible to hook into the macvlan driver from another kernel module. In particular, the goal is to extend it with the macvtap backend that provides a tun/tap compatible interface directly on the macvlan device. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 113 ++++++++++++++++++++------------------------- include/linux/if_macvlan.h | 70 ++++++++++++++++++++++++++++ 2 files changed, 119 insertions(+), 64 deletions(-) (limited to 'include') diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index d32e0bdfc5e..40faa368b07 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -39,31 +39,6 @@ struct macvlan_port { struct list_head vlans; }; -/** - * struct macvlan_rx_stats - MACVLAN percpu rx stats - * @rx_packets: number of received packets - * @rx_bytes: number of received bytes - * @multicast: number of received multicast packets - * @rx_errors: number of errors - */ -struct macvlan_rx_stats { - unsigned long rx_packets; - unsigned long rx_bytes; - unsigned long multicast; - unsigned long rx_errors; -}; - -struct macvlan_dev { - struct net_device *dev; - struct list_head list; - struct hlist_node hlist; - struct macvlan_port *port; - struct net_device *lowerdev; - struct macvlan_rx_stats *rx_stats; - enum macvlan_mode mode; -}; - - static struct macvlan_dev *macvlan_hash_lookup(const struct macvlan_port *port, const unsigned char *addr) { @@ -118,31 +93,17 @@ static int macvlan_addr_busy(const struct macvlan_port *port, return 0; } -static inline void macvlan_count_rx(const struct macvlan_dev *vlan, - unsigned int len, bool success, - bool multicast) -{ - struct macvlan_rx_stats *rx_stats; - - rx_stats = per_cpu_ptr(vlan->rx_stats, smp_processor_id()); - if (likely(success)) { - rx_stats->rx_packets++;; - rx_stats->rx_bytes += len; - if (multicast) - rx_stats->multicast++; - } else { - rx_stats->rx_errors++; - } -} -static int macvlan_broadcast_one(struct sk_buff *skb, struct net_device *dev, +static int macvlan_broadcast_one(struct sk_buff *skb, + const struct macvlan_dev *vlan, const struct ethhdr *eth, bool local) { + struct net_device *dev = vlan->dev; if (!skb) return NET_RX_DROP; if (local) - return dev_forward_skb(dev, skb); + return vlan->forward(dev, skb); skb->dev = dev; if (!compare_ether_addr_64bits(eth->h_dest, @@ -151,7 +112,7 @@ static int macvlan_broadcast_one(struct sk_buff *skb, struct net_device *dev, else skb->pkt_type = PACKET_MULTICAST; - return netif_rx(skb); + return vlan->receive(skb); } static void macvlan_broadcast(struct sk_buff *skb, @@ -175,7 +136,7 @@ static void macvlan_broadcast(struct sk_buff *skb, continue; nskb = skb_clone(skb, GFP_ATOMIC); - err = macvlan_broadcast_one(nskb, vlan->dev, eth, + err = macvlan_broadcast_one(nskb, vlan, eth, mode == MACVLAN_MODE_BRIDGE); macvlan_count_rx(vlan, skb->len + ETH_HLEN, err == NET_RX_SUCCESS, 1); @@ -238,7 +199,7 @@ static struct sk_buff *macvlan_handle_frame(struct sk_buff *skb) skb->dev = dev; skb->pkt_type = PACKET_HOST; - netif_rx(skb); + vlan->receive(skb); return NULL; } @@ -260,7 +221,7 @@ static int macvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev) dest = macvlan_hash_lookup(port, eth->h_dest); if (dest && dest->mode == MACVLAN_MODE_BRIDGE) { unsigned int length = skb->len + ETH_HLEN; - int ret = dev_forward_skb(dest->dev, skb); + int ret = dest->forward(dest->dev, skb); macvlan_count_rx(dest, length, ret == NET_RX_SUCCESS, 0); @@ -273,8 +234,8 @@ xmit_world: return dev_queue_xmit(skb); } -static netdev_tx_t macvlan_start_xmit(struct sk_buff *skb, - struct net_device *dev) +netdev_tx_t macvlan_start_xmit(struct sk_buff *skb, + struct net_device *dev) { int i = skb_get_queue_mapping(skb); struct netdev_queue *txq = netdev_get_tx_queue(dev, i); @@ -290,6 +251,7 @@ static netdev_tx_t macvlan_start_xmit(struct sk_buff *skb, return ret; } +EXPORT_SYMBOL_GPL(macvlan_start_xmit); static int macvlan_hard_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, const void *daddr, @@ -623,8 +585,11 @@ static int macvlan_get_tx_queues(struct net *net, return 0; } -static int macvlan_newlink(struct net *src_net, struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[]) +int macvlan_common_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[], + int (*receive)(struct sk_buff *skb), + int (*forward)(struct net_device *dev, + struct sk_buff *skb)) { struct macvlan_dev *vlan = netdev_priv(dev); struct macvlan_port *port; @@ -664,6 +629,8 @@ static int macvlan_newlink(struct net *src_net, struct net_device *dev, vlan->lowerdev = lowerdev; vlan->dev = dev; vlan->port = port; + vlan->receive = receive; + vlan->forward = forward; vlan->mode = MACVLAN_MODE_VEPA; if (data && data[IFLA_MACVLAN_MODE]) @@ -677,8 +644,17 @@ static int macvlan_newlink(struct net *src_net, struct net_device *dev, netif_stacked_transfer_operstate(lowerdev, dev); return 0; } +EXPORT_SYMBOL_GPL(macvlan_common_newlink); -static void macvlan_dellink(struct net_device *dev, struct list_head *head) +static int macvlan_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[]) +{ + return macvlan_common_newlink(src_net, dev, tb, data, + netif_rx, + dev_forward_skb); +} + +void macvlan_dellink(struct net_device *dev, struct list_head *head) { struct macvlan_dev *vlan = netdev_priv(dev); struct macvlan_port *port = vlan->port; @@ -689,6 +665,7 @@ static void macvlan_dellink(struct net_device *dev, struct list_head *head) if (list_empty(&port->vlans)) macvlan_port_destroy(port->dev); } +EXPORT_SYMBOL_GPL(macvlan_dellink); static int macvlan_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) @@ -720,19 +697,27 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = { [IFLA_MACVLAN_MODE] = { .type = NLA_U32 }, }; -static struct rtnl_link_ops macvlan_link_ops __read_mostly = { +int macvlan_link_register(struct rtnl_link_ops *ops) +{ + /* common fields */ + ops->priv_size = sizeof(struct macvlan_dev); + ops->get_tx_queues = macvlan_get_tx_queues; + ops->setup = macvlan_setup; + ops->validate = macvlan_validate; + ops->maxtype = IFLA_MACVLAN_MAX; + ops->policy = macvlan_policy; + ops->changelink = macvlan_changelink; + ops->get_size = macvlan_get_size; + ops->fill_info = macvlan_fill_info; + + return rtnl_link_register(ops); +}; +EXPORT_SYMBOL_GPL(macvlan_link_register); + +static struct rtnl_link_ops macvlan_link_ops = { .kind = "macvlan", - .priv_size = sizeof(struct macvlan_dev), - .get_tx_queues = macvlan_get_tx_queues, - .setup = macvlan_setup, - .validate = macvlan_validate, .newlink = macvlan_newlink, .dellink = macvlan_dellink, - .maxtype = IFLA_MACVLAN_MAX, - .policy = macvlan_policy, - .changelink = macvlan_changelink, - .get_size = macvlan_get_size, - .fill_info = macvlan_fill_info, }; static int macvlan_device_event(struct notifier_block *unused, @@ -761,7 +746,7 @@ static int macvlan_device_event(struct notifier_block *unused, break; case NETDEV_UNREGISTER: list_for_each_entry_safe(vlan, next, &port->vlans, list) - macvlan_dellink(vlan->dev, NULL); + vlan->dev->rtnl_link_ops->dellink(vlan->dev, NULL); break; } return NOTIFY_DONE; @@ -778,7 +763,7 @@ static int __init macvlan_init_module(void) register_netdevice_notifier(&macvlan_notifier_block); macvlan_handle_frame_hook = macvlan_handle_frame; - err = rtnl_link_register(&macvlan_link_ops); + err = macvlan_link_register(&macvlan_link_ops); if (err < 0) goto err1; return 0; diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index 5f200bac374..9a11544bb0b 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -1,6 +1,76 @@ #ifndef _LINUX_IF_MACVLAN_H #define _LINUX_IF_MACVLAN_H +#include +#include +#include +#include +#include + +struct macvlan_port; +struct macvtap_queue; + +/** + * struct macvlan_rx_stats - MACVLAN percpu rx stats + * @rx_packets: number of received packets + * @rx_bytes: number of received bytes + * @multicast: number of received multicast packets + * @rx_errors: number of errors + */ +struct macvlan_rx_stats { + unsigned long rx_packets; + unsigned long rx_bytes; + unsigned long multicast; + unsigned long rx_errors; +}; + +struct macvlan_dev { + struct net_device *dev; + struct list_head list; + struct hlist_node hlist; + struct macvlan_port *port; + struct net_device *lowerdev; + struct macvlan_rx_stats *rx_stats; + enum macvlan_mode mode; + int (*receive)(struct sk_buff *skb); + int (*forward)(struct net_device *dev, struct sk_buff *skb); +}; + +static inline void macvlan_count_rx(const struct macvlan_dev *vlan, + unsigned int len, bool success, + bool multicast) +{ + struct macvlan_rx_stats *rx_stats; + + rx_stats = per_cpu_ptr(vlan->rx_stats, smp_processor_id()); + if (likely(success)) { + rx_stats->rx_packets++;; + rx_stats->rx_bytes += len; + if (multicast) + rx_stats->multicast++; + } else { + rx_stats->rx_errors++; + } +} + +extern int macvlan_common_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[], + int (*receive)(struct sk_buff *skb), + int (*forward)(struct net_device *dev, + struct sk_buff *skb)); + +extern void macvlan_count_rx(const struct macvlan_dev *vlan, + unsigned int len, bool success, + bool multicast); + +extern void macvlan_dellink(struct net_device *dev, struct list_head *head); + +extern int macvlan_link_register(struct rtnl_link_ops *ops); + +extern netdev_tx_t macvlan_start_xmit(struct sk_buff *skb, + struct net_device *dev); + + extern struct sk_buff *(*macvlan_handle_frame_hook)(struct sk_buff *); #endif /* _LINUX_IF_MACVLAN_H */ -- cgit v1.2.3-70-g09d2 From 20d29d7a916a47bf533b5709437fe735b6b5b79e Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sat, 30 Jan 2010 12:24:26 +0000 Subject: net: macvtap driver In order to use macvlan with qemu and other tools that require a tap file descriptor, the macvtap driver adds a small backend with a character device with the same interface as the tun driver, with a minimum set of features. Macvtap interfaces are created in the same way as macvlan interfaces using ip link, but the netif is just used as a handle for configuration and accounting, while the data goes through the chardev. Each macvtap interface has its own character device, simplifying permission management significantly over the generic tun/tap driver. Cc: Patrick McHardy Cc: Stephen Hemminger Cc: David S. Miller" Cc: "Michael S. Tsirkin" Cc: Herbert Xu Cc: Or Gerlitz Cc: netdev@vger.kernel.org Cc: bridge@lists.linux-foundation.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/Kconfig | 12 + drivers/net/Makefile | 1 + drivers/net/macvtap.c | 581 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/if_macvlan.h | 1 + 4 files changed, 595 insertions(+) create mode 100644 drivers/net/macvtap.c (limited to 'include') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index cb0e534418e..411e2070311 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -90,6 +90,18 @@ config MACVLAN To compile this driver as a module, choose M here: the module will be called macvlan. +config MACVTAP + tristate "MAC-VLAN based tap driver (EXPERIMENTAL)" + depends on MACVLAN + help + This adds a specialized tap character device driver that is based + on the MAC-VLAN network interface, called macvtap. A macvtap device + can be added in the same way as a macvlan device, using 'type + macvlan', and then be accessed through the tap user space interface. + + To compile this driver as a module, choose M here: the module + will be called macvtap. + config EQUALIZER tristate "EQL (serial line load balancing) support" ---help--- diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 0b763cbe9b1..95958032cd3 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -169,6 +169,7 @@ obj-$(CONFIG_XEN_NETDEV_FRONTEND) += xen-netfront.o obj-$(CONFIG_DUMMY) += dummy.o obj-$(CONFIG_IFB) += ifb.o obj-$(CONFIG_MACVLAN) += macvlan.o +obj-$(CONFIG_MACVTAP) += macvtap.o obj-$(CONFIG_DE600) += de600.o obj-$(CONFIG_DE620) += de620.o obj-$(CONFIG_LANCE) += lance.o diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c new file mode 100644 index 00000000000..ad1f6ef8930 --- /dev/null +++ b/drivers/net/macvtap.c @@ -0,0 +1,581 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +/* + * A macvtap queue is the central object of this driver, it connects + * an open character device to a macvlan interface. There can be + * multiple queues on one interface, which map back to queues + * implemented in hardware on the underlying device. + * + * macvtap_proto is used to allocate queues through the sock allocation + * mechanism. + * + * TODO: multiqueue support is currently not implemented, even though + * macvtap is basically prepared for that. We will need to add this + * here as well as in virtio-net and qemu to get line rate on 10gbit + * adapters from a guest. + */ +struct macvtap_queue { + struct sock sk; + struct socket sock; + struct macvlan_dev *vlan; + struct file *file; +}; + +static struct proto macvtap_proto = { + .name = "macvtap", + .owner = THIS_MODULE, + .obj_size = sizeof (struct macvtap_queue), +}; + +/* + * Minor number matches netdev->ifindex, so need a potentially + * large value. This also makes it possible to split the + * tap functionality out again in the future by offering it + * from other drivers besides macvtap. As long as every device + * only has one tap, the interface numbers assure that the + * device nodes are unique. + */ +static unsigned int macvtap_major; +#define MACVTAP_NUM_DEVS 65536 +static struct class *macvtap_class; +static struct cdev macvtap_cdev; + +/* + * RCU usage: + * The macvtap_queue is referenced both from the chardev struct file + * and from the struct macvlan_dev using rcu_read_lock. + * + * We never actually update the contents of a macvtap_queue atomically + * with RCU but it is used for race-free destruction of a queue when + * either the file or the macvlan_dev goes away. Pointers back to + * the dev and the file are implicitly valid as long as the queue + * exists. + * + * The callbacks from macvlan are always done with rcu_read_lock held + * already, while in the file_operations, we get it ourselves. + * + * When destroying a queue, we remove the pointers from the file and + * from the dev and then synchronize_rcu to make sure no thread is + * still using the queue. There may still be references to the struct + * sock inside of the queue from outbound SKBs, but these never + * reference back to the file or the dev. The data structure is freed + * through __sk_free when both our references and any pending SKBs + * are gone. + * + * macvtap_lock is only used to prevent multiple concurrent open() + * calls to assign a new vlan->tap pointer. It could be moved into + * the macvlan_dev itself but is extremely rarely used. + */ +static DEFINE_SPINLOCK(macvtap_lock); + +/* + * Choose the next free queue, for now there is only one + */ +static int macvtap_set_queue(struct net_device *dev, struct file *file, + struct macvtap_queue *q) +{ + struct macvlan_dev *vlan = netdev_priv(dev); + int err = -EBUSY; + + spin_lock(&macvtap_lock); + if (rcu_dereference(vlan->tap)) + goto out; + + err = 0; + q->vlan = vlan; + rcu_assign_pointer(vlan->tap, q); + + q->file = file; + rcu_assign_pointer(file->private_data, q); + +out: + spin_unlock(&macvtap_lock); + return err; +} + +/* + * We must destroy each queue exactly once, when either + * the netdev or the file go away. + * + * Using the spinlock makes sure that we don't get + * to the queue again after destroying it. + * + * synchronize_rcu serializes with the packet flow + * that uses rcu_read_lock. + */ +static void macvtap_del_queue(struct macvtap_queue **qp) +{ + struct macvtap_queue *q; + + spin_lock(&macvtap_lock); + q = rcu_dereference(*qp); + if (!q) { + spin_unlock(&macvtap_lock); + return; + } + + rcu_assign_pointer(q->vlan->tap, NULL); + rcu_assign_pointer(q->file->private_data, NULL); + spin_unlock(&macvtap_lock); + + synchronize_rcu(); + sock_put(&q->sk); +} + +/* + * Since we only support one queue, just dereference the pointer. + */ +static struct macvtap_queue *macvtap_get_queue(struct net_device *dev, + struct sk_buff *skb) +{ + struct macvlan_dev *vlan = netdev_priv(dev); + + return rcu_dereference(vlan->tap); +} + +static void macvtap_del_queues(struct net_device *dev) +{ + struct macvlan_dev *vlan = netdev_priv(dev); + macvtap_del_queue(&vlan->tap); +} + +static inline struct macvtap_queue *macvtap_file_get_queue(struct file *file) +{ + rcu_read_lock_bh(); + return rcu_dereference(file->private_data); +} + +static inline void macvtap_file_put_queue(void) +{ + rcu_read_unlock_bh(); +} + +/* + * Forward happens for data that gets sent from one macvlan + * endpoint to another one in bridge mode. We just take + * the skb and put it into the receive queue. + */ +static int macvtap_forward(struct net_device *dev, struct sk_buff *skb) +{ + struct macvtap_queue *q = macvtap_get_queue(dev, skb); + if (!q) + return -ENOLINK; + + skb_queue_tail(&q->sk.sk_receive_queue, skb); + wake_up(q->sk.sk_sleep); + return 0; +} + +/* + * Receive is for data from the external interface (lowerdev), + * in case of macvtap, we can treat that the same way as + * forward, which macvlan cannot. + */ +static int macvtap_receive(struct sk_buff *skb) +{ + skb_push(skb, ETH_HLEN); + return macvtap_forward(skb->dev, skb); +} + +static int macvtap_newlink(struct net *src_net, + struct net_device *dev, + struct nlattr *tb[], + struct nlattr *data[]) +{ + struct device *classdev; + dev_t devt; + int err; + + err = macvlan_common_newlink(src_net, dev, tb, data, + macvtap_receive, macvtap_forward); + if (err) + goto out; + + devt = MKDEV(MAJOR(macvtap_major), dev->ifindex); + + classdev = device_create(macvtap_class, &dev->dev, devt, + dev, "tap%d", dev->ifindex); + if (IS_ERR(classdev)) { + err = PTR_ERR(classdev); + macvtap_del_queues(dev); + } + +out: + return err; +} + +static void macvtap_dellink(struct net_device *dev, + struct list_head *head) +{ + device_destroy(macvtap_class, + MKDEV(MAJOR(macvtap_major), dev->ifindex)); + + macvtap_del_queues(dev); + macvlan_dellink(dev, head); +} + +static struct rtnl_link_ops macvtap_link_ops __read_mostly = { + .kind = "macvtap", + .newlink = macvtap_newlink, + .dellink = macvtap_dellink, +}; + + +static void macvtap_sock_write_space(struct sock *sk) +{ + if (!sock_writeable(sk) || + !test_and_clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags)) + return; + + if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) + wake_up_interruptible_sync(sk->sk_sleep); +} + +static int macvtap_open(struct inode *inode, struct file *file) +{ + struct net *net = current->nsproxy->net_ns; + struct net_device *dev = dev_get_by_index(net, iminor(inode)); + struct macvtap_queue *q; + int err; + + err = -ENODEV; + if (!dev) + goto out; + + /* check if this is a macvtap device */ + err = -EINVAL; + if (dev->rtnl_link_ops != &macvtap_link_ops) + goto out; + + err = -ENOMEM; + q = (struct macvtap_queue *)sk_alloc(net, AF_UNSPEC, GFP_KERNEL, + &macvtap_proto); + if (!q) + goto out; + + init_waitqueue_head(&q->sock.wait); + q->sock.type = SOCK_RAW; + q->sock.state = SS_CONNECTED; + sock_init_data(&q->sock, &q->sk); + q->sk.sk_allocation = GFP_ATOMIC; /* for now */ + q->sk.sk_write_space = macvtap_sock_write_space; + + err = macvtap_set_queue(dev, file, q); + if (err) + sock_put(&q->sk); + +out: + if (dev) + dev_put(dev); + + return err; +} + +static int macvtap_release(struct inode *inode, struct file *file) +{ + macvtap_del_queue((struct macvtap_queue **)&file->private_data); + return 0; +} + +static unsigned int macvtap_poll(struct file *file, poll_table * wait) +{ + struct macvtap_queue *q = macvtap_file_get_queue(file); + unsigned int mask = POLLERR; + + if (!q) + goto out; + + mask = 0; + poll_wait(file, &q->sock.wait, wait); + + if (!skb_queue_empty(&q->sk.sk_receive_queue)) + mask |= POLLIN | POLLRDNORM; + + if (sock_writeable(&q->sk) || + (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock.flags) && + sock_writeable(&q->sk))) + mask |= POLLOUT | POLLWRNORM; + +out: + macvtap_file_put_queue(); + return mask; +} + +/* Get packet from user space buffer */ +static ssize_t macvtap_get_user(struct macvtap_queue *q, + const struct iovec *iv, size_t count, + int noblock) +{ + struct sk_buff *skb; + size_t len = count; + int err; + + if (unlikely(len < ETH_HLEN)) + return -EINVAL; + + skb = sock_alloc_send_skb(&q->sk, NET_IP_ALIGN + len, noblock, &err); + + if (!skb) { + macvlan_count_rx(q->vlan, 0, false, false); + return err; + } + + skb_reserve(skb, NET_IP_ALIGN); + skb_put(skb, count); + + if (skb_copy_datagram_from_iovec(skb, 0, iv, 0, len)) { + macvlan_count_rx(q->vlan, 0, false, false); + kfree_skb(skb); + return -EFAULT; + } + + skb_set_network_header(skb, ETH_HLEN); + + macvlan_start_xmit(skb, q->vlan->dev); + + return count; +} + +static ssize_t macvtap_aio_write(struct kiocb *iocb, const struct iovec *iv, + unsigned long count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + ssize_t result = -ENOLINK; + struct macvtap_queue *q = macvtap_file_get_queue(file); + + if (!q) + goto out; + + result = macvtap_get_user(q, iv, iov_length(iv, count), + file->f_flags & O_NONBLOCK); +out: + macvtap_file_put_queue(); + return result; +} + +/* Put packet to the user space buffer */ +static ssize_t macvtap_put_user(struct macvtap_queue *q, + const struct sk_buff *skb, + const struct iovec *iv, int len) +{ + struct macvlan_dev *vlan = q->vlan; + int ret; + + len = min_t(int, skb->len, len); + + ret = skb_copy_datagram_const_iovec(skb, 0, iv, 0, len); + + macvlan_count_rx(vlan, len, ret == 0, 0); + + return ret ? ret : len; +} + +static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv, + unsigned long count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct macvtap_queue *q = macvtap_file_get_queue(file); + + DECLARE_WAITQUEUE(wait, current); + struct sk_buff *skb; + ssize_t len, ret = 0; + + if (!q) { + ret = -ENOLINK; + goto out; + } + + len = iov_length(iv, count); + if (len < 0) { + ret = -EINVAL; + goto out; + } + + add_wait_queue(q->sk.sk_sleep, &wait); + while (len) { + current->state = TASK_INTERRUPTIBLE; + + /* Read frames from the queue */ + skb = skb_dequeue(&q->sk.sk_receive_queue); + if (!skb) { + if (file->f_flags & O_NONBLOCK) { + ret = -EAGAIN; + break; + } + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + /* Nothing to read, let's sleep */ + schedule(); + continue; + } + ret = macvtap_put_user(q, skb, iv, len); + kfree_skb(skb); + break; + } + + current->state = TASK_RUNNING; + remove_wait_queue(q->sk.sk_sleep, &wait); + +out: + macvtap_file_put_queue(); + return ret; +} + +/* + * provide compatibility with generic tun/tap interface + */ +static long macvtap_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + struct macvtap_queue *q; + void __user *argp = (void __user *)arg; + struct ifreq __user *ifr = argp; + unsigned int __user *up = argp; + unsigned int u; + char devname[IFNAMSIZ]; + + switch (cmd) { + case TUNSETIFF: + /* ignore the name, just look at flags */ + if (get_user(u, &ifr->ifr_flags)) + return -EFAULT; + if (u != (IFF_TAP | IFF_NO_PI)) + return -EINVAL; + return 0; + + case TUNGETIFF: + q = macvtap_file_get_queue(file); + if (!q) + return -ENOLINK; + memcpy(devname, q->vlan->dev->name, sizeof(devname)); + macvtap_file_put_queue(); + + if (copy_to_user(&ifr->ifr_name, q->vlan->dev->name, IFNAMSIZ) || + put_user((TUN_TAP_DEV | TUN_NO_PI), &ifr->ifr_flags)) + return -EFAULT; + return 0; + + case TUNGETFEATURES: + if (put_user((IFF_TAP | IFF_NO_PI), up)) + return -EFAULT; + return 0; + + case TUNSETSNDBUF: + if (get_user(u, up)) + return -EFAULT; + + q = macvtap_file_get_queue(file); + q->sk.sk_sndbuf = u; + macvtap_file_put_queue(); + return 0; + + case TUNSETOFFLOAD: + /* let the user check for future flags */ + if (arg & ~(TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | + TUN_F_TSO_ECN | TUN_F_UFO)) + return -EINVAL; + + /* TODO: add support for these, so far we don't + support any offload */ + if (arg & (TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | + TUN_F_TSO_ECN | TUN_F_UFO)) + return -EINVAL; + + return 0; + + default: + return -EINVAL; + } +} + +#ifdef CONFIG_COMPAT +static long macvtap_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return macvtap_ioctl(file, cmd, (unsigned long)compat_ptr(arg)); +} +#endif + +static const struct file_operations macvtap_fops = { + .owner = THIS_MODULE, + .open = macvtap_open, + .release = macvtap_release, + .aio_read = macvtap_aio_read, + .aio_write = macvtap_aio_write, + .poll = macvtap_poll, + .llseek = no_llseek, + .unlocked_ioctl = macvtap_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = macvtap_compat_ioctl, +#endif +}; + +static int macvtap_init(void) +{ + int err; + + err = alloc_chrdev_region(&macvtap_major, 0, + MACVTAP_NUM_DEVS, "macvtap"); + if (err) + goto out1; + + cdev_init(&macvtap_cdev, &macvtap_fops); + err = cdev_add(&macvtap_cdev, macvtap_major, MACVTAP_NUM_DEVS); + if (err) + goto out2; + + macvtap_class = class_create(THIS_MODULE, "macvtap"); + if (IS_ERR(macvtap_class)) { + err = PTR_ERR(macvtap_class); + goto out3; + } + + err = macvlan_link_register(&macvtap_link_ops); + if (err) + goto out4; + + return 0; + +out4: + class_unregister(macvtap_class); +out3: + cdev_del(&macvtap_cdev); +out2: + unregister_chrdev_region(macvtap_major, MACVTAP_NUM_DEVS); +out1: + return err; +} +module_init(macvtap_init); + +static void macvtap_exit(void) +{ + rtnl_link_unregister(&macvtap_link_ops); + class_unregister(macvtap_class); + cdev_del(&macvtap_cdev); + unregister_chrdev_region(macvtap_major, MACVTAP_NUM_DEVS); +} +module_exit(macvtap_exit); + +MODULE_ALIAS_RTNL_LINK("macvtap"); +MODULE_AUTHOR("Arnd Bergmann "); +MODULE_LICENSE("GPL"); diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index 9a11544bb0b..51f1512045e 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -34,6 +34,7 @@ struct macvlan_dev { enum macvlan_mode mode; int (*receive)(struct sk_buff *skb); int (*forward)(struct net_device *dev, struct sk_buff *skb); + struct macvtap_queue *tap; }; static inline void macvlan_count_rx(const struct macvlan_dev *vlan, -- cgit v1.2.3-70-g09d2 From 1621e0940294c20e302faf401f41204de7252e22 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 1 Feb 2010 09:44:19 +0000 Subject: net: CONFIG_COMPAT redux Ifdef out struct proto_ops::compat_ioctl struct proto_ops::compat_setsockopt struct proto_ops::compat_getsockopt to make structures smaller on COMPAT=n kernels. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/net.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/net.h b/include/linux/net.h index 5e8083cacc8..4157b5d42bd 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -174,18 +174,22 @@ struct proto_ops { struct poll_table_struct *wait); int (*ioctl) (struct socket *sock, unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT int (*compat_ioctl) (struct socket *sock, unsigned int cmd, unsigned long arg); +#endif int (*listen) (struct socket *sock, int len); int (*shutdown) (struct socket *sock, int flags); int (*setsockopt)(struct socket *sock, int level, int optname, char __user *optval, unsigned int optlen); int (*getsockopt)(struct socket *sock, int level, int optname, char __user *optval, int __user *optlen); +#ifdef CONFIG_COMPAT int (*compat_setsockopt)(struct socket *sock, int level, int optname, char __user *optval, unsigned int optlen); int (*compat_getsockopt)(struct socket *sock, int level, int optname, char __user *optval, int __user *optlen); +#endif int (*sendmsg) (struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t total_len); int (*recvmsg) (struct kiocb *iocb, struct socket *sock, -- cgit v1.2.3-70-g09d2 From 6683ece36e3531fc8c75f69e7165c5f20930be88 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 4 Feb 2010 10:22:25 -0800 Subject: net: use helpers to access mc list V2 This patch introduces the similar helpers as those already done for uc list. However multicast lists are no list_head lists but "mademanually". The three macros added by this patch will make the transition of mc_list to list_head smooth in two steps: 1) convert all drivers to use these macros (with the original iterator of type "struct dev_mc_list") 2) once all drivers are converted, convert list type and iterators to "struct netdev_hw_addr" in one patch. >From now on, drivers can (and should) use "netdev_for_each_mc_addr" to iterate over the addresses with iterator of type "struct netdev_hw_addr". Also macros "netdev_mc_count" and "netdev_mc_empty" to read list's length. This is the state which should be reached in all drivers. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 622ba5aa93c..e535700a3b7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -268,6 +268,12 @@ struct netdev_hw_addr_list { #define netdev_for_each_uc_addr(ha, dev) \ list_for_each_entry(ha, &dev->uc.list, list) +#define netdev_mc_count(dev) ((dev)->mc_count) +#define netdev_mc_empty(dev) (netdev_mc_count(dev) == 0) + +#define netdev_for_each_mc_addr(mclist, dev) \ + for (mclist = dev->mc_list; mclist; mclist = mclist->next) + struct hh_cache { struct hh_cache *hh_next; /* Next entry */ atomic_t hh_refcnt; /* number of users */ -- cgit v1.2.3-70-g09d2 From f8f76db1db369f3a130ac3fd33e2eee5f1610d9c Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 4 Feb 2010 10:23:02 -0800 Subject: libphy: add phy_find_first function Many drivers do this in them manually. Now they can use this function. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 16 ++++++++++++++++ include/linux/phy.h | 1 + 2 files changed, 17 insertions(+) (limited to 'include') diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index adbc0fded13..db1794546c5 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -276,6 +276,22 @@ int phy_device_register(struct phy_device *phydev) } EXPORT_SYMBOL(phy_device_register); +/** + * phy_find_first - finds the first PHY device on the bus + * @bus: the target MII bus + */ +struct phy_device *phy_find_first(struct mii_bus *bus) +{ + int addr; + + for (addr = 0; addr < PHY_MAX_ADDR; addr++) { + if (bus->phy_map[addr]) + return bus->phy_map[addr]; + } + return NULL; +} +EXPORT_SYMBOL(phy_find_first); + /** * phy_prepare_link - prepares the PHY layer to monitor link status * @phydev: target phy_device struct diff --git a/include/linux/phy.h b/include/linux/phy.h index 6a7eb402165..14d7fdf6a90 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -452,6 +452,7 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface); struct phy_device * phy_attach(struct net_device *dev, const char *bus_id, u32 flags, phy_interface_t interface); +struct phy_device *phy_find_first(struct mii_bus *bus); int phy_connect_direct(struct net_device *dev, struct phy_device *phydev, void (*handler)(struct net_device *), u32 flags, phy_interface_t interface); -- cgit v1.2.3-70-g09d2 From bfd5f4a3d605e0f6054df0b59fe0907ff7e696d3 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Thu, 4 Feb 2010 20:24:10 -0800 Subject: packet: Add GSO/csum offload support. This patch adds GSO/checksum offload to af_packet sockets using virtio_net_hdr. Based on Rusty's patch to add this support to tun. It allows GSO/checksum offload to be enabled when using raw socket backend with virtio_net. Adds PACKET_VNET_HDR socket option to prepend virtio_net_hdr in the receive path and process/skip virtio_net_hdr in the send path. This option is only allowed with SOCK_RAW sockets attached to ethernet type devices. v2 updates ---------- Michael's Comments - Perform length check in packet_snd() when GSO is off even when vnet_hdr is present. - Check for SKB_GSO_FCOE type and return -EINVAL - don't allow tx/rx ring when vnet_hdr is enabled. Herbert's Comments - Removed ethernet specific code. - protocol value is assumed to be passed in by the caller. Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/linux/if_packet.h | 1 + net/packet/af_packet.c | 187 +++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 177 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h index 4021d47cc43..aa57a5f993f 100644 --- a/include/linux/if_packet.h +++ b/include/linux/if_packet.h @@ -46,6 +46,7 @@ struct sockaddr_ll { #define PACKET_RESERVE 12 #define PACKET_TX_RING 13 #define PACKET_LOSS 14 +#define PACKET_VNET_HDR 15 struct tpacket_stats { unsigned int tp_packets; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 53633c5fdb1..178e2937bba 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -80,6 +80,7 @@ #include #include #include +#include #ifdef CONFIG_INET #include @@ -193,7 +194,8 @@ struct packet_sock { struct mutex pg_vec_lock; unsigned int running:1, /* prot_hook is attached*/ auxdata:1, - origdev:1; + origdev:1, + has_vnet_hdr:1; int ifindex; /* bound device */ __be16 num; struct packet_mclist *mclist; @@ -1056,6 +1058,30 @@ out: } #endif +static inline struct sk_buff *packet_alloc_skb(struct sock *sk, size_t prepad, + size_t reserve, size_t len, + size_t linear, int noblock, + int *err) +{ + struct sk_buff *skb; + + /* Under a page? Don't bother with paged skb. */ + if (prepad + len < PAGE_SIZE || !linear) + linear = len; + + skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock, + err); + if (!skb) + return NULL; + + skb_reserve(skb, reserve); + skb_put(skb, linear); + skb->data_len = len - linear; + skb->len += len - linear; + + return skb; +} + static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) { @@ -1066,14 +1092,17 @@ static int packet_snd(struct socket *sock, __be16 proto; unsigned char *addr; int ifindex, err, reserve = 0; + struct virtio_net_hdr vnet_hdr = { 0 }; + int offset = 0; + int vnet_hdr_len; + struct packet_sock *po = pkt_sk(sk); + unsigned short gso_type = 0; /* * Get and verify the address. */ if (saddr == NULL) { - struct packet_sock *po = pkt_sk(sk); - ifindex = po->ifindex; proto = po->num; addr = NULL; @@ -1100,25 +1129,74 @@ static int packet_snd(struct socket *sock, if (!(dev->flags & IFF_UP)) goto out_unlock; + if (po->has_vnet_hdr) { + vnet_hdr_len = sizeof(vnet_hdr); + + err = -EINVAL; + if (len < vnet_hdr_len) + goto out_unlock; + + len -= vnet_hdr_len; + + err = memcpy_fromiovec((void *)&vnet_hdr, msg->msg_iov, + vnet_hdr_len); + if (err < 0) + goto out_unlock; + + if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && + (vnet_hdr.csum_start + vnet_hdr.csum_offset + 2 > + vnet_hdr.hdr_len)) + vnet_hdr.hdr_len = vnet_hdr.csum_start + + vnet_hdr.csum_offset + 2; + + err = -EINVAL; + if (vnet_hdr.hdr_len > len) + goto out_unlock; + + if (vnet_hdr.gso_type != VIRTIO_NET_HDR_GSO_NONE) { + switch (vnet_hdr.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { + case VIRTIO_NET_HDR_GSO_TCPV4: + gso_type = SKB_GSO_TCPV4; + break; + case VIRTIO_NET_HDR_GSO_TCPV6: + gso_type = SKB_GSO_TCPV6; + break; + case VIRTIO_NET_HDR_GSO_UDP: + gso_type = SKB_GSO_UDP; + break; + default: + goto out_unlock; + } + + if (vnet_hdr.gso_type & VIRTIO_NET_HDR_GSO_ECN) + gso_type |= SKB_GSO_TCP_ECN; + + if (vnet_hdr.gso_size == 0) + goto out_unlock; + + } + } + err = -EMSGSIZE; - if (len > dev->mtu+reserve) + if (!gso_type && (len > dev->mtu+reserve)) goto out_unlock; - skb = sock_alloc_send_skb(sk, len + LL_ALLOCATED_SPACE(dev), - msg->msg_flags & MSG_DONTWAIT, &err); + err = -ENOBUFS; + skb = packet_alloc_skb(sk, LL_ALLOCATED_SPACE(dev), + LL_RESERVED_SPACE(dev), len, vnet_hdr.hdr_len, + msg->msg_flags & MSG_DONTWAIT, &err); if (skb == NULL) goto out_unlock; - skb_reserve(skb, LL_RESERVED_SPACE(dev)); - skb_reset_network_header(skb); + skb_set_network_header(skb, reserve); err = -EINVAL; if (sock->type == SOCK_DGRAM && - dev_hard_header(skb, dev, ntohs(proto), addr, NULL, len) < 0) + (offset = dev_hard_header(skb, dev, ntohs(proto), addr, NULL, len)) < 0) goto out_free; /* Returns -EFAULT on error */ - err = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len); + err = skb_copy_datagram_from_iovec(skb, offset, msg->msg_iov, 0, len); if (err) goto out_free; @@ -1127,6 +1205,25 @@ static int packet_snd(struct socket *sock, skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; + if (po->has_vnet_hdr) { + if (vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { + if (!skb_partial_csum_set(skb, vnet_hdr.csum_start, + vnet_hdr.csum_offset)) { + err = -EINVAL; + goto out_free; + } + } + + skb_shinfo(skb)->gso_size = vnet_hdr.gso_size; + skb_shinfo(skb)->gso_type = gso_type; + + /* Header must be checked, and gso_segs computed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + + len += vnet_hdr_len; + } + /* * Now send it */ @@ -1420,6 +1517,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, struct sk_buff *skb; int copied, err; struct sockaddr_ll *sll; + int vnet_hdr_len = 0; err = -EINVAL; if (flags & ~(MSG_PEEK|MSG_DONTWAIT|MSG_TRUNC|MSG_CMSG_COMPAT)) @@ -1451,6 +1549,48 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, if (skb == NULL) goto out; + if (pkt_sk(sk)->has_vnet_hdr) { + struct virtio_net_hdr vnet_hdr = { 0 }; + + err = -EINVAL; + vnet_hdr_len = sizeof(vnet_hdr); + if ((len -= vnet_hdr_len) < 0) + goto out_free; + + if (skb_is_gso(skb)) { + struct skb_shared_info *sinfo = skb_shinfo(skb); + + /* This is a hint as to how much should be linear. */ + vnet_hdr.hdr_len = skb_headlen(skb); + vnet_hdr.gso_size = sinfo->gso_size; + if (sinfo->gso_type & SKB_GSO_TCPV4) + vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4; + else if (sinfo->gso_type & SKB_GSO_TCPV6) + vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV6; + else if (sinfo->gso_type & SKB_GSO_UDP) + vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_UDP; + else if (sinfo->gso_type & SKB_GSO_FCOE) + goto out_free; + else + BUG(); + if (sinfo->gso_type & SKB_GSO_TCP_ECN) + vnet_hdr.gso_type |= VIRTIO_NET_HDR_GSO_ECN; + } else + vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_NONE; + + if (skb->ip_summed == CHECKSUM_PARTIAL) { + vnet_hdr.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; + vnet_hdr.csum_start = skb->csum_start - + skb_headroom(skb); + vnet_hdr.csum_offset = skb->csum_offset; + } /* else everything is zero */ + + err = memcpy_toiovec(msg->msg_iov, (void *)&vnet_hdr, + vnet_hdr_len); + if (err < 0) + goto out_free; + } + /* * If the address length field is there to be filled in, we fill * it in now. @@ -1502,7 +1642,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock, * Free or return the buffer as appropriate. Again this * hides all the races and re-entrancy issues from us. */ - err = (flags&MSG_TRUNC) ? skb->len : copied; + err = vnet_hdr_len + ((flags&MSG_TRUNC) ? skb->len : copied); out_free: skb_free_datagram(sk, skb); @@ -1740,6 +1880,8 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv if (optlen < sizeof(req)) return -EINVAL; + if (pkt_sk(sk)->has_vnet_hdr) + return -EINVAL; if (copy_from_user(&req, optval, sizeof(req))) return -EFAULT; return packet_set_ring(sk, &req, 0, optname == PACKET_TX_RING); @@ -1826,6 +1968,22 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv po->origdev = !!val; return 0; } + case PACKET_VNET_HDR: + { + int val; + + if (sock->type != SOCK_RAW) + return -EINVAL; + if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) + return -EBUSY; + if (optlen < sizeof(val)) + return -EINVAL; + if (copy_from_user(&val, optval, sizeof(val))) + return -EFAULT; + + po->has_vnet_hdr = !!val; + return 0; + } default: return -ENOPROTOOPT; } @@ -1874,6 +2032,13 @@ static int packet_getsockopt(struct socket *sock, int level, int optname, len = sizeof(int); val = po->origdev; + data = &val; + break; + case PACKET_VNET_HDR: + if (len > sizeof(int)) + len = sizeof(int); + val = po->has_vnet_hdr; + data = &val; break; #ifdef CONFIG_PACKET_MMAP -- cgit v1.2.3-70-g09d2 From 33e5a2f776e331dc8a4379b6efb660d38f182d96 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 3 Feb 2010 10:24:30 +0100 Subject: wireless: update radiotap parser Upstream radiotap has adopted the namespace proposal David Young made and I then took care of, for which I had adapted the radiotap parser as a library outside the kernel. This brings the in-kernel parser up to speed. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/cfg80211.h | 83 ++++++++--- include/net/ieee80211_radiotap.h | 4 + net/mac80211/tx.c | 9 +- net/wireless/radiotap.c | 305 +++++++++++++++++++++++++-------------- 4 files changed, 271 insertions(+), 130 deletions(-) (limited to 'include') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index a3f0a7ed31a..5b3569b2a74 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1561,37 +1561,82 @@ ieee80211_get_response_rate(struct ieee80211_supported_band *sband, * Documentation in Documentation/networking/radiotap-headers.txt */ +struct radiotap_align_size { + uint8_t align:4, size:4; +}; + +struct ieee80211_radiotap_namespace { + const struct radiotap_align_size *align_size; + int n_bits; + uint32_t oui; + uint8_t subns; +}; + +struct ieee80211_radiotap_vendor_namespaces { + const struct ieee80211_radiotap_namespace *ns; + int n_ns; +}; + /** * struct ieee80211_radiotap_iterator - tracks walk thru present radiotap args - * @rtheader: pointer to the radiotap header we are walking through - * @max_length: length of radiotap header in cpu byte ordering - * @this_arg_index: IEEE80211_RADIOTAP_... index of current arg - * @this_arg: pointer to current radiotap arg - * @arg_index: internal next argument index - * @arg: internal next argument pointer - * @next_bitmap: internal pointer to next present u32 - * @bitmap_shifter: internal shifter for curr u32 bitmap, b0 set == arg present + * @this_arg_index: index of current arg, valid after each successful call + * to ieee80211_radiotap_iterator_next() + * @this_arg: pointer to current radiotap arg; it is valid after each + * call to ieee80211_radiotap_iterator_next() but also after + * ieee80211_radiotap_iterator_init() where it will point to + * the beginning of the actual data portion + * @this_arg_size: length of the current arg, for convenience + * @current_namespace: pointer to the current namespace definition + * (or internally %NULL if the current namespace is unknown) + * @is_radiotap_ns: indicates whether the current namespace is the default + * radiotap namespace or not + * + * @overrides: override standard radiotap fields + * @n_overrides: number of overrides + * + * @_rtheader: pointer to the radiotap header we are walking through + * @_max_length: length of radiotap header in cpu byte ordering + * @_arg_index: next argument index + * @_arg: next argument pointer + * @_next_bitmap: internal pointer to next present u32 + * @_bitmap_shifter: internal shifter for curr u32 bitmap, b0 set == arg present + * @_vns: vendor namespace definitions + * @_next_ns_data: beginning of the next namespace's data + * @_reset_on_ext: internal; reset the arg index to 0 when going to the + * next bitmap word + * + * Describes the radiotap parser state. Fields prefixed with an underscore + * must not be used by users of the parser, only by the parser internally. */ struct ieee80211_radiotap_iterator { - struct ieee80211_radiotap_header *rtheader; - int max_length; + struct ieee80211_radiotap_header *_rtheader; + const struct ieee80211_radiotap_vendor_namespaces *_vns; + const struct ieee80211_radiotap_namespace *current_namespace; + + unsigned char *_arg, *_next_ns_data; + uint32_t *_next_bitmap; + + unsigned char *this_arg; int this_arg_index; - u8 *this_arg; + int this_arg_size; - int arg_index; - u8 *arg; - __le32 *next_bitmap; - u32 bitmap_shifter; + int is_radiotap_ns; + + int _max_length; + int _arg_index; + uint32_t _bitmap_shifter; + int _reset_on_ext; }; extern int ieee80211_radiotap_iterator_init( - struct ieee80211_radiotap_iterator *iterator, - struct ieee80211_radiotap_header *radiotap_header, - int max_length); + struct ieee80211_radiotap_iterator *iterator, + struct ieee80211_radiotap_header *radiotap_header, + int max_length, const struct ieee80211_radiotap_vendor_namespaces *vns); extern int ieee80211_radiotap_iterator_next( - struct ieee80211_radiotap_iterator *iterator); + struct ieee80211_radiotap_iterator *iterator); + extern const unsigned char rfc1042_header[6]; extern const unsigned char bridge_tunnel_header[6]; diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h index 9d3d86aaccb..af49f8ab7f8 100644 --- a/include/net/ieee80211_radiotap.h +++ b/include/net/ieee80211_radiotap.h @@ -198,6 +198,10 @@ enum ieee80211_radiotap_type { IEEE80211_RADIOTAP_TX_FLAGS = 15, IEEE80211_RADIOTAP_RTS_RETRIES = 16, IEEE80211_RADIOTAP_DATA_RETRIES = 17, + + /* valid in every it_present bitmap, even vendor namespaces */ + IEEE80211_RADIOTAP_RADIOTAP_NAMESPACE = 29, + IEEE80211_RADIOTAP_VENDOR_NAMESPACE = 30, IEEE80211_RADIOTAP_EXT = 31 }; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 85e382aa894..e392820a4c3 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1010,7 +1010,8 @@ static bool __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx, (struct ieee80211_radiotap_header *) skb->data; struct ieee80211_supported_band *sband; struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len); + int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len, + NULL); sband = tx->local->hw.wiphy->bands[tx->channel->band]; @@ -1046,7 +1047,7 @@ static bool __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx, * because it will be recomputed and added * on transmission */ - if (skb->len < (iterator.max_length + FCS_LEN)) + if (skb->len < (iterator._max_length + FCS_LEN)) return false; skb_trim(skb, skb->len - FCS_LEN); @@ -1073,10 +1074,10 @@ static bool __ieee80211_parse_tx_radiotap(struct ieee80211_tx_data *tx, /* * remove the radiotap header - * iterator->max_length was sanity-checked against + * iterator->_max_length was sanity-checked against * skb->len by iterator init */ - skb_pull(skb, iterator.max_length); + skb_pull(skb, iterator._max_length); return true; } diff --git a/net/wireless/radiotap.c b/net/wireless/radiotap.c index f591871a7b4..1332c445d1c 100644 --- a/net/wireless/radiotap.c +++ b/net/wireless/radiotap.c @@ -2,6 +2,16 @@ * Radiotap parser * * Copyright 2007 Andy Green + * Copyright 2009 Johannes Berg + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Alternatively, this software may be distributed under the terms of BSD + * license. + * + * See COPYING for more details. */ #include @@ -10,6 +20,35 @@ /* function prototypes and related defs are in include/net/cfg80211.h */ +static const struct radiotap_align_size rtap_namespace_sizes[] = { + [IEEE80211_RADIOTAP_TSFT] = { .align = 8, .size = 8, }, + [IEEE80211_RADIOTAP_FLAGS] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_RATE] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_CHANNEL] = { .align = 2, .size = 4, }, + [IEEE80211_RADIOTAP_FHSS] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_DBM_ANTSIGNAL] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_DBM_ANTNOISE] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_LOCK_QUALITY] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_TX_ATTENUATION] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_DB_TX_ATTENUATION] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_DBM_TX_POWER] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_ANTENNA] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_DB_ANTSIGNAL] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_DB_ANTNOISE] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_RX_FLAGS] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_TX_FLAGS] = { .align = 2, .size = 2, }, + [IEEE80211_RADIOTAP_RTS_RETRIES] = { .align = 1, .size = 1, }, + [IEEE80211_RADIOTAP_DATA_RETRIES] = { .align = 1, .size = 1, }, + /* + * add more here as they are defined in radiotap.h + */ +}; + +static const struct ieee80211_radiotap_namespace radiotap_ns = { + .n_bits = sizeof(rtap_namespace_sizes) / sizeof(rtap_namespace_sizes[0]), + .align_size = rtap_namespace_sizes, +}; + /** * ieee80211_radiotap_iterator_init - radiotap parser iterator initialization * @iterator: radiotap_iterator to initialize @@ -50,9 +89,9 @@ */ int ieee80211_radiotap_iterator_init( - struct ieee80211_radiotap_iterator *iterator, - struct ieee80211_radiotap_header *radiotap_header, - int max_length) + struct ieee80211_radiotap_iterator *iterator, + struct ieee80211_radiotap_header *radiotap_header, + int max_length, const struct ieee80211_radiotap_vendor_namespaces *vns) { /* Linux only supports version 0 radiotap format */ if (radiotap_header->it_version) @@ -62,19 +101,24 @@ int ieee80211_radiotap_iterator_init( if (max_length < get_unaligned_le16(&radiotap_header->it_len)) return -EINVAL; - iterator->rtheader = radiotap_header; - iterator->max_length = get_unaligned_le16(&radiotap_header->it_len); - iterator->arg_index = 0; - iterator->bitmap_shifter = get_unaligned_le32(&radiotap_header->it_present); - iterator->arg = (u8 *)radiotap_header + sizeof(*radiotap_header); - iterator->this_arg = NULL; + iterator->_rtheader = radiotap_header; + iterator->_max_length = get_unaligned_le16(&radiotap_header->it_len); + iterator->_arg_index = 0; + iterator->_bitmap_shifter = get_unaligned_le32(&radiotap_header->it_present); + iterator->_arg = (uint8_t *)radiotap_header + sizeof(*radiotap_header); + iterator->_reset_on_ext = 0; + iterator->_next_bitmap = &radiotap_header->it_present; + iterator->_next_bitmap++; + iterator->_vns = vns; + iterator->current_namespace = &radiotap_ns; + iterator->is_radiotap_ns = 1; /* find payload start allowing for extended bitmap(s) */ - if (unlikely(iterator->bitmap_shifter & (1<arg) & - (1 << IEEE80211_RADIOTAP_EXT)) { - iterator->arg += sizeof(u32); + if (iterator->_bitmap_shifter & (1<_arg) & + (1 << IEEE80211_RADIOTAP_EXT)) { + iterator->_arg += sizeof(uint32_t); /* * check for insanity where the present bitmaps @@ -82,12 +126,13 @@ int ieee80211_radiotap_iterator_init( * stated radiotap header length */ - if (((ulong)iterator->arg - - (ulong)iterator->rtheader) > iterator->max_length) + if ((unsigned long)iterator->_arg - + (unsigned long)iterator->_rtheader > + (unsigned long)iterator->_max_length) return -EINVAL; } - iterator->arg += sizeof(u32); + iterator->_arg += sizeof(uint32_t); /* * no need to check again for blowing past stated radiotap @@ -96,12 +141,36 @@ int ieee80211_radiotap_iterator_init( */ } + iterator->this_arg = iterator->_arg; + /* we are all initialized happily */ return 0; } EXPORT_SYMBOL(ieee80211_radiotap_iterator_init); +static void find_ns(struct ieee80211_radiotap_iterator *iterator, + uint32_t oui, uint8_t subns) +{ + int i; + + iterator->current_namespace = NULL; + + if (!iterator->_vns) + return; + + for (i = 0; i < iterator->_vns->n_ns; i++) { + if (iterator->_vns->ns[i].oui != oui) + continue; + if (iterator->_vns->ns[i].subns != subns) + continue; + + iterator->current_namespace = &iterator->_vns->ns[i]; + break; + } +} + + /** * ieee80211_radiotap_iterator_next - return next radiotap parser iterator arg @@ -127,99 +196,80 @@ EXPORT_SYMBOL(ieee80211_radiotap_iterator_init); */ int ieee80211_radiotap_iterator_next( - struct ieee80211_radiotap_iterator *iterator) + struct ieee80211_radiotap_iterator *iterator) { - - /* - * small length lookup table for all radiotap types we heard of - * starting from b0 in the bitmap, so we can walk the payload - * area of the radiotap header - * - * There is a requirement to pad args, so that args - * of a given length must begin at a boundary of that length - * -- but note that compound args are allowed (eg, 2 x u16 - * for IEEE80211_RADIOTAP_CHANNEL) so total arg length is not - * a reliable indicator of alignment requirement. - * - * upper nybble: content alignment for arg - * lower nybble: content length for arg - */ - - static const u8 rt_sizes[] = { - [IEEE80211_RADIOTAP_TSFT] = 0x88, - [IEEE80211_RADIOTAP_FLAGS] = 0x11, - [IEEE80211_RADIOTAP_RATE] = 0x11, - [IEEE80211_RADIOTAP_CHANNEL] = 0x24, - [IEEE80211_RADIOTAP_FHSS] = 0x22, - [IEEE80211_RADIOTAP_DBM_ANTSIGNAL] = 0x11, - [IEEE80211_RADIOTAP_DBM_ANTNOISE] = 0x11, - [IEEE80211_RADIOTAP_LOCK_QUALITY] = 0x22, - [IEEE80211_RADIOTAP_TX_ATTENUATION] = 0x22, - [IEEE80211_RADIOTAP_DB_TX_ATTENUATION] = 0x22, - [IEEE80211_RADIOTAP_DBM_TX_POWER] = 0x11, - [IEEE80211_RADIOTAP_ANTENNA] = 0x11, - [IEEE80211_RADIOTAP_DB_ANTSIGNAL] = 0x11, - [IEEE80211_RADIOTAP_DB_ANTNOISE] = 0x11, - [IEEE80211_RADIOTAP_RX_FLAGS] = 0x22, - [IEEE80211_RADIOTAP_TX_FLAGS] = 0x22, - [IEEE80211_RADIOTAP_RTS_RETRIES] = 0x11, - [IEEE80211_RADIOTAP_DATA_RETRIES] = 0x11, - /* - * add more here as they are defined in - * include/net/ieee80211_radiotap.h - */ - }; - - /* - * for every radiotap entry we can at - * least skip (by knowing the length)... - */ - - while (iterator->arg_index < sizeof(rt_sizes)) { + while (1) { int hit = 0; - int pad; + int pad, align, size, subns, vnslen; + uint32_t oui; - if (!(iterator->bitmap_shifter & 1)) + /* if no more EXT bits, that's it */ + if ((iterator->_arg_index % 32) == IEEE80211_RADIOTAP_EXT && + !(iterator->_bitmap_shifter & 1)) + return -ENOENT; + + if (!(iterator->_bitmap_shifter & 1)) goto next_entry; /* arg not present */ + /* get alignment/size of data */ + switch (iterator->_arg_index % 32) { + case IEEE80211_RADIOTAP_RADIOTAP_NAMESPACE: + case IEEE80211_RADIOTAP_EXT: + align = 1; + size = 0; + break; + case IEEE80211_RADIOTAP_VENDOR_NAMESPACE: + align = 2; + size = 6; + break; + default: + if (!iterator->current_namespace || + iterator->_arg_index >= iterator->current_namespace->n_bits) { + if (iterator->current_namespace == &radiotap_ns) + return -ENOENT; + align = 0; + } else { + align = iterator->current_namespace->align_size[iterator->_arg_index].align; + size = iterator->current_namespace->align_size[iterator->_arg_index].size; + } + if (!align) { + /* skip all subsequent data */ + iterator->_arg = iterator->_next_ns_data; + /* give up on this namespace */ + iterator->current_namespace = NULL; + goto next_entry; + } + break; + } + /* * arg is present, account for alignment padding - * 8-bit args can be at any alignment - * 16-bit args must start on 16-bit boundary - * 32-bit args must start on 32-bit boundary - * 64-bit args must start on 64-bit boundary * - * note that total arg size can differ from alignment of - * elements inside arg, so we use upper nybble of length - * table to base alignment on - * - * also note: these alignments are ** relative to the - * start of the radiotap header **. There is no guarantee + * Note that these alignments are relative to the start + * of the radiotap header. There is no guarantee * that the radiotap header itself is aligned on any * kind of boundary. * - * the above is why get_unaligned() is used to dereference - * multibyte elements from the radiotap area + * The above is why get_unaligned() is used to dereference + * multibyte elements from the radiotap area. */ - pad = (((ulong)iterator->arg) - - ((ulong)iterator->rtheader)) & - ((rt_sizes[iterator->arg_index] >> 4) - 1); + pad = ((unsigned long)iterator->_arg - + (unsigned long)iterator->_rtheader) & (align - 1); if (pad) - iterator->arg += - (rt_sizes[iterator->arg_index] >> 4) - pad; + iterator->_arg += align - pad; /* * this is what we will return to user, but we need to * move on first so next call has something fresh to test */ - iterator->this_arg_index = iterator->arg_index; - iterator->this_arg = iterator->arg; - hit = 1; + iterator->this_arg_index = iterator->_arg_index; + iterator->this_arg = iterator->_arg; + iterator->this_arg_size = size; /* internally move on the size of this arg */ - iterator->arg += rt_sizes[iterator->arg_index] & 0x0f; + iterator->_arg += size; /* * check for insanity where we are given a bitmap that @@ -228,32 +278,73 @@ int ieee80211_radiotap_iterator_next( * max_length on the last arg, never exceeding it. */ - if (((ulong)iterator->arg - (ulong)iterator->rtheader) > - iterator->max_length) + if ((unsigned long)iterator->_arg - + (unsigned long)iterator->_rtheader > + (unsigned long)iterator->_max_length) return -EINVAL; - next_entry: - iterator->arg_index++; - if (unlikely((iterator->arg_index & 31) == 0)) { - /* completed current u32 bitmap */ - if (iterator->bitmap_shifter & 1) { - /* b31 was set, there is more */ - /* move to next u32 bitmap */ - iterator->bitmap_shifter = - get_unaligned_le32(iterator->next_bitmap); - iterator->next_bitmap++; - } else - /* no more bitmaps: end */ - iterator->arg_index = sizeof(rt_sizes); - } else /* just try the next bit */ - iterator->bitmap_shifter >>= 1; + /* these special ones are valid in each bitmap word */ + switch (iterator->_arg_index % 32) { + case IEEE80211_RADIOTAP_VENDOR_NAMESPACE: + iterator->_bitmap_shifter >>= 1; + iterator->_arg_index++; + + iterator->_reset_on_ext = 1; + + vnslen = get_unaligned_le16(iterator->this_arg + 4); + iterator->_next_ns_data = iterator->_arg + vnslen; + oui = (*iterator->this_arg << 16) | + (*(iterator->this_arg + 1) << 8) | + *(iterator->this_arg + 2); + subns = *(iterator->this_arg + 3); + + find_ns(iterator, oui, subns); + + iterator->is_radiotap_ns = 0; + /* allow parsers to show this information */ + iterator->this_arg_index = + IEEE80211_RADIOTAP_VENDOR_NAMESPACE; + iterator->this_arg_size += vnslen; + if ((unsigned long)iterator->this_arg + + iterator->this_arg_size - + (unsigned long)iterator->_rtheader > + (unsigned long)(unsigned long)iterator->_max_length) + return -EINVAL; + hit = 1; + break; + case IEEE80211_RADIOTAP_RADIOTAP_NAMESPACE: + iterator->_bitmap_shifter >>= 1; + iterator->_arg_index++; + + iterator->_reset_on_ext = 1; + iterator->current_namespace = &radiotap_ns; + iterator->is_radiotap_ns = 1; + break; + case IEEE80211_RADIOTAP_EXT: + /* + * bit 31 was set, there is more + * -- move to next u32 bitmap + */ + iterator->_bitmap_shifter = + get_unaligned_le32(iterator->_next_bitmap); + iterator->_next_bitmap++; + if (iterator->_reset_on_ext) + iterator->_arg_index = 0; + else + iterator->_arg_index++; + iterator->_reset_on_ext = 0; + break; + default: + /* we've got a hit! */ + hit = 1; + next_entry: + iterator->_bitmap_shifter >>= 1; + iterator->_arg_index++; + } /* if we found a valid arg earlier, return it now */ if (hit) return 0; } - - /* we don't know how to handle any more args, we're done */ - return -ENOENT; } EXPORT_SYMBOL(ieee80211_radiotap_iterator_next); -- cgit v1.2.3-70-g09d2 From 34e895075e21be3e21e71d6317440d1ee7969ad0 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 3 Feb 2010 13:59:58 +0100 Subject: mac80211: allow station add/remove to sleep Many drivers would like to sleep during station addition and removal, and currently have a high complexity there from not being able to. This introduces two new callbacks sta_add() and sta_remove() that drivers can implement instead of using sta_notify() and that can sleep, and the new sta_add() callback is also allowed to fail. The reason we didn't do this previously is that the IBSS code wants to insert stations from the RX path, which is a tasklet, so cannot sleep. This patch will keep the station allocation in that path, but moves adding the station to the driver out of line. Since the addition can now fail, we can have IBSS peer structs the driver rejected -- in that case we still talk to the station but never tell the driver about it in the control.sta pointer. If there will ever be a driver that has a low limit on the number of stations and that cannot talk to any stations that are not known to it, we need to do come up with a new strategy of handling larger IBSSs, maybe quicker expiry or rejecting peers. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 21 +- net/mac80211/cfg.c | 23 +- net/mac80211/driver-ops.h | 34 +++ net/mac80211/driver-trace.h | 52 ++++ net/mac80211/ibss.c | 22 +- net/mac80211/ieee80211_i.h | 18 +- net/mac80211/mesh_plink.c | 17 +- net/mac80211/mlme.c | 14 +- net/mac80211/pm.c | 10 +- net/mac80211/rx.c | 4 +- net/mac80211/sta_info.c | 731 ++++++++++++++++++++++---------------------- net/mac80211/sta_info.h | 32 +- net/mac80211/tx.c | 2 +- net/mac80211/util.c | 16 +- 14 files changed, 526 insertions(+), 470 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 74ccf30fdf8..a19fac35259 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -814,7 +814,7 @@ enum set_key_cmd { * mac80211, any ieee80211_sta pointer you get access to must * either be protected by rcu_read_lock() explicitly or implicitly, * or you must take good care to not use such a pointer after a - * call to your sta_notify callback that removed it. + * call to your sta_remove callback that removed it. * * @addr: MAC address * @aid: AID we assigned to the station if we're an AP @@ -840,8 +840,8 @@ struct ieee80211_sta { * indicates addition and removal of a station to station table, * or if a associated station made a power state transition. * - * @STA_NOTIFY_ADD: a station was added to the station table - * @STA_NOTIFY_REMOVE: a station being removed from the station table + * @STA_NOTIFY_ADD: (DEPRECATED) a station was added to the station table + * @STA_NOTIFY_REMOVE: (DEPRECATED) a station being removed from the station table * @STA_NOTIFY_SLEEP: a station is now sleeping * @STA_NOTIFY_AWAKE: a sleeping station woke up */ @@ -1534,9 +1534,14 @@ enum ieee80211_ampdu_mlme_action { * @set_rts_threshold: Configuration of RTS threshold (if device needs it) * The callback can sleep. * - * @sta_notify: Notifies low level driver about addition, removal or power - * state transition of an associated station, AP, IBSS/WDS/mesh peer etc. - * Must be atomic. + * @sta_add: Notifies low level driver about addition of an associated station, + * AP, IBSS/WDS/mesh peer etc. This callback can sleep. + * + * @sta_remove: Notifies low level driver about removal of an associated + * station, AP, IBSS/WDS/mesh peer etc. This callback can sleep. + * + * @sta_notify: Notifies low level driver about power state transition of an + * associated station, AP, IBSS/WDS/mesh peer etc. Must be atomic. * * @conf_tx: Configure TX queue parameters (EDCF (aifs, cw_min, cw_max), * bursting) for a hardware TX queue. @@ -1635,6 +1640,10 @@ struct ieee80211_ops { void (*get_tkip_seq)(struct ieee80211_hw *hw, u8 hw_key_idx, u32 *iv32, u16 *iv16); int (*set_rts_threshold)(struct ieee80211_hw *hw, u32 value); + int (*sta_add)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta); + int (*sta_remove)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta); void (*sta_notify)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, enum sta_notify_cmd, struct ieee80211_sta *sta); int (*conf_tx)(struct ieee80211_hw *hw, u16 queue, diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index facf233843e..a362523d8eb 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -747,9 +747,7 @@ static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev, layer2_update = sdata->vif.type == NL80211_IFTYPE_AP_VLAN || sdata->vif.type == NL80211_IFTYPE_AP; - rcu_read_lock(); - - err = sta_info_insert(sta); + err = sta_info_insert_rcu(sta); if (err) { rcu_read_unlock(); return err; @@ -768,26 +766,13 @@ static int ieee80211_del_station(struct wiphy *wiphy, struct net_device *dev, { struct ieee80211_local *local = wiphy_priv(wiphy); struct ieee80211_sub_if_data *sdata; - struct sta_info *sta; sdata = IEEE80211_DEV_TO_SUB_IF(dev); - if (mac) { - rcu_read_lock(); - - sta = sta_info_get_bss(sdata, mac); - if (!sta) { - rcu_read_unlock(); - return -ENOENT; - } - - sta_info_unlink(&sta); - rcu_read_unlock(); - - sta_info_destroy(sta); - } else - sta_info_flush(local, sdata); + if (mac) + return sta_info_destroy_addr_bss(sdata, mac); + sta_info_flush(local, sdata); return 0; } diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 6c31f38ac7f..855e85b5506 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -243,6 +243,40 @@ static inline void drv_sta_notify(struct ieee80211_local *local, trace_drv_sta_notify(local, sdata, cmd, sta); } +static inline int drv_sta_add(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta) +{ + int ret = 0; + + might_sleep(); + + if (local->ops->sta_add) + ret = local->ops->sta_add(&local->hw, &sdata->vif, sta); + else if (local->ops->sta_notify) + local->ops->sta_notify(&local->hw, &sdata->vif, + STA_NOTIFY_ADD, sta); + + trace_drv_sta_add(local, sdata, sta, ret); + + return ret; +} + +static inline void drv_sta_remove(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta) +{ + might_sleep(); + + if (local->ops->sta_remove) + local->ops->sta_remove(&local->hw, &sdata->vif, sta); + else if (local->ops->sta_notify) + local->ops->sta_notify(&local->hw, &sdata->vif, + STA_NOTIFY_REMOVE, sta); + + trace_drv_sta_remove(local, sdata, sta); +} + static inline int drv_conf_tx(struct ieee80211_local *local, u16 queue, const struct ieee80211_tx_queue_params *params) { diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index 502424b2538..c984910bf27 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -545,6 +545,58 @@ TRACE_EVENT(drv_sta_notify, ) ); +TRACE_EVENT(drv_sta_add, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta, int ret), + + TP_ARGS(local, sdata, sta, ret), + + TP_STRUCT__entry( + LOCAL_ENTRY + VIF_ENTRY + STA_ENTRY + __field(int, ret) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + VIF_ASSIGN; + STA_ASSIGN; + __entry->ret = ret; + ), + + TP_printk( + LOCAL_PR_FMT VIF_PR_FMT STA_PR_FMT " ret:%d", + LOCAL_PR_ARG, VIF_PR_ARG, STA_PR_ARG, __entry->ret + ) +); + +TRACE_EVENT(drv_sta_remove, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta), + + TP_ARGS(local, sdata, sta), + + TP_STRUCT__entry( + LOCAL_ENTRY + VIF_ENTRY + STA_ENTRY + ), + + TP_fast_assign( + LOCAL_ASSIGN; + VIF_ASSIGN; + STA_ASSIGN; + ), + + TP_printk( + LOCAL_PR_FMT VIF_PR_FMT STA_PR_FMT, + LOCAL_PR_ARG, VIF_PR_ARG, STA_PR_ARG + ) +); + TRACE_EVENT(drv_conf_tx, TP_PROTO(struct ieee80211_local *local, u16 queue, const struct ieee80211_tx_queue_params *params, diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c index 85c4ba14c77..f3e94248674 100644 --- a/net/mac80211/ibss.c +++ b/net/mac80211/ibss.c @@ -275,10 +275,12 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, (unsigned long long) supp_rates, (unsigned long long) sta->sta.supp_rates[band]); #endif - } else - ieee80211_ibss_add_sta(sdata, mgmt->bssid, mgmt->sa, supp_rates); - - rcu_read_unlock(); + rcu_read_unlock(); + } else { + rcu_read_unlock(); + ieee80211_ibss_add_sta(sdata, mgmt->bssid, mgmt->sa, + supp_rates, GFP_KERNEL); + } } bss = ieee80211_bss_info_update(local, rx_status, mgmt, len, elems, @@ -368,7 +370,8 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, sdata->name, mgmt->bssid); #endif ieee80211_sta_join_ibss(sdata, bss); - ieee80211_ibss_add_sta(sdata, mgmt->bssid, mgmt->sa, supp_rates); + ieee80211_ibss_add_sta(sdata, mgmt->bssid, mgmt->sa, + supp_rates, GFP_KERNEL); } put_bss: @@ -381,7 +384,8 @@ static void ieee80211_rx_bss_info(struct ieee80211_sub_if_data *sdata, * must be callable in atomic context. */ struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata, - u8 *bssid,u8 *addr, u32 supp_rates) + u8 *bssid,u8 *addr, u32 supp_rates, + gfp_t gfp) { struct ieee80211_if_ibss *ifibss = &sdata->u.ibss; struct ieee80211_local *local = sdata->local; @@ -410,7 +414,7 @@ struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata, wiphy_name(local->hw.wiphy), addr, sdata->name); #endif - sta = sta_info_alloc(sdata, addr, GFP_ATOMIC); + sta = sta_info_alloc(sdata, addr, gfp); if (!sta) return NULL; @@ -422,9 +426,9 @@ struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata, rate_control_rate_init(sta); + /* If it fails, maybe we raced another insertion? */ if (sta_info_insert(sta)) - return NULL; - + return sta_info_get(sdata, addr); return sta; } diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 3067fbd69d6..a5911191f22 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -688,15 +688,18 @@ struct ieee80211_local { /* Station data */ /* - * The lock only protects the list, hash, timer and counter - * against manipulation, reads are done in RCU. Additionally, - * the lock protects each BSS's TIM bitmap. + * The mutex only protects the list and counter, + * reads are done in RCU. + * Additionally, the lock protects the hash table, + * the pending list and each BSS's TIM bitmap. */ + struct mutex sta_mtx; spinlock_t sta_lock; unsigned long num_sta; - struct list_head sta_list; + struct list_head sta_list, sta_pending_list; struct sta_info *sta_hash[STA_HASH_SIZE]; struct timer_list sta_cleanup; + struct work_struct sta_finish_work; int sta_generation; struct sk_buff_head pending[IEEE80211_MAX_QUEUES]; @@ -770,10 +773,6 @@ struct ieee80211_local { assoc_led_name[32], radio_led_name[32]; #endif -#ifdef CONFIG_MAC80211_DEBUGFS - struct work_struct sta_debugfs_add; -#endif - #ifdef CONFIG_MAC80211_DEBUG_COUNTERS /* TX/RX handler statistics */ unsigned int tx_handlers_drop; @@ -985,7 +984,8 @@ void ieee80211_ibss_setup_sdata(struct ieee80211_sub_if_data *sdata); ieee80211_rx_result ieee80211_ibss_rx_mgmt(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); struct sta_info *ieee80211_ibss_add_sta(struct ieee80211_sub_if_data *sdata, - u8 *bssid, u8 *addr, u32 supp_rates); + u8 *bssid, u8 *addr, u32 supp_rates, + gfp_t gfp); int ieee80211_ibss_join(struct ieee80211_sub_if_data *sdata, struct cfg80211_ibss_params *params); int ieee80211_ibss_leave(struct ieee80211_sub_if_data *sdata); diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index 7985e515089..bc4e20e57ff 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -102,7 +102,7 @@ static struct sta_info *mesh_plink_alloc(struct ieee80211_sub_if_data *sdata, if (local->num_sta >= MESH_MAX_PLINKS) return NULL; - sta = sta_info_alloc(sdata, hw_addr, GFP_ATOMIC); + sta = sta_info_alloc(sdata, hw_addr, GFP_KERNEL); if (!sta) return NULL; @@ -236,12 +236,12 @@ void mesh_neighbour_update(u8 *hw_addr, u32 rates, struct ieee80211_sub_if_data sta = sta_info_get(sdata, hw_addr); if (!sta) { + rcu_read_unlock(); + sta = mesh_plink_alloc(sdata, hw_addr, rates); - if (!sta) { - rcu_read_unlock(); + if (!sta) return; - } - if (sta_info_insert(sta)) { + if (sta_info_insert_rcu(sta)) { rcu_read_unlock(); return; } @@ -485,9 +485,11 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m } else if (!sta) { /* ftype == PLINK_OPEN */ u32 rates; + + rcu_read_unlock(); + if (!mesh_plink_free_count(sdata)) { mpl_dbg("Mesh plink error: no more free plinks\n"); - rcu_read_unlock(); return; } @@ -495,10 +497,9 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, struct ieee80211_m sta = mesh_plink_alloc(sdata, mgmt->sa, rates); if (!sta) { mpl_dbg("Mesh plink error: plink table full\n"); - rcu_read_unlock(); return; } - if (sta_info_insert(sta)) { + if (sta_info_insert_rcu(sta)) { rcu_read_unlock(); return; } diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index f437284830e..ac9429e8d72 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -822,19 +822,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata) changed |= BSS_CHANGED_BSSID; ieee80211_bss_info_change_notify(sdata, changed); - rcu_read_lock(); - - sta = sta_info_get(sdata, bssid); - if (!sta) { - rcu_read_unlock(); - return; - } - - sta_info_unlink(&sta); - - rcu_read_unlock(); - - sta_info_destroy(sta); + sta_info_destroy_addr(sdata, bssid); } void ieee80211_sta_rx_notify(struct ieee80211_sub_if_data *sdata, diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index 47f818959ad..0e64484e861 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -11,7 +11,6 @@ int __ieee80211_suspend(struct ieee80211_hw *hw) struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata; struct sta_info *sta; - unsigned long flags; ieee80211_scan_cancel(local); @@ -55,22 +54,21 @@ int __ieee80211_suspend(struct ieee80211_hw *hw) rcu_read_unlock(); /* remove STAs */ - spin_lock_irqsave(&local->sta_lock, flags); + mutex_lock(&local->sta_mtx); list_for_each_entry(sta, &local->sta_list, list) { - if (local->ops->sta_notify) { + if (sta->uploaded) { sdata = sta->sdata; if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) sdata = container_of(sdata->bss, struct ieee80211_sub_if_data, u.ap); - drv_sta_notify(local, sdata, STA_NOTIFY_REMOVE, - &sta->sta); + drv_sta_remove(local, sdata, &sta->sta); } mesh_plink_quiesce(sta); } - spin_unlock_irqrestore(&local->sta_lock, flags); + mutex_unlock(&local->sta_mtx); /* remove all interfaces */ list_for_each_entry(sdata, &local->interfaces, list) { diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 5709307fcb9..01dba761839 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2244,8 +2244,8 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, rate_idx = 0; /* TODO: HT rates */ else rate_idx = status->rate_idx; - rx->sta = ieee80211_ibss_add_sta(sdata, bssid, hdr->addr2, - BIT(rate_idx)); + rx->sta = ieee80211_ibss_add_sta(sdata, bssid, + hdr->addr2, BIT(rate_idx), GFP_ATOMIC); } break; case NL80211_IFTYPE_MESH_POINT: diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index f735826f055..211c475f73c 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -32,49 +32,33 @@ * for faster lookup and a list for iteration. They are managed using * RCU, i.e. access to the list and hash table is protected by RCU. * - * Upon allocating a STA info structure with sta_info_alloc(), the caller owns - * that structure. It must then either destroy it using sta_info_destroy() - * (which is pretty useless) or insert it into the hash table using - * sta_info_insert() which demotes the reference from ownership to a regular - * RCU-protected reference; if the function is called without protection by an - * RCU critical section the reference is instantly invalidated. Note that the - * caller may not do much with the STA info before inserting it, in particular, - * it may not start any mesh peer link management or add encryption keys. + * Upon allocating a STA info structure with sta_info_alloc(), the caller + * owns that structure. It must then insert it into the hash table using + * either sta_info_insert() or sta_info_insert_rcu(); only in the latter + * case (which acquires an rcu read section but must not be called from + * within one) will the pointer still be valid after the call. Note that + * the caller may not do much with the STA info before inserting it, in + * particular, it may not start any mesh peer link management or add + * encryption keys. * * When the insertion fails (sta_info_insert()) returns non-zero), the * structure will have been freed by sta_info_insert()! * - * sta entries are added by mac80211 when you establish a link with a + * Station entries are added by mac80211 when you establish a link with a * peer. This means different things for the different type of interfaces * we support. For a regular station this mean we add the AP sta when we * receive an assocation response from the AP. For IBSS this occurs when - * we receive a probe response or a beacon from target IBSS network. For - * WDS we add the sta for the peer imediately upon device open. When using - * AP mode we add stations for each respective station upon request from - * userspace through nl80211. + * get to know about a peer on the same IBSS. For WDS we add the sta for + * the peer imediately upon device open. When using AP mode we add stations + * for each respective station upon request from userspace through nl80211. * - * Because there are debugfs entries for each station, and adding those - * must be able to sleep, it is also possible to "pin" a station entry, - * that means it can be removed from the hash table but not be freed. - * See the comment in __sta_info_unlink() for more information, this is - * an internal capability only. + * In order to remove a STA info structure, various sta_info_destroy_*() + * calls are available. * - * In order to remove a STA info structure, the caller needs to first - * unlink it (sta_info_unlink()) from the list and hash tables and - * then destroy it; sta_info_destroy() will wait for an RCU grace period - * to elapse before actually freeing it. Due to the pinning and the - * possibility of multiple callers trying to remove the same STA info at - * the same time, sta_info_unlink() can clear the STA info pointer it is - * passed to indicate that the STA info is owned by somebody else now. - * - * If sta_info_unlink() did not clear the pointer then the caller owns - * the STA info structure now and is responsible of destroying it with - * a call to sta_info_destroy(). - * - * In all other cases, there is no concept of ownership on a STA entry, - * each structure is owned by the global hash table/list until it is - * removed. All users of the structure need to be RCU protected so that - * the structure won't be freed before they are done using it. + * There is no concept of ownership on a STA entry, each structure is + * owned by the global hash table/list until it is removed. All users of + * the structure need to be RCU protected so that the structure won't be + * freed before they are done using it. */ /* Caller must hold local->sta_lock */ @@ -185,101 +169,6 @@ static void __sta_info_free(struct ieee80211_local *local, kfree(sta); } -void sta_info_destroy(struct sta_info *sta) -{ - struct ieee80211_local *local; - struct sk_buff *skb; - int i; - - might_sleep(); - - if (!sta) - return; - - local = sta->local; - - cancel_work_sync(&sta->drv_unblock_wk); - - rate_control_remove_sta_debugfs(sta); - ieee80211_sta_debugfs_remove(sta); - -#ifdef CONFIG_MAC80211_MESH - if (ieee80211_vif_is_mesh(&sta->sdata->vif)) - mesh_plink_deactivate(sta); -#endif - - /* - * We have only unlinked the key, and actually destroying it - * may mean it is removed from hardware which requires that - * the key->sta pointer is still valid, so flush the key todo - * list here. - * - * ieee80211_key_todo() will synchronize_rcu() so after this - * nothing can reference this sta struct any more. - */ - ieee80211_key_todo(); - -#ifdef CONFIG_MAC80211_MESH - if (ieee80211_vif_is_mesh(&sta->sdata->vif)) - del_timer_sync(&sta->plink_timer); -#endif - - while ((skb = skb_dequeue(&sta->ps_tx_buf)) != NULL) { - local->total_ps_buffered--; - dev_kfree_skb_any(skb); - } - - while ((skb = skb_dequeue(&sta->tx_filtered)) != NULL) - dev_kfree_skb_any(skb); - - for (i = 0; i < STA_TID_NUM; i++) { - struct tid_ampdu_rx *tid_rx; - struct tid_ampdu_tx *tid_tx; - - spin_lock_bh(&sta->lock); - tid_rx = sta->ampdu_mlme.tid_rx[i]; - /* Make sure timer won't free the tid_rx struct, see below */ - if (tid_rx) - tid_rx->shutdown = true; - - spin_unlock_bh(&sta->lock); - - /* - * Outside spinlock - shutdown is true now so that the timer - * won't free tid_rx, we have to do that now. Can't let the - * timer do it because we have to sync the timer outside the - * lock that it takes itself. - */ - if (tid_rx) { - del_timer_sync(&tid_rx->session_timer); - kfree(tid_rx); - } - - /* - * No need to do such complications for TX agg sessions, the - * path leading to freeing the tid_tx struct goes via a call - * from the driver, and thus needs to look up the sta struct - * again, which cannot be found when we get here. Hence, we - * just need to delete the timer and free the aggregation - * info; we won't be telling the peer about it then but that - * doesn't matter if we're not talking to it again anyway. - */ - tid_tx = sta->ampdu_mlme.tid_tx[i]; - if (tid_tx) { - del_timer_sync(&tid_tx->addba_resp_timer); - /* - * STA removed while aggregation session being - * started? Bit odd, but purge frames anyway. - */ - skb_queue_purge(&tid_tx->pending); - kfree(tid_tx); - } - } - - __sta_info_free(local, sta); -} - - /* Caller must hold local->sta_lock */ static void sta_info_hash_add(struct ieee80211_local *local, struct sta_info *sta) @@ -376,7 +265,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata, return sta; } -int sta_info_insert(struct sta_info *sta) +static int sta_info_finish_insert(struct sta_info *sta, bool async) { struct ieee80211_local *local = sta->local; struct ieee80211_sub_if_data *sdata = sta->sdata; @@ -384,6 +273,91 @@ int sta_info_insert(struct sta_info *sta) unsigned long flags; int err = 0; + WARN_ON(!mutex_is_locked(&local->sta_mtx)); + + /* notify driver */ + if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + sdata = container_of(sdata->bss, + struct ieee80211_sub_if_data, + u.ap); + err = drv_sta_add(local, sdata, &sta->sta); + if (err) { + if (!async) + return err; + printk(KERN_DEBUG "%s: failed to add IBSS STA %pM to driver (%d)" + " - keeping it anyway.\n", + sdata->name, sta->sta.addr, err); + } else { + sta->uploaded = true; +#ifdef CONFIG_MAC80211_VERBOSE_DEBUG + if (async) + printk(KERN_DEBUG "%s: Finished adding IBSS STA %pM\n", + wiphy_name(local->hw.wiphy), sta->sta.addr); +#endif + } + + sdata = sta->sdata; + + if (!async) { + local->num_sta++; + local->sta_generation++; + smp_mb(); + + /* make the station visible */ + spin_lock_irqsave(&local->sta_lock, flags); + sta_info_hash_add(local, sta); + spin_unlock_irqrestore(&local->sta_lock, flags); + } + + list_add(&sta->list, &local->sta_list); + + ieee80211_sta_debugfs_add(sta); + rate_control_add_sta_debugfs(sta); + + sinfo.filled = 0; + sinfo.generation = local->sta_generation; + cfg80211_new_sta(sdata->dev, sta->sta.addr, &sinfo, GFP_KERNEL); + + + return 0; +} + +static void sta_info_finish_pending(struct ieee80211_local *local) +{ + struct sta_info *sta; + unsigned long flags; + + spin_lock_irqsave(&local->sta_lock, flags); + while (!list_empty(&local->sta_pending_list)) { + sta = list_first_entry(&local->sta_pending_list, + struct sta_info, list); + list_del(&sta->list); + spin_unlock_irqrestore(&local->sta_lock, flags); + + sta_info_finish_insert(sta, true); + + spin_lock_irqsave(&local->sta_lock, flags); + } + spin_unlock_irqrestore(&local->sta_lock, flags); +} + +static void sta_info_finish_work(struct work_struct *work) +{ + struct ieee80211_local *local = + container_of(work, struct ieee80211_local, sta_finish_work); + + mutex_lock(&local->sta_mtx); + sta_info_finish_pending(local); + mutex_unlock(&local->sta_mtx); +} + +int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU) +{ + struct ieee80211_local *local = sta->local; + struct ieee80211_sub_if_data *sdata = sta->sdata; + unsigned long flags; + int err = 0; + /* * Can't be a WARN_ON because it can be triggered through a race: * something inserts a STA (on one CPU) without holding the RTNL @@ -391,36 +365,87 @@ int sta_info_insert(struct sta_info *sta) */ if (unlikely(!ieee80211_sdata_running(sdata))) { err = -ENETDOWN; + rcu_read_lock(); goto out_free; } if (WARN_ON(compare_ether_addr(sta->sta.addr, sdata->vif.addr) == 0 || is_multicast_ether_addr(sta->sta.addr))) { err = -EINVAL; + rcu_read_lock(); goto out_free; } + /* + * In ad-hoc mode, we sometimes need to insert stations + * from tasklet context from the RX path. To avoid races, + * always do so in that case -- see the comment below. + */ + if (sdata->vif.type == NL80211_IFTYPE_ADHOC) { + spin_lock_irqsave(&local->sta_lock, flags); + /* check if STA exists already */ + if (sta_info_get_bss(sdata, sta->sta.addr)) { + spin_unlock_irqrestore(&local->sta_lock, flags); + rcu_read_lock(); + err = -EEXIST; + goto out_free; + } + + local->num_sta++; + local->sta_generation++; + smp_mb(); + sta_info_hash_add(local, sta); + + list_add_tail(&sta->list, &local->sta_pending_list); + + rcu_read_lock(); + spin_unlock_irqrestore(&local->sta_lock, flags); + +#ifdef CONFIG_MAC80211_VERBOSE_DEBUG + printk(KERN_DEBUG "%s: Added IBSS STA %pM\n", + wiphy_name(local->hw.wiphy), sta->sta.addr); +#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ + + ieee80211_queue_work(&local->hw, &local->sta_finish_work); + + return 0; + } + + /* + * On first glance, this will look racy, because the code + * below this point, which inserts a station with sleeping, + * unlocks the sta_lock between checking existence in the + * hash table and inserting into it. + * + * However, it is not racy against itself because it keeps + * the mutex locked. It still seems to race against the + * above code that atomically inserts the station... That, + * however, is not true because the above code can only + * be invoked for IBSS interfaces, and the below code will + * not be -- and the two do not race against each other as + * the hash table also keys off the interface. + */ + + might_sleep(); + + mutex_lock(&local->sta_mtx); + spin_lock_irqsave(&local->sta_lock, flags); /* check if STA exists already */ - if (sta_info_get(sdata, sta->sta.addr)) { + if (sta_info_get_bss(sdata, sta->sta.addr)) { spin_unlock_irqrestore(&local->sta_lock, flags); + rcu_read_lock(); err = -EEXIST; goto out_free; } - list_add(&sta->list, &local->sta_list); - local->sta_generation++; - local->num_sta++; - sta_info_hash_add(local, sta); - /* notify driver */ - if (local->ops->sta_notify) { - if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - sdata = container_of(sdata->bss, - struct ieee80211_sub_if_data, - u.ap); + spin_unlock_irqrestore(&local->sta_lock, flags); - drv_sta_notify(local, sdata, STA_NOTIFY_ADD, &sta->sta); - sdata = sta->sdata; + err = sta_info_finish_insert(sta, false); + if (err) { + mutex_unlock(&local->sta_mtx); + rcu_read_lock(); + goto out_free; } #ifdef CONFIG_MAC80211_VERBOSE_DEBUG @@ -428,22 +453,9 @@ int sta_info_insert(struct sta_info *sta) wiphy_name(local->hw.wiphy), sta->sta.addr); #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ - spin_unlock_irqrestore(&local->sta_lock, flags); - - sinfo.filled = 0; - sinfo.generation = local->sta_generation; - cfg80211_new_sta(sdata->dev, sta->sta.addr, &sinfo, GFP_ATOMIC); - -#ifdef CONFIG_MAC80211_DEBUGFS - /* - * Debugfs entry adding might sleep, so schedule process - * context task for adding entry for STAs that do not yet - * have one. - * NOTE: due to auto-freeing semantics this may only be done - * if the insertion is successful! - */ - schedule_work(&local->sta_debugfs_add); -#endif + /* move reference to rcu-protected */ + rcu_read_lock(); + mutex_unlock(&local->sta_mtx); if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_accept_plinks_update(sdata); @@ -455,6 +467,15 @@ int sta_info_insert(struct sta_info *sta) return err; } +int sta_info_insert(struct sta_info *sta) +{ + int err = sta_info_insert_rcu(sta); + + rcu_read_unlock(); + + return err; +} + static inline void __bss_tim_set(struct ieee80211_if_ap *bss, u16 aid) { /* @@ -523,108 +544,6 @@ void sta_info_clear_tim_bit(struct sta_info *sta) spin_unlock_irqrestore(&sta->local->sta_lock, flags); } -static void __sta_info_unlink(struct sta_info **sta) -{ - struct ieee80211_local *local = (*sta)->local; - struct ieee80211_sub_if_data *sdata = (*sta)->sdata; - /* - * pull caller's reference if we're already gone. - */ - if (sta_info_hash_del(local, *sta)) { - *sta = NULL; - return; - } - - if ((*sta)->key) { - ieee80211_key_free((*sta)->key); - WARN_ON((*sta)->key); - } - - list_del(&(*sta)->list); - (*sta)->dead = true; - - if (test_and_clear_sta_flags(*sta, - WLAN_STA_PS_STA | WLAN_STA_PS_DRIVER)) { - BUG_ON(!sdata->bss); - - atomic_dec(&sdata->bss->num_sta_ps); - __sta_info_clear_tim_bit(sdata->bss, *sta); - } - - local->num_sta--; - local->sta_generation++; - - if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - rcu_assign_pointer(sdata->u.vlan.sta, NULL); - - if (local->ops->sta_notify) { - if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - sdata = container_of(sdata->bss, - struct ieee80211_sub_if_data, - u.ap); - - drv_sta_notify(local, sdata, STA_NOTIFY_REMOVE, - &(*sta)->sta); - sdata = (*sta)->sdata; - } - - if (ieee80211_vif_is_mesh(&sdata->vif)) { - mesh_accept_plinks_update(sdata); -#ifdef CONFIG_MAC80211_MESH - del_timer(&(*sta)->plink_timer); -#endif - } - -#ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "%s: Removed STA %pM\n", - wiphy_name(local->hw.wiphy), (*sta)->sta.addr); -#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ - - /* - * Finally, pull caller's reference if the STA is pinned by the - * task that is adding the debugfs entries. In that case, we - * leave the STA "to be freed". - * - * The rules are not trivial, but not too complex either: - * (1) pin_status is only modified under the sta_lock - * (2) STAs may only be pinned under the RTNL so that - * sta_info_flush() is guaranteed to actually destroy - * all STAs that are active for a given interface, this - * is required for correctness because otherwise we - * could notify a driver that an interface is going - * away and only after that (!) notify it about a STA - * on that interface going away. - * (3) sta_info_debugfs_add_work() will set the status - * to PINNED when it found an item that needs a new - * debugfs directory created. In that case, that item - * must not be freed although all *RCU* users are done - * with it. Hence, we tell the caller of _unlink() - * that the item is already gone (as can happen when - * two tasks try to unlink/destroy at the same time) - * (4) We set the pin_status to DESTROY here when we - * find such an item. - * (5) sta_info_debugfs_add_work() will reset the pin_status - * from PINNED to NORMAL when it is done with the item, - * but will check for DESTROY before resetting it in - * which case it will free the item. - */ - if ((*sta)->pin_status == STA_INFO_PIN_STAT_PINNED) { - (*sta)->pin_status = STA_INFO_PIN_STAT_DESTROY; - *sta = NULL; - return; - } -} - -void sta_info_unlink(struct sta_info **sta) -{ - struct ieee80211_local *local = (*sta)->local; - unsigned long flags; - - spin_lock_irqsave(&local->sta_lock, flags); - __sta_info_unlink(sta); - spin_unlock_irqrestore(&local->sta_lock, flags); -} - static int sta_info_buffer_expired(struct sta_info *sta, struct sk_buff *skb) { @@ -681,109 +600,209 @@ static void sta_info_cleanup_expire_buffered(struct ieee80211_local *local, } } - -static void sta_info_cleanup(unsigned long data) +static int __must_check __sta_info_destroy(struct sta_info *sta) { - struct ieee80211_local *local = (struct ieee80211_local *) data; - struct sta_info *sta; + struct ieee80211_local *local; + struct ieee80211_sub_if_data *sdata; + struct sk_buff *skb; + unsigned long flags; + int ret, i; - rcu_read_lock(); - list_for_each_entry_rcu(sta, &local->sta_list, list) - sta_info_cleanup_expire_buffered(local, sta); - rcu_read_unlock(); + might_sleep(); - if (local->quiescing) - return; + if (!sta) + return -ENOENT; - local->sta_cleanup.expires = - round_jiffies(jiffies + STA_INFO_CLEANUP_INTERVAL); - add_timer(&local->sta_cleanup); -} + local = sta->local; + sdata = sta->sdata; -#ifdef CONFIG_MAC80211_DEBUGFS -/* - * See comment in __sta_info_unlink, - * caller must hold local->sta_lock. - */ -static void __sta_info_pin(struct sta_info *sta) -{ - WARN_ON(sta->pin_status != STA_INFO_PIN_STAT_NORMAL); - sta->pin_status = STA_INFO_PIN_STAT_PINNED; + spin_lock_irqsave(&local->sta_lock, flags); + ret = sta_info_hash_del(local, sta); + /* this might still be the pending list ... which is fine */ + if (!ret) + list_del(&sta->list); + spin_unlock_irqrestore(&local->sta_lock, flags); + if (ret) + return ret; + + if (sta->key) { + ieee80211_key_free(sta->key); + /* + * We have only unlinked the key, and actually destroying it + * may mean it is removed from hardware which requires that + * the key->sta pointer is still valid, so flush the key todo + * list here. + * + * ieee80211_key_todo() will synchronize_rcu() so after this + * nothing can reference this sta struct any more. + */ + ieee80211_key_todo(); + + WARN_ON(sta->key); + } + + sta->dead = true; + + if (test_and_clear_sta_flags(sta, + WLAN_STA_PS_STA | WLAN_STA_PS_DRIVER)) { + BUG_ON(!sdata->bss); + + atomic_dec(&sdata->bss->num_sta_ps); + __sta_info_clear_tim_bit(sdata->bss, sta); + } + + local->num_sta--; + local->sta_generation++; + + if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + rcu_assign_pointer(sdata->u.vlan.sta, NULL); + + if (sta->uploaded) { + if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + sdata = container_of(sdata->bss, + struct ieee80211_sub_if_data, + u.ap); + drv_sta_remove(local, sdata, &sta->sta); + sdata = sta->sdata; + } + +#ifdef CONFIG_MAC80211_MESH + if (ieee80211_vif_is_mesh(&sdata->vif)) { + mesh_accept_plinks_update(sdata); + del_timer(&sta->plink_timer); + } +#endif + +#ifdef CONFIG_MAC80211_VERBOSE_DEBUG + printk(KERN_DEBUG "%s: Removed STA %pM\n", + wiphy_name(local->hw.wiphy), sta->sta.addr); +#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ + cancel_work_sync(&sta->drv_unblock_wk); + + rate_control_remove_sta_debugfs(sta); + ieee80211_sta_debugfs_remove(sta); + +#ifdef CONFIG_MAC80211_MESH + if (ieee80211_vif_is_mesh(&sta->sdata->vif)) { + mesh_plink_deactivate(sta); + del_timer_sync(&sta->plink_timer); + } +#endif + + while ((skb = skb_dequeue(&sta->ps_tx_buf)) != NULL) { + local->total_ps_buffered--; + dev_kfree_skb_any(skb); + } + + while ((skb = skb_dequeue(&sta->tx_filtered)) != NULL) + dev_kfree_skb_any(skb); + + for (i = 0; i < STA_TID_NUM; i++) { + struct tid_ampdu_rx *tid_rx; + struct tid_ampdu_tx *tid_tx; + + spin_lock_bh(&sta->lock); + tid_rx = sta->ampdu_mlme.tid_rx[i]; + /* Make sure timer won't free the tid_rx struct, see below */ + if (tid_rx) + tid_rx->shutdown = true; + + spin_unlock_bh(&sta->lock); + + /* + * Outside spinlock - shutdown is true now so that the timer + * won't free tid_rx, we have to do that now. Can't let the + * timer do it because we have to sync the timer outside the + * lock that it takes itself. + */ + if (tid_rx) { + del_timer_sync(&tid_rx->session_timer); + kfree(tid_rx); + } + + /* + * No need to do such complications for TX agg sessions, the + * path leading to freeing the tid_tx struct goes via a call + * from the driver, and thus needs to look up the sta struct + * again, which cannot be found when we get here. Hence, we + * just need to delete the timer and free the aggregation + * info; we won't be telling the peer about it then but that + * doesn't matter if we're not talking to it again anyway. + */ + tid_tx = sta->ampdu_mlme.tid_tx[i]; + if (tid_tx) { + del_timer_sync(&tid_tx->addba_resp_timer); + /* + * STA removed while aggregation session being + * started? Bit odd, but purge frames anyway. + */ + skb_queue_purge(&tid_tx->pending); + kfree(tid_tx); + } + } + + __sta_info_free(local, sta); + + return 0; } -/* - * See comment in __sta_info_unlink, returns sta if it - * needs to be destroyed. - */ -static struct sta_info *__sta_info_unpin(struct sta_info *sta) +int sta_info_destroy_addr(struct ieee80211_sub_if_data *sdata, const u8 *addr) { - struct sta_info *ret = NULL; - unsigned long flags; + struct sta_info *sta; + int ret; - spin_lock_irqsave(&sta->local->sta_lock, flags); - WARN_ON(sta->pin_status != STA_INFO_PIN_STAT_DESTROY && - sta->pin_status != STA_INFO_PIN_STAT_PINNED); - if (sta->pin_status == STA_INFO_PIN_STAT_DESTROY) - ret = sta; - sta->pin_status = STA_INFO_PIN_STAT_NORMAL; - spin_unlock_irqrestore(&sta->local->sta_lock, flags); + mutex_lock(&sdata->local->sta_mtx); + sta = sta_info_get(sdata, addr); + ret = __sta_info_destroy(sta); + mutex_unlock(&sdata->local->sta_mtx); return ret; } -static void sta_info_debugfs_add_work(struct work_struct *work) +int sta_info_destroy_addr_bss(struct ieee80211_sub_if_data *sdata, + const u8 *addr) { - struct ieee80211_local *local = - container_of(work, struct ieee80211_local, sta_debugfs_add); - struct sta_info *sta, *tmp; - unsigned long flags; + struct sta_info *sta; + int ret; - /* We need to keep the RTNL across the whole pinned status. */ - rtnl_lock(); - while (1) { - sta = NULL; + mutex_lock(&sdata->local->sta_mtx); + sta = sta_info_get_bss(sdata, addr); + ret = __sta_info_destroy(sta); + mutex_unlock(&sdata->local->sta_mtx); - spin_lock_irqsave(&local->sta_lock, flags); - list_for_each_entry(tmp, &local->sta_list, list) { - /* - * debugfs.add_has_run will be set by - * ieee80211_sta_debugfs_add regardless - * of what else it does. - */ - if (!tmp->debugfs.add_has_run) { - sta = tmp; - __sta_info_pin(sta); - break; - } - } - spin_unlock_irqrestore(&local->sta_lock, flags); + return ret; +} - if (!sta) - break; +static void sta_info_cleanup(unsigned long data) +{ + struct ieee80211_local *local = (struct ieee80211_local *) data; + struct sta_info *sta; + + rcu_read_lock(); + list_for_each_entry_rcu(sta, &local->sta_list, list) + sta_info_cleanup_expire_buffered(local, sta); + rcu_read_unlock(); - ieee80211_sta_debugfs_add(sta); - rate_control_add_sta_debugfs(sta); + if (local->quiescing) + return; - sta = __sta_info_unpin(sta); - sta_info_destroy(sta); - } - rtnl_unlock(); + local->sta_cleanup.expires = + round_jiffies(jiffies + STA_INFO_CLEANUP_INTERVAL); + add_timer(&local->sta_cleanup); } -#endif void sta_info_init(struct ieee80211_local *local) { spin_lock_init(&local->sta_lock); + mutex_init(&local->sta_mtx); INIT_LIST_HEAD(&local->sta_list); + INIT_LIST_HEAD(&local->sta_pending_list); + INIT_WORK(&local->sta_finish_work, sta_info_finish_work); setup_timer(&local->sta_cleanup, sta_info_cleanup, (unsigned long)local); local->sta_cleanup.expires = round_jiffies(jiffies + STA_INFO_CLEANUP_INTERVAL); - -#ifdef CONFIG_MAC80211_DEBUGFS - INIT_WORK(&local->sta_debugfs_add, sta_info_debugfs_add_work); -#endif } int sta_info_start(struct ieee80211_local *local) @@ -795,16 +814,6 @@ int sta_info_start(struct ieee80211_local *local) void sta_info_stop(struct ieee80211_local *local) { del_timer(&local->sta_cleanup); -#ifdef CONFIG_MAC80211_DEBUGFS - /* - * Make sure the debugfs adding work isn't pending after this - * because we're about to be destroyed. It doesn't matter - * whether it ran or not since we're going to flush all STAs - * anyway. - */ - cancel_work_sync(&local->sta_debugfs_add); -#endif - sta_info_flush(local, NULL); } @@ -820,26 +829,19 @@ int sta_info_flush(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata) { struct sta_info *sta, *tmp; - LIST_HEAD(tmp_list); int ret = 0; - unsigned long flags; might_sleep(); - spin_lock_irqsave(&local->sta_lock, flags); + mutex_lock(&local->sta_mtx); + + sta_info_finish_pending(local); + list_for_each_entry_safe(sta, tmp, &local->sta_list, list) { - if (!sdata || sdata == sta->sdata) { - __sta_info_unlink(&sta); - if (sta) { - list_add_tail(&sta->list, &tmp_list); - ret++; - } - } + if (!sdata || sdata == sta->sdata) + WARN_ON(__sta_info_destroy(sta)); } - spin_unlock_irqrestore(&local->sta_lock, flags); - - list_for_each_entry_safe(sta, tmp, &tmp_list, list) - sta_info_destroy(sta); + mutex_unlock(&local->sta_mtx); return ret; } @@ -849,24 +851,17 @@ void ieee80211_sta_expire(struct ieee80211_sub_if_data *sdata, { struct ieee80211_local *local = sdata->local; struct sta_info *sta, *tmp; - LIST_HEAD(tmp_list); - unsigned long flags; - spin_lock_irqsave(&local->sta_lock, flags); + mutex_lock(&local->sta_mtx); list_for_each_entry_safe(sta, tmp, &local->sta_list, list) if (time_after(jiffies, sta->last_rx + exp_time)) { #ifdef CONFIG_MAC80211_IBSS_DEBUG printk(KERN_DEBUG "%s: expiring inactive STA %pM\n", sdata->name, sta->sta.addr); #endif - __sta_info_unlink(&sta); - if (sta) - list_add(&sta->list, &tmp_list); + WARN_ON(__sta_info_destroy(sta)); } - spin_unlock_irqrestore(&local->sta_lock, flags); - - list_for_each_entry_safe(sta, tmp, &tmp_list, list) - sta_info_destroy(sta); + mutex_unlock(&local->sta_mtx); } struct ieee80211_sta *ieee80211_find_sta_by_hw(struct ieee80211_hw *hw, diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 6f79bba5706..5ff611a3597 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -162,11 +162,6 @@ struct sta_ampdu_mlme { }; -/* see __sta_info_unlink */ -#define STA_INFO_PIN_STAT_NORMAL 0 -#define STA_INFO_PIN_STAT_PINNED 1 -#define STA_INFO_PIN_STAT_DESTROY 2 - /** * struct sta_info - STA information * @@ -187,7 +182,6 @@ struct sta_ampdu_mlme { * @flaglock: spinlock for flags accesses * @drv_unblock_wk: used for driver PS unblocking * @listen_interval: listen interval of this station, when we're acting as AP - * @pin_status: used internally for pinning a STA struct into memory * @flags: STA flags, see &enum ieee80211_sta_info_flags * @ps_tx_buf: buffer of frames to transmit to this station * when it leaves power saving state @@ -226,6 +220,7 @@ struct sta_ampdu_mlme { * @debugfs: debug filesystem info * @sta: station information we share with the driver * @dead: set to true when sta is unlinked + * @uploaded: set to true when sta is uploaded to the driver */ struct sta_info { /* General information, mostly static */ @@ -245,11 +240,7 @@ struct sta_info { bool dead; - /* - * for use by the internal lifetime management, - * see __sta_info_unlink - */ - u8 pin_status; + bool uploaded; /* * frequently updated, locked with own spinlock (flaglock), @@ -449,18 +440,19 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata, * Insert STA info into hash table/list, returns zero or a * -EEXIST if (if the same MAC address is already present). * - * Calling this without RCU protection makes the caller - * relinquish its reference to @sta. + * Calling the non-rcu version makes the caller relinquish, + * the _rcu version calls read_lock_rcu() and must be called + * without it held. */ int sta_info_insert(struct sta_info *sta); -/* - * Unlink a STA info from the hash table/list. - * This can NULL the STA pointer if somebody else - * has already unlinked it. - */ -void sta_info_unlink(struct sta_info **sta); +int sta_info_insert_rcu(struct sta_info *sta) __acquires(RCU); +int sta_info_insert_atomic(struct sta_info *sta); + +int sta_info_destroy_addr(struct ieee80211_sub_if_data *sdata, + const u8 *addr); +int sta_info_destroy_addr_bss(struct ieee80211_sub_if_data *sdata, + const u8 *addr); -void sta_info_destroy(struct sta_info *sta); void sta_info_set_tim_bit(struct sta_info *sta); void sta_info_clear_tim_bit(struct sta_info *sta); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index e392820a4c3..cbe53ed4fb0 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -571,7 +571,7 @@ ieee80211_tx_h_sta(struct ieee80211_tx_data *tx) { struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx->skb); - if (tx->sta) + if (tx->sta && tx->sta->uploaded) info->control.sta = &tx->sta->sta; return TX_CONTINUE; diff --git a/net/mac80211/util.c b/net/mac80211/util.c index ca170b417da..3af439a85b3 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -1082,7 +1082,6 @@ int ieee80211_reconfig(struct ieee80211_local *local) struct ieee80211_hw *hw = &local->hw; struct ieee80211_sub_if_data *sdata; struct sta_info *sta; - unsigned long flags; int res; if (local->suspended) @@ -1116,20 +1115,19 @@ int ieee80211_reconfig(struct ieee80211_local *local) } /* add STAs back */ - if (local->ops->sta_notify) { - spin_lock_irqsave(&local->sta_lock, flags); - list_for_each_entry(sta, &local->sta_list, list) { + mutex_lock(&local->sta_mtx); + list_for_each_entry(sta, &local->sta_list, list) { + if (sta->uploaded) { sdata = sta->sdata; if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN) sdata = container_of(sdata->bss, struct ieee80211_sub_if_data, u.ap); - drv_sta_notify(local, sdata, STA_NOTIFY_ADD, - &sta->sta); + WARN_ON(drv_sta_add(local, sdata, &sta->sta)); } - spin_unlock_irqrestore(&local->sta_lock, flags); } + mutex_unlock(&local->sta_mtx); /* Clear Suspend state so that ADDBA requests can be processed */ @@ -1219,10 +1217,10 @@ int ieee80211_reconfig(struct ieee80211_local *local) add_timer(&local->sta_cleanup); - spin_lock_irqsave(&local->sta_lock, flags); + mutex_lock(&local->sta_mtx); list_for_each_entry(sta, &local->sta_list, list) mesh_plink_restart(sta); - spin_unlock_irqrestore(&local->sta_lock, flags); + mutex_unlock(&local->sta_mtx); #else WARN_ON(1); #endif -- cgit v1.2.3-70-g09d2 From 349e6b7289f8a3d3d5d3b859e00b41f27d1211df Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Sun, 7 Feb 2010 10:22:01 +0200 Subject: mac80211: remove get_tx_stats() driver op get_tx_stats() driver operation is not currently used anywhere in mac80211 and there are no plans to use it in the not-so-near future. So it can go without anyone missing it. Signed-off-by: Kalle Valo Acked-by: Johannes Berg Signed-off-by: John W. Linville --- Documentation/DocBook/mac80211.tmpl | 1 - include/net/mac80211.h | 22 ---------------------- net/mac80211/driver-ops.h | 8 -------- net/mac80211/driver-trace.h | 23 ----------------------- 4 files changed, 54 deletions(-) (limited to 'include') diff --git a/Documentation/DocBook/mac80211.tmpl b/Documentation/DocBook/mac80211.tmpl index 971d1c0c83e..affb15a344a 100644 --- a/Documentation/DocBook/mac80211.tmpl +++ b/Documentation/DocBook/mac80211.tmpl @@ -234,7 +234,6 @@ usage should require reading the full document. Multiple queues and QoS support TBD !Finclude/net/mac80211.h ieee80211_tx_queue_params -!Finclude/net/mac80211.h ieee80211_tx_queue_stats diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a19fac35259..414d774028b 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -117,19 +117,6 @@ struct ieee80211_tx_queue_params { bool uapsd; }; -/** - * struct ieee80211_tx_queue_stats - transmit queue statistics - * - * @len: number of packets in queue - * @limit: queue length limit - * @count: number of frames sent - */ -struct ieee80211_tx_queue_stats { - unsigned int len; - unsigned int limit; - unsigned int count; -}; - struct ieee80211_low_level_stats { unsigned int dot11ACKFailureCount; unsigned int dot11RTSFailureCount; @@ -1548,13 +1535,6 @@ enum ieee80211_ampdu_mlme_action { * Returns a negative error code on failure. * The callback can sleep. * - * @get_tx_stats: Get statistics of the current TX queue status. This is used - * to get number of currently queued packets (queue length), maximum queue - * size (limit), and total number of packets sent using each TX queue - * (count). The 'stats' pointer points to an array that has hw->queues - * items. - * The callback must be atomic. - * * @get_tsf: Get the current TSF timer value from firmware/hardware. Currently, * this is only used for IBSS mode BSSID merging and debugging. Is not a * required function. @@ -1648,8 +1628,6 @@ struct ieee80211_ops { enum sta_notify_cmd, struct ieee80211_sta *sta); int (*conf_tx)(struct ieee80211_hw *hw, u16 queue, const struct ieee80211_tx_queue_params *params); - int (*get_tx_stats)(struct ieee80211_hw *hw, - struct ieee80211_tx_queue_stats *stats); u64 (*get_tsf)(struct ieee80211_hw *hw); void (*set_tsf)(struct ieee80211_hw *hw, u64 tsf); void (*reset_tsf)(struct ieee80211_hw *hw); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 855e85b5506..c3d844093a2 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -290,14 +290,6 @@ static inline int drv_conf_tx(struct ieee80211_local *local, u16 queue, return ret; } -static inline int drv_get_tx_stats(struct ieee80211_local *local, - struct ieee80211_tx_queue_stats *stats) -{ - int ret = local->ops->get_tx_stats(&local->hw, stats); - trace_drv_get_tx_stats(local, stats, ret); - return ret; -} - static inline u64 drv_get_tsf(struct ieee80211_local *local) { u64 ret = -1ULL; diff --git a/net/mac80211/driver-trace.h b/net/mac80211/driver-trace.h index c984910bf27..41baf730a5c 100644 --- a/net/mac80211/driver-trace.h +++ b/net/mac80211/driver-trace.h @@ -630,29 +630,6 @@ TRACE_EVENT(drv_conf_tx, ) ); -TRACE_EVENT(drv_get_tx_stats, - TP_PROTO(struct ieee80211_local *local, - struct ieee80211_tx_queue_stats *stats, - int ret), - - TP_ARGS(local, stats, ret), - - TP_STRUCT__entry( - LOCAL_ENTRY - __field(int, ret) - ), - - TP_fast_assign( - LOCAL_ASSIGN; - __entry->ret = ret; - ), - - TP_printk( - LOCAL_PR_FMT " ret:%d", - LOCAL_PR_ARG, __entry->ret - ) -); - TRACE_EVENT(drv_get_tsf, TP_PROTO(struct ieee80211_local *local, u64 ret), -- cgit v1.2.3-70-g09d2 From 375177bf35efc08e1bd37bbda4cc0c8cc4db8500 Mon Sep 17 00:00:00 2001 From: Vivek Natarajan Date: Tue, 9 Feb 2010 14:50:28 +0530 Subject: mac80211: Retry null data frame for power save. Even if the null data frame is not acked by the AP, mac80211 goes into power save. This might lead to loss of frames from the AP. Prevent this by restarting dynamic_ps_timer when ack is not received for null data frames. Cc: Johannes Berg Signed-off-by: Vivek Natarajan Signed-off-by: John W. Linville --- include/net/mac80211.h | 6 ++++++ net/mac80211/ieee80211_i.h | 1 + net/mac80211/mlme.c | 20 +++++++++++++++----- net/mac80211/status.c | 17 +++++++++++++++-- 4 files changed, 37 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 414d774028b..314e9817316 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -945,6 +945,11 @@ enum ieee80211_tkip_key_type { * Hardware supports Unscheduled Automatic Power Save Delivery * (U-APSD) in managed mode. The mode is configured with * conf_tx() operation. + * + * @IEEE80211_HW_REPORTS_TX_ACK_STATUS: + * Hardware can provide ack status reports of Tx frames to + * the stack. + * */ enum ieee80211_hw_flags { IEEE80211_HW_HAS_RATE_CONTROL = 1<<0, @@ -965,6 +970,7 @@ enum ieee80211_hw_flags { IEEE80211_HW_SUPPORTS_STATIC_SMPS = 1<<15, IEEE80211_HW_SUPPORTS_DYNAMIC_SMPS = 1<<16, IEEE80211_HW_SUPPORTS_UAPSD = 1<<17, + IEEE80211_HW_REPORTS_TX_ACK_STATUS = 1<<18, }; /** diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index a5911191f22..9dd98b674cb 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -316,6 +316,7 @@ enum ieee80211_sta_flags { IEEE80211_STA_CSA_RECEIVED = BIT(5), IEEE80211_STA_MFP_ENABLED = BIT(6), IEEE80211_STA_UAPSD_ENABLED = BIT(7), + IEEE80211_STA_NULLFUNC_ACKED = BIT(8), }; struct ieee80211_if_managed { diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 7a792147658..ee9443dc20f 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -434,8 +434,11 @@ static void ieee80211_enable_ps(struct ieee80211_local *local, } else { if (local->hw.flags & IEEE80211_HW_PS_NULLFUNC_STACK) ieee80211_send_nullfunc(local, sdata, 1); - conf->flags |= IEEE80211_CONF_PS; - ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_PS); + + if (!(local->hw.flags & IEEE80211_HW_REPORTS_TX_ACK_STATUS)) { + conf->flags |= IEEE80211_CONF_PS; + ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_PS); + } } } @@ -541,6 +544,7 @@ void ieee80211_dynamic_ps_enable_work(struct work_struct *work) container_of(work, struct ieee80211_local, dynamic_ps_enable_work); struct ieee80211_sub_if_data *sdata = local->ps_sdata; + struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; /* can only happen when PS was just disabled anyway */ if (!sdata) @@ -549,11 +553,16 @@ void ieee80211_dynamic_ps_enable_work(struct work_struct *work) if (local->hw.conf.flags & IEEE80211_CONF_PS) return; - if (local->hw.flags & IEEE80211_HW_PS_NULLFUNC_STACK) + if ((local->hw.flags & IEEE80211_HW_PS_NULLFUNC_STACK) && + (!(ifmgd->flags & IEEE80211_STA_NULLFUNC_ACKED))) ieee80211_send_nullfunc(local, sdata, 1); - local->hw.conf.flags |= IEEE80211_CONF_PS; - ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_PS); + if (!(local->hw.flags & IEEE80211_HW_REPORTS_TX_ACK_STATUS) || + (ifmgd->flags & IEEE80211_STA_NULLFUNC_ACKED)) { + ifmgd->flags &= ~IEEE80211_STA_NULLFUNC_ACKED; + local->hw.conf.flags |= IEEE80211_CONF_PS; + ieee80211_hw_config(local, IEEE80211_CONF_CHANGE_PS); + } } void ieee80211_dynamic_ps_timer(unsigned long data) @@ -1892,6 +1901,7 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, return -ENOMEM; ifmgd->flags &= ~IEEE80211_STA_DISABLE_11N; + ifmgd->flags &= ~IEEE80211_STA_NULLFUNC_ACKED; for (i = 0; i < req->crypto.n_ciphers_pairwise; i++) if (req->crypto.ciphers_pairwise[i] == WLAN_CIPHER_SUITE_WEP40 || diff --git a/net/mac80211/status.c b/net/mac80211/status.c index e57ad6b1d7e..ded98730c11 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -188,6 +188,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) rcu_read_lock(); sband = local->hw.wiphy->bands[info->band]; + fc = hdr->frame_control; for_each_sta_info(local, hdr->addr1, sta, tmp) { /* skip wrong virtual interface */ @@ -205,8 +206,6 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) return; } - fc = hdr->frame_control; - if ((info->flags & IEEE80211_TX_STAT_AMPDU_NO_BACK) && (ieee80211_is_data_qos(fc))) { u16 tid, ssn; @@ -275,6 +274,20 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) local->dot11FailedCount++; } + if (ieee80211_is_nullfunc(fc) && ieee80211_has_pm(fc) && + (local->hw.flags & IEEE80211_HW_REPORTS_TX_ACK_STATUS) && + !(info->flags & IEEE80211_TX_CTL_INJECTED) && + local->ps_sdata && !(local->scanning)) { + if (info->flags & IEEE80211_TX_STAT_ACK) { + local->ps_sdata->u.mgd.flags |= + IEEE80211_STA_NULLFUNC_ACKED; + ieee80211_queue_work(&local->hw, + &local->dynamic_ps_enable_work); + } else + mod_timer(&local->dynamic_ps_timer, jiffies + + msecs_to_jiffies(10)); + } + /* this was a transmitted frame, but now we want to reuse it */ skb_orphan(skb); -- cgit v1.2.3-70-g09d2 From 0a02604628c49037e1de2091d75462fd856b26ed Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 10 Feb 2010 15:00:32 +0100 Subject: netfilter: xtables: consistent struct compat_xt_counters definition There is compat_u64 type which deals with different u64 type alignment on different compat-capable platforms, so use it and removed some hardcoded assumptions. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 365fabe1b16..8b6c0e7d265 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -562,11 +562,7 @@ struct compat_xt_entry_target { * current task alignment */ struct compat_xt_counters { -#if defined(CONFIG_X86_64) || defined(CONFIG_IA64) - u_int32_t cnt[4]; -#else - u_int64_t cnt[2]; -#endif + compat_u64 pcnt, bcnt; /* Packet and byte counters */ }; struct compat_xt_counters_info { -- cgit v1.2.3-70-g09d2 From 42107f5009da223daa800d6da6904d77297ae829 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 10 Feb 2010 15:03:27 +0100 Subject: netfilter: xtables: symmetric COMPAT_XT_ALIGN definition Rewrite COMPAT_XT_ALIGN in terms of dummy structure hack. Compat counters logically have nothing to do with it. Use ALIGN() macro while I'm at it for same types. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 13 +++++++++---- net/netfilter/x_tables.c | 4 ++-- 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 8b6c0e7d265..9d671ebf060 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -93,8 +93,7 @@ struct _xt_align { __u64 u64; }; -#define XT_ALIGN(s) (((s) + (__alignof__(struct _xt_align)-1)) \ - & ~(__alignof__(struct _xt_align)-1)) +#define XT_ALIGN(s) ALIGN((s), __alignof__(struct _xt_align)) /* Standard return verdict, or do jump. */ #define XT_STANDARD_TARGET "" @@ -571,8 +570,14 @@ struct compat_xt_counters_info { struct compat_xt_counters counters[0]; }; -#define COMPAT_XT_ALIGN(s) (((s) + (__alignof__(struct compat_xt_counters)-1)) \ - & ~(__alignof__(struct compat_xt_counters)-1)) +struct _compat_xt_align { + __u8 u8; + __u16 u16; + __u32 u32; + compat_u64 u64; +}; + +#define COMPAT_XT_ALIGN(s) ALIGN((s), __alignof__(struct _compat_xt_align)) extern void xt_compat_lock(u_int8_t af); extern void xt_compat_unlock(u_int8_t af); diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index f01955cce31..5c564ff10a3 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -364,7 +364,7 @@ int xt_check_match(struct xt_mtchk_param *par, * ebt_among is exempt from centralized matchsize checking * because it uses a dynamic-size data set. */ - pr_err("%s_tables: %s match: invalid size %Zu != %u\n", + pr_err("%s_tables: %s match: invalid size %u != %u\n", xt_prefix[par->family], par->match->name, XT_ALIGN(par->match->matchsize), size); return -EINVAL; @@ -514,7 +514,7 @@ int xt_check_target(struct xt_tgchk_param *par, unsigned int size, u_int8_t proto, bool inv_proto) { if (XT_ALIGN(par->target->targetsize) != size) { - pr_err("%s_tables: %s target: invalid size %Zu != %u\n", + pr_err("%s_tables: %s target: invalid size %u != %u\n", xt_prefix[par->family], par->target->name, XT_ALIGN(par->target->targetsize), size); return -EINVAL; -- cgit v1.2.3-70-g09d2 From 2b95efe7f6bb750256a702cc32d33b0cb2cd8223 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 17 Jun 2009 13:57:48 +0200 Subject: netfilter: xtables: use xt_table for hook instantiation The respective xt_table structures already have most of the metadata needed for hook setup. Add a 'priority' field to struct xt_table so that xt_hook_link() can be called with a reduced number of arguments. So should we be having more tables in the future, it comes at no static cost (only runtime, as before) - space saved: 6807373->6806555. Signed-off-by: Jan Engelhardt --- include/linux/netfilter/x_tables.h | 4 +++ net/ipv4/netfilter/arptable_filter.c | 33 +++++---------------- net/ipv4/netfilter/iptable_filter.c | 33 +++++---------------- net/ipv4/netfilter/iptable_mangle.c | 47 +++++------------------------ net/ipv4/netfilter/iptable_raw.c | 27 +++++------------ net/ipv4/netfilter/iptable_security.c | 33 +++++---------------- net/ipv6/netfilter/ip6table_filter.c | 33 +++++---------------- net/ipv6/netfilter/ip6table_mangle.c | 48 +++++------------------------- net/ipv6/netfilter/ip6table_raw.c | 26 +++++----------- net/ipv6/netfilter/ip6table_security.c | 33 +++++---------------- net/netfilter/x_tables.c | 54 ++++++++++++++++++++++++++++++++++ 11 files changed, 121 insertions(+), 250 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 365fabe1b16..fdd3342c123 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -361,6 +361,7 @@ struct xt_table { struct module *me; u_int8_t af; /* address/protocol family */ + int priority; /* hook order */ /* A unique name... */ const char name[XT_TABLE_MAXNAMELEN]; @@ -522,6 +523,9 @@ static inline unsigned long ifname_compare_aligned(const char *_a, return ret; } +extern struct nf_hook_ops *xt_hook_link(const struct xt_table *, nf_hookfn *); +extern void xt_hook_unlink(const struct xt_table *, struct nf_hook_ops *); + #ifdef CONFIG_COMPAT #include diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index deeda9b2cf0..b361de0dac4 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -50,6 +50,7 @@ static const struct xt_table packet_filter = { .valid_hooks = FILTER_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_ARP, + .priority = NF_IP_PRI_FILTER, }; /* The work comes in here from netfilter.c */ @@ -63,29 +64,7 @@ arptable_filter_hook(unsigned int hook, struct sk_buff *skb, return arpt_do_table(skb, hook, in, out, net->ipv4.arptable_filter); } -static struct nf_hook_ops arpt_ops[] __read_mostly = { - { - .hook = arptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_ARP, - .hooknum = NF_ARP_IN, - .priority = NF_IP_PRI_FILTER, - }, - { - .hook = arptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_ARP, - .hooknum = NF_ARP_OUT, - .priority = NF_IP_PRI_FILTER, - }, - { - .hook = arptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_ARP, - .hooknum = NF_ARP_FORWARD, - .priority = NF_IP_PRI_FILTER, - }, -}; +static struct nf_hook_ops *arpfilter_ops __read_mostly; static int __net_init arptable_filter_net_init(struct net *net) { @@ -115,9 +94,11 @@ static int __init arptable_filter_init(void) if (ret < 0) return ret; - ret = nf_register_hooks(arpt_ops, ARRAY_SIZE(arpt_ops)); - if (ret < 0) + arpfilter_ops = xt_hook_link(&packet_filter, arptable_filter_hook); + if (IS_ERR(arpfilter_ops)) { + ret = PTR_ERR(arpfilter_ops); goto cleanup_table; + } return ret; cleanup_table: @@ -127,7 +108,7 @@ cleanup_table: static void __exit arptable_filter_fini(void) { - nf_unregister_hooks(arpt_ops, ARRAY_SIZE(arpt_ops)); + xt_hook_unlink(&packet_filter, arpfilter_ops); unregister_pernet_subsys(&arptable_filter_net_ops); } diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 1bfeaae6f62..c14bb85db1d 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -58,6 +58,7 @@ static const struct xt_table packet_filter = { .valid_hooks = FILTER_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, + .priority = NF_IP_PRI_FILTER, }; static unsigned int @@ -77,29 +78,7 @@ iptable_filter_hook(unsigned int hook, struct sk_buff *skb, return ipt_do_table(skb, hook, in, out, net->ipv4.iptable_filter); } -static struct nf_hook_ops ipt_ops[] __read_mostly = { - { - .hook = iptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP_PRI_FILTER, - }, - { - .hook = iptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP_PRI_FILTER, - }, - { - .hook = iptable_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP_PRI_FILTER, - }, -}; +static struct nf_hook_ops *filter_ops __read_mostly; /* Default to forward because I got too much mail already. */ static int forward = NF_ACCEPT; @@ -142,9 +121,11 @@ static int __init iptable_filter_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); - if (ret < 0) + filter_ops = xt_hook_link(&packet_filter, iptable_filter_hook); + if (IS_ERR(filter_ops)) { + ret = PTR_ERR(filter_ops); goto cleanup_table; + } return ret; @@ -155,7 +136,7 @@ static int __init iptable_filter_init(void) static void __exit iptable_filter_fini(void) { - nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); + xt_hook_unlink(&packet_filter, filter_ops); unregister_pernet_subsys(&iptable_filter_net_ops); } diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 4e699cd275c..2355a229f8e 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -69,6 +69,7 @@ static const struct xt_table packet_mangler = { .valid_hooks = MANGLE_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, + .priority = NF_IP_PRI_MANGLE, }; static unsigned int @@ -129,43 +130,7 @@ iptable_mangle_hook(unsigned int hook, dev_net(in)->ipv4.iptable_mangle); } -static struct nf_hook_ops ipt_ops[] __read_mostly = { - { - .hook = iptable_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_PRE_ROUTING, - .priority = NF_IP_PRI_MANGLE, - }, - { - .hook = iptable_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP_PRI_MANGLE, - }, - { - .hook = iptable_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP_PRI_MANGLE, - }, - { - .hook = iptable_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP_PRI_MANGLE, - }, - { - .hook = iptable_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_POST_ROUTING, - .priority = NF_IP_PRI_MANGLE, - }, -}; +static struct nf_hook_ops *mangle_ops __read_mostly; static int __net_init iptable_mangle_net_init(struct net *net) { @@ -196,9 +161,11 @@ static int __init iptable_mangle_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); - if (ret < 0) + mangle_ops = xt_hook_link(&packet_mangler, iptable_mangle_hook); + if (IS_ERR(mangle_ops)) { + ret = PTR_ERR(mangle_ops); goto cleanup_table; + } return ret; @@ -209,7 +176,7 @@ static int __init iptable_mangle_init(void) static void __exit iptable_mangle_fini(void) { - nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); + xt_hook_unlink(&packet_mangler, mangle_ops); unregister_pernet_subsys(&iptable_mangle_net_ops); } diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index d16e43777c3..62a99154f14 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -41,6 +41,7 @@ static const struct xt_table packet_raw = { .valid_hooks = RAW_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, + .priority = NF_IP_PRI_RAW, }; /* The work comes in here from netfilter.c. */ @@ -61,23 +62,7 @@ iptable_raw_hook(unsigned int hook, struct sk_buff *skb, return ipt_do_table(skb, hook, in, out, net->ipv4.iptable_raw); } -/* 'raw' is the very first table. */ -static struct nf_hook_ops ipt_ops[] __read_mostly = { - { - .hook = iptable_raw_hook, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_PRE_ROUTING, - .priority = NF_IP_PRI_RAW, - .owner = THIS_MODULE, - }, - { - .hook = iptable_raw_hook, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP_PRI_RAW, - .owner = THIS_MODULE, - }, -}; +static struct nf_hook_ops *rawtable_ops __read_mostly; static int __net_init iptable_raw_net_init(struct net *net) { @@ -108,9 +93,11 @@ static int __init iptable_raw_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); - if (ret < 0) + rawtable_ops = xt_hook_link(&packet_raw, iptable_raw_hook); + if (IS_ERR(rawtable_ops)) { + ret = PTR_ERR(rawtable_ops); goto cleanup_table; + } return ret; @@ -121,7 +108,7 @@ static int __init iptable_raw_init(void) static void __exit iptable_raw_fini(void) { - nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); + xt_hook_unlink(&packet_raw, rawtable_ops); unregister_pernet_subsys(&iptable_raw_net_ops); } diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index 324505aaaa7..b1bf3ca2c6c 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -62,6 +62,7 @@ static const struct xt_table security_table = { .valid_hooks = SECURITY_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, + .priority = NF_IP_PRI_SECURITY, }; static unsigned int @@ -82,29 +83,7 @@ iptable_security_hook(unsigned int hook, struct sk_buff *skb, return ipt_do_table(skb, hook, in, out, net->ipv4.iptable_security); } -static struct nf_hook_ops ipt_ops[] __read_mostly = { - { - .hook = iptable_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP_PRI_SECURITY, - }, - { - .hook = iptable_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP_PRI_SECURITY, - }, - { - .hook = iptable_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP_PRI_SECURITY, - }, -}; +static struct nf_hook_ops *sectbl_ops __read_mostly; static int __net_init iptable_security_net_init(struct net *net) { @@ -135,9 +114,11 @@ static int __init iptable_security_init(void) if (ret < 0) return ret; - ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); - if (ret < 0) + sectbl_ops = xt_hook_link(&security_table, iptable_security_hook); + if (IS_ERR(sectbl_ops)) { + ret = PTR_ERR(sectbl_ops); goto cleanup_table; + } return ret; @@ -148,7 +129,7 @@ cleanup_table: static void __exit iptable_security_fini(void) { - nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops)); + xt_hook_unlink(&security_table, sectbl_ops); unregister_pernet_subsys(&iptable_security_net_ops); } diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index 866f34ae236..6e95d0614ca 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -56,6 +56,7 @@ static const struct xt_table packet_filter = { .valid_hooks = FILTER_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV6, + .priority = NF_IP6_PRI_FILTER, }; /* The work comes in here from netfilter.c. */ @@ -69,29 +70,7 @@ ip6table_filter_hook(unsigned int hook, struct sk_buff *skb, return ip6t_do_table(skb, hook, in, out, net->ipv6.ip6table_filter); } -static struct nf_hook_ops ip6t_ops[] __read_mostly = { - { - .hook = ip6table_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP6_PRI_FILTER, - }, - { - .hook = ip6table_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP6_PRI_FILTER, - }, - { - .hook = ip6table_filter_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP6_PRI_FILTER, - }, -}; +static struct nf_hook_ops *filter_ops __read_mostly; /* Default to forward because I got too much mail already. */ static int forward = NF_ACCEPT; @@ -134,9 +113,11 @@ static int __init ip6table_filter_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); - if (ret < 0) + filter_ops = xt_hook_link(&packet_filter, ip6table_filter_hook); + if (IS_ERR(filter_ops)) { + ret = PTR_ERR(filter_ops); goto cleanup_table; + } return ret; @@ -147,7 +128,7 @@ static int __init ip6table_filter_init(void) static void __exit ip6table_filter_fini(void) { - nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); + xt_hook_unlink(&packet_filter, filter_ops); unregister_pernet_subsys(&ip6table_filter_net_ops); } diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 405ac1f7639..5023ac52ffe 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -62,6 +62,7 @@ static const struct xt_table packet_mangler = { .valid_hooks = MANGLE_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV6, + .priority = NF_IP6_PRI_MANGLE, }; static unsigned int @@ -122,44 +123,7 @@ ip6table_mangle_hook(unsigned int hook, struct sk_buff *skb, dev_net(in)->ipv6.ip6table_mangle); } -static struct nf_hook_ops ip6t_ops[] __read_mostly = { - { - .hook = ip6table_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_PRE_ROUTING, - .priority = NF_IP6_PRI_MANGLE, - }, - { - .hook = ip6table_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP6_PRI_MANGLE, - }, - { - .hook = ip6table_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP6_PRI_MANGLE, - }, - { - .hook = ip6table_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP6_PRI_MANGLE, - }, - { - .hook = ip6table_mangle_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_POST_ROUTING, - .priority = NF_IP6_PRI_MANGLE, - }, -}; - +static struct nf_hook_ops *mangle_ops __read_mostly; static int __net_init ip6table_mangle_net_init(struct net *net) { /* Register table */ @@ -189,9 +153,11 @@ static int __init ip6table_mangle_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); - if (ret < 0) + mangle_ops = xt_hook_link(&packet_mangler, ip6table_mangle_hook); + if (IS_ERR(mangle_ops)) { + ret = PTR_ERR(mangle_ops); goto cleanup_table; + } return ret; @@ -202,7 +168,7 @@ static int __init ip6table_mangle_init(void) static void __exit ip6table_mangle_fini(void) { - nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); + xt_hook_unlink(&packet_mangler, mangle_ops); unregister_pernet_subsys(&ip6table_mangle_net_ops); } diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 5451a36fbc2..3bfa6951164 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -40,6 +40,7 @@ static const struct xt_table packet_raw = { .valid_hooks = RAW_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV6, + .priority = NF_IP6_PRI_FIRST, }; /* The work comes in here from netfilter.c. */ @@ -53,22 +54,7 @@ ip6table_raw_hook(unsigned int hook, struct sk_buff *skb, return ip6t_do_table(skb, hook, in, out, net->ipv6.ip6table_raw); } -static struct nf_hook_ops ip6t_ops[] __read_mostly = { - { - .hook = ip6table_raw_hook, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_PRE_ROUTING, - .priority = NF_IP6_PRI_FIRST, - .owner = THIS_MODULE, - }, - { - .hook = ip6table_raw_hook, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP6_PRI_FIRST, - .owner = THIS_MODULE, - }, -}; +static struct nf_hook_ops *rawtable_ops __read_mostly; static int __net_init ip6table_raw_net_init(struct net *net) { @@ -99,9 +85,11 @@ static int __init ip6table_raw_init(void) return ret; /* Register hooks */ - ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); - if (ret < 0) + rawtable_ops = xt_hook_link(&packet_raw, ip6table_raw_hook); + if (IS_ERR(rawtable_ops)) { + ret = PTR_ERR(rawtable_ops); goto cleanup_table; + } return ret; @@ -112,7 +100,7 @@ static int __init ip6table_raw_init(void) static void __exit ip6table_raw_fini(void) { - nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); + xt_hook_unlink(&packet_raw, rawtable_ops); unregister_pernet_subsys(&ip6table_raw_net_ops); } diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index 841ea77f521..dd2200f17a6 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -61,6 +61,7 @@ static const struct xt_table security_table = { .valid_hooks = SECURITY_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV6, + .priority = NF_IP6_PRI_SECURITY, }; static unsigned int @@ -74,29 +75,7 @@ ip6table_security_hook(unsigned int hook, struct sk_buff *skb, return ip6t_do_table(skb, hook, in, out, net->ipv6.ip6table_security); } -static struct nf_hook_ops ip6t_ops[] __read_mostly = { - { - .hook = ip6table_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP6_PRI_SECURITY, - }, - { - .hook = ip6table_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_FORWARD, - .priority = NF_IP6_PRI_SECURITY, - }, - { - .hook = ip6table_security_hook, - .owner = THIS_MODULE, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_OUT, - .priority = NF_IP6_PRI_SECURITY, - }, -}; +static struct nf_hook_ops *sectbl_ops __read_mostly; static int __net_init ip6table_security_net_init(struct net *net) { @@ -127,9 +106,11 @@ static int __init ip6table_security_init(void) if (ret < 0) return ret; - ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); - if (ret < 0) + sectbl_ops = xt_hook_link(&security_table, ip6table_security_hook); + if (IS_ERR(sectbl_ops)) { + ret = PTR_ERR(sectbl_ops); goto cleanup_table; + } return ret; @@ -140,7 +121,7 @@ cleanup_table: static void __exit ip6table_security_fini(void) { - nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops)); + xt_hook_unlink(&security_table, sectbl_ops); unregister_pernet_subsys(&ip6table_security_net_ops); } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index f01955cce31..b51cb0d7234 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1091,6 +1091,60 @@ static const struct file_operations xt_target_ops = { #endif /* CONFIG_PROC_FS */ +/** + * xt_hook_link - set up hooks for a new table + * @table: table with metadata needed to set up hooks + * @fn: Hook function + * + * This function will take care of creating and registering the necessary + * Netfilter hooks for XT tables. + */ +struct nf_hook_ops *xt_hook_link(const struct xt_table *table, nf_hookfn *fn) +{ + unsigned int hook_mask = table->valid_hooks; + uint8_t i, num_hooks = hweight32(hook_mask); + uint8_t hooknum; + struct nf_hook_ops *ops; + int ret; + + ops = kmalloc(sizeof(*ops) * num_hooks, GFP_KERNEL); + if (ops == NULL) + return ERR_PTR(-ENOMEM); + + for (i = 0, hooknum = 0; i < num_hooks && hook_mask != 0; + hook_mask >>= 1, ++hooknum) { + if (!(hook_mask & 1)) + continue; + ops[i].hook = fn; + ops[i].owner = table->me; + ops[i].pf = table->af; + ops[i].hooknum = hooknum; + ops[i].priority = table->priority; + ++i; + } + + ret = nf_register_hooks(ops, num_hooks); + if (ret < 0) { + kfree(ops); + return ERR_PTR(ret); + } + + return ops; +} +EXPORT_SYMBOL_GPL(xt_hook_link); + +/** + * xt_hook_unlink - remove hooks for a table + * @ops: nf_hook_ops array as returned by nf_hook_link + * @hook_mask: the very same mask that was passed to nf_hook_link + */ +void xt_hook_unlink(const struct xt_table *table, struct nf_hook_ops *ops) +{ + nf_unregister_hooks(ops, hweight32(table->valid_hooks)); + kfree(ops); +} +EXPORT_SYMBOL_GPL(xt_hook_unlink); + int xt_proto_init(struct net *net, u_int8_t af) { #ifdef CONFIG_PROC_FS -- cgit v1.2.3-70-g09d2 From e3eaa9910b380530cfd2c0670fcd3f627674da8a Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 17 Jun 2009 22:14:54 +0200 Subject: netfilter: xtables: generate initial table on-demand The static initial tables are pretty large, and after the net namespace has been instantiated, they just hang around for nothing. This commit removes them and creates tables on-demand at runtime when needed. Size shrinks by 7735 bytes (x86_64). Signed-off-by: Jan Engelhardt --- include/linux/netfilter_arp/arp_tables.h | 1 + include/linux/netfilter_ipv4/ip_tables.h | 1 + include/linux/netfilter_ipv6/ip6_tables.h | 1 + net/ipv4/netfilter/arp_tables.c | 7 +++++ net/ipv4/netfilter/arptable_filter.c | 40 ++++++--------------------- net/ipv4/netfilter/ip_tables.c | 7 +++++ net/ipv4/netfilter/iptable_filter.c | 46 ++++++++----------------------- net/ipv4/netfilter/iptable_mangle.c | 46 +++++-------------------------- net/ipv4/netfilter/iptable_raw.c | 36 +++++------------------- net/ipv4/netfilter/iptable_security.c | 39 +++++--------------------- net/ipv4/netfilter/nf_nat_rule.c | 39 +++++--------------------- net/ipv6/netfilter/ip6_tables.c | 7 +++++ net/ipv6/netfilter/ip6table_filter.c | 46 ++++++++----------------------- net/ipv6/netfilter/ip6table_mangle.c | 45 +++++------------------------- net/ipv6/netfilter/ip6table_raw.c | 36 +++++------------------- net/ipv6/netfilter/ip6table_security.c | 39 +++++--------------------- net/netfilter/x_tables.c | 4 ++- net/netfilter/xt_repldata.h | 35 +++++++++++++++++++++++ 18 files changed, 141 insertions(+), 334 deletions(-) create mode 100644 net/netfilter/xt_repldata.h (limited to 'include') diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index f2336523a9d..0b33980611b 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -258,6 +258,7 @@ struct arpt_error { .target.errorname = "ERROR", \ } +extern void *arpt_alloc_initial_table(const struct xt_table *); extern struct xt_table *arpt_register_table(struct net *net, const struct xt_table *table, const struct arpt_replace *repl); diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 8d1f273d350..364973b4213 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -282,6 +282,7 @@ struct ipt_error { .target.errorname = "ERROR", \ } +extern void *ipt_alloc_initial_table(const struct xt_table *); extern unsigned int ipt_do_table(struct sk_buff *skb, unsigned int hook, const struct net_device *in, diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index d2952d2fa65..8031eb486a1 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -297,6 +297,7 @@ ip6t_get_target(struct ip6t_entry *e) #include extern void ip6t_init(void) __init; +extern void *ip6t_alloc_initial_table(const struct xt_table *); extern struct xt_table *ip6t_register_table(struct net *net, const struct xt_table *table, const struct ip6t_replace *repl); diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 90203e1b918..72723ea1054 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -27,6 +27,7 @@ #include #include +#include "../../netfilter/xt_repldata.h" MODULE_LICENSE("GPL"); MODULE_AUTHOR("David S. Miller "); @@ -58,6 +59,12 @@ do { \ #define ARP_NF_ASSERT(x) #endif +void *arpt_alloc_initial_table(const struct xt_table *info) +{ + return xt_alloc_initial_table(arpt, ARPT); +} +EXPORT_SYMBOL_GPL(arpt_alloc_initial_table); + static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap, const char *hdr_addr, int len) { diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index b361de0dac4..bfe26f32b93 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -6,6 +6,7 @@ */ #include +#include #include MODULE_LICENSE("GPL"); @@ -15,36 +16,6 @@ MODULE_DESCRIPTION("arptables filter table"); #define FILTER_VALID_HOOKS ((1 << NF_ARP_IN) | (1 << NF_ARP_OUT) | \ (1 << NF_ARP_FORWARD)) -static const struct -{ - struct arpt_replace repl; - struct arpt_standard entries[3]; - struct arpt_error term; -} initial_table __net_initdata = { - .repl = { - .name = "filter", - .valid_hooks = FILTER_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct arpt_standard) * 3 + sizeof(struct arpt_error), - .hook_entry = { - [NF_ARP_IN] = 0, - [NF_ARP_OUT] = sizeof(struct arpt_standard), - [NF_ARP_FORWARD] = 2 * sizeof(struct arpt_standard), - }, - .underflow = { - [NF_ARP_IN] = 0, - [NF_ARP_OUT] = sizeof(struct arpt_standard), - [NF_ARP_FORWARD] = 2 * sizeof(struct arpt_standard), - }, - }, - .entries = { - ARPT_STANDARD_INIT(NF_ACCEPT), /* ARP_IN */ - ARPT_STANDARD_INIT(NF_ACCEPT), /* ARP_OUT */ - ARPT_STANDARD_INIT(NF_ACCEPT), /* ARP_FORWARD */ - }, - .term = ARPT_ERROR_INIT, -}; - static const struct xt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, @@ -68,9 +39,14 @@ static struct nf_hook_ops *arpfilter_ops __read_mostly; static int __net_init arptable_filter_net_init(struct net *net) { - /* Register table */ + struct arpt_replace *repl; + + repl = arpt_alloc_initial_table(&packet_filter); + if (repl == NULL) + return -ENOMEM; net->ipv4.arptable_filter = - arpt_register_table(net, &packet_filter, &initial_table.repl); + arpt_register_table(net, &packet_filter, repl); + kfree(repl); if (IS_ERR(net->ipv4.arptable_filter)) return PTR_ERR(net->ipv4.arptable_filter); return 0; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 5bf7de1527a..2057b1bb617 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -28,6 +28,7 @@ #include #include #include +#include "../../netfilter/xt_repldata.h" MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); @@ -66,6 +67,12 @@ do { \ #define inline #endif +void *ipt_alloc_initial_table(const struct xt_table *info) +{ + return xt_alloc_initial_table(ipt, IPT); +} +EXPORT_SYMBOL_GPL(ipt_alloc_initial_table); + /* We keep a set of rules for each CPU, so we can avoid write-locking them in the softirq when updating the counters and therefore diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index c14bb85db1d..c8dc9800d62 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -23,36 +23,6 @@ MODULE_DESCRIPTION("iptables filter table"); (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT)) -static struct -{ - struct ipt_replace repl; - struct ipt_standard entries[3]; - struct ipt_error term; -} initial_table __net_initdata = { - .repl = { - .name = "filter", - .valid_hooks = FILTER_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error), - .hook_entry = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2, - }, - .underflow = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2, - }, - }, - .entries = { - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IPT_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IPT_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, @@ -86,9 +56,18 @@ module_param(forward, bool, 0000); static int __net_init iptable_filter_net_init(struct net *net) { - /* Register table */ + struct ipt_replace *repl; + + repl = ipt_alloc_initial_table(&packet_filter); + if (repl == NULL) + return -ENOMEM; + /* Entry 1 is the FORWARD hook */ + ((struct ipt_standard *)repl->entries)[1].target.verdict = + -forward - 1; + net->ipv4.iptable_filter = - ipt_register_table(net, &packet_filter, &initial_table.repl); + ipt_register_table(net, &packet_filter, repl); + kfree(repl); if (IS_ERR(net->ipv4.iptable_filter)) return PTR_ERR(net->ipv4.iptable_filter); return 0; @@ -113,9 +92,6 @@ static int __init iptable_filter_init(void) return -EINVAL; } - /* Entry 1 is the FORWARD hook */ - initial_table.entries[1].target.verdict = -forward - 1; - ret = register_pernet_subsys(&iptable_filter_net_ops); if (ret < 0) return ret; diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 2355a229f8e..58d7097baa3 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -27,43 +27,6 @@ MODULE_DESCRIPTION("iptables mangle table"); (1 << NF_INET_LOCAL_OUT) | \ (1 << NF_INET_POST_ROUTING)) -/* Ouch - five different hooks? Maybe this should be a config option..... -- BC */ -static const struct -{ - struct ipt_replace repl; - struct ipt_standard entries[5]; - struct ipt_error term; -} initial_table __net_initdata = { - .repl = { - .name = "mangle", - .valid_hooks = MANGLE_VALID_HOOKS, - .num_entries = 6, - .size = sizeof(struct ipt_standard) * 5 + sizeof(struct ipt_error), - .hook_entry = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_IN] = sizeof(struct ipt_standard), - [NF_INET_FORWARD] = sizeof(struct ipt_standard) * 2, - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 3, - [NF_INET_POST_ROUTING] = sizeof(struct ipt_standard) * 4, - }, - .underflow = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_IN] = sizeof(struct ipt_standard), - [NF_INET_FORWARD] = sizeof(struct ipt_standard) * 2, - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 3, - [NF_INET_POST_ROUTING] = sizeof(struct ipt_standard) * 4, - }, - }, - .entries = { - IPT_STANDARD_INIT(NF_ACCEPT), /* PRE_ROUTING */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IPT_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - IPT_STANDARD_INIT(NF_ACCEPT), /* POST_ROUTING */ - }, - .term = IPT_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_mangler = { .name = "mangle", .valid_hooks = MANGLE_VALID_HOOKS, @@ -134,9 +97,14 @@ static struct nf_hook_ops *mangle_ops __read_mostly; static int __net_init iptable_mangle_net_init(struct net *net) { - /* Register table */ + struct ipt_replace *repl; + + repl = ipt_alloc_initial_table(&packet_mangler); + if (repl == NULL) + return -ENOMEM; net->ipv4.iptable_mangle = - ipt_register_table(net, &packet_mangler, &initial_table.repl); + ipt_register_table(net, &packet_mangler, repl); + kfree(repl); if (IS_ERR(net->ipv4.iptable_mangle)) return PTR_ERR(net->ipv4.iptable_mangle); return 0; diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 62a99154f14..06fb9d11953 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -9,33 +9,6 @@ #define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT)) -static const struct -{ - struct ipt_replace repl; - struct ipt_standard entries[2]; - struct ipt_error term; -} initial_table __net_initdata = { - .repl = { - .name = "raw", - .valid_hooks = RAW_VALID_HOOKS, - .num_entries = 3, - .size = sizeof(struct ipt_standard) * 2 + sizeof(struct ipt_error), - .hook_entry = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) - }, - .underflow = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) - }, - }, - .entries = { - IPT_STANDARD_INIT(NF_ACCEPT), /* PRE_ROUTING */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IPT_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_raw = { .name = "raw", .valid_hooks = RAW_VALID_HOOKS, @@ -66,9 +39,14 @@ static struct nf_hook_ops *rawtable_ops __read_mostly; static int __net_init iptable_raw_net_init(struct net *net) { - /* Register table */ + struct ipt_replace *repl; + + repl = ipt_alloc_initial_table(&packet_raw); + if (repl == NULL) + return -ENOMEM; net->ipv4.iptable_raw = - ipt_register_table(net, &packet_raw, &initial_table.repl); + ipt_register_table(net, &packet_raw, repl); + kfree(repl); if (IS_ERR(net->ipv4.iptable_raw)) return PTR_ERR(net->ipv4.iptable_raw); return 0; diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index b1bf3ca2c6c..cce2f64e6f2 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -27,36 +27,6 @@ MODULE_DESCRIPTION("iptables security table, for MAC rules"); (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT) -static const struct -{ - struct ipt_replace repl; - struct ipt_standard entries[3]; - struct ipt_error term; -} initial_table __net_initdata = { - .repl = { - .name = "security", - .valid_hooks = SECURITY_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error), - .hook_entry = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2, - }, - .underflow = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2, - }, - }, - .entries = { - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IPT_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IPT_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table security_table = { .name = "security", .valid_hooks = SECURITY_VALID_HOOKS, @@ -87,9 +57,14 @@ static struct nf_hook_ops *sectbl_ops __read_mostly; static int __net_init iptable_security_net_init(struct net *net) { - net->ipv4.iptable_security = - ipt_register_table(net, &security_table, &initial_table.repl); + struct ipt_replace *repl; + repl = ipt_alloc_initial_table(&security_table); + if (repl == NULL) + return -ENOMEM; + net->ipv4.iptable_security = + ipt_register_table(net, &security_table, repl); + kfree(repl); if (IS_ERR(net->ipv4.iptable_security)) return PTR_ERR(net->ipv4.iptable_security); diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c index 85da34fdc75..ab74cc0535e 100644 --- a/net/ipv4/netfilter/nf_nat_rule.c +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -28,36 +28,6 @@ (1 << NF_INET_POST_ROUTING) | \ (1 << NF_INET_LOCAL_OUT)) -static const struct -{ - struct ipt_replace repl; - struct ipt_standard entries[3]; - struct ipt_error term; -} nat_initial_table __net_initdata = { - .repl = { - .name = "nat", - .valid_hooks = NAT_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error), - .hook_entry = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_POST_ROUTING] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2 - }, - .underflow = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_POST_ROUTING] = sizeof(struct ipt_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ipt_standard) * 2 - }, - }, - .entries = { - IPT_STANDARD_INIT(NF_ACCEPT), /* PRE_ROUTING */ - IPT_STANDARD_INIT(NF_ACCEPT), /* POST_ROUTING */ - IPT_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IPT_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table nat_table = { .name = "nat", .valid_hooks = NAT_VALID_HOOKS, @@ -186,8 +156,13 @@ static struct xt_target ipt_dnat_reg __read_mostly = { static int __net_init nf_nat_rule_net_init(struct net *net) { - net->ipv4.nat_table = ipt_register_table(net, &nat_table, - &nat_initial_table.repl); + struct ipt_replace *repl; + + repl = ipt_alloc_initial_table(&nat_table); + if (repl == NULL) + return -ENOMEM; + net->ipv4.nat_table = ipt_register_table(net, &nat_table, repl); + kfree(repl); if (IS_ERR(net->ipv4.nat_table)) return PTR_ERR(net->ipv4.nat_table); return 0; diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 4332f459148..dcd7825fe7b 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -29,6 +29,7 @@ #include #include #include +#include "../../netfilter/xt_repldata.h" MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); @@ -67,6 +68,12 @@ do { \ #define inline #endif +void *ip6t_alloc_initial_table(const struct xt_table *info) +{ + return xt_alloc_initial_table(ip6t, IP6T); +} +EXPORT_SYMBOL_GPL(ip6t_alloc_initial_table); + /* We keep a set of rules for each CPU, so we can avoid write-locking them in the softirq when updating the counters and therefore diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index 6e95d0614ca..36b72cafc22 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -21,36 +21,6 @@ MODULE_DESCRIPTION("ip6tables filter table"); (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT)) -static struct -{ - struct ip6t_replace repl; - struct ip6t_standard entries[3]; - struct ip6t_error term; -} initial_table __net_initdata = { - .repl = { - .name = "filter", - .valid_hooks = FILTER_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct ip6t_standard) * 3 + sizeof(struct ip6t_error), - .hook_entry = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ip6t_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2 - }, - .underflow = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ip6t_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2 - }, - }, - .entries = { - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IP6T_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, @@ -78,9 +48,18 @@ module_param(forward, bool, 0000); static int __net_init ip6table_filter_net_init(struct net *net) { - /* Register table */ + struct ip6t_replace *repl; + + repl = ip6t_alloc_initial_table(&packet_filter); + if (repl == NULL) + return -ENOMEM; + /* Entry 1 is the FORWARD hook */ + ((struct ip6t_standard *)repl->entries)[1].target.verdict = + -forward - 1; + net->ipv6.ip6table_filter = - ip6t_register_table(net, &packet_filter, &initial_table.repl); + ip6t_register_table(net, &packet_filter, repl); + kfree(repl); if (IS_ERR(net->ipv6.ip6table_filter)) return PTR_ERR(net->ipv6.ip6table_filter); return 0; @@ -105,9 +84,6 @@ static int __init ip6table_filter_init(void) return -EINVAL; } - /* Entry 1 is the FORWARD hook */ - initial_table.entries[1].target.verdict = -forward - 1; - ret = register_pernet_subsys(&ip6table_filter_net_ops); if (ret < 0) return ret; diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 5023ac52ffe..dc803b7e8e5 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -21,42 +21,6 @@ MODULE_DESCRIPTION("ip6tables mangle table"); (1 << NF_INET_LOCAL_OUT) | \ (1 << NF_INET_POST_ROUTING)) -static const struct -{ - struct ip6t_replace repl; - struct ip6t_standard entries[5]; - struct ip6t_error term; -} initial_table __net_initdata = { - .repl = { - .name = "mangle", - .valid_hooks = MANGLE_VALID_HOOKS, - .num_entries = 6, - .size = sizeof(struct ip6t_standard) * 5 + sizeof(struct ip6t_error), - .hook_entry = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_IN] = sizeof(struct ip6t_standard), - [NF_INET_FORWARD] = sizeof(struct ip6t_standard) * 2, - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 3, - [NF_INET_POST_ROUTING] = sizeof(struct ip6t_standard) * 4, - }, - .underflow = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_IN] = sizeof(struct ip6t_standard), - [NF_INET_FORWARD] = sizeof(struct ip6t_standard) * 2, - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 3, - [NF_INET_POST_ROUTING] = sizeof(struct ip6t_standard) * 4, - }, - }, - .entries = { - IP6T_STANDARD_INIT(NF_ACCEPT), /* PRE_ROUTING */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* POST_ROUTING */ - }, - .term = IP6T_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_mangler = { .name = "mangle", .valid_hooks = MANGLE_VALID_HOOKS, @@ -126,9 +90,14 @@ ip6table_mangle_hook(unsigned int hook, struct sk_buff *skb, static struct nf_hook_ops *mangle_ops __read_mostly; static int __net_init ip6table_mangle_net_init(struct net *net) { - /* Register table */ + struct ip6t_replace *repl; + + repl = ip6t_alloc_initial_table(&packet_mangler); + if (repl == NULL) + return -ENOMEM; net->ipv6.ip6table_mangle = - ip6t_register_table(net, &packet_mangler, &initial_table.repl); + ip6t_register_table(net, &packet_mangler, repl); + kfree(repl); if (IS_ERR(net->ipv6.ip6table_mangle)) return PTR_ERR(net->ipv6.ip6table_mangle); return 0; diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index 3bfa6951164..aef31a29de9 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -8,33 +8,6 @@ #define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT)) -static const struct -{ - struct ip6t_replace repl; - struct ip6t_standard entries[2]; - struct ip6t_error term; -} initial_table __net_initdata = { - .repl = { - .name = "raw", - .valid_hooks = RAW_VALID_HOOKS, - .num_entries = 3, - .size = sizeof(struct ip6t_standard) * 2 + sizeof(struct ip6t_error), - .hook_entry = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) - }, - .underflow = { - [NF_INET_PRE_ROUTING] = 0, - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) - }, - }, - .entries = { - IP6T_STANDARD_INIT(NF_ACCEPT), /* PRE_ROUTING */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IP6T_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table packet_raw = { .name = "raw", .valid_hooks = RAW_VALID_HOOKS, @@ -58,9 +31,14 @@ static struct nf_hook_ops *rawtable_ops __read_mostly; static int __net_init ip6table_raw_net_init(struct net *net) { - /* Register table */ + struct ip6t_replace *repl; + + repl = ip6t_alloc_initial_table(&packet_raw); + if (repl == NULL) + return -ENOMEM; net->ipv6.ip6table_raw = - ip6t_register_table(net, &packet_raw, &initial_table.repl); + ip6t_register_table(net, &packet_raw, repl); + kfree(repl); if (IS_ERR(net->ipv6.ip6table_raw)) return PTR_ERR(net->ipv6.ip6table_raw); return 0; diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c index dd2200f17a6..0824d865aa9 100644 --- a/net/ipv6/netfilter/ip6table_security.c +++ b/net/ipv6/netfilter/ip6table_security.c @@ -26,36 +26,6 @@ MODULE_DESCRIPTION("ip6tables security table, for MAC rules"); (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT) -static const struct -{ - struct ip6t_replace repl; - struct ip6t_standard entries[3]; - struct ip6t_error term; -} initial_table __net_initdata = { - .repl = { - .name = "security", - .valid_hooks = SECURITY_VALID_HOOKS, - .num_entries = 4, - .size = sizeof(struct ip6t_standard) * 3 + sizeof(struct ip6t_error), - .hook_entry = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ip6t_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2, - }, - .underflow = { - [NF_INET_LOCAL_IN] = 0, - [NF_INET_FORWARD] = sizeof(struct ip6t_standard), - [NF_INET_LOCAL_OUT] = sizeof(struct ip6t_standard) * 2, - }, - }, - .entries = { - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_IN */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* FORWARD */ - IP6T_STANDARD_INIT(NF_ACCEPT), /* LOCAL_OUT */ - }, - .term = IP6T_ERROR_INIT, /* ERROR */ -}; - static const struct xt_table security_table = { .name = "security", .valid_hooks = SECURITY_VALID_HOOKS, @@ -79,9 +49,14 @@ static struct nf_hook_ops *sectbl_ops __read_mostly; static int __net_init ip6table_security_net_init(struct net *net) { - net->ipv6.ip6table_security = - ip6t_register_table(net, &security_table, &initial_table.repl); + struct ip6t_replace *repl; + repl = ip6t_alloc_initial_table(&security_table); + if (repl == NULL) + return -ENOMEM; + net->ipv6.ip6table_security = + ip6t_register_table(net, &security_table, repl); + kfree(repl); if (IS_ERR(net->ipv6.ip6table_security)) return PTR_ERR(net->ipv6.ip6table_security); diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index b51cb0d7234..dc2e05cb54c 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -26,7 +26,9 @@ #include #include - +#include +#include +#include MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte "); diff --git a/net/netfilter/xt_repldata.h b/net/netfilter/xt_repldata.h new file mode 100644 index 00000000000..6efe4e5a81c --- /dev/null +++ b/net/netfilter/xt_repldata.h @@ -0,0 +1,35 @@ +/* + * Today's hack: quantum tunneling in structs + * + * 'entries' and 'term' are never anywhere referenced by word in code. In fact, + * they serve as the hanging-off data accessed through repl.data[]. + */ + +#define xt_alloc_initial_table(type, typ2) ({ \ + unsigned int hook_mask = info->valid_hooks; \ + unsigned int nhooks = hweight32(hook_mask); \ + unsigned int bytes = 0, hooknum = 0, i = 0; \ + struct { \ + struct type##_replace repl; \ + struct type##_standard entries[nhooks]; \ + struct type##_error term; \ + } *tbl = kzalloc(sizeof(*tbl), GFP_KERNEL); \ + if (tbl == NULL) \ + return NULL; \ + strncpy(tbl->repl.name, info->name, sizeof(tbl->repl.name)); \ + tbl->term = (struct type##_error)typ2##_ERROR_INIT; \ + tbl->repl.valid_hooks = hook_mask; \ + tbl->repl.num_entries = nhooks + 1; \ + tbl->repl.size = nhooks * sizeof(struct type##_standard) + \ + sizeof(struct type##_error); \ + for (; hook_mask != 0; hook_mask >>= 1, ++hooknum) { \ + if (!(hook_mask & 1)) \ + continue; \ + tbl->repl.hook_entry[hooknum] = bytes; \ + tbl->repl.underflow[hooknum] = bytes; \ + tbl->entries[i++] = (struct type##_standard) \ + typ2##_STANDARD_INIT(NF_ACCEPT); \ + bytes += sizeof(struct type##_standard); \ + } \ + tbl; \ +}) -- cgit v1.2.3-70-g09d2 From 66655de6d132b726be64c324bc3f9ea366d20697 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 8 Feb 2010 23:18:22 +0000 Subject: seq_file: Add helpers for iteration over a hlist Some places in kernel need to iterate over a hlist in seq_file, so provide some common helpers. Signed-off-by: Li Zefan Signed-off-by: David S. Miller --- fs/seq_file.c | 59 +++++++++++++++++++++++++++++++++++++++++++++--- include/linux/seq_file.h | 11 +++++++++ 2 files changed, 67 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/fs/seq_file.c b/fs/seq_file.c index eae7d9dbf3f..f65b16f02da 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -674,7 +674,6 @@ struct list_head *seq_list_start(struct list_head *head, loff_t pos) return NULL; } - EXPORT_SYMBOL(seq_list_start); struct list_head *seq_list_start_head(struct list_head *head, loff_t pos) @@ -684,7 +683,6 @@ struct list_head *seq_list_start_head(struct list_head *head, loff_t pos) return seq_list_start(head, pos - 1); } - EXPORT_SYMBOL(seq_list_start_head); struct list_head *seq_list_next(void *v, struct list_head *head, loff_t *ppos) @@ -695,5 +693,60 @@ struct list_head *seq_list_next(void *v, struct list_head *head, loff_t *ppos) ++*ppos; return lh == head ? NULL : lh; } - EXPORT_SYMBOL(seq_list_next); + +/** + * seq_hlist_start - start an iteration of a hlist + * @head: the head of the hlist + * @pos: the start position of the sequence + * + * Called at seq_file->op->start(). + */ +struct hlist_node *seq_hlist_start(struct hlist_head *head, loff_t pos) +{ + struct hlist_node *node; + + hlist_for_each(node, head) + if (pos-- == 0) + return node; + return NULL; +} +EXPORT_SYMBOL(seq_hlist_start); + +/** + * seq_hlist_start_head - start an iteration of a hlist + * @head: the head of the hlist + * @pos: the start position of the sequence + * + * Called at seq_file->op->start(). Call this function if you want to + * print a header at the top of the output. + */ +struct hlist_node *seq_hlist_start_head(struct hlist_head *head, loff_t pos) +{ + if (!pos) + return SEQ_START_TOKEN; + + return seq_hlist_start(head, pos - 1); +} +EXPORT_SYMBOL(seq_hlist_start_head); + +/** + * seq_hlist_next - move to the next position of the hlist + * @v: the current iterator + * @head: the head of the hlist + * @pos: the current posision + * + * Called at seq_file->op->next(). + */ +struct hlist_node *seq_hlist_next(void *v, struct hlist_head *head, + loff_t *ppos) +{ + struct hlist_node *node = v; + + ++*ppos; + if (v == SEQ_START_TOKEN) + return head->first; + else + return node->next; +} +EXPORT_SYMBOL(seq_hlist_next); diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 8366d8f12e5..c95bcdc18f4 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -135,4 +135,15 @@ extern struct list_head *seq_list_start_head(struct list_head *head, extern struct list_head *seq_list_next(void *v, struct list_head *head, loff_t *ppos); +/* + * Helpers for iteration over hlist_head-s in seq_files + */ + +extern struct hlist_node *seq_hlist_start(struct hlist_head *head, + loff_t pos); +extern struct hlist_node *seq_hlist_start_head(struct hlist_head *head, + loff_t pos); +extern struct hlist_node *seq_hlist_next(void *v, struct hlist_head *head, + loff_t *ppos); + #endif -- cgit v1.2.3-70-g09d2 From c4146644a56b1f213c4c5567c75771883bec33c7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 8 Feb 2010 23:18:45 +0000 Subject: net: add a wrapper sk_entry() Signed-off-by: Li Zefan Signed-off-by: David S. Miller --- include/net/sock.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 3f1a4804bb3..c8d400063c1 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -317,6 +317,11 @@ struct sock { /* * Hashed lists helper routines */ +static inline struct sock *sk_entry(const struct hlist_node *node) +{ + return hlist_entry(node, struct sock, sk_node); +} + static inline struct sock *__sk_head(const struct hlist_head *head) { return hlist_entry(head->first, struct sock, sk_node); -- cgit v1.2.3-70-g09d2 From 15682bc488d4af8c9bb998844a94281025e0a333 Mon Sep 17 00:00:00 2001 From: Peter P Waskiewicz Jr Date: Wed, 10 Feb 2010 20:03:05 -0800 Subject: ethtool: Introduce n-tuple filter programming support This patchset enables the ethtool layer to program n-tuple filters to an underlying device. The idea is to allow capable hardware to have static rules applied that can assist steering flows into appropriate queues. Hardware that is known to support these types of filters today are ixgbe and niu. Signed-off-by: Peter P Waskiewicz Jr Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- include/linux/ethtool.h | 50 +++++++ include/linux/netdevice.h | 3 + net/core/dev.c | 5 + net/core/ethtool.c | 329 +++++++++++++++++++++++++++++++++++++++++++++- 4 files changed, 386 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index ef4a2d84d92..a3cac53a076 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -14,6 +14,7 @@ #define _LINUX_ETHTOOL_H #include +#include /* This should work for both 32 and 64 bit userland. */ struct ethtool_cmd { @@ -242,6 +243,7 @@ enum ethtool_stringset { ETH_SS_TEST = 0, ETH_SS_STATS, ETH_SS_PRIV_FLAGS, + ETH_SS_NTUPLE_FILTERS, }; /* for passing string sets for data tagging */ @@ -290,6 +292,7 @@ struct ethtool_perm_addr { */ enum ethtool_flags { ETH_FLAG_LRO = (1 << 15), /* LRO is enabled */ + ETH_FLAG_NTUPLE = (1 << 27), /* N-tuple filters enabled */ }; /* The following structures are for supporting RX network flow @@ -363,6 +366,35 @@ struct ethtool_rxnfc { __u32 rule_locs[0]; }; +struct ethtool_rx_ntuple_flow_spec { + __u32 flow_type; + union { + struct ethtool_tcpip4_spec tcp_ip4_spec; + struct ethtool_tcpip4_spec udp_ip4_spec; + struct ethtool_tcpip4_spec sctp_ip4_spec; + struct ethtool_ah_espip4_spec ah_ip4_spec; + struct ethtool_ah_espip4_spec esp_ip4_spec; + struct ethtool_rawip4_spec raw_ip4_spec; + struct ethtool_ether_spec ether_spec; + struct ethtool_usrip4_spec usr_ip4_spec; + __u8 hdata[64]; + } h_u, m_u; /* entry, mask */ + + __u16 vlan_tag; + __u16 vlan_tag_mask; + __u64 data; /* user-defined flow spec data */ + __u64 data_mask; /* user-defined flow spec mask */ + + /* signed to distinguish between queue and actions (DROP) */ + __s32 action; +#define ETHTOOL_RXNTUPLE_ACTION_DROP -1 +}; + +struct ethtool_rx_ntuple { + __u32 cmd; + struct ethtool_rx_ntuple_flow_spec fs; +}; + #define ETHTOOL_FLASH_MAX_FILENAME 128 enum ethtool_flash_op_type { ETHTOOL_FLASH_ALL_REGIONS = 0, @@ -377,6 +409,18 @@ struct ethtool_flash { #ifdef __KERNEL__ +struct ethtool_rx_ntuple_flow_spec_container { + struct ethtool_rx_ntuple_flow_spec fs; + struct list_head list; +}; + +struct ethtool_rx_ntuple_list { +#define ETHTOOL_MAX_NTUPLE_LIST_ENTRY 1024 +#define ETHTOOL_MAX_NTUPLE_STRING_PER_ENTRY 14 + struct list_head list; + unsigned int count; +}; + struct net_device; /* Some generic methods drivers may use in their ethtool_ops */ @@ -394,6 +438,7 @@ u32 ethtool_op_get_ufo(struct net_device *dev); int ethtool_op_set_ufo(struct net_device *dev, u32 data); u32 ethtool_op_get_flags(struct net_device *dev); int ethtool_op_set_flags(struct net_device *dev, u32 data); +void ethtool_ntuple_flush(struct net_device *dev); /** * ðtool_ops - Alter and report network device settings @@ -500,6 +545,8 @@ struct ethtool_ops { int (*set_rxnfc)(struct net_device *, struct ethtool_rxnfc *); int (*flash_device)(struct net_device *, struct ethtool_flash *); int (*reset)(struct net_device *, u32 *); + int (*set_rx_ntuple)(struct net_device *, struct ethtool_rx_ntuple *); + int (*get_rx_ntuple)(struct net_device *, u32 stringset, void *); }; #endif /* __KERNEL__ */ @@ -559,6 +606,9 @@ struct ethtool_ops { #define ETHTOOL_FLASHDEV 0x00000033 /* Flash firmware to device */ #define ETHTOOL_RESET 0x00000034 /* Reset hardware */ +#define ETHTOOL_SRXNTUPLE 0x00000035 /* Add an n-tuple filter to device */ +#define ETHTOOL_GRXNTUPLE 0x00000036 /* Get n-tuple filters from device */ + /* compatibility with older code */ #define SPARC_ETH_GSET ETHTOOL_GSET #define SPARC_ETH_SSET ETHTOOL_SSET diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e535700a3b7..cdf53a8d9ff 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -746,6 +746,7 @@ struct net_device { #define NETIF_F_FCOE_CRC (1 << 24) /* FCoE CRC32 */ #define NETIF_F_SCTP_CSUM (1 << 25) /* SCTP checksum offload */ #define NETIF_F_FCOE_MTU (1 << 26) /* Supports max FCoE MTU, 2158 bytes*/ +#define NETIF_F_NTUPLE (1 << 27) /* N-tuple filters supported */ /* Segmentation offload features */ #define NETIF_F_GSO_SHIFT 16 @@ -954,6 +955,8 @@ struct net_device { /* max exchange id for FCoE LRO by ddp */ unsigned int fcoe_ddp_xid; #endif + /* n-tuple filter list attached to this device */ + struct ethtool_rx_ntuple_list ethtool_ntuple_list; }; #define to_net_dev(d) container_of(d, struct net_device, dev) diff --git a/net/core/dev.c b/net/core/dev.c index 94c1eeed25e..ae75f25ac0a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5419,6 +5419,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, netdev_init_queues(dev); + INIT_LIST_HEAD(&dev->ethtool_ntuple_list.list); + dev->ethtool_ntuple_list.count = 0; INIT_LIST_HEAD(&dev->napi_list); INIT_LIST_HEAD(&dev->unreg_list); INIT_LIST_HEAD(&dev->link_watch_list); @@ -5455,6 +5457,9 @@ void free_netdev(struct net_device *dev) /* Flush device addresses */ dev_addr_flush(dev); + /* Clear ethtool n-tuple list */ + ethtool_ntuple_flush(dev); + list_for_each_entry_safe(p, n, &dev->napi_list, dev_list) netif_napi_del(p); diff --git a/net/core/ethtool.c b/net/core/ethtool.c index d8aee584e8d..6ec73d3983a 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -120,7 +120,7 @@ int ethtool_op_set_ufo(struct net_device *dev, u32 data) * NETIF_F_xxx values in include/linux/netdevice.h */ static const u32 flags_dup_features = - ETH_FLAG_LRO; + (ETH_FLAG_LRO | ETH_FLAG_NTUPLE); u32 ethtool_op_get_flags(struct net_device *dev) { @@ -139,9 +139,26 @@ int ethtool_op_set_flags(struct net_device *dev, u32 data) else dev->features &= ~NETIF_F_LRO; + if (data & ETH_FLAG_NTUPLE) + dev->features |= NETIF_F_NTUPLE; + else + dev->features &= ~NETIF_F_NTUPLE; + return 0; } +void ethtool_ntuple_flush(struct net_device *dev) +{ + struct ethtool_rx_ntuple_flow_spec_container *fsc, *f; + + list_for_each_entry_safe(fsc, f, &dev->ethtool_ntuple_list.list, list) { + list_del(&fsc->list); + kfree(fsc); + } + dev->ethtool_ntuple_list.count = 0; +} +EXPORT_SYMBOL(ethtool_ntuple_flush); + /* Handlers for each ethtool command */ static int ethtool_get_settings(struct net_device *dev, void __user *useraddr) @@ -266,6 +283,307 @@ err_out: return ret; } +static int __rx_ntuple_filter_add(struct ethtool_rx_ntuple_list *list, + struct ethtool_rx_ntuple_flow_spec *spec) +{ + struct ethtool_rx_ntuple_flow_spec_container *fsc; + + /* don't add filters forever */ + if (list->count >= ETHTOOL_MAX_NTUPLE_LIST_ENTRY) + return 0; + + fsc = kmalloc(sizeof(*fsc), GFP_ATOMIC); + if (!fsc) + return -ENOMEM; + + /* Copy the whole filter over */ + fsc->fs.flow_type = spec->flow_type; + memcpy(&fsc->fs.h_u, &spec->h_u, sizeof(spec->h_u)); + memcpy(&fsc->fs.m_u, &spec->m_u, sizeof(spec->m_u)); + + fsc->fs.vlan_tag = spec->vlan_tag; + fsc->fs.vlan_tag_mask = spec->vlan_tag_mask; + fsc->fs.data = spec->data; + fsc->fs.data_mask = spec->data_mask; + fsc->fs.action = spec->action; + + /* add to the list */ + list_add_tail_rcu(&fsc->list, &list->list); + list->count++; + + return 0; +} + +static int ethtool_set_rx_ntuple(struct net_device *dev, void __user *useraddr) +{ + struct ethtool_rx_ntuple cmd; + const struct ethtool_ops *ops = dev->ethtool_ops; + int ret; + + if (!ops->set_rx_ntuple) + return -EOPNOTSUPP; + + if (!(dev->features & NETIF_F_NTUPLE)) + return -EINVAL; + + if (copy_from_user(&cmd, useraddr, sizeof(cmd))) + return -EFAULT; + + ret = ops->set_rx_ntuple(dev, &cmd); + + /* + * Cache filter in dev struct for GET operation only if + * the underlying driver doesn't have its own GET operation, and + * only if the filter was added successfully. + */ + if (!ops->get_rx_ntuple && !ret) + if (__rx_ntuple_filter_add(&dev->ethtool_ntuple_list, &cmd.fs)) + return -ENOMEM; + + return ret; +} + +static int ethtool_get_rx_ntuple(struct net_device *dev, void __user *useraddr) +{ + struct ethtool_gstrings gstrings; + const struct ethtool_ops *ops = dev->ethtool_ops; + struct ethtool_rx_ntuple_flow_spec_container *fsc; + u8 *data; + char *p; + int ret, i, num_strings = 0; + + if (!ops->get_sset_count) + return -EOPNOTSUPP; + + if (copy_from_user(&gstrings, useraddr, sizeof(gstrings))) + return -EFAULT; + + ret = ops->get_sset_count(dev, gstrings.string_set); + if (ret < 0) + return ret; + + gstrings.len = ret; + + data = kmalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER); + if (!data) + return -ENOMEM; + + if (ops->get_rx_ntuple) { + /* driver-specific filter grab */ + ret = ops->get_rx_ntuple(dev, gstrings.string_set, data); + goto copy; + } + + /* default ethtool filter grab */ + i = 0; + p = (char *)data; + list_for_each_entry(fsc, &dev->ethtool_ntuple_list.list, list) { + sprintf(p, "Filter %d:\n", i); + p += ETH_GSTRING_LEN; + num_strings++; + + switch (fsc->fs.flow_type) { + case TCP_V4_FLOW: + sprintf(p, "\tFlow Type: TCP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case UDP_V4_FLOW: + sprintf(p, "\tFlow Type: UDP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case SCTP_V4_FLOW: + sprintf(p, "\tFlow Type: SCTP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case AH_ESP_V4_FLOW: + sprintf(p, "\tFlow Type: AH ESP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case ESP_V4_FLOW: + sprintf(p, "\tFlow Type: ESP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case IP_USER_FLOW: + sprintf(p, "\tFlow Type: Raw IP\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case IPV4_FLOW: + sprintf(p, "\tFlow Type: IPv4\n"); + p += ETH_GSTRING_LEN; + num_strings++; + break; + default: + sprintf(p, "\tFlow Type: Unknown\n"); + p += ETH_GSTRING_LEN; + num_strings++; + goto unknown_filter; + }; + + /* now the rest of the filters */ + switch (fsc->fs.flow_type) { + case TCP_V4_FLOW: + case UDP_V4_FLOW: + case SCTP_V4_FLOW: + sprintf(p, "\tSrc IP addr: 0x%x\n", + fsc->fs.h_u.tcp_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSrc IP mask: 0x%x\n", + fsc->fs.m_u.tcp_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP addr: 0x%x\n", + fsc->fs.h_u.tcp_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP mask: 0x%x\n", + fsc->fs.m_u.tcp_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSrc Port: %d, mask: 0x%x\n", + fsc->fs.h_u.tcp_ip4_spec.psrc, + fsc->fs.m_u.tcp_ip4_spec.psrc); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest Port: %d, mask: 0x%x\n", + fsc->fs.h_u.tcp_ip4_spec.pdst, + fsc->fs.m_u.tcp_ip4_spec.pdst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tTOS: %d, mask: 0x%x\n", + fsc->fs.h_u.tcp_ip4_spec.tos, + fsc->fs.m_u.tcp_ip4_spec.tos); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case AH_ESP_V4_FLOW: + case ESP_V4_FLOW: + sprintf(p, "\tSrc IP addr: 0x%x\n", + fsc->fs.h_u.ah_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSrc IP mask: 0x%x\n", + fsc->fs.m_u.ah_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP addr: 0x%x\n", + fsc->fs.h_u.ah_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP mask: 0x%x\n", + fsc->fs.m_u.ah_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSPI: %d, mask: 0x%x\n", + fsc->fs.h_u.ah_ip4_spec.spi, + fsc->fs.m_u.ah_ip4_spec.spi); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tTOS: %d, mask: 0x%x\n", + fsc->fs.h_u.ah_ip4_spec.tos, + fsc->fs.m_u.ah_ip4_spec.tos); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case IP_USER_FLOW: + sprintf(p, "\tSrc IP addr: 0x%x\n", + fsc->fs.h_u.raw_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSrc IP mask: 0x%x\n", + fsc->fs.m_u.raw_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP addr: 0x%x\n", + fsc->fs.h_u.raw_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP mask: 0x%x\n", + fsc->fs.m_u.raw_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + break; + case IPV4_FLOW: + sprintf(p, "\tSrc IP addr: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tSrc IP mask: 0x%x\n", + fsc->fs.m_u.usr_ip4_spec.ip4src); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP addr: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tDest IP mask: 0x%x\n", + fsc->fs.m_u.usr_ip4_spec.ip4dst); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tL4 bytes: 0x%x, mask: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.l4_4_bytes, + fsc->fs.m_u.usr_ip4_spec.l4_4_bytes); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tTOS: %d, mask: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.tos, + fsc->fs.m_u.usr_ip4_spec.tos); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tIP Version: %d, mask: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.ip_ver, + fsc->fs.m_u.usr_ip4_spec.ip_ver); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tProtocol: %d, mask: 0x%x\n", + fsc->fs.h_u.usr_ip4_spec.proto, + fsc->fs.m_u.usr_ip4_spec.proto); + p += ETH_GSTRING_LEN; + num_strings++; + break; + }; + sprintf(p, "\tVLAN: %d, mask: 0x%x\n", + fsc->fs.vlan_tag, fsc->fs.vlan_tag_mask); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tUser-defined: 0x%Lx\n", fsc->fs.data); + p += ETH_GSTRING_LEN; + num_strings++; + sprintf(p, "\tUser-defined mask: 0x%Lx\n", fsc->fs.data_mask); + p += ETH_GSTRING_LEN; + num_strings++; + if (fsc->fs.action == ETHTOOL_RXNTUPLE_ACTION_DROP) + sprintf(p, "\tAction: Drop\n"); + else + sprintf(p, "\tAction: Direct to queue %d\n", + fsc->fs.action); + p += ETH_GSTRING_LEN; + num_strings++; +unknown_filter: + i++; + } +copy: + /* indicate to userspace how many strings we actually have */ + gstrings.len = num_strings; + ret = -EFAULT; + if (copy_to_user(useraddr, &gstrings, sizeof(gstrings))) + goto out; + useraddr += sizeof(gstrings); + if (copy_to_user(useraddr, data, gstrings.len * ETH_GSTRING_LEN)) + goto out; + ret = 0; + +out: + kfree(data); + return ret; +} + static int ethtool_get_regs(struct net_device *dev, char __user *useraddr) { struct ethtool_regs regs; @@ -313,6 +631,9 @@ static int ethtool_reset(struct net_device *dev, char __user *useraddr) if (copy_from_user(&reset, useraddr, sizeof(reset))) return -EFAULT; + /* Clear ethtool n-tuple list */ + ethtool_ntuple_flush(dev); + ret = dev->ethtool_ops->reset(dev, &reset.data); if (ret) return ret; @@ -1112,6 +1433,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_RESET: rc = ethtool_reset(dev, useraddr); break; + case ETHTOOL_SRXNTUPLE: + rc = ethtool_set_rx_ntuple(dev, useraddr); + break; + case ETHTOOL_GRXNTUPLE: + rc = ethtool_get_rx_ntuple(dev, useraddr); + break; default: rc = -EOPNOTSUPP; } -- cgit v1.2.3-70-g09d2 From b87921bdf25485afd8f5a5f25e86b5acef32a9cf Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:22:48 +0100 Subject: netfilter: nf_conntrack: show helper and class in /proc/net/nf_conntrack_expect Make the output a bit more informative by showing the helper an expectation belongs to and the expectation class. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_expect.h | 1 + net/netfilter/nf_conntrack_expect.c | 9 +++++++++ net/netfilter/nf_conntrack_sip.c | 3 +++ 3 files changed, 13 insertions(+) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 9a2b9cb5227..917e170fa75 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -66,6 +66,7 @@ static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp) struct nf_conntrack_expect_policy { unsigned int max_expected; unsigned int timeout; + const char *name; }; #define NF_CT_EXPECT_CLASS_DEFAULT 0 diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 2f25ff61098..33b85f834c0 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -500,6 +500,7 @@ static void exp_seq_stop(struct seq_file *seq, void *v) static int exp_seq_show(struct seq_file *s, void *v) { struct nf_conntrack_expect *expect; + struct nf_conntrack_helper *helper; struct hlist_node *n = v; char *delim = ""; @@ -525,6 +526,14 @@ static int exp_seq_show(struct seq_file *s, void *v) if (expect->flags & NF_CT_EXPECT_INACTIVE) seq_printf(s, "%sINACTIVE", delim); + helper = rcu_dereference(nfct_help(expect->master)->helper); + if (helper) { + seq_printf(s, "%s%s", expect->flags ? " " : "", helper->name); + if (helper->expect_policy[expect->class].name) + seq_printf(s, "/%s", + helper->expect_policy[expect->class].name); + } + return seq_putc(s, '\n'); } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 023966b569b..419c5cabb33 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1305,14 +1305,17 @@ static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1] = { [SIP_EXPECT_SIGNALLING] = { + .name = "signalling", .max_expected = 1, .timeout = 3 * 60, }, [SIP_EXPECT_AUDIO] = { + .name = "audio", .max_expected = 2 * IP_CT_DIR_MAX, .timeout = 3 * 60, }, [SIP_EXPECT_VIDEO] = { + .name = "video", .max_expected = 2 * IP_CT_DIR_MAX, .timeout = 3 * 60, }, -- cgit v1.2.3-70-g09d2 From 3b6b9fab42fe98358d70735cf98d43fc18dc79c9 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:23:53 +0100 Subject: netfilter: nf_conntrack_sip: pass data offset to NAT functions When using TCP multiple SIP messages might be present in a single packet. A following patch will parse them by setting the dptr to the beginning of each message. The NAT helper needs to reload the dptr value after mangling the packet however, so it needs to know the offset of the message to the beginning of the packet. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_sip.h | 14 ++-- net/ipv4/netfilter/nf_nat_sip.c | 101 +++++++++++++++-------------- net/netfilter/nf_conntrack_sip.c | 82 ++++++++++++----------- 3 files changed, 105 insertions(+), 92 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 23aa2ec6b7b..2c6950b8bf7 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -34,10 +34,10 @@ struct sdp_media_type { struct sip_handler { const char *method; unsigned int len; - int (*request)(struct sk_buff *skb, + int (*request)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq); - int (*response)(struct sk_buff *skb, + int (*response)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq, unsigned int code); }; @@ -100,33 +100,39 @@ enum sdp_header_types { }; extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, + unsigned int dataoff, const char **dptr, unsigned int *datalen); extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, + unsigned int dataoff, const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *exp, unsigned int matchoff, unsigned int matchlen); extern unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, - const char **dptr, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, enum sdp_header_types type, enum sdp_header_types term, const union nf_inet_addr *addr); extern unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, + unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int matchoff, unsigned int matchlen, u_int16_t port); extern unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, - const char **dptr, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, const union nf_inet_addr *addr); extern unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb, + unsigned int dataoff, const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *rtp_exp, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 07d61a57613..2454ea5abb7 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -29,7 +29,7 @@ MODULE_DESCRIPTION("SIP NAT helper"); MODULE_ALIAS("ip_nat_sip"); -static unsigned int mangle_packet(struct sk_buff *skb, +static unsigned int mangle_packet(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int matchoff, unsigned int matchlen, const char *buffer, unsigned int buflen) @@ -42,12 +42,12 @@ static unsigned int mangle_packet(struct sk_buff *skb, return 0; /* Reload data pointer and adjust datalen value */ - *dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr); + *dptr = skb->data + dataoff; *datalen += buflen - matchlen; return 1; } -static int map_addr(struct sk_buff *skb, +static int map_addr(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int matchoff, unsigned int matchlen, union nf_inet_addr *addr, __be16 port) @@ -76,11 +76,11 @@ static int map_addr(struct sk_buff *skb, buflen = sprintf(buffer, "%pI4:%u", &newaddr, ntohs(newport)); - return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + return mangle_packet(skb, dataoff, dptr, datalen, matchoff, matchlen, buffer, buflen); } -static int map_sip_addr(struct sk_buff *skb, +static int map_sip_addr(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, enum sip_header_types type) { @@ -93,16 +93,17 @@ static int map_sip_addr(struct sk_buff *skb, if (ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, type, NULL, &matchoff, &matchlen, &addr, &port) <= 0) return 1; - return map_addr(skb, dptr, datalen, matchoff, matchlen, &addr, port); + return map_addr(skb, dataoff, dptr, datalen, matchoff, matchlen, + &addr, port); } -static unsigned int ip_nat_sip(struct sk_buff *skb, +static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - unsigned int dataoff, matchoff, matchlen; + unsigned int coff, matchoff, matchlen; union nf_inet_addr addr; __be16 port; int request, in_header; @@ -112,7 +113,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, if (ct_sip_parse_request(ct, *dptr, *datalen, &matchoff, &matchlen, &addr, &port) > 0 && - !map_addr(skb, dptr, datalen, matchoff, matchlen, + !map_addr(skb, dataoff, dptr, datalen, matchoff, matchlen, &addr, port)) return NF_DROP; request = 1; @@ -138,7 +139,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, goto next; } - if (!map_addr(skb, dptr, datalen, matchoff, matchlen, + if (!map_addr(skb, dataoff, dptr, datalen, matchoff, matchlen, &addr, port)) return NF_DROP; @@ -153,8 +154,8 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, addr.ip != ct->tuplehash[!dir].tuple.dst.u3.ip) { buflen = sprintf(buffer, "%pI4", &ct->tuplehash[!dir].tuple.dst.u3.ip); - if (!mangle_packet(skb, dptr, datalen, poff, plen, - buffer, buflen)) + if (!mangle_packet(skb, dataoff, dptr, datalen, + poff, plen, buffer, buflen)) return NF_DROP; } @@ -167,8 +168,8 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, addr.ip != ct->tuplehash[!dir].tuple.src.u3.ip) { buflen = sprintf(buffer, "%pI4", &ct->tuplehash[!dir].tuple.src.u3.ip); - if (!mangle_packet(skb, dptr, datalen, poff, plen, - buffer, buflen)) + if (!mangle_packet(skb, dataoff, dptr, datalen, + poff, plen, buffer, buflen)) return NF_DROP; } @@ -181,27 +182,27 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, htons(n) != ct->tuplehash[!dir].tuple.src.u.udp.port) { __be16 p = ct->tuplehash[!dir].tuple.src.u.udp.port; buflen = sprintf(buffer, "%u", ntohs(p)); - if (!mangle_packet(skb, dptr, datalen, poff, plen, - buffer, buflen)) + if (!mangle_packet(skb, dataoff, dptr, datalen, + poff, plen, buffer, buflen)) return NF_DROP; } } next: /* Translate Contact headers */ - dataoff = 0; + coff = 0; in_header = 0; - while (ct_sip_parse_header_uri(ct, *dptr, &dataoff, *datalen, + while (ct_sip_parse_header_uri(ct, *dptr, &coff, *datalen, SIP_HDR_CONTACT, &in_header, &matchoff, &matchlen, &addr, &port) > 0) { - if (!map_addr(skb, dptr, datalen, matchoff, matchlen, + if (!map_addr(skb, dataoff, dptr, datalen, matchoff, matchlen, &addr, port)) return NF_DROP; } - if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM) || - !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO)) + if (!map_sip_addr(skb, dataoff, dptr, datalen, SIP_HDR_FROM) || + !map_sip_addr(skb, dataoff, dptr, datalen, SIP_HDR_TO)) return NF_DROP; return NF_ACCEPT; } @@ -232,7 +233,7 @@ static void ip_nat_sip_expected(struct nf_conn *ct, } } -static unsigned int ip_nat_sip_expect(struct sk_buff *skb, +static unsigned int ip_nat_sip_expect(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *exp, unsigned int matchoff, @@ -279,8 +280,8 @@ static unsigned int ip_nat_sip_expect(struct sk_buff *skb, if (exp->tuple.dst.u3.ip != exp->saved_ip || exp->tuple.dst.u.udp.port != exp->saved_proto.udp.port) { buflen = sprintf(buffer, "%pI4:%u", &newip, port); - if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen, - buffer, buflen)) + if (!mangle_packet(skb, dataoff, dptr, datalen, + matchoff, matchlen, buffer, buflen)) goto err; } return NF_ACCEPT; @@ -290,7 +291,7 @@ err: return NF_DROP; } -static int mangle_content_len(struct sk_buff *skb, +static int mangle_content_len(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; @@ -312,12 +313,13 @@ static int mangle_content_len(struct sk_buff *skb, return 0; buflen = sprintf(buffer, "%u", c_len); - return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + return mangle_packet(skb, dataoff, dptr, datalen, matchoff, matchlen, buffer, buflen); } -static int mangle_sdp_packet(struct sk_buff *skb, const char **dptr, - unsigned int dataoff, unsigned int *datalen, +static int mangle_sdp_packet(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, enum sdp_header_types type, enum sdp_header_types term, char *buffer, int buflen) @@ -326,16 +328,16 @@ static int mangle_sdp_packet(struct sk_buff *skb, const char **dptr, struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchlen, matchoff; - if (ct_sip_get_sdp_header(ct, *dptr, dataoff, *datalen, type, term, + if (ct_sip_get_sdp_header(ct, *dptr, sdpoff, *datalen, type, term, &matchoff, &matchlen) <= 0) return -ENOENT; - return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + return mangle_packet(skb, dataoff, dptr, datalen, matchoff, matchlen, buffer, buflen) ? 0 : -EINVAL; } -static unsigned int ip_nat_sdp_addr(struct sk_buff *skb, const char **dptr, - unsigned int dataoff, - unsigned int *datalen, +static unsigned int ip_nat_sdp_addr(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, enum sdp_header_types type, enum sdp_header_types term, const union nf_inet_addr *addr) @@ -344,16 +346,15 @@ static unsigned int ip_nat_sdp_addr(struct sk_buff *skb, const char **dptr, unsigned int buflen; buflen = sprintf(buffer, "%pI4", &addr->ip); - if (mangle_sdp_packet(skb, dptr, dataoff, datalen, type, term, + if (mangle_sdp_packet(skb, dataoff, dptr, datalen, sdpoff, type, term, buffer, buflen)) return 0; - return mangle_content_len(skb, dptr, datalen); + return mangle_content_len(skb, dataoff, dptr, datalen); } -static unsigned int ip_nat_sdp_port(struct sk_buff *skb, - const char **dptr, - unsigned int *datalen, +static unsigned int ip_nat_sdp_port(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, unsigned int matchoff, unsigned int matchlen, u_int16_t port) @@ -362,16 +363,16 @@ static unsigned int ip_nat_sdp_port(struct sk_buff *skb, unsigned int buflen; buflen = sprintf(buffer, "%u", port); - if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen, + if (!mangle_packet(skb, dataoff, dptr, datalen, matchoff, matchlen, buffer, buflen)) return 0; - return mangle_content_len(skb, dptr, datalen); + return mangle_content_len(skb, dataoff, dptr, datalen); } -static unsigned int ip_nat_sdp_session(struct sk_buff *skb, const char **dptr, - unsigned int dataoff, - unsigned int *datalen, +static unsigned int ip_nat_sdp_session(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, const union nf_inet_addr *addr) { char buffer[sizeof("nnn.nnn.nnn.nnn")]; @@ -379,12 +380,12 @@ static unsigned int ip_nat_sdp_session(struct sk_buff *skb, const char **dptr, /* Mangle session description owner and contact addresses */ buflen = sprintf(buffer, "%pI4", &addr->ip); - if (mangle_sdp_packet(skb, dptr, dataoff, datalen, + if (mangle_sdp_packet(skb, dataoff, dptr, datalen, sdpoff, SDP_HDR_OWNER_IP4, SDP_HDR_MEDIA, buffer, buflen)) return 0; - switch (mangle_sdp_packet(skb, dptr, dataoff, datalen, + switch (mangle_sdp_packet(skb, dataoff, dptr, datalen, sdpoff, SDP_HDR_CONNECTION_IP4, SDP_HDR_MEDIA, buffer, buflen)) { case 0: @@ -401,14 +402,13 @@ static unsigned int ip_nat_sdp_session(struct sk_buff *skb, const char **dptr, return 0; } - return mangle_content_len(skb, dptr, datalen); + return mangle_content_len(skb, dataoff, dptr, datalen); } /* So, this packet has hit the connection tracking matching code. Mangle it, and change the expectation to match the new version. */ -static unsigned int ip_nat_sdp_media(struct sk_buff *skb, - const char **dptr, - unsigned int *datalen, +static unsigned int ip_nat_sdp_media(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *rtp_exp, struct nf_conntrack_expect *rtcp_exp, unsigned int mediaoff, @@ -456,7 +456,8 @@ static unsigned int ip_nat_sdp_media(struct sk_buff *skb, /* Update media port. */ if (rtp_exp->tuple.dst.u.udp.port != rtp_exp->saved_proto.udp.port && - !ip_nat_sdp_port(skb, dptr, datalen, mediaoff, medialen, port)) + !ip_nat_sdp_port(skb, dataoff, dptr, datalen, + mediaoff, medialen, port)) goto err2; return NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 0ca2f2b5c2f..0ec37d6a2df 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -50,12 +50,13 @@ module_param(sip_direct_media, int, 0600); MODULE_PARM_DESC(sip_direct_media, "Expect Media streams between signalling " "endpoints only (default 1)"); -unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, +unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_hook); unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, + unsigned int dataoff, const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *exp, @@ -63,17 +64,17 @@ unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, unsigned int matchlen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook); -unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, +unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, - unsigned int dataoff, unsigned int *datalen, + unsigned int sdpoff, enum sdp_header_types type, enum sdp_header_types term, const union nf_inet_addr *addr) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_addr_hook); -unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, +unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int matchoff, @@ -82,14 +83,15 @@ unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, EXPORT_SYMBOL_GPL(nf_nat_sdp_port_hook); unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, - const char **dptr, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int sdpoff, const union nf_inet_addr *addr) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_session_hook); -unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb, +unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, struct nf_conntrack_expect *rtp_exp, @@ -729,7 +731,7 @@ static void flush_expectations(struct nf_conn *ct, bool media) spin_unlock_bh(&nf_conntrack_lock); } -static int set_expected_rtp_rtcp(struct sk_buff *skb, +static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, union nf_inet_addr *daddr, __be16 port, enum sip_expectation_classes class, @@ -806,7 +808,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, if (direct_rtp) { nf_nat_sdp_port = rcu_dereference(nf_nat_sdp_port_hook); if (nf_nat_sdp_port && - !nf_nat_sdp_port(skb, dptr, datalen, + !nf_nat_sdp_port(skb, dataoff, dptr, datalen, mediaoff, medialen, ntohs(rtp_port))) goto err1; } @@ -828,7 +830,8 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook); if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK && !direct_rtp) - ret = nf_nat_sdp_media(skb, dptr, datalen, rtp_exp, rtcp_exp, + ret = nf_nat_sdp_media(skb, dataoff, dptr, datalen, + rtp_exp, rtcp_exp, mediaoff, medialen, daddr); else { if (nf_ct_expect_related(rtp_exp) == 0) { @@ -867,7 +870,7 @@ static const struct sdp_media_type *sdp_media_type(const char *dptr, return NULL; } -static int process_sdp(struct sk_buff *skb, +static int process_sdp(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq) { @@ -942,7 +945,7 @@ static int process_sdp(struct sk_buff *skb, else return NF_DROP; - ret = set_expected_rtp_rtcp(skb, dptr, datalen, + ret = set_expected_rtp_rtcp(skb, dataoff, dptr, datalen, &rtp_addr, htons(port), t->class, mediaoff, medialen); if (ret != NF_ACCEPT) @@ -950,8 +953,9 @@ static int process_sdp(struct sk_buff *skb, /* Update media connection address if present */ if (maddr_len && nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) { - ret = nf_nat_sdp_addr(skb, dptr, mediaoff, datalen, - c_hdr, SDP_HDR_MEDIA, &rtp_addr); + ret = nf_nat_sdp_addr(skb, dataoff, dptr, datalen, + mediaoff, c_hdr, SDP_HDR_MEDIA, + &rtp_addr); if (ret != NF_ACCEPT) return ret; } @@ -961,14 +965,15 @@ static int process_sdp(struct sk_buff *skb, /* Update session connection and owner addresses */ nf_nat_sdp_session = rcu_dereference(nf_nat_sdp_session_hook); if (nf_nat_sdp_session && ct->status & IPS_NAT_MASK) - ret = nf_nat_sdp_session(skb, dptr, sdpoff, datalen, &rtp_addr); + ret = nf_nat_sdp_session(skb, dataoff, dptr, datalen, sdpoff, + &rtp_addr); if (ret == NF_ACCEPT && i > 0) help->help.ct_sip_info.invite_cseq = cseq; return ret; } -static int process_invite_response(struct sk_buff *skb, +static int process_invite_response(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq, unsigned int code) { @@ -978,13 +983,13 @@ static int process_invite_response(struct sk_buff *skb, if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) - return process_sdp(skb, dptr, datalen, cseq); + return process_sdp(skb, dataoff, dptr, datalen, cseq); else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); return NF_ACCEPT; } -static int process_update_response(struct sk_buff *skb, +static int process_update_response(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq, unsigned int code) { @@ -994,13 +999,13 @@ static int process_update_response(struct sk_buff *skb, if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) - return process_sdp(skb, dptr, datalen, cseq); + return process_sdp(skb, dataoff, dptr, datalen, cseq); else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); return NF_ACCEPT; } -static int process_prack_response(struct sk_buff *skb, +static int process_prack_response(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq, unsigned int code) { @@ -1010,13 +1015,13 @@ static int process_prack_response(struct sk_buff *skb, if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) - return process_sdp(skb, dptr, datalen, cseq); + return process_sdp(skb, dataoff, dptr, datalen, cseq); else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); return NF_ACCEPT; } -static int process_bye_request(struct sk_buff *skb, +static int process_bye_request(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq) { @@ -1031,7 +1036,7 @@ static int process_bye_request(struct sk_buff *skb, * signalling connections. The expectation is marked inactive and is activated * when receiving a response indicating success from the registrar. */ -static int process_register_request(struct sk_buff *skb, +static int process_register_request(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq) { @@ -1101,7 +1106,7 @@ static int process_register_request(struct sk_buff *skb, nf_nat_sip_expect = rcu_dereference(nf_nat_sip_expect_hook); if (nf_nat_sip_expect && ct->status & IPS_NAT_MASK) - ret = nf_nat_sip_expect(skb, dptr, datalen, exp, + ret = nf_nat_sip_expect(skb, dataoff, dptr, datalen, exp, matchoff, matchlen); else { if (nf_ct_expect_related(exp) != 0) @@ -1117,7 +1122,7 @@ store_cseq: return ret; } -static int process_register_response(struct sk_buff *skb, +static int process_register_response(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq, unsigned int code) { @@ -1127,7 +1132,7 @@ static int process_register_response(struct sk_buff *skb, enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); union nf_inet_addr addr; __be16 port; - unsigned int matchoff, matchlen, dataoff = 0; + unsigned int matchoff, matchlen, coff = 0; unsigned int expires = 0; int in_contact = 0, ret; @@ -1154,7 +1159,7 @@ static int process_register_response(struct sk_buff *skb, while (1) { unsigned int c_expires = expires; - ret = ct_sip_parse_header_uri(ct, *dptr, &dataoff, *datalen, + ret = ct_sip_parse_header_uri(ct, *dptr, &coff, *datalen, SIP_HDR_CONTACT, &in_contact, &matchoff, &matchlen, &addr, &port); @@ -1193,13 +1198,13 @@ static const struct sip_handler sip_handlers[] = { SIP_HANDLER("REGISTER", process_register_request, process_register_response), }; -static int process_sip_response(struct sk_buff *skb, +static int process_sip_response(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - unsigned int matchoff, matchlen; - unsigned int code, cseq, dataoff, i; + unsigned int matchoff, matchlen, matchend; + unsigned int code, cseq, i; if (*datalen < strlen("SIP/2.0 200")) return NF_ACCEPT; @@ -1213,7 +1218,7 @@ static int process_sip_response(struct sk_buff *skb, cseq = simple_strtoul(*dptr + matchoff, NULL, 10); if (!cseq) return NF_DROP; - dataoff = matchoff + matchlen + 1; + matchend = matchoff + matchlen + 1; for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) { const struct sip_handler *handler; @@ -1221,15 +1226,16 @@ static int process_sip_response(struct sk_buff *skb, handler = &sip_handlers[i]; if (handler->response == NULL) continue; - if (*datalen < dataoff + handler->len || - strnicmp(*dptr + dataoff, handler->method, handler->len)) + if (*datalen < matchend + handler->len || + strnicmp(*dptr + matchend, handler->method, handler->len)) continue; - return handler->response(skb, dptr, datalen, cseq, code); + return handler->response(skb, dataoff, dptr, datalen, + cseq, code); } return NF_ACCEPT; } -static int process_sip_request(struct sk_buff *skb, +static int process_sip_request(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; @@ -1254,7 +1260,7 @@ static int process_sip_request(struct sk_buff *skb, if (!cseq) return NF_DROP; - return handler->request(skb, dptr, datalen, cseq); + return handler->request(skb, dataoff, dptr, datalen, cseq); } return NF_ACCEPT; } @@ -1288,13 +1294,13 @@ static int sip_help(struct sk_buff *skb, return NF_ACCEPT; if (strnicmp(dptr, "SIP/2.0 ", strlen("SIP/2.0 ")) != 0) - ret = process_sip_request(skb, &dptr, &datalen); + ret = process_sip_request(skb, dataoff, &dptr, &datalen); else - ret = process_sip_response(skb, &dptr, &datalen); + ret = process_sip_response(skb, dataoff, &dptr, &datalen); if (ret == NF_ACCEPT && ct->status & IPS_NAT_MASK) { nf_nat_sip = rcu_dereference(nf_nat_sip_hook); - if (nf_nat_sip && !nf_nat_sip(skb, &dptr, &datalen)) + if (nf_nat_sip && !nf_nat_sip(skb, dataoff, &dptr, &datalen)) ret = NF_DROP; } -- cgit v1.2.3-70-g09d2 From f5b321bd37fbec9188feb1f721ab46a5ac0b35da Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:26:19 +0100 Subject: netfilter: nf_conntrack_sip: add TCP support Add TCP support, which is mandated by RFC3261 for all SIP elements. SIP over TCP is similar to UDP, except that messages are delimited by Content-Length: headers and multiple messages may appear in one packet. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_sip.h | 3 +- net/ipv4/netfilter/nf_nat_sip.c | 2 +- net/netfilter/nf_conntrack_sip.c | 205 +++++++++++++++++++++++++---- 3 files changed, 179 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 2c6950b8bf7..fa9bb898145 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -84,7 +84,8 @@ enum sip_header_types { SIP_HDR_FROM, SIP_HDR_TO, SIP_HDR_CONTACT, - SIP_HDR_VIA, + SIP_HDR_VIA_UDP, + SIP_HDR_VIA_TCP, SIP_HDR_EXPIRES, SIP_HDR_CONTENT_LENGTH, }; diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 2454ea5abb7..b232e4040dc 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -122,7 +122,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, /* Translate topmost Via header and parameters */ if (ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, - SIP_HDR_VIA, NULL, &matchoff, &matchlen, + SIP_HDR_VIA_UDP, NULL, &matchoff, &matchlen, &addr, &port) > 0) { unsigned int matchend, poff, plen, buflen, n; char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 0ec37d6a2df..1cc75c5a822 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -287,7 +288,8 @@ static const struct sip_header ct_sip_hdrs[] = { [SIP_HDR_FROM] = SIP_HDR("From", "f", "sip:", skp_epaddr_len), [SIP_HDR_TO] = SIP_HDR("To", "t", "sip:", skp_epaddr_len), [SIP_HDR_CONTACT] = SIP_HDR("Contact", "m", "sip:", skp_epaddr_len), - [SIP_HDR_VIA] = SIP_HDR("Via", "v", "UDP ", epaddr_len), + [SIP_HDR_VIA_UDP] = SIP_HDR("Via", "v", "UDP ", epaddr_len), + [SIP_HDR_VIA_TCP] = SIP_HDR("Via", "v", "TCP ", epaddr_len), [SIP_HDR_EXPIRES] = SIP_HDR("Expires", NULL, NULL, digits_len), [SIP_HDR_CONTENT_LENGTH] = SIP_HDR("Content-Length", "l", NULL, digits_len), }; @@ -519,6 +521,33 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_parse_header_uri); +static int ct_sip_parse_param(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + const char *name, + unsigned int *matchoff, unsigned int *matchlen) +{ + const char *limit = dptr + datalen; + const char *start; + const char *end; + + limit = ct_sip_header_search(dptr + dataoff, limit, ",", strlen(",")); + if (!limit) + limit = dptr + datalen; + + start = ct_sip_header_search(dptr + dataoff, limit, name, strlen(name)); + if (!start) + return 0; + start += strlen(name); + + end = ct_sip_header_search(start, limit, ";", strlen(";")); + if (!end) + end = limit; + + *matchoff = start - dptr; + *matchlen = end - start; + return 1; +} + /* Parse address from header parameter and return address, offset and length */ int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, @@ -577,6 +606,29 @@ int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_parse_numerical_param); +static int ct_sip_parse_transport(struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + u8 *proto) +{ + unsigned int matchoff, matchlen; + + if (ct_sip_parse_param(ct, dptr, dataoff, datalen, "transport=", + &matchoff, &matchlen)) { + if (!strnicmp(dptr + matchoff, "TCP", strlen("TCP"))) + *proto = IPPROTO_TCP; + else if (!strnicmp(dptr + matchoff, "UDP", strlen("UDP"))) + *proto = IPPROTO_UDP; + else + return 0; + + if (*proto != nf_ct_protonum(ct)) + return 0; + } else + *proto = nf_ct_protonum(ct); + + return 1; +} + /* SDP header parsing: a SDP session description contains an ordered set of * headers, starting with a section containing general session parameters, * optionally followed by multiple media descriptions. @@ -685,7 +737,7 @@ static int ct_sip_parse_sdp_addr(const struct nf_conn *ct, const char *dptr, static int refresh_signalling_expectation(struct nf_conn *ct, union nf_inet_addr *addr, - __be16 port, + u8 proto, __be16 port, unsigned int expires) { struct nf_conn_help *help = nfct_help(ct); @@ -697,6 +749,7 @@ static int refresh_signalling_expectation(struct nf_conn *ct, hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) { if (exp->class != SIP_EXPECT_SIGNALLING || !nf_inet_addr_cmp(&exp->tuple.dst.u3, addr) || + exp->tuple.dst.protonum != proto || exp->tuple.dst.u.udp.port != port) continue; if (!del_timer(&exp->timeout)) @@ -1048,6 +1101,7 @@ static int process_register_request(struct sk_buff *skb, unsigned int dataoff, struct nf_conntrack_expect *exp; union nf_inet_addr *saddr, daddr; __be16 port; + u8 proto; unsigned int expires = 0; int ret; typeof(nf_nat_sip_expect_hook) nf_nat_sip_expect; @@ -1080,6 +1134,10 @@ static int process_register_request(struct sk_buff *skb, unsigned int dataoff, if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.src.u3, &daddr)) return NF_ACCEPT; + if (ct_sip_parse_transport(ct, *dptr, matchoff + matchlen, *datalen, + &proto) == 0) + return NF_ACCEPT; + if (ct_sip_parse_numerical_param(ct, *dptr, matchoff + matchlen, *datalen, "expires=", NULL, NULL, &expires) < 0) @@ -1099,7 +1157,7 @@ static int process_register_request(struct sk_buff *skb, unsigned int dataoff, saddr = &ct->tuplehash[!dir].tuple.src.u3; nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct), - saddr, &daddr, IPPROTO_UDP, NULL, &port); + saddr, &daddr, proto, NULL, &port); exp->timeout.expires = sip_timeout * HZ; exp->helper = nfct_help(ct)->helper; exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE; @@ -1132,6 +1190,7 @@ static int process_register_response(struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); union nf_inet_addr addr; __be16 port; + u8 proto; unsigned int matchoff, matchlen, coff = 0; unsigned int expires = 0; int in_contact = 0, ret; @@ -1172,6 +1231,10 @@ static int process_register_response(struct sk_buff *skb, unsigned int dataoff, if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.dst.u3, &addr)) continue; + if (ct_sip_parse_transport(ct, *dptr, matchoff + matchlen, + *datalen, &proto) == 0) + continue; + ret = ct_sip_parse_numerical_param(ct, *dptr, matchoff + matchlen, *datalen, "expires=", @@ -1180,7 +1243,8 @@ static int process_register_response(struct sk_buff *skb, unsigned int dataoff, return NF_DROP; if (c_expires == 0) break; - if (refresh_signalling_expectation(ct, &addr, port, c_expires)) + if (refresh_signalling_expectation(ct, &addr, proto, port, + c_expires)) return NF_ACCEPT; } @@ -1265,50 +1329,123 @@ static int process_sip_request(struct sk_buff *skb, unsigned int dataoff, return NF_ACCEPT; } -static int sip_help(struct sk_buff *skb, - unsigned int protoff, - struct nf_conn *ct, - enum ip_conntrack_info ctinfo) +static int process_sip_msg(struct sk_buff *skb, struct nf_conn *ct, + unsigned int dataoff, const char **dptr, + unsigned int *datalen) +{ + typeof(nf_nat_sip_hook) nf_nat_sip; + int ret; + + if (strnicmp(*dptr, "SIP/2.0 ", strlen("SIP/2.0 ")) != 0) + ret = process_sip_request(skb, dataoff, dptr, datalen); + else + ret = process_sip_response(skb, dataoff, dptr, datalen); + + if (ret == NF_ACCEPT && ct->status & IPS_NAT_MASK) { + nf_nat_sip = rcu_dereference(nf_nat_sip_hook); + if (nf_nat_sip && !nf_nat_sip(skb, dataoff, dptr, datalen)) + ret = NF_DROP; + } + + return ret; +} + +static int sip_help_tcp(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) { + struct tcphdr *th, _tcph; unsigned int dataoff, datalen; - const char *dptr; + unsigned int matchoff, matchlen, clen; + unsigned int msglen, origlen; + const char *dptr, *end; + s16 diff, tdiff = 0; int ret; - typeof(nf_nat_sip_hook) nf_nat_sip; + + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) + return NF_ACCEPT; /* No Data ? */ - dataoff = protoff + sizeof(struct udphdr); + th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return NF_ACCEPT; + dataoff = protoff + th->doff * 4; if (dataoff >= skb->len) return NF_ACCEPT; nf_ct_refresh(ct, skb, sip_timeout * HZ); - if (!skb_is_nonlinear(skb)) - dptr = skb->data + dataoff; - else { + if (skb_is_nonlinear(skb)) { pr_debug("Copy of skbuff not supported yet.\n"); return NF_ACCEPT; } + dptr = skb->data + dataoff; datalen = skb->len - dataoff; if (datalen < strlen("SIP/2.0 200")) return NF_ACCEPT; - if (strnicmp(dptr, "SIP/2.0 ", strlen("SIP/2.0 ")) != 0) - ret = process_sip_request(skb, dataoff, &dptr, &datalen); - else - ret = process_sip_response(skb, dataoff, &dptr, &datalen); + while (1) { + if (ct_sip_get_header(ct, dptr, 0, datalen, + SIP_HDR_CONTENT_LENGTH, + &matchoff, &matchlen) <= 0) + break; - if (ret == NF_ACCEPT && ct->status & IPS_NAT_MASK) { - nf_nat_sip = rcu_dereference(nf_nat_sip_hook); - if (nf_nat_sip && !nf_nat_sip(skb, dataoff, &dptr, &datalen)) - ret = NF_DROP; + clen = simple_strtoul(dptr + matchoff, (char **)&end, 10); + if (dptr + matchoff == end) + break; + + if (end + strlen("\r\n\r\n") > dptr + datalen) + break; + if (end[0] != '\r' || end[1] != '\n' || + end[2] != '\r' || end[3] != '\n') + break; + end += strlen("\r\n\r\n") + clen; + + msglen = origlen = end - dptr; + + ret = process_sip_msg(skb, ct, dataoff, &dptr, &msglen); + if (ret != NF_ACCEPT) + break; + diff = msglen - origlen; + tdiff += diff; + + dataoff += msglen; + dptr += msglen; + datalen = datalen + diff - msglen; } return ret; } -static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; -static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; +static int sip_help_udp(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + unsigned int dataoff, datalen; + const char *dptr; + + /* No Data ? */ + dataoff = protoff + sizeof(struct udphdr); + if (dataoff >= skb->len) + return NF_ACCEPT; + + nf_ct_refresh(ct, skb, sip_timeout * HZ); + + if (skb_is_nonlinear(skb)) { + pr_debug("Copy of skbuff not supported yet.\n"); + return NF_ACCEPT; + } + + dptr = skb->data + dataoff; + datalen = skb->len - dataoff; + if (datalen < strlen("SIP/2.0 200")) + return NF_ACCEPT; + + return process_sip_msg(skb, ct, dataoff, &dptr, &datalen); +} + +static struct nf_conntrack_helper sip[MAX_PORTS][4] __read_mostly; +static char sip_names[MAX_PORTS][4][sizeof("sip-65535")] __read_mostly; static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1] = { [SIP_EXPECT_SIGNALLING] = { @@ -1333,7 +1470,7 @@ static void nf_conntrack_sip_fini(void) int i, j; for (i = 0; i < ports_c; i++) { - for (j = 0; j < 2; j++) { + for (j = 0; j < ARRAY_SIZE(sip[i]); j++) { if (sip[i][j].me == NULL) continue; nf_conntrack_helper_unregister(&sip[i][j]); @@ -1353,14 +1490,24 @@ static int __init nf_conntrack_sip_init(void) memset(&sip[i], 0, sizeof(sip[i])); sip[i][0].tuple.src.l3num = AF_INET; - sip[i][1].tuple.src.l3num = AF_INET6; - for (j = 0; j < 2; j++) { - sip[i][j].tuple.dst.protonum = IPPROTO_UDP; + sip[i][0].tuple.dst.protonum = IPPROTO_UDP; + sip[i][0].help = sip_help_udp; + sip[i][1].tuple.src.l3num = AF_INET; + sip[i][1].tuple.dst.protonum = IPPROTO_TCP; + sip[i][1].help = sip_help_tcp; + + sip[i][2].tuple.src.l3num = AF_INET6; + sip[i][2].tuple.dst.protonum = IPPROTO_UDP; + sip[i][2].help = sip_help_udp; + sip[i][3].tuple.src.l3num = AF_INET6; + sip[i][3].tuple.dst.protonum = IPPROTO_TCP; + sip[i][3].help = sip_help_tcp; + + for (j = 0; j < ARRAY_SIZE(sip[i]); j++) { sip[i][j].tuple.src.u.udp.port = htons(ports[i]); sip[i][j].expect_policy = sip_exp_policy; sip[i][j].expect_class_max = SIP_EXPECT_MAX; sip[i][j].me = THIS_MODULE; - sip[i][j].help = sip_help; tmpname = &sip_names[i][j][0]; if (ports[i] == SIP_PORT) -- cgit v1.2.3-70-g09d2 From 010c0b9f34a4c567b431f8b49a58b7332ed42e47 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:27:09 +0100 Subject: netfilter: nf_nat: support mangling a single TCP packet multiple times nf_nat_mangle_tcp_packet() can currently only handle a single mangling per window because it only maintains two sequence adjustment positions: the one before the last adjustment and the one after. This patch makes sequence number adjustment tracking in nf_nat_mangle_tcp_packet() optional and allows a helper to manually update the offsets after the packet has been fully handled. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_helper.h | 32 +++++++++++++++++++++------- net/ipv4/netfilter/nf_nat_helper.c | 39 +++++++++++++++++++++-------------- 2 files changed, 48 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h index 4222220920a..02bb6c29dc3 100644 --- a/include/net/netfilter/nf_nat_helper.h +++ b/include/net/netfilter/nf_nat_helper.h @@ -7,13 +7,27 @@ struct sk_buff; /* These return true or false. */ -extern int nf_nat_mangle_tcp_packet(struct sk_buff *skb, - struct nf_conn *ct, - enum ip_conntrack_info ctinfo, - unsigned int match_offset, - unsigned int match_len, - const char *rep_buffer, - unsigned int rep_len); +extern int __nf_nat_mangle_tcp_packet(struct sk_buff *skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len, bool adjust); + +static inline int nf_nat_mangle_tcp_packet(struct sk_buff *skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + return __nf_nat_mangle_tcp_packet(skb, ct, ctinfo, + match_offset, match_len, + rep_buffer, rep_len, true); +} + extern int nf_nat_mangle_udp_packet(struct sk_buff *skb, struct nf_conn *ct, enum ip_conntrack_info ctinfo, @@ -21,6 +35,10 @@ extern int nf_nat_mangle_udp_packet(struct sk_buff *skb, unsigned int match_len, const char *rep_buffer, unsigned int rep_len); + +extern void nf_nat_set_seq_adjust(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + __be32 seq, s16 off); extern int nf_nat_seq_adjust(struct sk_buff *skb, struct nf_conn *ct, enum ip_conntrack_info ctinfo); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c index 7f10a6be019..4b6af4bb1f5 100644 --- a/net/ipv4/netfilter/nf_nat_helper.c +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -141,6 +141,17 @@ static int enlarge_skb(struct sk_buff *skb, unsigned int extra) return 1; } +void nf_nat_set_seq_adjust(struct nf_conn *ct, enum ip_conntrack_info ctinfo, + __be32 seq, s16 off) +{ + if (!off) + return; + set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); + adjust_tcp_sequence(ntohl(seq), off, ct, ctinfo); + nf_conntrack_event_cache(IPCT_NATSEQADJ, ct); +} +EXPORT_SYMBOL_GPL(nf_nat_set_seq_adjust); + /* Generic function for mangling variable-length address changes inside * NATed TCP connections (like the PORT XXX,XXX,XXX,XXX,XXX,XXX * command in FTP). @@ -149,14 +160,13 @@ static int enlarge_skb(struct sk_buff *skb, unsigned int extra) * skb enlargement, ... * * */ -int -nf_nat_mangle_tcp_packet(struct sk_buff *skb, - struct nf_conn *ct, - enum ip_conntrack_info ctinfo, - unsigned int match_offset, - unsigned int match_len, - const char *rep_buffer, - unsigned int rep_len) +int __nf_nat_mangle_tcp_packet(struct sk_buff *skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len, bool adjust) { struct rtable *rt = skb_rtable(skb); struct iphdr *iph; @@ -202,16 +212,13 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb, inet_proto_csum_replace2(&tcph->check, skb, htons(oldlen), htons(datalen), 1); - if (rep_len != match_len) { - set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); - adjust_tcp_sequence(ntohl(tcph->seq), - (int)rep_len - (int)match_len, - ct, ctinfo); - nf_conntrack_event_cache(IPCT_NATSEQADJ, ct); - } + if (adjust && rep_len != match_len) + nf_nat_set_seq_adjust(ct, ctinfo, tcph->seq, + (int)rep_len - (int)match_len); + return 1; } -EXPORT_SYMBOL(nf_nat_mangle_tcp_packet); +EXPORT_SYMBOL(__nf_nat_mangle_tcp_packet); /* Generic function for mangling variable-length address changes inside * NATed UDP connections (like the CONNECT DATA XXXXX MESG XXXXX INDEX XXXXX -- cgit v1.2.3-70-g09d2 From 48f8ac26537c1b7b1a2422f5232f45d06c945348 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:29:38 +0100 Subject: netfilter: nf_nat_sip: add TCP support Add support for mangling TCP SIP packets. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_sip.h | 1 + net/ipv4/netfilter/nf_nat_sip.c | 53 ++++++++++++++++++++++++++---- net/netfilter/nf_conntrack_sip.c | 10 ++++++ 3 files changed, 58 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index fa9bb898145..cd84d6f44d1 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -104,6 +104,7 @@ extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen); +extern void (*nf_nat_sip_seq_adjust_hook)(struct sk_buff *skb, s16 off); extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index b232e4040dc..11b538deaae 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -1,4 +1,4 @@ -/* SIP extension for UDP NAT alteration. +/* SIP extension for NAT alteration. * * (C) 2005 by Christian Hentschel * based on RR's ip_nat_ftp.c and other modules. @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -36,10 +37,27 @@ static unsigned int mangle_packet(struct sk_buff *skb, unsigned int dataoff, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - - if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, matchoff, matchlen, - buffer, buflen)) - return 0; + struct tcphdr *th; + unsigned int baseoff; + + if (nf_ct_protonum(ct) == IPPROTO_TCP) { + th = (struct tcphdr *)(skb->data + ip_hdrlen(skb)); + baseoff = ip_hdrlen(skb) + th->doff * 4; + matchoff += dataoff - baseoff; + + if (!__nf_nat_mangle_tcp_packet(skb, ct, ctinfo, + matchoff, matchlen, + buffer, buflen, false)) + return 0; + } else { + baseoff = ip_hdrlen(skb) + sizeof(struct udphdr); + matchoff += dataoff - baseoff; + + if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, + matchoff, matchlen, + buffer, buflen)) + return 0; + } /* Reload data pointer and adjust datalen value */ *dptr = skb->data + dataoff; @@ -104,6 +122,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); unsigned int coff, matchoff, matchlen; + enum sip_header_types hdr; union nf_inet_addr addr; __be16 port; int request, in_header; @@ -120,9 +139,14 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, } else request = 0; + if (nf_ct_protonum(ct) == IPPROTO_TCP) + hdr = SIP_HDR_VIA_TCP; + else + hdr = SIP_HDR_VIA_UDP; + /* Translate topmost Via header and parameters */ if (ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, - SIP_HDR_VIA_UDP, NULL, &matchoff, &matchlen, + hdr, NULL, &matchoff, &matchlen, &addr, &port) > 0) { unsigned int matchend, poff, plen, buflen, n; char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; @@ -204,9 +228,23 @@ next: if (!map_sip_addr(skb, dataoff, dptr, datalen, SIP_HDR_FROM) || !map_sip_addr(skb, dataoff, dptr, datalen, SIP_HDR_TO)) return NF_DROP; + return NF_ACCEPT; } +static void ip_nat_sip_seq_adjust(struct sk_buff *skb, s16 off) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + const struct tcphdr *th; + + if (nf_ct_protonum(ct) != IPPROTO_TCP || off == 0) + return; + + th = (struct tcphdr *)(skb->data + ip_hdrlen(skb)); + nf_nat_set_seq_adjust(ct, ctinfo, th->seq, off); +} + /* Handles expected signalling connections and media streams */ static void ip_nat_sip_expected(struct nf_conn *ct, struct nf_conntrack_expect *exp) @@ -472,6 +510,7 @@ err1: static void __exit nf_nat_sip_fini(void) { rcu_assign_pointer(nf_nat_sip_hook, NULL); + rcu_assign_pointer(nf_nat_sip_seq_adjust_hook, NULL); rcu_assign_pointer(nf_nat_sip_expect_hook, NULL); rcu_assign_pointer(nf_nat_sdp_addr_hook, NULL); rcu_assign_pointer(nf_nat_sdp_port_hook, NULL); @@ -483,12 +522,14 @@ static void __exit nf_nat_sip_fini(void) static int __init nf_nat_sip_init(void) { BUG_ON(nf_nat_sip_hook != NULL); + BUG_ON(nf_nat_sip_seq_adjust_hook != NULL); BUG_ON(nf_nat_sip_expect_hook != NULL); BUG_ON(nf_nat_sdp_addr_hook != NULL); BUG_ON(nf_nat_sdp_port_hook != NULL); BUG_ON(nf_nat_sdp_session_hook != NULL); BUG_ON(nf_nat_sdp_media_hook != NULL); rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); + rcu_assign_pointer(nf_nat_sip_seq_adjust_hook, ip_nat_sip_seq_adjust); rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect); rcu_assign_pointer(nf_nat_sdp_addr_hook, ip_nat_sdp_addr); rcu_assign_pointer(nf_nat_sdp_port_hook, ip_nat_sdp_port); diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 1cc75c5a822..3bb3aaff76e 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -56,6 +56,9 @@ unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, unsigned int dataoff, unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_hook); +void (*nf_nat_sip_seq_adjust_hook)(struct sk_buff *skb, s16 off) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sip_seq_adjust_hook); + unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, unsigned int dataoff, const char **dptr, @@ -1360,6 +1363,7 @@ static int sip_help_tcp(struct sk_buff *skb, unsigned int protoff, const char *dptr, *end; s16 diff, tdiff = 0; int ret; + typeof(nf_nat_sip_seq_adjust_hook) nf_nat_sip_seq_adjust; if (ctinfo != IP_CT_ESTABLISHED && ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) @@ -1415,6 +1419,12 @@ static int sip_help_tcp(struct sk_buff *skb, unsigned int protoff, datalen = datalen + diff - msglen; } + if (ret == NF_ACCEPT && ct->status & IPS_NAT_MASK) { + nf_nat_sip_seq_adjust = rcu_dereference(nf_nat_sip_seq_adjust_hook); + if (nf_nat_sip_seq_adjust) + nf_nat_sip_seq_adjust(skb, tdiff); + } + return ret; } -- cgit v1.2.3-70-g09d2 From 9d288dffe3a276e1f06ba556845c456d696c5a4f Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 Feb 2010 12:30:21 +0100 Subject: netfilter: nf_conntrack_sip: add T.38 FAX support Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_sip.h | 1 + include/net/netfilter/nf_conntrack.h | 2 +- net/netfilter/nf_conntrack_sip.c | 28 +++++++++++++++++++++++----- 3 files changed, 25 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index cd84d6f44d1..ff8cfbcf3b8 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -14,6 +14,7 @@ enum sip_expectation_classes { SIP_EXPECT_SIGNALLING, SIP_EXPECT_AUDIO, SIP_EXPECT_VIDEO, + SIP_EXPECT_IMAGE, __SIP_EXPECT_MAX }; #define SIP_EXPECT_MAX (__SIP_EXPECT_MAX - 1) diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 5043d61c99a..5b7d8835523 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -70,7 +70,7 @@ union nf_conntrack_help { struct nf_conntrack_helper; /* Must be kept in sync with the classes defined by helpers */ -#define NF_CT_MAX_EXPECT_CLASSES 3 +#define NF_CT_MAX_EXPECT_CLASSES 4 /* nf_conn feature for connections that have a helper */ struct nf_conn_help { diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 3bb3aaff76e..fbe8ff5a420 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -907,6 +907,7 @@ err1: static const struct sdp_media_type sdp_media_types[] = { SDP_MEDIA_TYPE("audio ", SIP_EXPECT_AUDIO), SDP_MEDIA_TYPE("video ", SIP_EXPECT_VIDEO), + SDP_MEDIA_TYPE("image ", SIP_EXPECT_IMAGE), }; static const struct sdp_media_type *sdp_media_type(const char *dptr, @@ -932,7 +933,6 @@ static int process_sdp(struct sk_buff *skb, unsigned int dataoff, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - struct nf_conn_help *help = nfct_help(ct); unsigned int matchoff, matchlen; unsigned int mediaoff, medialen; unsigned int sdpoff; @@ -1024,9 +1024,6 @@ static int process_sdp(struct sk_buff *skb, unsigned int dataoff, ret = nf_nat_sdp_session(skb, dataoff, dptr, datalen, sdpoff, &rtp_addr); - if (ret == NF_ACCEPT && i > 0) - help->help.ct_sip_info.invite_cseq = cseq; - return ret; } static int process_invite_response(struct sk_buff *skb, unsigned int dataoff, @@ -1077,6 +1074,22 @@ static int process_prack_response(struct sk_buff *skb, unsigned int dataoff, return NF_ACCEPT; } +static int process_invite_request(struct sk_buff *skb, unsigned int dataoff, + const char **dptr, unsigned int *datalen, + unsigned int cseq) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); + unsigned int ret; + + flush_expectations(ct, true); + ret = process_sdp(skb, dataoff, dptr, datalen, cseq); + if (ret == NF_ACCEPT) + help->help.ct_sip_info.invite_cseq = cseq; + return ret; +} + static int process_bye_request(struct sk_buff *skb, unsigned int dataoff, const char **dptr, unsigned int *datalen, unsigned int cseq) @@ -1257,7 +1270,7 @@ flush: } static const struct sip_handler sip_handlers[] = { - SIP_HANDLER("INVITE", process_sdp, process_invite_response), + SIP_HANDLER("INVITE", process_invite_request, process_invite_response), SIP_HANDLER("UPDATE", process_sdp, process_update_response), SIP_HANDLER("ACK", process_sdp, NULL), SIP_HANDLER("PRACK", process_sdp, process_prack_response), @@ -1473,6 +1486,11 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1 .max_expected = 2 * IP_CT_DIR_MAX, .timeout = 3 * 60, }, + [SIP_EXPECT_IMAGE] = { + .name = "image", + .max_expected = IP_CT_DIR_MAX, + .timeout = 3 * 60, + }, }; static void nf_conntrack_sip_fini(void) -- cgit v1.2.3-70-g09d2 From 857b409a48bdc33e824dff2d730e271b964e78bd Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 12 Feb 2010 06:24:46 +0100 Subject: netfilter: nf_conntrack: elegantly simplify nf_ct_exp_net() Remove #ifdef at nf_ct_exp_net() by using nf_ct_net(). Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_expect.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 917e170fa75..4b47ec19ef3 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -56,11 +56,7 @@ struct nf_conntrack_expect { static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp) { -#ifdef CONFIG_NET_NS - return exp->master->ct_net; /* by definition */ -#else - return &init_net; -#endif + return nf_ct_net(exp->master); } struct nf_conntrack_expect_policy { -- cgit v1.2.3-70-g09d2 From 2bec5a369ee79576a3eea2c23863325089785a2c Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 8 Feb 2010 05:19:03 +0000 Subject: ipv6: fib: fix crash when changing large fib while dumping it When the fib size exceeds what can be dumped in a single skb, the dump is suspended and resumed once the last skb has been received by userspace. When the fib is changed while the dump is suspended, the walker might contain stale pointers, causing a crash when the dump is resumed. BUG: unable to handle kernel NULL pointer dereference at 0000000000000018 IP: [] fib6_walk_continue+0xbb/0x124 [ipv6] PGD 5347a067 PUD 65c7067 PMD 0 Oops: 0000 [#1] PREEMPT SMP ... RIP: 0010:[] [] fib6_walk_continue+0xbb/0x124 [ipv6] ... Call Trace: [] ? mutex_spin_on_owner+0x59/0x71 [] inet6_dump_fib+0x11b/0x1b9 [ipv6] [] netlink_dump+0x5b/0x19e [] ? consume_skb+0x28/0x2a [] netlink_recvmsg+0x1ab/0x2c6 [] ? netlink_unicast+0xfa/0x151 [] __sock_recvmsg+0x6d/0x79 [] sock_recvmsg+0xca/0xe3 [] ? autoremove_wake_function+0x0/0x38 [] ? radix_tree_lookup_slot+0xe/0x10 [] ? find_get_page+0x90/0xa5 [] ? filemap_fault+0x201/0x34f [] ? fget_light+0x2f/0xac [] ? verify_iovec+0x4f/0x94 [] sys_recvmsg+0x14d/0x223 Store the serial number when beginning to walk the fib and reload pointers when continuing to walk after a change occured. Similar to other dumping functions, this might cause unrelated entries to be missed when entries are deleted. Tested-by: Ben Greear Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 2 ++ net/ipv6/ip6_fib.c | 29 +++++++++++++++++++++++++++-- 2 files changed, 29 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 257808188ad..8f279ddb359 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -129,6 +129,8 @@ struct fib6_walker_t { struct rt6_info *leaf; unsigned char state; unsigned char prune; + unsigned int skip; + unsigned int count; int (*func)(struct fib6_walker_t *); void *args; }; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index f626ea2b304..77e122f53ea 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -319,12 +319,26 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb, w->root = &table->tb6_root; if (cb->args[4] == 0) { + w->count = 0; + w->skip = 0; + read_lock_bh(&table->tb6_lock); res = fib6_walk(w); read_unlock_bh(&table->tb6_lock); - if (res > 0) + if (res > 0) { cb->args[4] = 1; + cb->args[5] = w->root->fn_sernum; + } } else { + if (cb->args[5] != w->root->fn_sernum) { + /* Begin at the root if the tree changed */ + cb->args[5] = w->root->fn_sernum; + w->state = FWS_INIT; + w->node = w->root; + w->skip = w->count; + } else + w->skip = 0; + read_lock_bh(&table->tb6_lock); res = fib6_walk_continue(w); read_unlock_bh(&table->tb6_lock); @@ -1250,9 +1264,18 @@ static int fib6_walk_continue(struct fib6_walker_t *w) w->leaf = fn->leaf; case FWS_C: if (w->leaf && fn->fn_flags&RTN_RTINFO) { - int err = w->func(w); + int err; + + if (w->count < w->skip) { + w->count++; + continue; + } + + err = w->func(w); if (err) return err; + + w->count++; continue; } w->state = FWS_U; @@ -1346,6 +1369,8 @@ static void fib6_clean_tree(struct net *net, struct fib6_node *root, c.w.root = root; c.w.func = fib6_clean_node; c.w.prune = prune; + c.w.count = 0; + c.w.skip = 0; c.func = func; c.arg = arg; c.net = net; -- cgit v1.2.3-70-g09d2 From 571ba42303813106d533bf6bda929d8e289f51bf Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 9 Feb 2010 11:49:47 +0000 Subject: netdevice.h: Add netdev_printk helpers like dev_printk These netdev_printk routines take a struct net_device * and emit dev_printk logging messages adding "%s: " ... netdev->dev.parent to the dev_printk format and arguments. This can create some uniformity in the output message log. These helpers should not be used until a successful alloc_netdev. Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/linux/netdevice.h | 71 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 71 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index cdf53a8d9ff..a51228ac933 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2095,6 +2095,77 @@ static inline u32 dev_ethtool_get_flags(struct net_device *dev) return 0; return dev->ethtool_ops->get_flags(dev); } + +/* Logging, debugging and troubleshooting/diagnostic helpers. */ + +/* netdev_printk helpers, similar to dev_printk */ + +static inline const char *netdev_name(const struct net_device *dev) +{ + if (dev->reg_state != NETREG_REGISTERED) + return "(unregistered net_device)"; + return dev->name; +} + +#define netdev_printk(level, netdev, format, args...) \ + dev_printk(level, (netdev)->dev.parent, \ + "%s: " format, \ + netdev_name(netdev), ##args) + +#define netdev_emerg(dev, format, args...) \ + netdev_printk(KERN_EMERG, dev, format, ##args) +#define netdev_alert(dev, format, args...) \ + netdev_printk(KERN_ALERT, dev, format, ##args) +#define netdev_crit(dev, format, args...) \ + netdev_printk(KERN_CRIT, dev, format, ##args) +#define netdev_err(dev, format, args...) \ + netdev_printk(KERN_ERR, dev, format, ##args) +#define netdev_warn(dev, format, args...) \ + netdev_printk(KERN_WARNING, dev, format, ##args) +#define netdev_notice(dev, format, args...) \ + netdev_printk(KERN_NOTICE, dev, format, ##args) +#define netdev_info(dev, format, args...) \ + netdev_printk(KERN_INFO, dev, format, ##args) + +#if defined(DEBUG) +#define netdev_dbg(__dev, format, args...) \ + netdev_printk(KERN_DEBUG, __dev, format, ##args) +#elif defined(CONFIG_DYNAMIC_DEBUG) +#define netdev_dbg(__dev, format, args...) \ +do { \ + dynamic_dev_dbg((__dev)->dev.parent, "%s: " format, \ + netdev_name(__dev), ##args); \ +} while (0) +#else +#define netdev_dbg(__dev, format, args...) \ +({ \ + if (0) \ + netdev_printk(KERN_DEBUG, __dev, format, ##args); \ + 0; \ +}) +#endif + +#if defined(VERBOSE_DEBUG) +#define netdev_vdbg netdev_dbg +#else + +#define netdev_vdbg(dev, format, args...) \ +({ \ + if (0) \ + netdev_printk(KERN_DEBUG, dev, format, ##args); \ + 0; \ +}) +#endif + +/* + * netdev_WARN() acts like dev_printk(), but with the key difference + * of using a WARN/WARN_ON to get the message out, including the + * file/line information and a backtrace. + */ +#define netdev_WARN(dev, format, args...) \ + WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args); + + #endif /* __KERNEL__ */ #endif /* _LINUX_NETDEVICE_H */ -- cgit v1.2.3-70-g09d2 From b3d95c5c93d4b57eaea0ad3f582b08a6b5fb3eb1 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 9 Feb 2010 11:49:49 +0000 Subject: include/linux/netdevice.h: Add netif_printk helpers Add macros to test a private structure for msg_enable bits and the netif_msg_##bit to test and call netdev_printk if set Simplifies logic in callers and adds message logging consistency Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/linux/netdevice.h | 53 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a51228ac933..1412dde16d0 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2165,6 +2165,59 @@ do { \ #define netdev_WARN(dev, format, args...) \ WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args); +/* netif printk helpers, similar to netdev_printk */ + +#define netif_printk(priv, type, level, dev, fmt, args...) \ +do { \ + if (netif_msg_##type(priv)) \ + netdev_printk(level, (dev), fmt, ##args); \ +} while (0) + +#define netif_emerg(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_EMERG, dev, fmt, ##args) +#define netif_alert(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_ALERT, dev, fmt, ##args) +#define netif_crit(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_CRIT, dev, fmt, ##args) +#define netif_err(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_ERR, dev, fmt, ##args) +#define netif_warn(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_WARNING, dev, fmt, ##args) +#define netif_notice(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_NOTICE, dev, fmt, ##args) +#define netif_info(priv, type, dev, fmt, args...) \ + netif_printk(priv, type, KERN_INFO, (dev), fmt, ##args) + +#if defined(DEBUG) +#define netif_dbg(priv, type, dev, format, args...) \ + netif_printk(priv, type, KERN_DEBUG, dev, format, ##args) +#elif defined(CONFIG_DYNAMIC_DEBUG) +#define netif_dbg(priv, type, netdev, format, args...) \ +do { \ + if (netif_msg_##type(priv)) \ + dynamic_dev_dbg((netdev)->dev.parent, \ + "%s: " format, \ + netdev_name(netdev), ##args); \ +} while (0) +#else +#define netif_dbg(priv, type, dev, format, args...) \ +({ \ + if (0) \ + netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \ + 0; \ +}) +#endif + +#if defined(VERBOSE_DEBUG) +#define netif_vdbg netdev_dbg +#else +#define netif_vdbg(priv, type, dev, format, args...) \ +({ \ + if (0) \ + netif_printk(KERN_DEBUG, dev, format, ##args); \ + 0; \ +}) +#endif #endif /* __KERNEL__ */ -- cgit v1.2.3-70-g09d2 From a63374631e7192a64648ecc6672619a7abf9ebcd Mon Sep 17 00:00:00 2001 From: jamal Date: Tue, 9 Feb 2010 13:21:17 +0000 Subject: xfrm: use proper kernel types kernel side should use uxx instead of __uxx types Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/xfrm.h | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index fcee547ca7e..0beb413c01c 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -317,8 +317,8 @@ extern void xfrm_state_delete_tunnel(struct xfrm_state *x); struct xfrm_type { char *description; struct module *owner; - __u8 proto; - __u8 flags; + u8 proto; + u8 flags; #define XFRM_TYPE_NON_FRAGMENT 1 #define XFRM_TYPE_REPLAY_PROT 2 #define XFRM_TYPE_LOCAL_COADDR 4 @@ -434,24 +434,24 @@ struct xfrm_tmpl { unsigned short encap_family; - __u32 reqid; + u32 reqid; /* Mode: transport, tunnel etc. */ - __u8 mode; + u8 mode; /* Sharing mode: unique, this session only, this user only etc. */ - __u8 share; + u8 share; /* May skip this transfomration if no SA is found */ - __u8 optional; + u8 optional; /* Skip aalgos/ealgos/calgos checks. */ - __u8 allalgs; + u8 allalgs; /* Bit mask of algos allowed for acquisition */ - __u32 aalgos; - __u32 ealgos; - __u32 calgos; + u32 aalgos; + u32 ealgos; + u32 calgos; }; #define XFRM_MAX_DEPTH 6 @@ -770,7 +770,7 @@ static __inline__ int addr_match(void *token1, void *token2, int prefixlen) int pdw; int pbi; - pdw = prefixlen >> 5; /* num of whole __u32 in prefix */ + pdw = prefixlen >> 5; /* num of whole u32 in prefix */ pbi = prefixlen & 0x1f; /* num of bits in incomplete u32 in prefix */ if (pdw) @@ -1259,7 +1259,7 @@ struct xfrm_algo_desc { /* XFRM tunnel handlers. */ struct xfrm_tunnel { int (*handler)(struct sk_buff *skb); - int (*err_handler)(struct sk_buff *skb, __u32 info); + int (*err_handler)(struct sk_buff *skb, u32 info); struct xfrm_tunnel *next; int priority; @@ -1500,7 +1500,7 @@ static inline int xfrm_addr_cmp(xfrm_address_t *a, xfrm_address_t *b, switch (family) { default: case AF_INET: - return (__force __u32)a->a4 - (__force __u32)b->a4; + return (__force u32)a->a4 - (__force u32)b->a4; case AF_INET6: return ipv6_addr_cmp((struct in6_addr *)a, (struct in6_addr *)b); -- cgit v1.2.3-70-g09d2 From fb8a0d9d1bfd1e4355f307e86a6da7209eefd5f3 Mon Sep 17 00:00:00 2001 From: "Williams, Mitch A" Date: Wed, 10 Feb 2010 01:43:04 +0000 Subject: pci: Add SR-IOV convenience functions and macros Add and export pci_num_vf to allow other subsystems to determine how many virtual function devices are associated with an SR-IOV physical function device. Add macros dev_is_pci, dev_is_ps, and dev_num_vf to make it easier for non-PCI specific code to determine SR-IOV capabilities. Signed-off-by: Mitch Williams Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- drivers/pci/iov.c | 15 +++++++++++++++ include/linux/pci.h | 11 +++++++++++ 2 files changed, 26 insertions(+) (limited to 'include') diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index b2a448e19fe..3e5ab2bf6a5 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -706,6 +706,21 @@ irqreturn_t pci_sriov_migration(struct pci_dev *dev) } EXPORT_SYMBOL_GPL(pci_sriov_migration); +/** + * pci_num_vf - return number of VFs associated with a PF device_release_driver + * @dev: the PCI device + * + * Returns number of VFs, or 0 if SR-IOV is not enabled. + */ +int pci_num_vf(struct pci_dev *dev) +{ + if (!dev || !dev->is_physfn) + return 0; + else + return dev->sriov->nr_virtfn; +} +EXPORT_SYMBOL_GPL(pci_num_vf); + static int ats_alloc_one(struct pci_dev *dev, int ps) { int pos; diff --git a/include/linux/pci.h b/include/linux/pci.h index 174e5392e51..59a98e2ee2c 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -612,6 +612,9 @@ extern void pci_remove_bus_device(struct pci_dev *dev); extern void pci_stop_bus_device(struct pci_dev *dev); void pci_setup_cardbus(struct pci_bus *bus); extern void pci_sort_breadthfirst(void); +#define dev_is_pci(d) ((d)->bus == &pci_bus_type) +#define dev_is_pf(d) ((dev_is_pci(d) ? to_pci_dev(d)->is_physfn : false)) +#define dev_num_vf(d) ((dev_is_pci(d) ? pci_num_vf(to_pci_dev(d)) : 0)) /* Generic PCI functions exported to card drivers */ @@ -1129,6 +1132,9 @@ static inline struct pci_dev *pci_get_bus_and_slot(unsigned int bus, unsigned int devfn) { return NULL; } +#define dev_is_pci(d) (false) +#define dev_is_pf(d) (false) +#define dev_num_vf(d) (0) #endif /* CONFIG_PCI */ /* Include architecture-dependent settings and functions */ @@ -1286,6 +1292,7 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar); extern int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn); extern void pci_disable_sriov(struct pci_dev *dev); extern irqreturn_t pci_sriov_migration(struct pci_dev *dev); +extern int pci_num_vf(struct pci_dev *dev); #else static inline int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn) { @@ -1298,6 +1305,10 @@ static inline irqreturn_t pci_sriov_migration(struct pci_dev *dev) { return IRQ_NONE; } +static inline int pci_num_vf(struct pci_dev *dev) +{ + return 0; +} #endif #if defined(CONFIG_HOTPLUG_PCI) || defined(CONFIG_HOTPLUG_PCI_MODULE) -- cgit v1.2.3-70-g09d2 From b280da8d54b8d82b52f368a8703b7ada6c1744d5 Mon Sep 17 00:00:00 2001 From: "Williams, Mitch A" Date: Wed, 10 Feb 2010 01:43:24 +0000 Subject: if_link: Add SR-IOV configuration methods Add SR-IOV VF management methods to IFLA_LINKINFO. This allows userspace to use rtnetlink to configure VF network devices. Signed-off-by: Mitch Williams Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- include/linux/if_link.h | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) (limited to 'include') diff --git a/include/linux/if_link.h b/include/linux/if_link.h index 6674791622c..c9bf92cd765 100644 --- a/include/linux/if_link.h +++ b/include/linux/if_link.h @@ -78,6 +78,11 @@ enum { #define IFLA_LINKINFO IFLA_LINKINFO IFLA_NET_NS_PID, IFLA_IFALIAS, + IFLA_NUM_VF, /* Number of VFs if device is SR-IOV PF */ + IFLA_VF_MAC, /* Hardware queue specific attributes */ + IFLA_VF_VLAN, + IFLA_VF_TX_RATE, /* TX Bandwidth Allocation */ + IFLA_VFINFO, __IFLA_MAX }; @@ -196,4 +201,29 @@ enum macvlan_mode { MACVLAN_MODE_BRIDGE = 4, /* talk to bridge ports directly */ }; +/* SR-IOV virtual function managment section */ + +struct ifla_vf_mac { + __u32 vf; + __u8 mac[32]; /* MAX_ADDR_LEN */ +}; + +struct ifla_vf_vlan { + __u32 vf; + __u32 vlan; /* 0 - 4095, 0 disables VLAN filter */ + __u32 qos; +}; + +struct ifla_vf_tx_rate { + __u32 vf; + __u32 rate; /* Max TX bandwidth in Mbps, 0 disables throttling */ +}; + +struct ifla_vf_info { + __u32 vf; + __u8 mac[32]; + __u32 vlan; + __u32 qos; + __u32 tx_rate; +}; #endif /* _LINUX_IF_LINK_H */ -- cgit v1.2.3-70-g09d2 From 95c26df829ba4a25f40595cadb4c9433883cbe28 Mon Sep 17 00:00:00 2001 From: "Williams, Mitch A" Date: Wed, 10 Feb 2010 01:43:46 +0000 Subject: net: Add netdev ops for SR-IOV configuration Add netdev ops for configuring SR-IOV VF devices through the PF driver. Signed-off-by: Mitch Williams Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- include/linux/netdevice.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1412dde16d0..682d02521bb 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -28,6 +28,7 @@ #include #include #include +#include #ifdef __KERNEL__ #include @@ -621,6 +622,13 @@ struct netdev_queue { * this function is called when a VLAN id is unregistered. * * void (*ndo_poll_controller)(struct net_device *dev); + * + * SR-IOV management functions. + * int (*ndo_set_vf_mac)(struct net_device *dev, int vf, u8* mac); + * int (*ndo_set_vf_vlan)(struct net_device *dev, int vf, u16 vlan, u8 qos); + * int (*ndo_set_vf_tx_rate)(struct net_device *dev, int vf, int rate); + * int (*ndo_get_vf_config)(struct net_device *dev, + * int vf, struct ifla_vf_info *ivf); */ #define HAVE_NET_DEVICE_OPS struct net_device_ops { @@ -660,6 +668,15 @@ struct net_device_ops { #ifdef CONFIG_NET_POLL_CONTROLLER void (*ndo_poll_controller)(struct net_device *dev); #endif + int (*ndo_set_vf_mac)(struct net_device *dev, + int queue, u8 *mac); + int (*ndo_set_vf_vlan)(struct net_device *dev, + int queue, u16 vlan, u8 qos); + int (*ndo_set_vf_tx_rate)(struct net_device *dev, + int vf, int rate); + int (*ndo_get_vf_config)(struct net_device *dev, + int vf, + struct ifla_vf_info *ivf); #if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE) int (*ndo_fcoe_enable)(struct net_device *dev); int (*ndo_fcoe_disable)(struct net_device *dev); -- cgit v1.2.3-70-g09d2 From 1a5778aa000ebfec7f07eed0ffa2852ffb5d16bb Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 14 Feb 2010 22:35:47 -0800 Subject: net: Fix first line of kernel-doc for a few functions The function name must be followed by a space, hypen, space, and a short description. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- include/net/sock.h | 2 +- net/sunrpc/rpc_pipe.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ae836fded53..ba0f8e3a9cd 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -738,7 +738,7 @@ static inline struct sk_buff *skb_unshare(struct sk_buff *skb, } /** - * skb_peek + * skb_peek - peek at the head of an &sk_buff_head * @list_: list to peek at * * Peek an &sk_buff. Unlike most other operations you _MUST_ @@ -759,7 +759,7 @@ static inline struct sk_buff *skb_peek(struct sk_buff_head *list_) } /** - * skb_peek_tail + * skb_peek_tail - peek at the tail of an &sk_buff_head * @list_: list to peek at * * Peek an &sk_buff. Unlike most other operations you _MUST_ diff --git a/include/net/sock.h b/include/net/sock.h index c8d400063c1..580d51fa28e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1049,7 +1049,7 @@ extern void sk_common_release(struct sock *sk); extern void sock_init_data(struct socket *sock, struct sock *sk); /** - * sk_filter_release: Release a socket filter + * sk_filter_release - release a socket filter * @fp: filter to remove * * Remove a filter from a socket and release its resources. diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 49278f83036..9ea45383480 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -78,7 +78,7 @@ rpc_timeout_upcall_queue(struct work_struct *work) } /** - * rpc_queue_upcall + * rpc_queue_upcall - queue an upcall message to userspace * @inode: inode of upcall pipe on which to queue given message * @msg: message to queue * -- cgit v1.2.3-70-g09d2 From a1467085dcad8214bbb1d7edafbaa295cbd8c0e7 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 14 Feb 2010 22:38:54 -0800 Subject: ethtool: Fix includes build break Based upon a patch by Oliver Hartkopp . Signed-off-by: David S. Miller --- include/linux/ethtool.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index a3cac53a076..cca1c3de140 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -14,7 +14,6 @@ #define _LINUX_ETHTOOL_H #include -#include /* This should work for both 32 and 64 bit userland. */ struct ethtool_cmd { @@ -409,6 +408,8 @@ struct ethtool_flash { #ifdef __KERNEL__ +#include + struct ethtool_rx_ntuple_flow_spec_container { struct ethtool_rx_ntuple_flow_spec fs; struct list_head list; -- cgit v1.2.3-70-g09d2 From 23f3733d440b918ccb7746718f77256334cf6176 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Fri, 5 Jun 2009 17:31:46 +0200 Subject: netfilter: reduce NF_HOOK by one argument No changes in vmlinux filesize. Signed-off-by: Jan Engelhardt --- include/linux/netfilter.h | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 78f33d22368..2f22816a551 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -163,11 +163,8 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh, - int cond) + int (*okfn)(struct sk_buff *), int thresh) { - if (!cond) - return 1; #ifndef CONFIG_NETFILTER_DEBUG if (list_empty(&nf_hooks[pf][hook])) return 1; @@ -179,7 +176,7 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *)) { - return nf_hook_thresh(pf, hook, skb, indev, outdev, okfn, INT_MIN, 1); + return nf_hook_thresh(pf, hook, skb, indev, outdev, okfn, INT_MIN); } /* Activate hook; either okfn or kfree_skb called, unless a hook @@ -206,13 +203,13 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, #define NF_HOOK_THRESH(pf, hook, skb, indev, outdev, okfn, thresh) \ ({int __ret; \ -if ((__ret=nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, thresh, 1)) == 1)\ +if ((__ret=nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, thresh)) == 1)\ __ret = (okfn)(skb); \ __ret;}) #define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) \ ({int __ret; \ -if ((__ret=nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, INT_MIN, cond)) == 1)\ +if ((cond) || (__ret = nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, INT_MIN)) == 1)\ __ret = (okfn)(skb); \ __ret;}) @@ -328,8 +325,7 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh, - int cond) + int (*okfn)(struct sk_buff *), int thresh) { return okfn(skb); } -- cgit v1.2.3-70-g09d2 From 2249065f4b22b493bae2caf549b86f175f33188e Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Sat, 13 Jun 2009 04:13:26 +0200 Subject: netfilter: get rid of the grossness in netfilter.h GCC is now smart enough to follow the inline trail correctly. vmlinux size remain the same. Signed-off-by: Jan Engelhardt --- include/linux/netfilter.h | 45 ++++++++++++++++++++++++++++----------------- 1 file changed, 28 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 2f22816a551..70079454ffd 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -196,25 +196,36 @@ static inline int nf_hook(u_int8_t pf, unsigned int hook, struct sk_buff *skb, coders :) */ -/* This is gross, but inline doesn't cut it for avoiding the function - call in fast path: gcc doesn't inline (needs value tracking?). --RR */ - -/* HX: It's slightly less gross now. */ - -#define NF_HOOK_THRESH(pf, hook, skb, indev, outdev, okfn, thresh) \ -({int __ret; \ -if ((__ret=nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, thresh)) == 1)\ - __ret = (okfn)(skb); \ -__ret;}) +static inline int +NF_HOOK_THRESH(uint8_t pf, unsigned int hook, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *), int thresh) +{ + int ret = nf_hook_thresh(pf, hook, skb, in, out, okfn, thresh); + if (ret == 1) + ret = okfn(skb); + return ret; +} -#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) \ -({int __ret; \ -if ((cond) || (__ret = nf_hook_thresh(pf, hook, (skb), indev, outdev, okfn, INT_MIN)) == 1)\ - __ret = (okfn)(skb); \ -__ret;}) +static inline int +NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *), bool cond) +{ + int ret = 1; + if (cond || + (ret = nf_hook_thresh(pf, hook, skb, in, out, okfn, INT_MIN) == 1)) + ret = okfn(skb); + return ret; +} -#define NF_HOOK(pf, hook, skb, indev, outdev, okfn) \ - NF_HOOK_THRESH(pf, hook, skb, indev, outdev, okfn, INT_MIN) +static inline int +NF_HOOK(uint8_t pf, unsigned int hook, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return NF_HOOK_THRESH(pf, hook, skb, in, out, okfn, INT_MIN); +} /* Call setsockopt() */ int nf_setsockopt(struct sock *sk, u_int8_t pf, int optval, char __user *opt, -- cgit v1.2.3-70-g09d2 From 739674fb7febf116e7d647031fab16989a08a965 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Fri, 26 Jun 2009 08:23:19 +0200 Subject: netfilter: xtables: constify args in compat copying functions Signed-off-by: Jan Engelhardt --- include/linux/netfilter/x_tables.h | 12 ++++++------ net/ipv4/netfilter/arp_tables.c | 4 ++-- net/ipv4/netfilter/ip_tables.c | 4 ++-- net/ipv4/netfilter/ipt_ULOG.c | 4 ++-- net/ipv6/netfilter/ip6_tables.c | 4 ++-- net/netfilter/x_tables.c | 8 ++++---- net/netfilter/xt_hashlimit.c | 4 ++-- net/netfilter/xt_limit.c | 4 ++-- 8 files changed, 22 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index a0a144a6c6d..c61758f4be3 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -286,8 +286,8 @@ struct xt_match { void (*destroy)(const struct xt_mtdtor_param *); #ifdef CONFIG_COMPAT /* Called when userspace align differs from kernel space one */ - void (*compat_from_user)(void *dst, void *src); - int (*compat_to_user)(void __user *dst, void *src); + void (*compat_from_user)(void *dst, const void *src); + int (*compat_to_user)(void __user *dst, const void *src); #endif /* Set this to THIS_MODULE if you are a module, otherwise NULL */ struct module *me; @@ -328,8 +328,8 @@ struct xt_target { void (*destroy)(const struct xt_tgdtor_param *); #ifdef CONFIG_COMPAT /* Called when userspace align differs from kernel space one */ - void (*compat_from_user)(void *dst, void *src); - int (*compat_to_user)(void __user *dst, void *src); + void (*compat_from_user)(void *dst, const void *src); + int (*compat_to_user)(void __user *dst, const void *src); #endif /* Set this to THIS_MODULE if you are a module, otherwise NULL */ struct module *me; @@ -593,13 +593,13 @@ extern short xt_compat_calc_jump(u_int8_t af, unsigned int offset); extern int xt_compat_match_offset(const struct xt_match *match); extern int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr, unsigned int *size); -extern int xt_compat_match_to_user(struct xt_entry_match *m, +extern int xt_compat_match_to_user(const struct xt_entry_match *m, void __user **dstptr, unsigned int *size); extern int xt_compat_target_offset(const struct xt_target *target); extern void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr, unsigned int *size); -extern int xt_compat_target_to_user(struct xt_entry_target *t, +extern int xt_compat_target_to_user(const struct xt_entry_target *t, void __user **dstptr, unsigned int *size); #endif /* CONFIG_COMPAT */ diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 72723ea1054..2303dc92a27 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -842,7 +842,7 @@ static int copy_entries_to_user(unsigned int total_size, } #ifdef CONFIG_COMPAT -static void compat_standard_from_user(void *dst, void *src) +static void compat_standard_from_user(void *dst, const void *src) { int v = *(compat_int_t *)src; @@ -851,7 +851,7 @@ static void compat_standard_from_user(void *dst, void *src) memcpy(dst, &v, sizeof(v)); } -static int compat_standard_to_user(void __user *dst, void *src) +static int compat_standard_to_user(void __user *dst, const void *src) { compat_int_t cv = *(int *)src; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 2057b1bb617..2a4f745ce36 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1047,7 +1047,7 @@ copy_entries_to_user(unsigned int total_size, } #ifdef CONFIG_COMPAT -static void compat_standard_from_user(void *dst, void *src) +static void compat_standard_from_user(void *dst, const void *src) { int v = *(compat_int_t *)src; @@ -1056,7 +1056,7 @@ static void compat_standard_from_user(void *dst, void *src) memcpy(dst, &v, sizeof(v)); } -static int compat_standard_to_user(void __user *dst, void *src) +static int compat_standard_to_user(void __user *dst, const void *src) { compat_int_t cv = *(int *)src; diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 399061c3fd7..09a5d3f7cc4 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -338,7 +338,7 @@ struct compat_ipt_ulog_info { char prefix[ULOG_PREFIX_LEN]; }; -static void ulog_tg_compat_from_user(void *dst, void *src) +static void ulog_tg_compat_from_user(void *dst, const void *src) { const struct compat_ipt_ulog_info *cl = src; struct ipt_ulog_info l = { @@ -351,7 +351,7 @@ static void ulog_tg_compat_from_user(void *dst, void *src) memcpy(dst, &l, sizeof(l)); } -static int ulog_tg_compat_to_user(void __user *dst, void *src) +static int ulog_tg_compat_to_user(void __user *dst, const void *src) { const struct ipt_ulog_info *l = src; struct compat_ipt_ulog_info cl = { diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index dcd7825fe7b..3ff4fd50e96 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1079,7 +1079,7 @@ copy_entries_to_user(unsigned int total_size, } #ifdef CONFIG_COMPAT -static void compat_standard_from_user(void *dst, void *src) +static void compat_standard_from_user(void *dst, const void *src) { int v = *(compat_int_t *)src; @@ -1088,7 +1088,7 @@ static void compat_standard_from_user(void *dst, void *src) memcpy(dst, &v, sizeof(v)); } -static int compat_standard_to_user(void __user *dst, void *src) +static int compat_standard_to_user(void __user *dst, const void *src) { compat_int_t cv = *(int *)src; diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 12503199826..69c56287d51 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -485,8 +485,8 @@ int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr, } EXPORT_SYMBOL_GPL(xt_compat_match_from_user); -int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr, - unsigned int *size) +int xt_compat_match_to_user(const struct xt_entry_match *m, + void __user **dstptr, unsigned int *size) { const struct xt_match *match = m->u.kernel.match; struct compat_xt_entry_match __user *cm = *dstptr; @@ -588,8 +588,8 @@ void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr, } EXPORT_SYMBOL_GPL(xt_compat_target_from_user); -int xt_compat_target_to_user(struct xt_entry_target *t, void __user **dstptr, - unsigned int *size) +int xt_compat_target_to_user(const struct xt_entry_target *t, + void __user **dstptr, unsigned int *size) { const struct xt_target *target = t->u.kernel.target; struct compat_xt_entry_target __user *ct = *dstptr; diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index 017c95966aa..e47fb805ffb 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -775,7 +775,7 @@ struct compat_xt_hashlimit_info { compat_uptr_t master; }; -static void hashlimit_mt_compat_from_user(void *dst, void *src) +static void hashlimit_mt_compat_from_user(void *dst, const void *src) { int off = offsetof(struct compat_xt_hashlimit_info, hinfo); @@ -783,7 +783,7 @@ static void hashlimit_mt_compat_from_user(void *dst, void *src) memset(dst + off, 0, sizeof(struct compat_xt_hashlimit_info) - off); } -static int hashlimit_mt_compat_to_user(void __user *dst, void *src) +static int hashlimit_mt_compat_to_user(void __user *dst, const void *src) { int off = offsetof(struct compat_xt_hashlimit_info, hinfo); diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c index 2773be6a71d..a0ca5339af4 100644 --- a/net/netfilter/xt_limit.c +++ b/net/netfilter/xt_limit.c @@ -148,7 +148,7 @@ struct compat_xt_rateinfo { /* To keep the full "prev" timestamp, the upper 32 bits are stored in the * master pointer, which does not need to be preserved. */ -static void limit_mt_compat_from_user(void *dst, void *src) +static void limit_mt_compat_from_user(void *dst, const void *src) { const struct compat_xt_rateinfo *cm = src; struct xt_rateinfo m = { @@ -162,7 +162,7 @@ static void limit_mt_compat_from_user(void *dst, void *src) memcpy(dst, &m, sizeof(m)); } -static int limit_mt_compat_to_user(void __user *dst, void *src) +static int limit_mt_compat_to_user(void __user *dst, const void *src) { const struct xt_rateinfo *m = src; struct compat_xt_rateinfo cm = { -- cgit v1.2.3-70-g09d2 From 8fea97ec1772bbf553d89187340ef624d548e115 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 15 Feb 2010 17:45:08 +0100 Subject: netfilter: nf_conntrack: pass template to l4proto ->error() handler The error handlers might need the template to get the conntrack zone introduced in the next patches to perform a conntrack lookup. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l4proto.h | 4 ++-- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 3 ++- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 3 ++- net/netfilter/nf_conntrack_core.c | 3 ++- net/netfilter/nf_conntrack_proto_dccp.c | 5 +++-- net/netfilter/nf_conntrack_proto_tcp.c | 2 +- net/netfilter/nf_conntrack_proto_udp.c | 4 ++-- net/netfilter/nf_conntrack_proto_udplite.c | 2 +- 8 files changed, 15 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index ca6dcf3445a..e3d3ee3c06a 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -49,8 +49,8 @@ struct nf_conntrack_l4proto { /* Called when a conntrack entry is destroyed */ void (*destroy)(struct nf_conn *ct); - int (*error)(struct net *net, struct sk_buff *skb, unsigned int dataoff, - enum ip_conntrack_info *ctinfo, + int (*error)(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, + unsigned int dataoff, enum ip_conntrack_info *ctinfo, u_int8_t pf, unsigned int hooknum); /* Print out the per-protocol part of the tuple. Return like seq_* */ diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 7afd39b5b78..327826a968a 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -163,7 +163,8 @@ icmp_error_message(struct net *net, struct sk_buff *skb, /* Small and modified version of icmp_rcv */ static int -icmp_error(struct net *net, struct sk_buff *skb, unsigned int dataoff, +icmp_error(struct net *net, struct nf_conn *tmpl, + struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, u_int8_t pf, unsigned int hooknum) { const struct icmphdr *icmph; diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index c7b8bd1d798..d772dc21857 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -179,7 +179,8 @@ icmpv6_error_message(struct net *net, } static int -icmpv6_error(struct net *net, struct sk_buff *skb, unsigned int dataoff, +icmpv6_error(struct net *net, struct nf_conn *tmpl, + struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, u_int8_t pf, unsigned int hooknum) { const struct icmp6hdr *icmp6h; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index bd831410a39..65351ed5d81 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -799,7 +799,8 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, * inverse of the return code tells to the netfilter * core what to do with the packet. */ if (l4proto->error != NULL) { - ret = l4proto->error(net, skb, dataoff, &ctinfo, pf, hooknum); + ret = l4proto->error(net, tmpl, skb, dataoff, &ctinfo, + pf, hooknum); if (ret <= 0) { NF_CT_STAT_INC_ATOMIC(net, error); NF_CT_STAT_INC_ATOMIC(net, invalid); diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index dd375500dcc..9a281554937 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -561,8 +561,9 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, return NF_ACCEPT; } -static int dccp_error(struct net *net, struct sk_buff *skb, - unsigned int dataoff, enum ip_conntrack_info *ctinfo, +static int dccp_error(struct net *net, struct nf_conn *tmpl, + struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, u_int8_t pf, unsigned int hooknum) { struct dccp_hdr _dh, *dh; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index ad118053971..9dd8cd4fb6e 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -760,7 +760,7 @@ static const u8 tcp_valid_flags[(TH_FIN|TH_SYN|TH_RST|TH_ACK|TH_URG) + 1] = }; /* Protect conntrack agaist broken packets. Code taken from ipt_unclean.c. */ -static int tcp_error(struct net *net, +static int tcp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index 8d38f9a4bed..8289088b821 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -91,8 +91,8 @@ static bool udp_new(struct nf_conn *ct, const struct sk_buff *skb, return true; } -static int udp_error(struct net *net, struct sk_buff *skb, unsigned int dataoff, - enum ip_conntrack_info *ctinfo, +static int udp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, + unsigned int dataoff, enum ip_conntrack_info *ctinfo, u_int8_t pf, unsigned int hooknum) { diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c index 0b1bc9ba667..263b5a72588 100644 --- a/net/netfilter/nf_conntrack_proto_udplite.c +++ b/net/netfilter/nf_conntrack_proto_udplite.c @@ -89,7 +89,7 @@ static bool udplite_new(struct nf_conn *ct, const struct sk_buff *skb, return true; } -static int udplite_error(struct net *net, +static int udplite_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, -- cgit v1.2.3-70-g09d2 From 5d0aa2ccd4699a01cfdf14886191c249d7b45a01 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 15 Feb 2010 18:13:33 +0100 Subject: netfilter: nf_conntrack: add support for "conntrack zones" Normally, each connection needs a unique identity. Conntrack zones allow to specify a numerical zone using the CT target, connections in different zones can use the same identity. Example: iptables -t raw -A PREROUTING -i veth0 -j CT --zone 1 iptables -t raw -A OUTPUT -o veth1 -j CT --zone 1 Signed-off-by: Patrick McHardy --- include/linux/netfilter/xt_CT.h | 2 +- include/net/ip.h | 3 + include/net/ipv6.h | 3 + include/net/netfilter/nf_conntrack.h | 5 +- include/net/netfilter/nf_conntrack_core.h | 3 +- include/net/netfilter/nf_conntrack_expect.h | 9 +- include/net/netfilter/nf_conntrack_extend.h | 2 + include/net/netfilter/nf_conntrack_zones.h | 23 ++++++ net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 3 +- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 8 +- net/ipv4/netfilter/nf_defrag_ipv4.c | 12 ++- net/ipv4/netfilter/nf_nat_core.c | 24 +++--- net/ipv4/netfilter/nf_nat_pptp.c | 3 +- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 12 ++- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 8 +- net/netfilter/Kconfig | 13 +++ net/netfilter/nf_conntrack_core.c | 109 +++++++++++++++++++------ net/netfilter/nf_conntrack_expect.c | 21 +++-- net/netfilter/nf_conntrack_h323_main.c | 3 +- net/netfilter/nf_conntrack_netlink.c | 20 ++--- net/netfilter/nf_conntrack_pptp.c | 14 ++-- net/netfilter/nf_conntrack_sip.c | 3 +- net/netfilter/nf_conntrack_standalone.c | 6 ++ net/netfilter/xt_CT.c | 8 +- net/netfilter/xt_connlimit.c | 4 +- 25 files changed, 236 insertions(+), 85 deletions(-) create mode 100644 include/net/netfilter/nf_conntrack_zones.h (limited to 'include') diff --git a/include/linux/netfilter/xt_CT.h b/include/linux/netfilter/xt_CT.h index 7fd0effe131..1b564106891 100644 --- a/include/linux/netfilter/xt_CT.h +++ b/include/linux/netfilter/xt_CT.h @@ -5,7 +5,7 @@ struct xt_ct_target_info { u_int16_t flags; - u_int16_t __unused; + u_int16_t zone; u_int32_t ct_events; u_int32_t exp_events; char helper[16]; diff --git a/include/net/ip.h b/include/net/ip.h index fb63371c07a..7bc47873e3f 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -352,8 +352,11 @@ enum ip_defrag_users { IP_DEFRAG_LOCAL_DELIVER, IP_DEFRAG_CALL_RA_CHAIN, IP_DEFRAG_CONNTRACK_IN, + __IP_DEFRAG_CONNTRACK_IN_END = IP_DEFRAG_CONNTRACK_IN + USHORT_MAX, IP_DEFRAG_CONNTRACK_OUT, + __IP_DEFRAG_CONNTRACK_OUT_END = IP_DEFRAG_CONNTRACK_OUT + USHORT_MAX, IP_DEFRAG_CONNTRACK_BRIDGE_IN, + __IP_DEFRAG_CONNTRACK_BRIDGE_IN = IP_DEFRAG_CONNTRACK_BRIDGE_IN + USHORT_MAX, IP_DEFRAG_VS_IN, IP_DEFRAG_VS_OUT, IP_DEFRAG_VS_FWD diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 299bbf5adfb..639ec53ea08 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -355,8 +355,11 @@ struct inet_frag_queue; enum ip6_defrag_users { IP6_DEFRAG_LOCAL_DELIVER, IP6_DEFRAG_CONNTRACK_IN, + __IP6_DEFRAG_CONNTRACK_IN = IP6_DEFRAG_CONNTRACK_IN + USHORT_MAX, IP6_DEFRAG_CONNTRACK_OUT, + __IP6_DEFRAG_CONNTRACK_OUT = IP6_DEFRAG_CONNTRACK_OUT + USHORT_MAX, IP6_DEFRAG_CONNTRACK_BRIDGE_IN, + __IP6_DEFRAG_CONNTRACK_BRIDGE_IN = IP6_DEFRAG_CONNTRACK_BRIDGE_IN + USHORT_MAX, }; struct ip6_create_arg { diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 5b7d8835523..bde095f7e84 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -198,7 +198,8 @@ extern void *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced, int null extern void nf_ct_free_hashtable(void *hash, int vmalloced, unsigned int size); extern struct nf_conntrack_tuple_hash * -__nf_conntrack_find(struct net *net, const struct nf_conntrack_tuple *tuple); +__nf_conntrack_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple); extern void nf_conntrack_hash_insert(struct nf_conn *ct); extern void nf_ct_delete_from_lists(struct nf_conn *ct); @@ -267,7 +268,7 @@ extern void nf_ct_iterate_cleanup(struct net *net, int (*iter)(struct nf_conn *i, void *data), void *data); extern void nf_conntrack_free(struct nf_conn *ct); extern struct nf_conn * -nf_conntrack_alloc(struct net *net, +nf_conntrack_alloc(struct net *net, u16 zone, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, gfp_t gfp); diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 5a449b44ba3..dffde8e6920 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -49,7 +49,8 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, /* Find a connection corresponding to a tuple. */ extern struct nf_conntrack_tuple_hash * -nf_conntrack_find_get(struct net *net, const struct nf_conntrack_tuple *tuple); +nf_conntrack_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple); extern int __nf_conntrack_confirm(struct sk_buff *skb); diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 4b47ec19ef3..11e815084fc 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -74,13 +74,16 @@ int nf_conntrack_expect_init(struct net *net); void nf_conntrack_expect_fini(struct net *net); struct nf_conntrack_expect * -__nf_ct_expect_find(struct net *net, const struct nf_conntrack_tuple *tuple); +__nf_ct_expect_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * -nf_ct_expect_find_get(struct net *net, const struct nf_conntrack_tuple *tuple); +nf_ct_expect_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * -nf_ct_find_expectation(struct net *net, const struct nf_conntrack_tuple *tuple); +nf_ct_find_expectation(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple); void nf_ct_unlink_expect(struct nf_conntrack_expect *exp); void nf_ct_remove_expectations(struct nf_conn *ct); diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h index e192dc17c58..2d2a1f9a61d 100644 --- a/include/net/netfilter/nf_conntrack_extend.h +++ b/include/net/netfilter/nf_conntrack_extend.h @@ -8,6 +8,7 @@ enum nf_ct_ext_id { NF_CT_EXT_NAT, NF_CT_EXT_ACCT, NF_CT_EXT_ECACHE, + NF_CT_EXT_ZONE, NF_CT_EXT_NUM, }; @@ -15,6 +16,7 @@ enum nf_ct_ext_id { #define NF_CT_EXT_NAT_TYPE struct nf_conn_nat #define NF_CT_EXT_ACCT_TYPE struct nf_conn_counter #define NF_CT_EXT_ECACHE_TYPE struct nf_conntrack_ecache +#define NF_CT_EXT_ZONE_TYPE struct nf_conntrack_zone /* Extensions: optional stuff which isn't permanently in struct. */ struct nf_ct_ext { diff --git a/include/net/netfilter/nf_conntrack_zones.h b/include/net/netfilter/nf_conntrack_zones.h new file mode 100644 index 00000000000..0bbb2bd51e8 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_zones.h @@ -0,0 +1,23 @@ +#ifndef _NF_CONNTRACK_ZONES_H +#define _NF_CONNTRACK_ZONES_H + +#include + +#define NF_CT_DEFAULT_ZONE 0 + +struct nf_conntrack_zone { + u16 id; +}; + +static inline u16 nf_ct_zone(const struct nf_conn *ct) +{ +#ifdef CONFIG_NF_CONNTRACK_ZONES + struct nf_conntrack_zone *nf_ct_zone; + nf_ct_zone = nf_ct_ext_find(ct, NF_CT_EXT_ZONE); + if (nf_ct_zone) + return nf_ct_zone->id; +#endif + return NF_CT_DEFAULT_ZONE; +} + +#endif /* _NF_CONNTRACK_ZONES_H */ diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index d1ea38a7c49..2bb1f87051c 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -266,7 +267,7 @@ getorigdst(struct sock *sk, int optval, void __user *user, int *len) return -EINVAL; } - h = nf_conntrack_find_get(sock_net(sk), &tuple); + h = nf_conntrack_find_get(sock_net(sk), NF_CT_DEFAULT_ZONE, &tuple); if (h) { struct sockaddr_in sin; struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 327826a968a..7404bde9599 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -18,6 +18,7 @@ #include #include #include +#include #include static unsigned int nf_ct_icmp_timeout __read_mostly = 30*HZ; @@ -114,13 +115,14 @@ static bool icmp_new(struct nf_conn *ct, const struct sk_buff *skb, /* Returns conntrack if it dealt with ICMP, and filled in skb fields */ static int -icmp_error_message(struct net *net, struct sk_buff *skb, +icmp_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, enum ip_conntrack_info *ctinfo, unsigned int hooknum) { struct nf_conntrack_tuple innertuple, origtuple; const struct nf_conntrack_l4proto *innerproto; const struct nf_conntrack_tuple_hash *h; + u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE; NF_CT_ASSERT(skb->nfct == NULL); @@ -146,7 +148,7 @@ icmp_error_message(struct net *net, struct sk_buff *skb, *ctinfo = IP_CT_RELATED; - h = nf_conntrack_find_get(net, &innertuple); + h = nf_conntrack_find_get(net, zone, &innertuple); if (!h) { pr_debug("icmp_error_message: no match\n"); return -NF_ACCEPT; @@ -209,7 +211,7 @@ icmp_error(struct net *net, struct nf_conn *tmpl, icmph->type != ICMP_REDIRECT) return NF_ACCEPT; - return icmp_error_message(net, skb, ctinfo, hooknum); + return icmp_error_message(net, tmpl, skb, ctinfo, hooknum); } #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index f6f46686cbc..d498a704d45 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -16,6 +16,7 @@ #include #include +#include #include #include @@ -39,15 +40,20 @@ static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) static enum ip_defrag_users nf_ct_defrag_user(unsigned int hooknum, struct sk_buff *skb) { + u16 zone = NF_CT_DEFAULT_ZONE; + + if (skb->nfct) + zone = nf_ct_zone((struct nf_conn *)skb->nfct); + #ifdef CONFIG_BRIDGE_NETFILTER if (skb->nf_bridge && skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING) - return IP_DEFRAG_CONNTRACK_BRIDGE_IN; + return IP_DEFRAG_CONNTRACK_BRIDGE_IN + zone; #endif if (hooknum == NF_INET_PRE_ROUTING) - return IP_DEFRAG_CONNTRACK_IN; + return IP_DEFRAG_CONNTRACK_IN + zone; else - return IP_DEFRAG_CONNTRACK_OUT; + return IP_DEFRAG_CONNTRACK_OUT + zone; } static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 26066a2327a..4595281c286 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -30,6 +30,7 @@ #include #include #include +#include static DEFINE_SPINLOCK(nf_nat_lock); @@ -69,13 +70,14 @@ EXPORT_SYMBOL_GPL(nf_nat_proto_put); /* We keep an extra hash for each conntrack, for fast searching. */ static inline unsigned int -hash_by_src(const struct net *net, const struct nf_conntrack_tuple *tuple) +hash_by_src(const struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { unsigned int hash; /* Original src, to ensure we map it consistently if poss. */ hash = jhash_3words((__force u32)tuple->src.u3.ip, - (__force u32)tuple->src.u.all, + (__force u32)tuple->src.u.all ^ zone, tuple->dst.protonum, 0); return ((u64)hash * net->ipv4.nat_htable_size) >> 32; } @@ -139,12 +141,12 @@ same_src(const struct nf_conn *ct, /* Only called for SRC manip */ static int -find_appropriate_src(struct net *net, +find_appropriate_src(struct net *net, u16 zone, const struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple *result, const struct nf_nat_range *range) { - unsigned int h = hash_by_src(net, tuple); + unsigned int h = hash_by_src(net, zone, tuple); const struct nf_conn_nat *nat; const struct nf_conn *ct; const struct hlist_node *n; @@ -152,7 +154,7 @@ find_appropriate_src(struct net *net, rcu_read_lock(); hlist_for_each_entry_rcu(nat, n, &net->ipv4.nat_bysource[h], bysource) { ct = nat->ct; - if (same_src(ct, tuple)) { + if (same_src(ct, tuple) && nf_ct_zone(ct) == zone) { /* Copy source part from reply tuple. */ nf_ct_invert_tuplepr(result, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); @@ -175,7 +177,7 @@ find_appropriate_src(struct net *net, the ip with the lowest src-ip/dst-ip/proto usage. */ static void -find_best_ips_proto(struct nf_conntrack_tuple *tuple, +find_best_ips_proto(u16 zone, struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, const struct nf_conn *ct, enum nf_nat_manip_type maniptype) @@ -209,7 +211,7 @@ find_best_ips_proto(struct nf_conntrack_tuple *tuple, maxip = ntohl(range->max_ip); j = jhash_2words((__force u32)tuple->src.u3.ip, range->flags & IP_NAT_RANGE_PERSISTENT ? - 0 : (__force u32)tuple->dst.u3.ip, 0); + 0 : (__force u32)tuple->dst.u3.ip ^ zone, 0); j = ((u64)j * (maxip - minip + 1)) >> 32; *var_ipp = htonl(minip + j); } @@ -229,6 +231,7 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple, { struct net *net = nf_ct_net(ct); const struct nf_nat_protocol *proto; + u16 zone = nf_ct_zone(ct); /* 1) If this srcip/proto/src-proto-part is currently mapped, and that same mapping gives a unique tuple within the given @@ -239,7 +242,7 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple, manips not an issue. */ if (maniptype == IP_NAT_MANIP_SRC && !(range->flags & IP_NAT_RANGE_PROTO_RANDOM)) { - if (find_appropriate_src(net, orig_tuple, tuple, range)) { + if (find_appropriate_src(net, zone, orig_tuple, tuple, range)) { pr_debug("get_unique_tuple: Found current src map\n"); if (!nf_nat_used_tuple(tuple, ct)) return; @@ -249,7 +252,7 @@ get_unique_tuple(struct nf_conntrack_tuple *tuple, /* 2) Select the least-used IP/proto combination in the given range. */ *tuple = *orig_tuple; - find_best_ips_proto(tuple, range, ct, maniptype); + find_best_ips_proto(zone, tuple, range, ct, maniptype); /* 3) The per-protocol part of the manip is made to map into the range to make a unique tuple. */ @@ -327,7 +330,8 @@ nf_nat_setup_info(struct nf_conn *ct, if (have_to_hash) { unsigned int srchash; - srchash = hash_by_src(net, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + srchash = hash_by_src(net, nf_ct_zone(ct), + &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); spin_lock_bh(&nf_nat_lock); /* nf_conntrack_alter_reply might re-allocate exntension aera */ nat = nfct_nat(ct); diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c index 9eb171056c6..4c060038d29 100644 --- a/net/ipv4/netfilter/nf_nat_pptp.c +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -74,7 +75,7 @@ static void pptp_nat_expected(struct nf_conn *ct, pr_debug("trying to unexpect other dir: "); nf_ct_dump_tuple_ip(&t); - other_exp = nf_ct_expect_find_get(net, &t); + other_exp = nf_ct_expect_find_get(net, nf_ct_zone(ct), &t); if (other_exp) { nf_ct_unexpect_related(other_exp); nf_ct_expect_put(other_exp); diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index 55ce22e5de4..996c3f41fec 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -191,15 +192,20 @@ out: static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, struct sk_buff *skb) { + u16 zone = NF_CT_DEFAULT_ZONE; + + if (skb->nfct) + zone = nf_ct_zone((struct nf_conn *)skb->nfct); + #ifdef CONFIG_BRIDGE_NETFILTER if (skb->nf_bridge && skb->nf_bridge->mask & BRNF_NF_BRIDGE_PREROUTING) - return IP6_DEFRAG_CONNTRACK_BRIDGE_IN; + return IP6_DEFRAG_CONNTRACK_BRIDGE_IN + zone; #endif if (hooknum == NF_INET_PRE_ROUTING) - return IP6_DEFRAG_CONNTRACK_IN; + return IP6_DEFRAG_CONNTRACK_IN + zone; else - return IP6_DEFRAG_CONNTRACK_OUT; + return IP6_DEFRAG_CONNTRACK_OUT + zone; } diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index d772dc21857..9be81776415 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -128,7 +129,7 @@ static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb, } static int -icmpv6_error_message(struct net *net, +icmpv6_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int icmp6off, enum ip_conntrack_info *ctinfo, @@ -137,6 +138,7 @@ icmpv6_error_message(struct net *net, struct nf_conntrack_tuple intuple, origtuple; const struct nf_conntrack_tuple_hash *h; const struct nf_conntrack_l4proto *inproto; + u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE; NF_CT_ASSERT(skb->nfct == NULL); @@ -163,7 +165,7 @@ icmpv6_error_message(struct net *net, *ctinfo = IP_CT_RELATED; - h = nf_conntrack_find_get(net, &intuple); + h = nf_conntrack_find_get(net, zone, &intuple); if (!h) { pr_debug("icmpv6_error: no match\n"); return -NF_ACCEPT; @@ -216,7 +218,7 @@ icmpv6_error(struct net *net, struct nf_conn *tmpl, if (icmp6h->icmp6_type >= 128) return NF_ACCEPT; - return icmpv6_error_message(net, skb, dataoff, ctinfo, hooknum); + return icmpv6_error_message(net, tmpl, skb, dataoff, ctinfo, hooknum); } #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 4469d45261f..18d77b5c351 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -83,6 +83,19 @@ config NF_CONNTRACK_SECMARK If unsure, say 'N'. +config NF_CONNTRACK_ZONES + bool 'Connection tracking zones' + depends on NETFILTER_ADVANCED + depends on NETFILTER_XT_TARGET_CT + help + This option enables support for connection tracking zones. + Normally, each connection needs to have a unique system wide + identity. Connection tracking zones allow to have multiple + connections using the same identity, as long as they are + contained in different zones. + + If unsure, say `N'. + config NF_CONNTRACK_EVENTS bool "Connection tracking events" depends on NETFILTER_ADVANCED diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 65351ed5d81..0c9bbe93cc1 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -42,6 +42,7 @@ #include #include #include +#include #include #include @@ -68,7 +69,7 @@ static int nf_conntrack_hash_rnd_initted; static unsigned int nf_conntrack_hash_rnd; static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, - unsigned int size, unsigned int rnd) + u16 zone, unsigned int size, unsigned int rnd) { unsigned int n; u_int32_t h; @@ -79,16 +80,16 @@ static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, */ n = (sizeof(tuple->src) + sizeof(tuple->dst.u3)) / sizeof(u32); h = jhash2((u32 *)tuple, n, - rnd ^ (((__force __u16)tuple->dst.u.all << 16) | - tuple->dst.protonum)); + zone ^ rnd ^ (((__force __u16)tuple->dst.u.all << 16) | + tuple->dst.protonum)); return ((u64)h * size) >> 32; } -static inline u_int32_t hash_conntrack(const struct net *net, +static inline u_int32_t hash_conntrack(const struct net *net, u16 zone, const struct nf_conntrack_tuple *tuple) { - return __hash_conntrack(tuple, net->ct.htable_size, + return __hash_conntrack(tuple, zone, net->ct.htable_size, nf_conntrack_hash_rnd); } @@ -292,11 +293,12 @@ static void death_by_timeout(unsigned long ul_conntrack) * - Caller must lock nf_conntrack_lock before calling this function */ struct nf_conntrack_tuple_hash * -__nf_conntrack_find(struct net *net, const struct nf_conntrack_tuple *tuple) +__nf_conntrack_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; struct hlist_nulls_node *n; - unsigned int hash = hash_conntrack(net, tuple); + unsigned int hash = hash_conntrack(net, zone, tuple); /* Disable BHs the entire time since we normally need to disable them * at least once for the stats anyway. @@ -304,7 +306,8 @@ __nf_conntrack_find(struct net *net, const struct nf_conntrack_tuple *tuple) local_bh_disable(); begin: hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnnode) { - if (nf_ct_tuple_equal(tuple, &h->tuple)) { + if (nf_ct_tuple_equal(tuple, &h->tuple) && + nf_ct_zone(nf_ct_tuplehash_to_ctrack(h)) == zone) { NF_CT_STAT_INC(net, found); local_bh_enable(); return h; @@ -326,21 +329,23 @@ EXPORT_SYMBOL_GPL(__nf_conntrack_find); /* Find a connection corresponding to a tuple. */ struct nf_conntrack_tuple_hash * -nf_conntrack_find_get(struct net *net, const struct nf_conntrack_tuple *tuple) +nf_conntrack_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; rcu_read_lock(); begin: - h = __nf_conntrack_find(net, tuple); + h = __nf_conntrack_find(net, zone, tuple); if (h) { ct = nf_ct_tuplehash_to_ctrack(h); if (unlikely(nf_ct_is_dying(ct) || !atomic_inc_not_zero(&ct->ct_general.use))) h = NULL; else { - if (unlikely(!nf_ct_tuple_equal(tuple, &h->tuple))) { + if (unlikely(!nf_ct_tuple_equal(tuple, &h->tuple) || + nf_ct_zone(ct) != zone)) { nf_ct_put(ct); goto begin; } @@ -368,9 +373,11 @@ void nf_conntrack_hash_insert(struct nf_conn *ct) { struct net *net = nf_ct_net(ct); unsigned int hash, repl_hash; + u16 zone; - hash = hash_conntrack(net, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - repl_hash = hash_conntrack(net, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + zone = nf_ct_zone(ct); + hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); __nf_conntrack_hash_insert(ct, hash, repl_hash); } @@ -387,6 +394,7 @@ __nf_conntrack_confirm(struct sk_buff *skb) struct hlist_nulls_node *n; enum ip_conntrack_info ctinfo; struct net *net; + u16 zone; ct = nf_ct_get(skb, &ctinfo); net = nf_ct_net(ct); @@ -398,8 +406,9 @@ __nf_conntrack_confirm(struct sk_buff *skb) if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) return NF_ACCEPT; - hash = hash_conntrack(net, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - repl_hash = hash_conntrack(net, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + zone = nf_ct_zone(ct); + hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(net, zone, &ct->tuplehash[IP_CT_DIR_REPLY].tuple); /* We're not in hash table, and we refuse to set up related connections for unconfirmed conns. But packet copies and @@ -418,11 +427,13 @@ __nf_conntrack_confirm(struct sk_buff *skb) not in the hash. If there is, we lost race. */ hlist_nulls_for_each_entry(h, n, &net->ct.hash[hash], hnnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, - &h->tuple)) + &h->tuple) && + zone == nf_ct_zone(nf_ct_tuplehash_to_ctrack(h))) goto out; hlist_nulls_for_each_entry(h, n, &net->ct.hash[repl_hash], hnnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple, - &h->tuple)) + &h->tuple) && + zone == nf_ct_zone(nf_ct_tuplehash_to_ctrack(h))) goto out; /* Remove from unconfirmed list */ @@ -469,15 +480,19 @@ nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, struct net *net = nf_ct_net(ignored_conntrack); struct nf_conntrack_tuple_hash *h; struct hlist_nulls_node *n; - unsigned int hash = hash_conntrack(net, tuple); + struct nf_conn *ct; + u16 zone = nf_ct_zone(ignored_conntrack); + unsigned int hash = hash_conntrack(net, zone, tuple); /* Disable BHs the entire time since we need to disable them at * least once for the stats anyway. */ rcu_read_lock_bh(); hlist_nulls_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnnode) { - if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack && - nf_ct_tuple_equal(tuple, &h->tuple)) { + ct = nf_ct_tuplehash_to_ctrack(h); + if (ct != ignored_conntrack && + nf_ct_tuple_equal(tuple, &h->tuple) && + nf_ct_zone(ct) == zone) { NF_CT_STAT_INC(net, found); rcu_read_unlock_bh(); return 1; @@ -540,7 +555,7 @@ static noinline int early_drop(struct net *net, unsigned int hash) return dropped; } -struct nf_conn *nf_conntrack_alloc(struct net *net, +struct nf_conn *nf_conntrack_alloc(struct net *net, u16 zone, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, gfp_t gfp) @@ -558,7 +573,7 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, if (nf_conntrack_max && unlikely(atomic_read(&net->ct.count) > nf_conntrack_max)) { - unsigned int hash = hash_conntrack(net, orig); + unsigned int hash = hash_conntrack(net, zone, orig); if (!early_drop(net, hash)) { atomic_dec(&net->ct.count); if (net_ratelimit()) @@ -595,13 +610,28 @@ struct nf_conn *nf_conntrack_alloc(struct net *net, #ifdef CONFIG_NET_NS ct->ct_net = net; #endif - +#ifdef CONFIG_NF_CONNTRACK_ZONES + if (zone) { + struct nf_conntrack_zone *nf_ct_zone; + + nf_ct_zone = nf_ct_ext_add(ct, NF_CT_EXT_ZONE, GFP_ATOMIC); + if (!nf_ct_zone) + goto out_free; + nf_ct_zone->id = zone; + } +#endif /* * changes to lookup keys must be done before setting refcnt to 1 */ smp_wmb(); atomic_set(&ct->ct_general.use, 1); return ct; + +#ifdef CONFIG_NF_CONNTRACK_ZONES +out_free: + kmem_cache_free(net->ct.nf_conntrack_cachep, ct); + return ERR_PTR(-ENOMEM); +#endif } EXPORT_SYMBOL_GPL(nf_conntrack_alloc); @@ -631,13 +661,14 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, struct nf_conntrack_tuple repl_tuple; struct nf_conntrack_ecache *ecache; struct nf_conntrack_expect *exp; + u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE; if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, l4proto)) { pr_debug("Can't invert tuple.\n"); return NULL; } - ct = nf_conntrack_alloc(net, tuple, &repl_tuple, GFP_ATOMIC); + ct = nf_conntrack_alloc(net, zone, tuple, &repl_tuple, GFP_ATOMIC); if (IS_ERR(ct)) { pr_debug("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)ct; @@ -657,7 +688,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, GFP_ATOMIC); spin_lock_bh(&nf_conntrack_lock); - exp = nf_ct_find_expectation(net, tuple); + exp = nf_ct_find_expectation(net, zone, tuple); if (exp) { pr_debug("conntrack: expectation arrives ct=%p exp=%p\n", ct, exp); @@ -713,6 +744,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl, struct nf_conntrack_tuple tuple; struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; + u16 zone = tmpl ? nf_ct_zone(tmpl) : NF_CT_DEFAULT_ZONE; if (!nf_ct_get_tuple(skb, skb_network_offset(skb), dataoff, l3num, protonum, &tuple, l3proto, @@ -722,7 +754,7 @@ resolve_normal_ct(struct net *net, struct nf_conn *tmpl, } /* look for tuple match */ - h = nf_conntrack_find_get(net, &tuple); + h = nf_conntrack_find_get(net, zone, &tuple); if (!h) { h = init_conntrack(net, tmpl, &tuple, l3proto, l4proto, skb, dataoff); @@ -958,6 +990,14 @@ bool __nf_ct_kill_acct(struct nf_conn *ct, } EXPORT_SYMBOL_GPL(__nf_ct_kill_acct); +#ifdef CONFIG_NF_CONNTRACK_ZONES +static struct nf_ct_ext_type nf_ct_zone_extend __read_mostly = { + .len = sizeof(struct nf_conntrack_zone), + .align = __alignof__(struct nf_conntrack_zone), + .id = NF_CT_EXT_ZONE, +}; +#endif + #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) #include @@ -1139,6 +1179,9 @@ static void nf_conntrack_cleanup_init_net(void) nf_conntrack_helper_fini(); nf_conntrack_proto_fini(); +#ifdef CONFIG_NF_CONNTRACK_ZONES + nf_ct_extend_unregister(&nf_ct_zone_extend); +#endif } static void nf_conntrack_cleanup_net(struct net *net) @@ -1214,6 +1257,7 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) unsigned int hashsize, old_size; struct hlist_nulls_head *hash, *old_hash; struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; if (current->nsproxy->net_ns != &init_net) return -EOPNOTSUPP; @@ -1240,8 +1284,10 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) while (!hlist_nulls_empty(&init_net.ct.hash[i])) { h = hlist_nulls_entry(init_net.ct.hash[i].first, struct nf_conntrack_tuple_hash, hnnode); + ct = nf_ct_tuplehash_to_ctrack(h); hlist_nulls_del_rcu(&h->hnnode); - bucket = __hash_conntrack(&h->tuple, hashsize, + bucket = __hash_conntrack(&h->tuple, nf_ct_zone(ct), + hashsize, nf_conntrack_hash_rnd); hlist_nulls_add_head_rcu(&h->hnnode, &hash[bucket]); } @@ -1299,6 +1345,11 @@ static int nf_conntrack_init_init_net(void) if (ret < 0) goto err_helper; +#ifdef CONFIG_NF_CONNTRACK_ZONES + ret = nf_ct_extend_register(&nf_ct_zone_extend); + if (ret < 0) + goto err_extend; +#endif /* Set up fake conntrack: to never be deleted, not in any hashes */ #ifdef CONFIG_NET_NS nf_conntrack_untracked.ct_net = &init_net; @@ -1309,6 +1360,10 @@ static int nf_conntrack_init_init_net(void) return 0; +#ifdef CONFIG_NF_CONNTRACK_ZONES +err_extend: + nf_conntrack_helper_fini(); +#endif err_helper: nf_conntrack_proto_fini(); err_proto: diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 6182fb1b55d..acb29ccaa41 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -27,6 +27,7 @@ #include #include #include +#include unsigned int nf_ct_expect_hsize __read_mostly; EXPORT_SYMBOL_GPL(nf_ct_expect_hsize); @@ -84,7 +85,8 @@ static unsigned int nf_ct_expect_dst_hash(const struct nf_conntrack_tuple *tuple } struct nf_conntrack_expect * -__nf_ct_expect_find(struct net *net, const struct nf_conntrack_tuple *tuple) +__nf_ct_expect_find(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; struct hlist_node *n; @@ -95,7 +97,8 @@ __nf_ct_expect_find(struct net *net, const struct nf_conntrack_tuple *tuple) h = nf_ct_expect_dst_hash(tuple); hlist_for_each_entry_rcu(i, n, &net->ct.expect_hash[h], hnode) { - if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) && + nf_ct_zone(i->master) == zone) return i; } return NULL; @@ -104,12 +107,13 @@ EXPORT_SYMBOL_GPL(__nf_ct_expect_find); /* Just find a expectation corresponding to a tuple. */ struct nf_conntrack_expect * -nf_ct_expect_find_get(struct net *net, const struct nf_conntrack_tuple *tuple) +nf_ct_expect_find_get(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; rcu_read_lock(); - i = __nf_ct_expect_find(net, tuple); + i = __nf_ct_expect_find(net, zone, tuple); if (i && !atomic_inc_not_zero(&i->use)) i = NULL; rcu_read_unlock(); @@ -121,7 +125,8 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_find_get); /* If an expectation for this connection is found, it gets delete from * global list then returned. */ struct nf_conntrack_expect * -nf_ct_find_expectation(struct net *net, const struct nf_conntrack_tuple *tuple) +nf_ct_find_expectation(struct net *net, u16 zone, + const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i, *exp = NULL; struct hlist_node *n; @@ -133,7 +138,8 @@ nf_ct_find_expectation(struct net *net, const struct nf_conntrack_tuple *tuple) h = nf_ct_expect_dst_hash(tuple); hlist_for_each_entry(i, n, &net->ct.expect_hash[h], hnode) { if (!(i->flags & NF_CT_EXPECT_INACTIVE) && - nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) { + nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) && + nf_ct_zone(i->master) == zone) { exp = i; break; } @@ -204,7 +210,8 @@ static inline int expect_matches(const struct nf_conntrack_expect *a, { return a->master == b->master && a->class == b->class && nf_ct_tuple_equal(&a->tuple, &b->tuple) && - nf_ct_tuple_mask_equal(&a->mask, &b->mask); + nf_ct_tuple_mask_equal(&a->mask, &b->mask) && + nf_ct_zone(a->master) == nf_ct_zone(b->master); } /* Generally a bad idea to call this: could have matched already. */ diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 66369490230..a1c8dd917e1 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -29,6 +29,7 @@ #include #include #include +#include #include /* Parameters */ @@ -1216,7 +1217,7 @@ static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, tuple.dst.u.tcp.port = port; tuple.dst.protonum = IPPROTO_TCP; - exp = __nf_ct_expect_find(net, &tuple); + exp = __nf_ct_expect_find(net, nf_ct_zone(ct), &tuple); if (exp && exp->master == ct) return exp; return NULL; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index db35edac307..51089cfe116 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -811,7 +811,7 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - h = nf_conntrack_find_get(net, &tuple); + h = nf_conntrack_find_get(net, 0, &tuple); if (!h) return -ENOENT; @@ -872,7 +872,7 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - h = nf_conntrack_find_get(net, &tuple); + h = nf_conntrack_find_get(net, 0, &tuple); if (!h) return -ENOENT; @@ -1221,7 +1221,7 @@ ctnetlink_create_conntrack(struct net *net, int err = -EINVAL; struct nf_conntrack_helper *helper; - ct = nf_conntrack_alloc(net, otuple, rtuple, GFP_ATOMIC); + ct = nf_conntrack_alloc(net, 0, otuple, rtuple, GFP_ATOMIC); if (IS_ERR(ct)) return ERR_PTR(-ENOMEM); @@ -1325,7 +1325,7 @@ ctnetlink_create_conntrack(struct net *net, if (err < 0) goto err2; - master_h = nf_conntrack_find_get(net, &master); + master_h = nf_conntrack_find_get(net, 0, &master); if (master_h == NULL) { err = -ENOENT; goto err2; @@ -1374,9 +1374,9 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, spin_lock_bh(&nf_conntrack_lock); if (cda[CTA_TUPLE_ORIG]) - h = __nf_conntrack_find(net, &otuple); + h = __nf_conntrack_find(net, 0, &otuple); else if (cda[CTA_TUPLE_REPLY]) - h = __nf_conntrack_find(net, &rtuple); + h = __nf_conntrack_find(net, 0, &rtuple); if (h == NULL) { err = -ENOENT; @@ -1714,7 +1714,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = nf_ct_expect_find_get(net, &tuple); + exp = nf_ct_expect_find_get(net, 0, &tuple); if (!exp) return -ENOENT; @@ -1770,7 +1770,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = nf_ct_expect_find_get(net, &tuple); + exp = nf_ct_expect_find_get(net, 0, &tuple); if (!exp) return -ENOENT; @@ -1855,7 +1855,7 @@ ctnetlink_create_expect(struct net *net, const struct nlattr * const cda[], return err; /* Look for master conntrack of this expectation */ - h = nf_conntrack_find_get(net, &master_tuple); + h = nf_conntrack_find_get(net, 0, &master_tuple); if (!h) return -ENOENT; ct = nf_ct_tuplehash_to_ctrack(h); @@ -1912,7 +1912,7 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, return err; spin_lock_bh(&nf_conntrack_lock); - exp = __nf_ct_expect_find(net, &tuple); + exp = __nf_ct_expect_find(net, 0, &tuple); if (!exp) { spin_unlock_bh(&nf_conntrack_lock); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 3807ac7faf4..088944824e1 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -123,7 +124,7 @@ static void pptp_expectfn(struct nf_conn *ct, pr_debug("trying to unexpect other dir: "); nf_ct_dump_tuple(&inv_t); - exp_other = nf_ct_expect_find_get(net, &inv_t); + exp_other = nf_ct_expect_find_get(net, nf_ct_zone(ct), &inv_t); if (exp_other) { /* delete other expectation. */ pr_debug("found\n"); @@ -136,17 +137,18 @@ static void pptp_expectfn(struct nf_conn *ct, rcu_read_unlock(); } -static int destroy_sibling_or_exp(struct net *net, +static int destroy_sibling_or_exp(struct net *net, struct nf_conn *ct, const struct nf_conntrack_tuple *t) { const struct nf_conntrack_tuple_hash *h; struct nf_conntrack_expect *exp; struct nf_conn *sibling; + u16 zone = nf_ct_zone(ct); pr_debug("trying to timeout ct or exp for tuple "); nf_ct_dump_tuple(t); - h = nf_conntrack_find_get(net, t); + h = nf_conntrack_find_get(net, zone, t); if (h) { sibling = nf_ct_tuplehash_to_ctrack(h); pr_debug("setting timeout of conntrack %p to 0\n", sibling); @@ -157,7 +159,7 @@ static int destroy_sibling_or_exp(struct net *net, nf_ct_put(sibling); return 1; } else { - exp = nf_ct_expect_find_get(net, t); + exp = nf_ct_expect_find_get(net, zone, t); if (exp) { pr_debug("unexpect_related of expect %p\n", exp); nf_ct_unexpect_related(exp); @@ -182,7 +184,7 @@ static void pptp_destroy_siblings(struct nf_conn *ct) t.dst.protonum = IPPROTO_GRE; t.src.u.gre.key = help->help.ct_pptp_info.pns_call_id; t.dst.u.gre.key = help->help.ct_pptp_info.pac_call_id; - if (!destroy_sibling_or_exp(net, &t)) + if (!destroy_sibling_or_exp(net, ct, &t)) pr_debug("failed to timeout original pns->pac ct/exp\n"); /* try reply (pac->pns) tuple */ @@ -190,7 +192,7 @@ static void pptp_destroy_siblings(struct nf_conn *ct) t.dst.protonum = IPPROTO_GRE; t.src.u.gre.key = help->help.ct_pptp_info.pac_call_id; t.dst.u.gre.key = help->help.ct_pptp_info.pns_call_id; - if (!destroy_sibling_or_exp(net, &t)) + if (!destroy_sibling_or_exp(net, ct, &t)) pr_debug("failed to timeout reply pac->pns ct/exp\n"); } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index fbe8ff5a420..8dd75d90efc 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -23,6 +23,7 @@ #include #include #include +#include #include MODULE_LICENSE("GPL"); @@ -836,7 +837,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int dataoff, rcu_read_lock(); do { - exp = __nf_ct_expect_find(net, &tuple); + exp = __nf_ct_expect_find(net, nf_ct_zone(ct), &tuple); if (!exp || exp->master == ct || nfct_help(exp->master)->helper != nfct_help(ct)->helper || diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index e310f1561bb..24a42efe62e 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -26,6 +26,7 @@ #include #include #include +#include MODULE_LICENSE("GPL"); @@ -171,6 +172,11 @@ static int ct_seq_show(struct seq_file *s, void *v) goto release; #endif +#ifdef CONFIG_NF_CONNTRACK_ZONES + if (seq_printf(s, "zone=%u ", nf_ct_zone(ct))) + goto release; +#endif + if (seq_printf(s, "use=%u\n", atomic_read(&ct->ct_general.use))) goto release; diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index 8183a054256..61c50fa8470 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -16,6 +16,7 @@ #include #include #include +#include static unsigned int xt_ct_target(struct sk_buff *skb, const struct xt_target_param *par) @@ -69,11 +70,16 @@ static bool xt_ct_tg_check(const struct xt_tgchk_param *par) goto out; } +#ifndef CONFIG_NF_CONNTRACK_ZONES + if (info->zone) + goto err1; +#endif + if (nf_ct_l3proto_try_module_get(par->family) < 0) goto err1; memset(&t, 0, sizeof(t)); - ct = nf_conntrack_alloc(par->net, &t, &t, GFP_KERNEL); + ct = nf_conntrack_alloc(par->net, info->zone, &t, &t, GFP_KERNEL); if (IS_ERR(ct)) goto err2; diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 0d9d18ea2b0..26997ce90e4 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -28,6 +28,7 @@ #include #include #include +#include /* we will save the tuples of all connections we care about */ struct xt_connlimit_conn { @@ -114,7 +115,8 @@ static int count_them(struct net *net, /* check the saved connections */ list_for_each_entry_safe(conn, tmp, hash, list) { - found = nf_conntrack_find_get(net, &conn->tuple); + found = nf_conntrack_find_get(net, NF_CT_DEFAULT_ZONE, + &conn->tuple); found_ct = NULL; if (found != NULL) -- cgit v1.2.3-70-g09d2 From ef00f89f1eb7e056aab9dfe068521e6f2320c94a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 15 Feb 2010 18:14:57 +0100 Subject: netfilter: ctnetlink: add zone support Parse and dump the conntrack zone in ctnetlink. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink_conntrack.h | 2 + net/netfilter/nf_conntrack_netlink.c | 92 +++++++++++++++++++++------ 2 files changed, 75 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h index ed4ef8d0b11..9ed534c991b 100644 --- a/include/linux/netfilter/nfnetlink_conntrack.h +++ b/include/linux/netfilter/nfnetlink_conntrack.h @@ -40,6 +40,7 @@ enum ctattr_type { CTA_NAT_SEQ_ADJ_ORIG, CTA_NAT_SEQ_ADJ_REPLY, CTA_SECMARK, + CTA_ZONE, __CTA_MAX }; #define CTA_MAX (__CTA_MAX - 1) @@ -159,6 +160,7 @@ enum ctattr_expect { CTA_EXPECT_TIMEOUT, CTA_EXPECT_ID, CTA_EXPECT_HELP_NAME, + CTA_EXPECT_ZONE, __CTA_EXPECT_MAX }; #define CTA_EXPECT_MAX (__CTA_EXPECT_MAX - 1) diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 51089cfe116..8b05f364b2f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -39,6 +39,7 @@ #include #include #include +#include #ifdef CONFIG_NF_NAT_NEEDED #include #include @@ -379,6 +380,9 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, goto nla_put_failure; nla_nest_end(skb, nest_parms); + if (nf_ct_zone(ct)) + NLA_PUT_BE16(skb, CTA_ZONE, htons(nf_ct_zone(ct))); + if (ctnetlink_dump_status(skb, ct) < 0 || ctnetlink_dump_timeout(skb, ct) < 0 || ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || @@ -517,6 +521,9 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) goto nla_put_failure; nla_nest_end(skb, nest_parms); + if (nf_ct_zone(ct)) + NLA_PUT_BE16(skb, CTA_ZONE, htons(nf_ct_zone(ct))); + if (ctnetlink_dump_id(skb, ct) < 0) goto nla_put_failure; @@ -750,6 +757,21 @@ ctnetlink_parse_tuple(const struct nlattr * const cda[], return 0; } +static int +ctnetlink_parse_zone(const struct nlattr *attr, u16 *zone) +{ + if (attr) +#ifdef CONFIG_NF_CONNTRACK_ZONES + *zone = ntohs(nla_get_be16(attr)); +#else + return -EOPNOTSUPP; +#endif + else + *zone = 0; + + return 0; +} + static const struct nla_policy help_nla_policy[CTA_HELP_MAX+1] = { [CTA_HELP_NAME] = { .type = NLA_NUL_STRING }, }; @@ -781,6 +803,7 @@ static const struct nla_policy ct_nla_policy[CTA_MAX+1] = { [CTA_ID] = { .type = NLA_U32 }, [CTA_NAT_DST] = { .type = NLA_NESTED }, [CTA_TUPLE_MASTER] = { .type = NLA_NESTED }, + [CTA_ZONE] = { .type = NLA_U16 }, }; static int @@ -794,7 +817,12 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, struct nf_conn *ct; struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; - int err = 0; + u16 zone; + int err; + + err = ctnetlink_parse_zone(cda[CTA_ZONE], &zone); + if (err < 0) + return err; if (cda[CTA_TUPLE_ORIG]) err = ctnetlink_parse_tuple(cda, &tuple, CTA_TUPLE_ORIG, u3); @@ -811,7 +839,7 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - h = nf_conntrack_find_get(net, 0, &tuple); + h = nf_conntrack_find_get(net, zone, &tuple); if (!h) return -ENOENT; @@ -856,12 +884,17 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, struct sk_buff *skb2 = NULL; struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; - int err = 0; + u16 zone; + int err; if (nlh->nlmsg_flags & NLM_F_DUMP) return netlink_dump_start(ctnl, skb, nlh, ctnetlink_dump_table, ctnetlink_done); + err = ctnetlink_parse_zone(cda[CTA_ZONE], &zone); + if (err < 0) + return err; + if (cda[CTA_TUPLE_ORIG]) err = ctnetlink_parse_tuple(cda, &tuple, CTA_TUPLE_ORIG, u3); else if (cda[CTA_TUPLE_REPLY]) @@ -872,7 +905,7 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - h = nf_conntrack_find_get(net, 0, &tuple); + h = nf_conntrack_find_get(net, zone, &tuple); if (!h) return -ENOENT; @@ -1211,7 +1244,7 @@ ctnetlink_change_conntrack(struct nf_conn *ct, } static struct nf_conn * -ctnetlink_create_conntrack(struct net *net, +ctnetlink_create_conntrack(struct net *net, u16 zone, const struct nlattr * const cda[], struct nf_conntrack_tuple *otuple, struct nf_conntrack_tuple *rtuple, @@ -1221,7 +1254,7 @@ ctnetlink_create_conntrack(struct net *net, int err = -EINVAL; struct nf_conntrack_helper *helper; - ct = nf_conntrack_alloc(net, 0, otuple, rtuple, GFP_ATOMIC); + ct = nf_conntrack_alloc(net, zone, otuple, rtuple, GFP_ATOMIC); if (IS_ERR(ct)) return ERR_PTR(-ENOMEM); @@ -1325,7 +1358,7 @@ ctnetlink_create_conntrack(struct net *net, if (err < 0) goto err2; - master_h = nf_conntrack_find_get(net, 0, &master); + master_h = nf_conntrack_find_get(net, zone, &master); if (master_h == NULL) { err = -ENOENT; goto err2; @@ -1358,7 +1391,12 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, struct nf_conntrack_tuple_hash *h = NULL; struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; - int err = 0; + u16 zone; + int err; + + err = ctnetlink_parse_zone(cda[CTA_ZONE], &zone); + if (err < 0) + return err; if (cda[CTA_TUPLE_ORIG]) { err = ctnetlink_parse_tuple(cda, &otuple, CTA_TUPLE_ORIG, u3); @@ -1374,9 +1412,9 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, spin_lock_bh(&nf_conntrack_lock); if (cda[CTA_TUPLE_ORIG]) - h = __nf_conntrack_find(net, 0, &otuple); + h = __nf_conntrack_find(net, zone, &otuple); else if (cda[CTA_TUPLE_REPLY]) - h = __nf_conntrack_find(net, 0, &rtuple); + h = __nf_conntrack_find(net, zone, &rtuple); if (h == NULL) { err = -ENOENT; @@ -1384,7 +1422,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, struct nf_conn *ct; enum ip_conntrack_events events; - ct = ctnetlink_create_conntrack(net, cda, &otuple, + ct = ctnetlink_create_conntrack(net, zone, cda, &otuple, &rtuple, u3); if (IS_ERR(ct)) { err = PTR_ERR(ct); @@ -1698,7 +1736,8 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, struct sk_buff *skb2; struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; - int err = 0; + u16 zone; + int err; if (nlh->nlmsg_flags & NLM_F_DUMP) { return netlink_dump_start(ctnl, skb, nlh, @@ -1706,6 +1745,10 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, ctnetlink_exp_done); } + err = ctnetlink_parse_zone(cda[CTA_EXPECT_ZONE], &zone); + if (err < 0) + return err; + if (cda[CTA_EXPECT_MASTER]) err = ctnetlink_parse_tuple(cda, &tuple, CTA_EXPECT_MASTER, u3); else @@ -1714,7 +1757,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = nf_ct_expect_find_get(net, 0, &tuple); + exp = nf_ct_expect_find_get(net, zone, &tuple); if (!exp) return -ENOENT; @@ -1761,16 +1804,21 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, struct hlist_node *n, *next; u_int8_t u3 = nfmsg->nfgen_family; unsigned int i; + u16 zone; int err; if (cda[CTA_EXPECT_TUPLE]) { /* delete a single expect by tuple */ + err = ctnetlink_parse_zone(cda[CTA_EXPECT_ZONE], &zone); + if (err < 0) + return err; + err = ctnetlink_parse_tuple(cda, &tuple, CTA_EXPECT_TUPLE, u3); if (err < 0) return err; /* bump usage count to 2 */ - exp = nf_ct_expect_find_get(net, 0, &tuple); + exp = nf_ct_expect_find_get(net, zone, &tuple); if (!exp) return -ENOENT; @@ -1832,7 +1880,8 @@ ctnetlink_change_expect(struct nf_conntrack_expect *x, } static int -ctnetlink_create_expect(struct net *net, const struct nlattr * const cda[], +ctnetlink_create_expect(struct net *net, u16 zone, + const struct nlattr * const cda[], u_int8_t u3, u32 pid, int report) { @@ -1855,7 +1904,7 @@ ctnetlink_create_expect(struct net *net, const struct nlattr * const cda[], return err; /* Look for master conntrack of this expectation */ - h = nf_conntrack_find_get(net, 0, &master_tuple); + h = nf_conntrack_find_get(net, zone, &master_tuple); if (!h) return -ENOENT; ct = nf_ct_tuplehash_to_ctrack(h); @@ -1900,25 +1949,30 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, struct nf_conntrack_expect *exp; struct nfgenmsg *nfmsg = nlmsg_data(nlh); u_int8_t u3 = nfmsg->nfgen_family; - int err = 0; + u16 zone; + int err; if (!cda[CTA_EXPECT_TUPLE] || !cda[CTA_EXPECT_MASK] || !cda[CTA_EXPECT_MASTER]) return -EINVAL; + err = ctnetlink_parse_zone(cda[CTA_EXPECT_ZONE], &zone); + if (err < 0) + return err; + err = ctnetlink_parse_tuple(cda, &tuple, CTA_EXPECT_TUPLE, u3); if (err < 0) return err; spin_lock_bh(&nf_conntrack_lock); - exp = __nf_ct_expect_find(net, 0, &tuple); + exp = __nf_ct_expect_find(net, zone, &tuple); if (!exp) { spin_unlock_bh(&nf_conntrack_lock); err = -ENOENT; if (nlh->nlmsg_flags & NLM_F_CREATE) { - err = ctnetlink_create_expect(net, cda, + err = ctnetlink_create_expect(net, zone, cda, u3, NETLINK_CB(skb).pid, nlmsg_report(nlh)); -- cgit v1.2.3-70-g09d2 From 3e5e524ffb5fcf2447eb5dd9f8e54ad22dd9baa7 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 15 Feb 2010 18:17:10 +0100 Subject: netfilter: CONFIG_COMPAT: allow delta to exceed 32767 with 32 bit userland and 64 bit kernels, it is unlikely but possible that insertion of new rules fails even tough there are only about 2000 iptables rules. This happens because the compat delta is using a short int. Easily reproducible via "iptables -m limit" ; after about 2050 rules inserting new ones fails with -ELOOP. Note that compat_delta included 2 bytes of padding on x86_64, so structure size remains the same. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 2 +- net/netfilter/x_tables.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index c61758f4be3..a18119fb88f 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -588,7 +588,7 @@ extern void xt_compat_unlock(u_int8_t af); extern int xt_compat_add_offset(u_int8_t af, unsigned int offset, short delta); extern void xt_compat_flush_offsets(u_int8_t af); -extern short xt_compat_calc_jump(u_int8_t af, unsigned int offset); +extern int xt_compat_calc_jump(u_int8_t af, unsigned int offset); extern int xt_compat_match_offset(const struct xt_match *match); extern int xt_compat_match_from_user(struct xt_entry_match *m, diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 69c56287d51..0a12cedfe9e 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -39,7 +39,7 @@ MODULE_DESCRIPTION("{ip,ip6,arp,eb}_tables backend module"); struct compat_delta { struct compat_delta *next; unsigned int offset; - short delta; + int delta; }; struct xt_af { @@ -439,10 +439,10 @@ void xt_compat_flush_offsets(u_int8_t af) } EXPORT_SYMBOL_GPL(xt_compat_flush_offsets); -short xt_compat_calc_jump(u_int8_t af, unsigned int offset) +int xt_compat_calc_jump(u_int8_t af, unsigned int offset) { struct compat_delta *tmp; - short delta; + int delta; for (tmp = xt[af].compat_offsets, delta = 0; tmp; tmp = tmp->next) if (tmp->offset < offset) -- cgit v1.2.3-70-g09d2 From 026331c4d9b526561ea96f95fac4bfc52b69e316 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Mon, 15 Feb 2010 12:53:10 +0200 Subject: cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 41 ++++++- include/net/cfg80211.h | 47 +++++++- include/net/mac80211.h | 6 +- net/mac80211/cfg.c | 12 ++- net/mac80211/ieee80211_i.h | 6 +- net/mac80211/mlme.c | 35 ++++++ net/mac80211/rx.c | 42 ++++++-- net/mac80211/status.c | 7 +- net/wireless/core.c | 4 + net/wireless/core.h | 9 ++ net/wireless/mlme.c | 166 +++++++++++++++++++++++++++++ net/wireless/nl80211.c | 260 ++++++++++++++++++++++++++++++++++++++++++++- net/wireless/nl80211.h | 8 ++ 13 files changed, 625 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 127a7301576..8e6384f8fda 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -3,7 +3,7 @@ /* * 802.11 netlink interface public header * - * Copyright 2006, 2007, 2008 Johannes Berg + * Copyright 2006-2010 Johannes Berg * Copyright 2008 Michael Wu * Copyright 2008 Luis Carlos Cobo * Copyright 2008 Michael Buesch @@ -299,6 +299,31 @@ * rate selection. %NL80211_ATTR_IFINDEX is used to specify the interface * and @NL80211_ATTR_TX_RATES the set of allowed rates. * + * @NL80211_CMD_REGISTER_ACTION: Register for receiving certain action frames + * (via @NL80211_CMD_ACTION) for processing in userspace. This command + * requires an interface index and a match attribute containing the first + * few bytes of the frame that should match, e.g. a single byte for only + * a category match or four bytes for vendor frames including the OUI. + * The registration cannot be dropped, but is removed automatically + * when the netlink socket is closed. Multiple registrations can be made. + * @NL80211_CMD_ACTION: Action frame TX request and RX notification. This + * command is used both as a request to transmit an Action frame and as an + * event indicating reception of an Action frame that was not processed in + * kernel code, but is for us (i.e., which may need to be processed in a + * user space application). %NL80211_ATTR_FRAME is used to specify the + * frame contents (including header). %NL80211_ATTR_WIPHY_FREQ (and + * optionally %NL80211_ATTR_WIPHY_CHANNEL_TYPE) is used to indicate on + * which channel the frame is to be transmitted or was received. This + * channel has to be the current channel (remain-on-channel or the + * operational channel). When called, this operation returns a cookie + * (%NL80211_ATTR_COOKIE) that will be included with the TX status event + * pertaining to the TX request. + * @NL80211_CMD_ACTION_TX_STATUS: Report TX status of an Action frame + * transmitted with %NL80211_CMD_ACTION. %NL80211_ATTR_COOKIE identifies + * the TX command and %NL80211_ATTR_FRAME includes the contents of the + * frame. %NL80211_ATTR_ACK flag is included if the recipient acknowledged + * the frame. + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -387,6 +412,10 @@ enum nl80211_commands { NL80211_CMD_SET_TX_BITRATE_MASK, + NL80211_CMD_REGISTER_ACTION, + NL80211_CMD_ACTION, + NL80211_CMD_ACTION_TX_STATUS, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -653,6 +682,12 @@ enum nl80211_commands { * rates based on negotiated supported rates information. This attribute * is used with %NL80211_CMD_SET_TX_BITRATE_MASK. * + * @NL80211_ATTR_FRAME_MATCH: A binary attribute which typically must contain + * at least one byte, currently used with @NL80211_CMD_REGISTER_ACTION. + * + * @NL80211_ATTR_ACK: Flag attribute indicating that the frame was + * acknowledged by the recipient. + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -798,6 +833,10 @@ enum nl80211_attrs { NL80211_ATTR_TX_RATES, + NL80211_ATTR_FRAME_MATCH, + + NL80211_ATTR_ACK, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 5b3569b2a74..7188934b64d 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -3,7 +3,7 @@ /* * 802.11 device and configuration interface * - * Copyright 2006-2009 Johannes Berg + * Copyright 2006-2010 Johannes Berg * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -998,6 +998,7 @@ struct cfg80211_pmksa { * @cancel_remain_on_channel: Cancel an on-going remain-on-channel operation. * This allows the operation to be terminated prior to timeout based on * the duration value. + * @action: Transmit an action frame * * @testmode_cmd: run a test mode command * @@ -1144,6 +1145,11 @@ struct cfg80211_ops { struct net_device *dev, u64 cookie); + int (*action)(struct wiphy *wiphy, struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie); + /* some temporary stuff to finish wext */ int (*set_power_mgmt)(struct wiphy *wiphy, struct net_device *dev, bool enabled, int timeout); @@ -1445,6 +1451,8 @@ struct cfg80211_cached_keys; * set by driver (if supported) on add_interface BEFORE registering the * netdev and may otherwise be used by driver read-only, will be update * by cfg80211 on change_interface + * @action_registrations: list of registrations for action frames + * @action_registrations_lock: lock for the list */ struct wireless_dev { struct wiphy *wiphy; @@ -1454,6 +1462,9 @@ struct wireless_dev { struct list_head list; struct net_device *netdev; + struct list_head action_registrations; + spinlock_t action_registrations_lock; + struct mutex mtx; struct work_struct cleanup_work; @@ -2291,4 +2302,38 @@ void cfg80211_remain_on_channel_expired(struct net_device *dev, void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr, struct station_info *sinfo, gfp_t gfp); +/** + * cfg80211_rx_action - notification of received, unprocessed Action frame + * @dev: network device + * @freq: Frequency on which the frame was received in MHz + * @buf: Action frame (header + body) + * @len: length of the frame data + * @gfp: context flags + * Returns %true if a user space application is responsible for rejecting the + * unrecognized Action frame; %false if no such application is registered + * (i.e., the driver is responsible for rejecting the unrecognized Action + * frame) + * + * This function is called whenever an Action frame is received for a station + * mode interface, but is not processed in kernel. + */ +bool cfg80211_rx_action(struct net_device *dev, int freq, const u8 *buf, + size_t len, gfp_t gfp); + +/** + * cfg80211_action_tx_status - notification of TX status for Action frame + * @dev: network device + * @cookie: Cookie returned by cfg80211_ops::action() + * @buf: Action frame (header + body) + * @len: length of the frame data + * @ack: Whether frame was acknowledged + * @gfp: context flags + * + * This function is called whenever an Action frame was requested to be + * transmitted with cfg80211_ops::action() to report the TX status of the + * transmission attempt. + */ +void cfg80211_action_tx_status(struct net_device *dev, u64 cookie, + const u8 *buf, size_t len, bool ack, gfp_t gfp); + #endif /* __NET_CFG80211_H */ diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 314e9817316..80eb7cc42ce 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -3,7 +3,7 @@ * * Copyright 2002-2005, Devicescape Software, Inc. * Copyright 2006-2007 Jiri Benc - * Copyright 2007-2008 Johannes Berg + * Copyright 2007-2010 Johannes Berg * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -264,6 +264,9 @@ struct ieee80211_bss_conf { * be modified again (no seqno assignment, crypto, etc.) * @IEEE80211_TX_INTFL_HAS_RADIOTAP: This frame was injected and still * has a radiotap header at skb->data. + * @IEEE80211_TX_INTFL_NL80211_FRAME_TX: Frame was requested through nl80211 + * MLME command (internal to mac80211 to figure out whether to send TX + * status to user space) */ enum mac80211_tx_control_flags { IEEE80211_TX_CTL_REQ_TX_STATUS = BIT(0), @@ -286,6 +289,7 @@ enum mac80211_tx_control_flags { IEEE80211_TX_CTL_MORE_FRAMES = BIT(18), IEEE80211_TX_INTFL_RETRANSMISSION = BIT(19), IEEE80211_TX_INTFL_HAS_RADIOTAP = BIT(20), + IEEE80211_TX_INTFL_NL80211_FRAME_TX = BIT(21), }; /** diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index e1731b7c252..b7116ef84a3 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1,7 +1,7 @@ /* * mac80211 configuration hooks for cfg80211 * - * Copyright 2006, 2007 Johannes Berg + * Copyright 2006-2010 Johannes Berg * * This file is GPLv2 as found in COPYING. */ @@ -1448,6 +1448,15 @@ static int ieee80211_cancel_remain_on_channel(struct wiphy *wiphy, return ieee80211_wk_cancel_remain_on_channel(sdata, cookie); } +static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie) +{ + return ieee80211_mgd_action(IEEE80211_DEV_TO_SUB_IF(dev), chan, + channel_type, buf, len, cookie); +} + struct cfg80211_ops mac80211_config_ops = { .add_virtual_intf = ieee80211_add_iface, .del_virtual_intf = ieee80211_del_iface, @@ -1496,4 +1505,5 @@ struct cfg80211_ops mac80211_config_ops = { .set_bitrate_mask = ieee80211_set_bitrate_mask, .remain_on_channel = ieee80211_remain_on_channel, .cancel_remain_on_channel = ieee80211_cancel_remain_on_channel, + .action = ieee80211_action, }; diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 9dd98b674cb..241533e1bc0 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2,7 +2,7 @@ * Copyright 2002-2005, Instant802 Networks, Inc. * Copyright 2005, Devicescape Software, Inc. * Copyright 2006-2007 Jiri Benc - * Copyright 2007-2008 Johannes Berg + * Copyright 2007-2010 Johannes Berg * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -966,6 +966,10 @@ int ieee80211_mgd_deauth(struct ieee80211_sub_if_data *sdata, int ieee80211_mgd_disassoc(struct ieee80211_sub_if_data *sdata, struct cfg80211_disassoc_request *req, void *cookie); +int ieee80211_mgd_action(struct ieee80211_sub_if_data *sdata, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie); ieee80211_rx_result ieee80211_sta_rx_mgmt(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); void ieee80211_send_pspoll(struct ieee80211_local *local, diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index bfc4a507001..41812a15eea 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2084,3 +2084,38 @@ int ieee80211_mgd_disassoc(struct ieee80211_sub_if_data *sdata, return 0; } + +int ieee80211_mgd_action(struct ieee80211_sub_if_data *sdata, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie) +{ + struct ieee80211_local *local = sdata->local; + struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; + struct sk_buff *skb; + + /* Check that we are on the requested channel for transmission */ + if ((chan != local->tmp_channel || + channel_type != local->tmp_channel_type) && + (chan != local->oper_channel || + channel_type != local->oper_channel_type)) + return -EBUSY; + + skb = dev_alloc_skb(local->hw.extra_tx_headroom + len); + if (!skb) + return -ENOMEM; + skb_reserve(skb, local->hw.extra_tx_headroom); + + memcpy(skb_put(skb, len), buf, len); + + if (!(ifmgd->flags & IEEE80211_STA_MFP_ENABLED)) + IEEE80211_SKB_CB(skb)->flags |= + IEEE80211_TX_INTFL_DONT_ENCRYPT; + IEEE80211_SKB_CB(skb)->flags |= IEEE80211_TX_INTFL_NL80211_FRAME_TX | + IEEE80211_TX_CTL_REQ_TX_STATUS; + skb->dev = sdata->dev; + ieee80211_tx_skb(sdata, skb); + + *cookie = (unsigned long) skb; + return 0; +} diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index a177472adc1..a6080d8d72b 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1856,28 +1856,25 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) struct ieee80211_sub_if_data *sdata = rx->sdata; struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data; struct sk_buff *nskb; + struct ieee80211_rx_status *status; int len = rx->skb->len; if (!ieee80211_is_action(mgmt->frame_control)) return RX_CONTINUE; - if (!rx->sta) + /* drop too small frames */ + if (len < IEEE80211_MIN_ACTION_SIZE) return RX_DROP_UNUSABLE; - if (!(rx->flags & IEEE80211_RX_RA_MATCH)) + if (!rx->sta && mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) return RX_DROP_UNUSABLE; - if (ieee80211_drop_unencrypted(rx, mgmt->frame_control)) + if (!(rx->flags & IEEE80211_RX_RA_MATCH)) return RX_DROP_UNUSABLE; - /* drop too small frames */ - if (len < IEEE80211_MIN_ACTION_SIZE) + if (ieee80211_drop_unencrypted(rx, mgmt->frame_control)) return RX_DROP_UNUSABLE; - /* return action frames that have *only* category */ - if (len < IEEE80211_MIN_ACTION_SIZE + 1) - goto return_frame; - switch (mgmt->u.action.category) { case WLAN_CATEGORY_BACK: /* @@ -1891,6 +1888,10 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) sdata->vif.type != NL80211_IFTYPE_AP) break; + /* verify action_code is present */ + if (len < IEEE80211_MIN_ACTION_SIZE + 1) + break; + switch (mgmt->u.action.u.addba_req.action_code) { case WLAN_ACTION_ADDBA_REQ: if (len < (IEEE80211_MIN_ACTION_SIZE + @@ -1919,6 +1920,10 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) if (sdata->vif.type != NL80211_IFTYPE_STATION) break; + /* verify action_code is present */ + if (len < IEEE80211_MIN_ACTION_SIZE + 1) + break; + switch (mgmt->u.action.u.measurement.action_code) { case WLAN_ACTION_SPCT_MSR_REQ: if (len < (IEEE80211_MIN_ACTION_SIZE + @@ -1954,7 +1959,7 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) } break; } - return_frame: + /* * For AP mode, hostapd is responsible for handling any action * frames that we didn't handle, including returning unknown @@ -1966,6 +1971,20 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) sdata->vif.type == NL80211_IFTYPE_AP_VLAN) return RX_DROP_MONITOR; + /* + * Getting here means the kernel doesn't know how to handle + * it, but maybe userspace does ... include returned frames + * so userspace can register for those to know whether ones + * it transmitted were processed or returned. + */ + status = IEEE80211_SKB_RXCB(rx->skb); + + if (sdata->vif.type == NL80211_IFTYPE_STATION && + cfg80211_rx_action(rx->sdata->dev, status->freq, + rx->skb->data, rx->skb->len, + GFP_ATOMIC)) + goto handled; + /* do not return rejected action frames */ if (mgmt->u.action.category & 0x80) return RX_DROP_UNUSABLE; @@ -1985,7 +2004,8 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) } handled: - rx->sta->rx_packets++; + if (rx->sta) + rx->sta->rx_packets++; dev_kfree_skb(rx->skb); return RX_QUEUED; } diff --git a/net/mac80211/status.c b/net/mac80211/status.c index ded98730c11..56d5b9a6ec5 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -2,7 +2,7 @@ * Copyright 2002-2005, Instant802 Networks, Inc. * Copyright 2005-2006, Devicescape Software, Inc. * Copyright 2006-2007 Jiri Benc - * Copyright 2008-2009 Johannes Berg + * Copyright 2008-2010 Johannes Berg * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -288,6 +288,11 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb) msecs_to_jiffies(10)); } + if (info->flags & IEEE80211_TX_INTFL_NL80211_FRAME_TX) + cfg80211_action_tx_status( + skb->dev, (unsigned long) skb, skb->data, skb->len, + !!(info->flags & IEEE80211_TX_STAT_ACK), GFP_ATOMIC); + /* this was a transmitted frame, but now we want to reuse it */ skb_orphan(skb); diff --git a/net/wireless/core.c b/net/wireless/core.c index 71b6b3a9cf1..51908dc2ea0 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -677,6 +677,9 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, INIT_WORK(&wdev->cleanup_work, wdev_cleanup_work); INIT_LIST_HEAD(&wdev->event_list); spin_lock_init(&wdev->event_lock); + INIT_LIST_HEAD(&wdev->action_registrations); + spin_lock_init(&wdev->action_registrations_lock); + mutex_lock(&rdev->devlist_mtx); list_add_rcu(&wdev->list, &rdev->netdev_list); rdev->devlist_generation++; @@ -792,6 +795,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, sysfs_remove_link(&dev->dev.kobj, "phy80211"); list_del_rcu(&wdev->list); rdev->devlist_generation++; + cfg80211_mlme_purge_actions(wdev); #ifdef CONFIG_CFG80211_WEXT kfree(wdev->wext.keys); #endif diff --git a/net/wireless/core.h b/net/wireless/core.h index c326a667022..d52da913145 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -329,6 +329,15 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *resp_ie, size_t resp_ie_len, u16 status, bool wextev, struct cfg80211_bss *bss); +int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid, + const u8 *match_data, int match_len); +void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid); +void cfg80211_mlme_purge_actions(struct wireless_dev *wdev); +int cfg80211_mlme_action(struct cfg80211_registered_device *rdev, + struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie); /* SME */ int __cfg80211_connect(struct cfg80211_registered_device *rdev, diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index 94d151f6f73..62bc8855e12 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -728,3 +728,169 @@ void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr, nl80211_send_sta_event(rdev, dev, mac_addr, sinfo, gfp); } EXPORT_SYMBOL(cfg80211_new_sta); + +struct cfg80211_action_registration { + struct list_head list; + + u32 nlpid; + + int match_len; + + u8 match[]; +}; + +int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid, + const u8 *match_data, int match_len) +{ + struct cfg80211_action_registration *reg, *nreg; + int err = 0; + + nreg = kzalloc(sizeof(*reg) + match_len, GFP_KERNEL); + if (!nreg) + return -ENOMEM; + + spin_lock_bh(&wdev->action_registrations_lock); + + list_for_each_entry(reg, &wdev->action_registrations, list) { + int mlen = min(match_len, reg->match_len); + + if (memcmp(reg->match, match_data, mlen) == 0) { + err = -EALREADY; + break; + } + } + + if (err) { + kfree(nreg); + goto out; + } + + memcpy(nreg->match, match_data, match_len); + nreg->match_len = match_len; + nreg->nlpid = snd_pid; + list_add(&nreg->list, &wdev->action_registrations); + + out: + spin_unlock_bh(&wdev->action_registrations_lock); + return err; +} + +void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid) +{ + struct cfg80211_action_registration *reg, *tmp; + + spin_lock_bh(&wdev->action_registrations_lock); + + list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) { + if (reg->nlpid == nlpid) { + list_del(®->list); + kfree(reg); + } + } + + spin_unlock_bh(&wdev->action_registrations_lock); +} + +void cfg80211_mlme_purge_actions(struct wireless_dev *wdev) +{ + struct cfg80211_action_registration *reg, *tmp; + + spin_lock_bh(&wdev->action_registrations_lock); + + list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) { + list_del(®->list); + kfree(reg); + } + + spin_unlock_bh(&wdev->action_registrations_lock); +} + +int cfg80211_mlme_action(struct cfg80211_registered_device *rdev, + struct net_device *dev, + struct ieee80211_channel *chan, + enum nl80211_channel_type channel_type, + const u8 *buf, size_t len, u64 *cookie) +{ + struct wireless_dev *wdev = dev->ieee80211_ptr; + const struct ieee80211_mgmt *mgmt; + + if (rdev->ops->action == NULL) + return -EOPNOTSUPP; + if (len < 24 + 1) + return -EINVAL; + + mgmt = (const struct ieee80211_mgmt *) buf; + if (!ieee80211_is_action(mgmt->frame_control)) + return -EINVAL; + if (mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) { + /* Verify that we are associated with the destination AP */ + if (!wdev->current_bss || + memcmp(wdev->current_bss->pub.bssid, mgmt->bssid, + ETH_ALEN) != 0 || + memcmp(wdev->current_bss->pub.bssid, mgmt->da, + ETH_ALEN) != 0) + return -ENOTCONN; + } + + if (memcmp(mgmt->sa, dev->dev_addr, ETH_ALEN) != 0) + return -EINVAL; + + /* Transmit the Action frame as requested by user space */ + return rdev->ops->action(&rdev->wiphy, dev, chan, channel_type, + buf, len, cookie); +} + +bool cfg80211_rx_action(struct net_device *dev, int freq, const u8 *buf, + size_t len, gfp_t gfp) +{ + struct wireless_dev *wdev = dev->ieee80211_ptr; + struct wiphy *wiphy = wdev->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + struct cfg80211_action_registration *reg; + const u8 *action_data; + int action_data_len; + bool result = false; + + /* frame length - min size excluding category */ + action_data_len = len - (IEEE80211_MIN_ACTION_SIZE - 1); + + /* action data starts with category */ + action_data = buf + IEEE80211_MIN_ACTION_SIZE - 1; + + spin_lock_bh(&wdev->action_registrations_lock); + + list_for_each_entry(reg, &wdev->action_registrations, list) { + if (reg->match_len > action_data_len) + continue; + + if (memcmp(reg->match, action_data, reg->match_len)) + continue; + + /* found match! */ + + /* Indicate the received Action frame to user space */ + if (nl80211_send_action(rdev, dev, reg->nlpid, freq, + buf, len, gfp)) + continue; + + result = true; + break; + } + + spin_unlock_bh(&wdev->action_registrations_lock); + + return result; +} +EXPORT_SYMBOL(cfg80211_rx_action); + +void cfg80211_action_tx_status(struct net_device *dev, u64 cookie, + const u8 *buf, size_t len, bool ack, gfp_t gfp) +{ + struct wireless_dev *wdev = dev->ieee80211_ptr; + struct wiphy *wiphy = wdev->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + + /* Indicate TX status of the Action frame to user space */ + nl80211_send_action_tx_status(rdev, dev, cookie, buf, len, ack, gfp); +} +EXPORT_SYMBOL(cfg80211_action_tx_status); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index a95ab9e4c19..32811208135 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1,7 +1,7 @@ /* * This is the new netlink-based wireless configuration interface. * - * Copyright 2006-2009 Johannes Berg + * Copyright 2006-2010 Johannes Berg */ #include @@ -145,6 +145,9 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { [NL80211_ATTR_DURATION] = { .type = NLA_U32 }, [NL80211_ATTR_COOKIE] = { .type = NLA_U64 }, [NL80211_ATTR_TX_RATES] = { .type = NLA_NESTED }, + [NL80211_ATTR_FRAME] = { .type = NLA_BINARY, + .len = IEEE80211_MAX_DATA_LEN }, + [NL80211_ATTR_FRAME_MATCH] = { .type = NLA_BINARY, }, }; /* policy for the attributes */ @@ -577,6 +580,7 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, CMD(flush_pmksa, FLUSH_PMKSA); CMD(remain_on_channel, REMAIN_ON_CHANNEL); CMD(set_bitrate_mask, SET_TX_BITRATE_MASK); + CMD(action, ACTION); if (dev->wiphy.flags & WIPHY_FLAG_NETNS_OK) { i++; NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS); @@ -4526,6 +4530,139 @@ static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb, return err; } +static int nl80211_register_action(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct net_device *dev; + int err; + + if (!info->attrs[NL80211_ATTR_FRAME_MATCH]) + return -EINVAL; + + if (nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH]) < 1) + return -EINVAL; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { + err = -EOPNOTSUPP; + goto out; + } + + /* not much point in registering if we can't reply */ + if (!rdev->ops->action) { + err = -EOPNOTSUPP; + goto out; + } + + err = cfg80211_mlme_register_action(dev->ieee80211_ptr, info->snd_pid, + nla_data(info->attrs[NL80211_ATTR_FRAME_MATCH]), + nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH])); + out: + cfg80211_unlock_rdev(rdev); + dev_put(dev); + unlock_rtnl: + rtnl_unlock(); + return err; +} + +static int nl80211_action(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct net_device *dev; + struct ieee80211_channel *chan; + enum nl80211_channel_type channel_type = NL80211_CHAN_NO_HT; + u32 freq; + int err; + void *hdr; + u64 cookie; + struct sk_buff *msg; + + if (!info->attrs[NL80211_ATTR_FRAME] || + !info->attrs[NL80211_ATTR_WIPHY_FREQ]) + return -EINVAL; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + if (!rdev->ops->action) { + err = -EOPNOTSUPP; + goto out; + } + + if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) { + err = -EOPNOTSUPP; + goto out; + } + + if (!netif_running(dev)) { + err = -ENETDOWN; + goto out; + } + + if (info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) { + channel_type = nla_get_u32( + info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]); + if (channel_type != NL80211_CHAN_NO_HT && + channel_type != NL80211_CHAN_HT20 && + channel_type != NL80211_CHAN_HT40PLUS && + channel_type != NL80211_CHAN_HT40MINUS) + err = -EINVAL; + goto out; + } + + freq = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ]); + chan = rdev_freq_to_chan(rdev, freq, channel_type); + if (chan == NULL) { + err = -EINVAL; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + err = -ENOMEM; + goto out; + } + + hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, + NL80211_CMD_ACTION); + + if (IS_ERR(hdr)) { + err = PTR_ERR(hdr); + goto free_msg; + } + err = cfg80211_mlme_action(rdev, dev, chan, channel_type, + nla_data(info->attrs[NL80211_ATTR_FRAME]), + nla_len(info->attrs[NL80211_ATTR_FRAME]), + &cookie); + if (err) + goto free_msg; + + NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); + + genlmsg_end(msg, hdr); + err = genlmsg_reply(msg, info); + goto out; + + nla_put_failure: + err = -ENOBUFS; + free_msg: + nlmsg_free(msg); + out: + cfg80211_unlock_rdev(rdev); + dev_put(dev); +unlock_rtnl: + rtnl_unlock(); + return err; +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -4806,6 +4943,18 @@ static struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, + { + .cmd = NL80211_CMD_REGISTER_ACTION, + .doit = nl80211_register_action, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_ACTION, + .doit = nl80211_action, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, }; static struct genl_multicast_group nl80211_mlme_mcgrp = { @@ -5478,6 +5627,110 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, nl80211_mlme_mcgrp.id, gfp); } +int nl80211_send_action(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u32 nlpid, + int freq, const u8 *buf, size_t len, gfp_t gfp) +{ + struct sk_buff *msg; + void *hdr; + int err; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp); + if (!msg) + return -ENOMEM; + + hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION); + if (!hdr) { + nlmsg_free(msg); + return -ENOMEM; + } + + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx); + NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex); + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY_FREQ, freq); + NLA_PUT(msg, NL80211_ATTR_FRAME, len, buf); + + err = genlmsg_end(msg, hdr); + if (err < 0) { + nlmsg_free(msg); + return err; + } + + err = genlmsg_unicast(wiphy_net(&rdev->wiphy), msg, nlpid); + if (err < 0) + return err; + return 0; + + nla_put_failure: + genlmsg_cancel(msg, hdr); + nlmsg_free(msg); + return -ENOBUFS; +} + +void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + const u8 *buf, size_t len, bool ack, + gfp_t gfp) +{ + struct sk_buff *msg; + void *hdr; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp); + if (!msg) + return; + + hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION_TX_STATUS); + if (!hdr) { + nlmsg_free(msg); + return; + } + + NLA_PUT_U32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx); + NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex); + NLA_PUT(msg, NL80211_ATTR_FRAME, len, buf); + NLA_PUT_U64(msg, NL80211_ATTR_COOKIE, cookie); + if (ack) + NLA_PUT_FLAG(msg, NL80211_ATTR_ACK); + + if (genlmsg_end(msg, hdr) < 0) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast(msg, 0, nl80211_mlme_mcgrp.id, gfp); + return; + + nla_put_failure: + genlmsg_cancel(msg, hdr); + nlmsg_free(msg); +} + +static int nl80211_netlink_notify(struct notifier_block * nb, + unsigned long state, + void *_notify) +{ + struct netlink_notify *notify = _notify; + struct cfg80211_registered_device *rdev; + struct wireless_dev *wdev; + + if (state != NETLINK_URELEASE) + return NOTIFY_DONE; + + rcu_read_lock(); + + list_for_each_entry_rcu(rdev, &cfg80211_rdev_list, list) + list_for_each_entry_rcu(wdev, &rdev->netdev_list, list) + cfg80211_mlme_unregister_actions(wdev, notify->pid); + + rcu_read_unlock(); + + return NOTIFY_DONE; +} + +static struct notifier_block nl80211_netlink_notifier = { + .notifier_call = nl80211_netlink_notify, +}; + /* initialisation/exit functions */ int nl80211_init(void) @@ -5511,6 +5764,10 @@ int nl80211_init(void) goto err_out; #endif + err = netlink_register_notifier(&nl80211_netlink_notifier); + if (err) + goto err_out; + return 0; err_out: genl_unregister_family(&nl80211_fam); @@ -5519,5 +5776,6 @@ int nl80211_init(void) void nl80211_exit(void) { + netlink_unregister_notifier(&nl80211_netlink_notifier); genl_unregister_family(&nl80211_fam); } diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index 14855b8fb43..4ca511102c6 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -74,4 +74,12 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev, struct net_device *dev, const u8 *mac_addr, struct station_info *sinfo, gfp_t gfp); +int nl80211_send_action(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u32 nlpid, int freq, + const u8 *buf, size_t len, gfp_t gfp); +void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev, + struct net_device *netdev, u64 cookie, + const u8 *buf, size_t len, bool ack, + gfp_t gfp); + #endif /* __NET_WIRELESS_NL80211_H */ -- cgit v1.2.3-70-g09d2 From 02291680ffba92e5b5865bc0c5e7d1f3056b80ec Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 14 Feb 2010 03:25:51 +0000 Subject: net ipv4: Decouple ipv4 interface parameters from binary sysctl numbers Stop using the binary sysctl enumeartion in sysctl.h as an index into a per interface array. This leads to unnecessary binary sysctl number allocation, and a fragility in data structure and implementation because of unnecessary coupling. Signed-off-by: Eric W. Biederman Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 42 ++++++++++++++++++++++++++++++++++++------ include/linux/sysctl.h | 4 ---- net/ipv4/devinet.c | 26 +++++++++++++------------- 3 files changed, 49 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index cf257809771..2be1a1a2beb 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -10,10 +10,40 @@ #include #include +enum +{ + IPV4_DEVCONF_FORWARDING=1, + IPV4_DEVCONF_MC_FORWARDING, + IPV4_DEVCONF_PROXY_ARP, + IPV4_DEVCONF_ACCEPT_REDIRECTS, + IPV4_DEVCONF_SECURE_REDIRECTS, + IPV4_DEVCONF_SEND_REDIRECTS, + IPV4_DEVCONF_SHARED_MEDIA, + IPV4_DEVCONF_RP_FILTER, + IPV4_DEVCONF_ACCEPT_SOURCE_ROUTE, + IPV4_DEVCONF_BOOTP_RELAY, + IPV4_DEVCONF_LOG_MARTIANS, + IPV4_DEVCONF_TAG, + IPV4_DEVCONF_ARPFILTER, + IPV4_DEVCONF_MEDIUM_ID, + IPV4_DEVCONF_NOXFRM, + IPV4_DEVCONF_NOPOLICY, + IPV4_DEVCONF_FORCE_IGMP_VERSION, + IPV4_DEVCONF_ARP_ANNOUNCE, + IPV4_DEVCONF_ARP_IGNORE, + IPV4_DEVCONF_PROMOTE_SECONDARIES, + IPV4_DEVCONF_ARP_ACCEPT, + IPV4_DEVCONF_ARP_NOTIFY, + IPV4_DEVCONF_ACCEPT_LOCAL, + IPV4_DEVCONF_SRC_VMARK, + IPV4_DEVCONF_PROXY_ARP_PVLAN, + __IPV4_DEVCONF_MAX +}; + struct ipv4_devconf { void *sysctl; - int data[__NET_IPV4_CONF_MAX - 1]; - DECLARE_BITMAP(state, __NET_IPV4_CONF_MAX - 1); + int data[__IPV4_DEVCONF_MAX - 1]; + DECLARE_BITMAP(state, __IPV4_DEVCONF_MAX - 1); }; struct in_device { @@ -40,7 +70,7 @@ struct in_device { struct rcu_head rcu_head; }; -#define IPV4_DEVCONF(cnf, attr) ((cnf).data[NET_IPV4_CONF_ ## attr - 1]) +#define IPV4_DEVCONF(cnf, attr) ((cnf).data[IPV4_DEVCONF_ ## attr - 1]) #define IPV4_DEVCONF_ALL(net, attr) \ IPV4_DEVCONF((*(net)->ipv4.devconf_all), attr) @@ -60,13 +90,13 @@ static inline void ipv4_devconf_set(struct in_device *in_dev, int index, static inline void ipv4_devconf_setall(struct in_device *in_dev) { - bitmap_fill(in_dev->cnf.state, __NET_IPV4_CONF_MAX - 1); + bitmap_fill(in_dev->cnf.state, __IPV4_DEVCONF_MAX - 1); } #define IN_DEV_CONF_GET(in_dev, attr) \ - ipv4_devconf_get((in_dev), NET_IPV4_CONF_ ## attr) + ipv4_devconf_get((in_dev), IPV4_DEVCONF_ ## attr) #define IN_DEV_CONF_SET(in_dev, attr, val) \ - ipv4_devconf_set((in_dev), NET_IPV4_CONF_ ## attr, (val)) + ipv4_devconf_set((in_dev), IPV4_DEVCONF_ ## attr, (val)) #define IN_DEV_ANDCONF(in_dev, attr) \ (IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr) && \ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 9f236cdcf3f..7c74e919cab 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -481,10 +481,6 @@ enum NET_IPV4_CONF_PROMOTE_SECONDARIES=20, NET_IPV4_CONF_ARP_ACCEPT=21, NET_IPV4_CONF_ARP_NOTIFY=22, - NET_IPV4_CONF_ACCEPT_LOCAL=23, - NET_IPV4_CONF_SRC_VMARK=24, - NET_IPV4_CONF_PROXY_ARP_PVLAN=25, - __NET_IPV4_CONF_MAX }; /* /proc/sys/net/ipv4/netfilter */ diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index cd71a390839..b1eddee9bf9 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -64,20 +64,20 @@ static struct ipv4_devconf ipv4_devconf = { .data = { - [NET_IPV4_CONF_ACCEPT_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SEND_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SECURE_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SHARED_MEDIA - 1] = 1, + [IPV4_DEVCONF_ACCEPT_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SEND_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SECURE_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SHARED_MEDIA - 1] = 1, }, }; static struct ipv4_devconf ipv4_devconf_dflt = { .data = { - [NET_IPV4_CONF_ACCEPT_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SEND_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SECURE_REDIRECTS - 1] = 1, - [NET_IPV4_CONF_SHARED_MEDIA - 1] = 1, - [NET_IPV4_CONF_ACCEPT_SOURCE_ROUTE - 1] = 1, + [IPV4_DEVCONF_ACCEPT_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SEND_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SECURE_REDIRECTS - 1] = 1, + [IPV4_DEVCONF_SHARED_MEDIA - 1] = 1, + [IPV4_DEVCONF_ACCEPT_SOURCE_ROUTE - 1] = 1, }, }; @@ -1360,7 +1360,7 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write, { \ .procname = name, \ .data = ipv4_devconf.data + \ - NET_IPV4_CONF_ ## attr - 1, \ + IPV4_DEVCONF_ ## attr - 1, \ .maxlen = sizeof(int), \ .mode = mval, \ .proc_handler = proc, \ @@ -1381,7 +1381,7 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write, static struct devinet_sysctl_table { struct ctl_table_header *sysctl_header; - struct ctl_table devinet_vars[__NET_IPV4_CONF_MAX]; + struct ctl_table devinet_vars[__IPV4_DEVCONF_MAX]; char *dev_name; } devinet_sysctl = { .devinet_vars = { @@ -1503,7 +1503,7 @@ static struct ctl_table ctl_forward_entry[] = { { .procname = "ip_forward", .data = &ipv4_devconf.data[ - NET_IPV4_CONF_FORWARDING - 1], + IPV4_DEVCONF_FORWARDING - 1], .maxlen = sizeof(int), .mode = 0644, .proc_handler = devinet_sysctl_forward, @@ -1547,7 +1547,7 @@ static __net_init int devinet_init_net(struct net *net) if (tbl == NULL) goto err_alloc_ctl; - tbl[0].data = &all->data[NET_IPV4_CONF_FORWARDING - 1]; + tbl[0].data = &all->data[IPV4_DEVCONF_FORWARDING - 1]; tbl[0].extra1 = all; tbl[0].extra2 = net; #endif -- cgit v1.2.3-70-g09d2 From 54716e3beb0ab20c49471348dfe399a71bfc8fd3 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 14 Feb 2010 03:27:03 +0000 Subject: net neigh: Decouple per interface neighbour table controls from binary sysctls Stop computing the number of neighbour table settings we have by counting the number of binary sysctls. This behaviour was silly and meant that we could not add another neighbour table setting without also adding another binary sysctl. Don't pass the binary sysctl path for neighour table entries into neigh_sysctl_register. These parameters are no longer used and so are just dead code. Signed-off-by: Eric W. Biederman Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 - include/net/neighbour.h | 1 - net/core/neighbour.c | 7 ++++--- net/ipv4/arp.c | 3 +-- net/ipv4/devinet.c | 3 +-- net/ipv6/addrconf.c | 3 +-- net/ipv6/ndisc.c | 3 +-- 7 files changed, 8 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 7c74e919cab..f66014c90c9 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -596,7 +596,6 @@ enum { NET_NEIGH_GC_THRESH3=16, NET_NEIGH_RETRANS_TIME_MS=17, NET_NEIGH_REACHABLE_TIME_MS=18, - __NET_NEIGH_MAX }; /* /proc/sys/net/dccp */ diff --git a/include/net/neighbour.h b/include/net/neighbour.h index b0173202cad..7834f470d14 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -251,7 +251,6 @@ extern void neigh_seq_stop(struct seq_file *, void *); extern int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, - int p_id, int pdev_id, char *p_name, proc_handler *proc_handler); extern void neigh_sysctl_unregister(struct neigh_parms *p); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index f2efd72da79..d102f6d9abd 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2556,9 +2556,11 @@ EXPORT_SYMBOL(neigh_app_ns); #ifdef CONFIG_SYSCTL +#define NEIGH_VARS_MAX 19 + static struct neigh_sysctl_table { struct ctl_table_header *sysctl_header; - struct ctl_table neigh_vars[__NET_NEIGH_MAX]; + struct ctl_table neigh_vars[NEIGH_VARS_MAX]; char *dev_name; } neigh_sysctl_template __read_mostly = { .neigh_vars = { @@ -2675,8 +2677,7 @@ static struct neigh_sysctl_table { }; int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, - int p_id, int pdev_id, char *p_name, - proc_handler *handler) + char *p_name, proc_handler *handler) { struct neigh_sysctl_table *t; const char *dev_name_source = NULL; diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 1940b4df769..c4dd1354280 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1284,8 +1284,7 @@ void __init arp_init(void) dev_add_pack(&arp_packet_type); arp_proc_init(); #ifdef CONFIG_SYSCTL - neigh_sysctl_register(NULL, &arp_tbl.parms, NET_IPV4, - NET_IPV4_NEIGH, "ipv4", NULL); + neigh_sysctl_register(NULL, &arp_tbl.parms, "ipv4", NULL); #endif register_netdevice_notifier(&arp_netdev_notifier); } diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index b1eddee9bf9..014982b6156 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1487,8 +1487,7 @@ static void __devinet_sysctl_unregister(struct ipv4_devconf *cnf) static void devinet_sysctl_register(struct in_device *idev) { - neigh_sysctl_register(idev->dev, idev->arp_parms, NET_IPV4, - NET_IPV4_NEIGH, "ipv4", NULL); + neigh_sysctl_register(idev->dev, idev->arp_parms, "ipv4", NULL); __devinet_sysctl_register(dev_net(idev->dev), idev->dev->name, &idev->cnf); } diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 764ad37ca07..c79cbff5437 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4416,8 +4416,7 @@ static void __addrconf_sysctl_unregister(struct ipv6_devconf *p) static void addrconf_sysctl_register(struct inet6_dev *idev) { - neigh_sysctl_register(idev->dev, idev->nd_parms, NET_IPV6, - NET_IPV6_NEIGH, "ipv6", + neigh_sysctl_register(idev->dev, idev->nd_parms, "ipv6", &ndisc_ifinfo_sysctl_change); __addrconf_sysctl_register(dev_net(idev->dev), idev->dev->name, idev, &idev->cnf); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 2dfec6bb8ad..8bcc4b7db3b 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1820,8 +1820,7 @@ int __init ndisc_init(void) neigh_table_init(&nd_tbl); #ifdef CONFIG_SYSCTL - err = neigh_sysctl_register(NULL, &nd_tbl.parms, NET_IPV6, - NET_IPV6_NEIGH, "ipv6", + err = neigh_sysctl_register(NULL, &nd_tbl.parms, "ipv6", &ndisc_ifinfo_sysctl_change); if (err) goto out_unregister_pernet; -- cgit v1.2.3-70-g09d2 From 522530311b35ec8fc4785062441dd2d63967ac55 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 8 Feb 2010 10:12:10 +0000 Subject: CAPI: Call a controller 'controller', not 'card' At least for our internal use, fix the misnomers that refer to a CAPI controller as 'card'. No functional changes. Signed-off-by: Jan Kiszka Signed-off-by: David S. Miller --- drivers/isdn/capi/kcapi.c | 320 +++++++++++++++++++++-------------------- drivers/isdn/capi/kcapi.h | 8 +- drivers/isdn/capi/kcapi_proc.c | 17 ++- include/linux/isdn/capilli.h | 2 +- 4 files changed, 178 insertions(+), 169 deletions(-) (limited to 'include') diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index ef564ee1c06..f37c13b4ddc 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -67,24 +67,24 @@ static DEFINE_RWLOCK(application_lock); static DEFINE_MUTEX(controller_mutex); struct capi20_appl *capi_applications[CAPI_MAXAPPL]; -struct capi_ctr *capi_cards[CAPI_MAXCONTR]; +struct capi_ctr *capi_controller[CAPI_MAXCONTR]; -static int ncards; +static int ncontrollers; /* -------- controller ref counting -------------------------------------- */ static inline struct capi_ctr * -capi_ctr_get(struct capi_ctr *card) +capi_ctr_get(struct capi_ctr *ctr) { - if (!try_module_get(card->owner)) + if (!try_module_get(ctr->owner)) return NULL; - return card; + return ctr; } static inline void -capi_ctr_put(struct capi_ctr *card) +capi_ctr_put(struct capi_ctr *ctr) { - module_put(card->owner); + module_put(ctr->owner); } /* ------------------------------------------------------------- */ @@ -94,7 +94,7 @@ static inline struct capi_ctr *get_capi_ctr_by_nr(u16 contr) if (contr - 1 >= CAPI_MAXCONTR) return NULL; - return capi_cards[contr - 1]; + return capi_controller[contr - 1]; } static inline struct capi20_appl *get_capi_appl_by_nr(u16 applid) @@ -144,46 +144,48 @@ static inline int capi_subcmd_valid(u8 subcmd) /* ------------------------------------------------------------ */ -static void register_appl(struct capi_ctr *card, u16 applid, capi_register_params *rparam) +static void +register_appl(struct capi_ctr *ctr, u16 applid, capi_register_params *rparam) { - card = capi_ctr_get(card); + ctr = capi_ctr_get(ctr); - if (card) - card->register_appl(card, applid, rparam); + if (ctr) + ctr->register_appl(ctr, applid, rparam); else - printk(KERN_WARNING "%s: cannot get card resources\n", __func__); + printk(KERN_WARNING "%s: cannot get controller resources\n", + __func__); } -static void release_appl(struct capi_ctr *card, u16 applid) +static void release_appl(struct capi_ctr *ctr, u16 applid) { DBG("applid %#x", applid); - card->release_appl(card, applid); - capi_ctr_put(card); + ctr->release_appl(ctr, applid); + capi_ctr_put(ctr); } /* -------- KCI_CONTRUP --------------------------------------- */ static void notify_up(u32 contr) { - struct capi_ctr *card = get_capi_ctr_by_nr(contr); + struct capi_ctr *ctr = get_capi_ctr_by_nr(contr); struct capi20_appl *ap; u16 applid; if (showcapimsgs & 1) { printk(KERN_DEBUG "kcapi: notify up contr %d\n", contr); } - if (!card) { + if (!ctr) { printk(KERN_WARNING "%s: invalid contr %d\n", __func__, contr); return; } for (applid = 1; applid <= CAPI_MAXAPPL; applid++) { ap = get_capi_appl_by_nr(applid); if (!ap || ap->release_in_progress) continue; - register_appl(card, applid, &ap->rparam); + register_appl(ctr, applid, &ap->rparam); if (ap->callback && !ap->release_in_progress) - ap->callback(KCI_CONTRUP, contr, &card->profile); + ap->callback(KCI_CONTRUP, contr, &ctr->profile); } } @@ -269,14 +271,15 @@ static void recv_handler(struct work_struct *work) /** * capi_ctr_handle_message() - handle incoming CAPI message - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * @appl: application ID. * @skb: message. * * Called by hardware driver to pass a CAPI message to the application. */ -void capi_ctr_handle_message(struct capi_ctr * card, u16 appl, struct sk_buff *skb) +void capi_ctr_handle_message(struct capi_ctr *ctr, u16 appl, + struct sk_buff *skb) { struct capi20_appl *ap; int showctl = 0; @@ -284,43 +287,45 @@ void capi_ctr_handle_message(struct capi_ctr * card, u16 appl, struct sk_buff *s unsigned long flags; _cdebbuf *cdb; - if (card->cardstate != CARD_RUNNING) { + if (ctr->state != CAPI_CTR_RUNNING) { cdb = capi_message2str(skb->data); if (cdb) { printk(KERN_INFO "kcapi: controller [%03d] not active, got: %s", - card->cnr, cdb->buf); + ctr->cnr, cdb->buf); cdebbuf_free(cdb); } else printk(KERN_INFO "kcapi: controller [%03d] not active, cannot trace\n", - card->cnr); + ctr->cnr); goto error; } cmd = CAPIMSG_COMMAND(skb->data); subcmd = CAPIMSG_SUBCOMMAND(skb->data); if (cmd == CAPI_DATA_B3 && subcmd == CAPI_IND) { - card->nrecvdatapkt++; - if (card->traceflag > 2) showctl |= 2; + ctr->nrecvdatapkt++; + if (ctr->traceflag > 2) + showctl |= 2; } else { - card->nrecvctlpkt++; - if (card->traceflag) showctl |= 2; + ctr->nrecvctlpkt++; + if (ctr->traceflag) + showctl |= 2; } - showctl |= (card->traceflag & 1); + showctl |= (ctr->traceflag & 1); if (showctl & 2) { if (showctl & 1) { printk(KERN_DEBUG "kcapi: got [%03d] id#%d %s len=%u\n", - card->cnr, CAPIMSG_APPID(skb->data), + ctr->cnr, CAPIMSG_APPID(skb->data), capi_cmd2str(cmd, subcmd), CAPIMSG_LEN(skb->data)); } else { cdb = capi_message2str(skb->data); if (cdb) { printk(KERN_DEBUG "kcapi: got [%03d] %s\n", - card->cnr, cdb->buf); + ctr->cnr, cdb->buf); cdebbuf_free(cdb); } else printk(KERN_DEBUG "kcapi: got [%03d] id#%d %s len=%u, cannot trace\n", - card->cnr, CAPIMSG_APPID(skb->data), + ctr->cnr, CAPIMSG_APPID(skb->data), capi_cmd2str(cmd, subcmd), CAPIMSG_LEN(skb->data)); } @@ -356,74 +361,75 @@ EXPORT_SYMBOL(capi_ctr_handle_message); /** * capi_ctr_ready() - signal CAPI controller ready - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to signal that the controller is up and running. */ -void capi_ctr_ready(struct capi_ctr * card) +void capi_ctr_ready(struct capi_ctr *ctr) { - card->cardstate = CARD_RUNNING; + ctr->state = CAPI_CTR_RUNNING; - printk(KERN_NOTICE "kcapi: card [%03d] \"%s\" ready.\n", - card->cnr, card->name); + printk(KERN_NOTICE "kcapi: controller [%03d] \"%s\" ready.\n", + ctr->cnr, ctr->name); - notify_push(KCI_CONTRUP, card->cnr, 0, 0); + notify_push(KCI_CONTRUP, ctr->cnr, 0, 0); } EXPORT_SYMBOL(capi_ctr_ready); /** * capi_ctr_down() - signal CAPI controller not ready - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to signal that the controller is down and * unavailable for use. */ -void capi_ctr_down(struct capi_ctr * card) +void capi_ctr_down(struct capi_ctr *ctr) { u16 appl; DBG(""); - if (card->cardstate == CARD_DETECTED) + if (ctr->state == CAPI_CTR_DETECTED) return; - card->cardstate = CARD_DETECTED; + ctr->state = CAPI_CTR_DETECTED; - memset(card->manu, 0, sizeof(card->manu)); - memset(&card->version, 0, sizeof(card->version)); - memset(&card->profile, 0, sizeof(card->profile)); - memset(card->serial, 0, sizeof(card->serial)); + memset(ctr->manu, 0, sizeof(ctr->manu)); + memset(&ctr->version, 0, sizeof(ctr->version)); + memset(&ctr->profile, 0, sizeof(ctr->profile)); + memset(ctr->serial, 0, sizeof(ctr->serial)); for (appl = 1; appl <= CAPI_MAXAPPL; appl++) { struct capi20_appl *ap = get_capi_appl_by_nr(appl); if (!ap || ap->release_in_progress) continue; - capi_ctr_put(card); + capi_ctr_put(ctr); } - printk(KERN_NOTICE "kcapi: card [%03d] down.\n", card->cnr); + printk(KERN_NOTICE "kcapi: controller [%03d] down.\n", ctr->cnr); - notify_push(KCI_CONTRDOWN, card->cnr, 0, 0); + notify_push(KCI_CONTRDOWN, ctr->cnr, 0, 0); } EXPORT_SYMBOL(capi_ctr_down); /** * capi_ctr_suspend_output() - suspend controller - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to stop data flow. */ -void capi_ctr_suspend_output(struct capi_ctr *card) +void capi_ctr_suspend_output(struct capi_ctr *ctr) { - if (!card->blocked) { - printk(KERN_DEBUG "kcapi: card [%03d] suspend\n", card->cnr); - card->blocked = 1; + if (!ctr->blocked) { + printk(KERN_DEBUG "kcapi: controller [%03d] suspend\n", + ctr->cnr); + ctr->blocked = 1; } } @@ -431,16 +437,17 @@ EXPORT_SYMBOL(capi_ctr_suspend_output); /** * capi_ctr_resume_output() - resume controller - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to resume data flow. */ -void capi_ctr_resume_output(struct capi_ctr *card) +void capi_ctr_resume_output(struct capi_ctr *ctr) { - if (card->blocked) { - printk(KERN_DEBUG "kcapi: card [%03d] resume\n", card->cnr); - card->blocked = 0; + if (ctr->blocked) { + printk(KERN_DEBUG "kcapi: controller [%03d] resumed\n", + ctr->cnr); + ctr->blocked = 0; } } @@ -450,21 +457,20 @@ EXPORT_SYMBOL(capi_ctr_resume_output); /** * attach_capi_ctr() - register CAPI controller - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to register a controller with the CAPI subsystem. * Return value: 0 on success, error code < 0 on error */ -int -attach_capi_ctr(struct capi_ctr *card) +int attach_capi_ctr(struct capi_ctr *ctr) { int i; mutex_lock(&controller_mutex); for (i = 0; i < CAPI_MAXCONTR; i++) { - if (capi_cards[i] == NULL) + if (!capi_controller[i]) break; } if (i == CAPI_MAXCONTR) { @@ -472,25 +478,25 @@ attach_capi_ctr(struct capi_ctr *card) printk(KERN_ERR "kcapi: out of controller slots\n"); return -EBUSY; } - capi_cards[i] = card; + capi_controller[i] = ctr; mutex_unlock(&controller_mutex); - card->nrecvctlpkt = 0; - card->nrecvdatapkt = 0; - card->nsentctlpkt = 0; - card->nsentdatapkt = 0; - card->cnr = i + 1; - card->cardstate = CARD_DETECTED; - card->blocked = 0; - card->traceflag = showcapimsgs; - - sprintf(card->procfn, "capi/controllers/%d", card->cnr); - card->procent = proc_create_data(card->procfn, 0, NULL, card->proc_fops, card); - - ncards++; - printk(KERN_NOTICE "kcapi: Controller [%03d]: %s attached\n", - card->cnr, card->name); + ctr->nrecvctlpkt = 0; + ctr->nrecvdatapkt = 0; + ctr->nsentctlpkt = 0; + ctr->nsentdatapkt = 0; + ctr->cnr = i + 1; + ctr->state = CAPI_CTR_DETECTED; + ctr->blocked = 0; + ctr->traceflag = showcapimsgs; + + sprintf(ctr->procfn, "capi/controllers/%d", ctr->cnr); + ctr->procent = proc_create_data(ctr->procfn, 0, NULL, ctr->proc_fops, ctr); + + ncontrollers++; + printk(KERN_NOTICE "kcapi: controller [%03d]: %s attached\n", + ctr->cnr, ctr->name); return 0; } @@ -498,27 +504,27 @@ EXPORT_SYMBOL(attach_capi_ctr); /** * detach_capi_ctr() - unregister CAPI controller - * @card: controller descriptor structure. + * @ctr: controller descriptor structure. * * Called by hardware driver to remove the registration of a controller * with the CAPI subsystem. * Return value: 0 on success, error code < 0 on error */ -int detach_capi_ctr(struct capi_ctr *card) +int detach_capi_ctr(struct capi_ctr *ctr) { - if (card->cardstate != CARD_DETECTED) - capi_ctr_down(card); + if (ctr->state != CAPI_CTR_DETECTED) + capi_ctr_down(ctr); - ncards--; + ncontrollers--; - if (card->procent) { - remove_proc_entry(card->procfn, NULL); - card->procent = NULL; + if (ctr->procent) { + remove_proc_entry(ctr->procfn, NULL); + ctr->procent = NULL; } - capi_cards[card->cnr - 1] = NULL; - printk(KERN_NOTICE "kcapi: Controller [%03d]: %s unregistered\n", - card->cnr, card->name); + capi_controller[ctr->cnr - 1] = NULL; + printk(KERN_NOTICE "kcapi: controller [%03d]: %s unregistered\n", + ctr->cnr, ctr->name); return 0; } @@ -576,7 +582,8 @@ u16 capi20_isinstalled(void) { int i; for (i = 0; i < CAPI_MAXCONTR; i++) { - if (capi_cards[i] && capi_cards[i]->cardstate == CARD_RUNNING) + if (capi_controller[i] && + capi_controller[i]->state == CAPI_CTR_RUNNING) return CAPI_NOERROR; } return CAPI_REGNOTINSTALLED; @@ -635,9 +642,10 @@ u16 capi20_register(struct capi20_appl *ap) mutex_lock(&controller_mutex); for (i = 0; i < CAPI_MAXCONTR; i++) { - if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING) + if (!capi_controller[i] || + capi_controller[i]->state != CAPI_CTR_RUNNING) continue; - register_appl(capi_cards[i], applid, &ap->rparam); + register_appl(capi_controller[i], applid, &ap->rparam); } mutex_unlock(&controller_mutex); @@ -674,9 +682,10 @@ u16 capi20_release(struct capi20_appl *ap) mutex_lock(&controller_mutex); for (i = 0; i < CAPI_MAXCONTR; i++) { - if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING) + if (!capi_controller[i] || + capi_controller[i]->state != CAPI_CTR_RUNNING) continue; - release_appl(capi_cards[i], ap->applid); + release_appl(capi_controller[i], ap->applid); } mutex_unlock(&controller_mutex); @@ -703,13 +712,13 @@ EXPORT_SYMBOL(capi20_release); u16 capi20_put_message(struct capi20_appl *ap, struct sk_buff *skb) { - struct capi_ctr *card; + struct capi_ctr *ctr; int showctl = 0; u8 cmd, subcmd; DBG("applid %#x", ap->applid); - if (ncards == 0) + if (ncontrollers == 0) return CAPI_REGNOTINSTALLED; if ((ap->applid == 0) || ap->release_in_progress) return CAPI_ILLAPPNR; @@ -717,28 +726,30 @@ u16 capi20_put_message(struct capi20_appl *ap, struct sk_buff *skb) || !capi_cmd_valid(CAPIMSG_COMMAND(skb->data)) || !capi_subcmd_valid(CAPIMSG_SUBCOMMAND(skb->data))) return CAPI_ILLCMDORSUBCMDORMSGTOSMALL; - card = get_capi_ctr_by_nr(CAPIMSG_CONTROLLER(skb->data)); - if (!card || card->cardstate != CARD_RUNNING) { - card = get_capi_ctr_by_nr(1); // XXX why? - if (!card || card->cardstate != CARD_RUNNING) + ctr = get_capi_ctr_by_nr(CAPIMSG_CONTROLLER(skb->data)); + if (!ctr || ctr->state != CAPI_CTR_RUNNING) { + ctr = get_capi_ctr_by_nr(1); /* XXX why? */ + if (!ctr || ctr->state != CAPI_CTR_RUNNING) return CAPI_REGNOTINSTALLED; } - if (card->blocked) + if (ctr->blocked) return CAPI_SENDQUEUEFULL; cmd = CAPIMSG_COMMAND(skb->data); subcmd = CAPIMSG_SUBCOMMAND(skb->data); if (cmd == CAPI_DATA_B3 && subcmd== CAPI_REQ) { - card->nsentdatapkt++; + ctr->nsentdatapkt++; ap->nsentdatapkt++; - if (card->traceflag > 2) showctl |= 2; + if (ctr->traceflag > 2) + showctl |= 2; } else { - card->nsentctlpkt++; + ctr->nsentctlpkt++; ap->nsentctlpkt++; - if (card->traceflag) showctl |= 2; + if (ctr->traceflag) + showctl |= 2; } - showctl |= (card->traceflag & 1); + showctl |= (ctr->traceflag & 1); if (showctl & 2) { if (showctl & 1) { printk(KERN_DEBUG "kcapi: put [%03d] id#%d %s len=%u\n", @@ -761,7 +772,7 @@ u16 capi20_put_message(struct capi20_appl *ap, struct sk_buff *skb) CAPIMSG_LEN(skb->data)); } } - return card->send_message(card, skb); + return ctr->send_message(ctr, skb); } EXPORT_SYMBOL(capi20_put_message); @@ -778,16 +789,16 @@ EXPORT_SYMBOL(capi20_put_message); u16 capi20_get_manufacturer(u32 contr, u8 *buf) { - struct capi_ctr *card; + struct capi_ctr *ctr; if (contr == 0) { strlcpy(buf, capi_manufakturer, CAPI_MANUFACTURER_LEN); return CAPI_NOERROR; } - card = get_capi_ctr_by_nr(contr); - if (!card || card->cardstate != CARD_RUNNING) + ctr = get_capi_ctr_by_nr(contr); + if (!ctr || ctr->state != CAPI_CTR_RUNNING) return CAPI_REGNOTINSTALLED; - strlcpy(buf, card->manu, CAPI_MANUFACTURER_LEN); + strlcpy(buf, ctr->manu, CAPI_MANUFACTURER_LEN); return CAPI_NOERROR; } @@ -805,17 +816,17 @@ EXPORT_SYMBOL(capi20_get_manufacturer); u16 capi20_get_version(u32 contr, struct capi_version *verp) { - struct capi_ctr *card; + struct capi_ctr *ctr; if (contr == 0) { *verp = driver_version; return CAPI_NOERROR; } - card = get_capi_ctr_by_nr(contr); - if (!card || card->cardstate != CARD_RUNNING) + ctr = get_capi_ctr_by_nr(contr); + if (!ctr || ctr->state != CAPI_CTR_RUNNING) return CAPI_REGNOTINSTALLED; - memcpy((void *) verp, &card->version, sizeof(capi_version)); + memcpy(verp, &ctr->version, sizeof(capi_version)); return CAPI_NOERROR; } @@ -833,17 +844,17 @@ EXPORT_SYMBOL(capi20_get_version); u16 capi20_get_serial(u32 contr, u8 *serial) { - struct capi_ctr *card; + struct capi_ctr *ctr; if (contr == 0) { strlcpy(serial, driver_serial, CAPI_SERIAL_LEN); return CAPI_NOERROR; } - card = get_capi_ctr_by_nr(contr); - if (!card || card->cardstate != CARD_RUNNING) + ctr = get_capi_ctr_by_nr(contr); + if (!ctr || ctr->state != CAPI_CTR_RUNNING) return CAPI_REGNOTINSTALLED; - strlcpy((void *) serial, card->serial, CAPI_SERIAL_LEN); + strlcpy(serial, ctr->serial, CAPI_SERIAL_LEN); return CAPI_NOERROR; } @@ -861,18 +872,17 @@ EXPORT_SYMBOL(capi20_get_serial); u16 capi20_get_profile(u32 contr, struct capi_profile *profp) { - struct capi_ctr *card; + struct capi_ctr *ctr; if (contr == 0) { - profp->ncontroller = ncards; + profp->ncontroller = ncontrollers; return CAPI_NOERROR; } - card = get_capi_ctr_by_nr(contr); - if (!card || card->cardstate != CARD_RUNNING) + ctr = get_capi_ctr_by_nr(contr); + if (!ctr || ctr->state != CAPI_CTR_RUNNING) return CAPI_REGNOTINSTALLED; - memcpy((void *) profp, &card->profile, - sizeof(struct capi_profile)); + memcpy(profp, &ctr->profile, sizeof(struct capi_profile)); return CAPI_NOERROR; } @@ -885,7 +895,7 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) avmb1_extcarddef cdef; avmb1_resetdef rdef; capicardparams cparams; - struct capi_ctr *card; + struct capi_ctr *ctr; struct capi_driver *driver = NULL; capiloaddata ldata; struct list_head *l; @@ -958,26 +968,26 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) sizeof(avmb1_loadandconfigdef))) return -EFAULT; } - card = get_capi_ctr_by_nr(ldef.contr); - if (!card) + ctr = get_capi_ctr_by_nr(ldef.contr); + if (!ctr) return -EINVAL; - card = capi_ctr_get(card); - if (!card) + ctr = capi_ctr_get(ctr); + if (!ctr) return -ESRCH; - if (card->load_firmware == NULL) { + if (ctr->load_firmware == NULL) { printk(KERN_DEBUG "kcapi: load: no load function\n"); - capi_ctr_put(card); + capi_ctr_put(ctr); return -ESRCH; } if (ldef.t4file.len <= 0) { printk(KERN_DEBUG "kcapi: load: invalid parameter: length of t4file is %d ?\n", ldef.t4file.len); - capi_ctr_put(card); + capi_ctr_put(ctr); return -EINVAL; } if (ldef.t4file.data == NULL) { printk(KERN_DEBUG "kcapi: load: invalid parameter: dataptr is 0\n"); - capi_ctr_put(card); + capi_ctr_put(ctr); return -EINVAL; } @@ -988,46 +998,46 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) ldata.configuration.data = ldef.t4config.data; ldata.configuration.len = ldef.t4config.len; - if (card->cardstate != CARD_DETECTED) { + if (ctr->state != CAPI_CTR_DETECTED) { printk(KERN_INFO "kcapi: load: contr=%d not in detect state\n", ldef.contr); - capi_ctr_put(card); + capi_ctr_put(ctr); return -EBUSY; } - card->cardstate = CARD_LOADING; + ctr->state = CAPI_CTR_LOADING; - retval = card->load_firmware(card, &ldata); + retval = ctr->load_firmware(ctr, &ldata); if (retval) { - card->cardstate = CARD_DETECTED; - capi_ctr_put(card); + ctr->state = CAPI_CTR_DETECTED; + capi_ctr_put(ctr); return retval; } - while (card->cardstate != CARD_RUNNING) { + while (ctr->state != CAPI_CTR_RUNNING) { msleep_interruptible(100); /* 0.1 sec */ if (signal_pending(current)) { - capi_ctr_put(card); + capi_ctr_put(ctr); return -EINTR; } } - capi_ctr_put(card); + capi_ctr_put(ctr); return 0; case AVMB1_RESETCARD: if (copy_from_user(&rdef, data, sizeof(avmb1_resetdef))) return -EFAULT; - card = get_capi_ctr_by_nr(rdef.contr); - if (!card) + ctr = get_capi_ctr_by_nr(rdef.contr); + if (!ctr) return -ESRCH; - if (card->cardstate == CARD_DETECTED) + if (ctr->state == CAPI_CTR_DETECTED) return 0; - card->reset_ctr(card); + ctr->reset_ctr(ctr); - while (card->cardstate > CARD_DETECTED) { + while (ctr->state > CAPI_CTR_DETECTED) { msleep_interruptible(100); /* 0.1 sec */ @@ -1052,7 +1062,7 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) int capi20_manufacturer(unsigned int cmd, void __user *data) { - struct capi_ctr *card; + struct capi_ctr *ctr; switch (cmd) { #ifdef AVMB1_COMPAT @@ -1070,13 +1080,13 @@ int capi20_manufacturer(unsigned int cmd, void __user *data) if (copy_from_user(&fdef, data, sizeof(kcapi_flagdef))) return -EFAULT; - card = get_capi_ctr_by_nr(fdef.contr); - if (!card) + ctr = get_capi_ctr_by_nr(fdef.contr); + if (!ctr) return -ESRCH; - card->traceflag = fdef.flag; + ctr->traceflag = fdef.flag; printk(KERN_INFO "kcapi: contr [%03d] set trace=%d\n", - card->cnr, card->traceflag); + ctr->cnr, ctr->traceflag); return 0; } case KCAPI_CMD_ADDCARD: diff --git a/drivers/isdn/capi/kcapi.h b/drivers/isdn/capi/kcapi.h index 244711f7f83..f62c53bd6a9 100644 --- a/drivers/isdn/capi/kcapi.h +++ b/drivers/isdn/capi/kcapi.h @@ -24,16 +24,16 @@ printk(KERN_DEBUG "%s: " format "\n" , __func__ , ## arg); \ #endif enum { - CARD_DETECTED = 1, - CARD_LOADING = 2, - CARD_RUNNING = 3, + CAPI_CTR_DETECTED = 1, + CAPI_CTR_LOADING = 2, + CAPI_CTR_RUNNING = 3, }; extern struct list_head capi_drivers; extern rwlock_t capi_drivers_list_lock; extern struct capi20_appl *capi_applications[CAPI_MAXAPPL]; -extern struct capi_ctr *capi_cards[CAPI_MAXCONTR]; +extern struct capi_ctr *capi_controller[CAPI_MAXCONTR]; #ifdef CONFIG_PROC_FS diff --git a/drivers/isdn/capi/kcapi_proc.c b/drivers/isdn/capi/kcapi_proc.c index 09d4db764d2..59fd16a4ad8 100644 --- a/drivers/isdn/capi/kcapi_proc.c +++ b/drivers/isdn/capi/kcapi_proc.c @@ -15,13 +15,12 @@ #include #include -static char * -cardstate2str(unsigned short cardstate) +static char *state2str(unsigned short state) { - switch (cardstate) { - case CARD_DETECTED: return "detected"; - case CARD_LOADING: return "loading"; - case CARD_RUNNING: return "running"; + switch (state) { + case CAPI_CTR_DETECTED: return "detected"; + case CAPI_CTR_LOADING: return "loading"; + case CAPI_CTR_RUNNING: return "running"; default: return "???"; } } @@ -38,7 +37,7 @@ cardstate2str(unsigned short cardstate) static void *controller_start(struct seq_file *seq, loff_t *pos) { if (*pos < CAPI_MAXCONTR) - return &capi_cards[*pos]; + return &capi_controller[*pos]; return NULL; } @@ -47,7 +46,7 @@ static void *controller_next(struct seq_file *seq, void *v, loff_t *pos) { ++*pos; if (*pos < CAPI_MAXCONTR) - return &capi_cards[*pos]; + return &capi_controller[*pos]; return NULL; } @@ -65,7 +64,7 @@ static int controller_show(struct seq_file *seq, void *v) seq_printf(seq, "%d %-10s %-8s %-16s %s\n", ctr->cnr, ctr->driver_name, - cardstate2str(ctr->cardstate), + state2str(ctr->state), ctr->name, ctr->procinfo ? ctr->procinfo(ctr) : ""); diff --git a/include/linux/isdn/capilli.h b/include/linux/isdn/capilli.h index d3e5e9da0c8..856f38eddd7 100644 --- a/include/linux/isdn/capilli.h +++ b/include/linux/isdn/capilli.h @@ -66,7 +66,7 @@ struct capi_ctr { unsigned long nsentdatapkt; int cnr; /* controller number */ - volatile unsigned short cardstate; /* controller state */ + volatile unsigned short state; /* controller state */ volatile int blocked; /* output blocked */ int traceflag; /* capi trace */ -- cgit v1.2.3-70-g09d2 From ef69bb2ec6036945da1d3d3f07b75253f484f693 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 8 Feb 2010 10:12:13 +0000 Subject: CAPI: Rework controller state notifier Another step towards proper locking: Rework the callback provided to capidrv for controller state changes. This is so far attached to an application, which would require us to hold the corresponding lock across notification calls. But there is no direct relation between a controller up/down event and an application, so let's decouple them and provide a notifier call chain for those events instead. This notifier chain is first of all used internally. Here we request the highest priority to unsure that housekeeping work is done before any other notifications. The chain is exported via [un]register_capictr_notifier to our only user, capidrv, to replace the racy and unfixable capi20_set_callback. Signed-off-by: Jan Kiszka Signed-off-by: David S. Miller --- drivers/isdn/capi/capidrv.c | 22 ++++++--- drivers/isdn/capi/kcapi.c | 112 ++++++++++++++++++++------------------------ include/linux/kernelcapi.h | 17 ++----- 3 files changed, 72 insertions(+), 79 deletions(-) (limited to 'include') diff --git a/drivers/isdn/capi/capidrv.c b/drivers/isdn/capi/capidrv.c index 7d8899ad579..bf55ed5f38e 100644 --- a/drivers/isdn/capi/capidrv.c +++ b/drivers/isdn/capi/capidrv.c @@ -2210,19 +2210,24 @@ static int capidrv_delcontr(u16 contr) } -static void lower_callback(unsigned int cmd, u32 contr, void *data) +static int +lower_callback(struct notifier_block *nb, unsigned long val, void *v) { + capi_profile profile; + u32 contr = (long)v; - switch (cmd) { - case KCI_CONTRUP: + switch (val) { + case CAPICTR_UP: printk(KERN_INFO "capidrv: controller %hu up\n", contr); - (void) capidrv_addcontr(contr, (capi_profile *) data); + if (capi20_get_profile(contr, &profile) == CAPI_NOERROR) + (void) capidrv_addcontr(contr, &profile); break; - case KCI_CONTRDOWN: + case CAPICTR_DOWN: printk(KERN_INFO "capidrv: controller %hu down\n", contr); (void) capidrv_delcontr(contr); break; } + return NOTIFY_OK; } /* @@ -2262,6 +2267,10 @@ static void __exit proc_exit(void) remove_proc_entry("capi/capidrv", NULL); } +static struct notifier_block capictr_nb = { + .notifier_call = lower_callback, +}; + static int __init capidrv_init(void) { capi_profile profile; @@ -2278,7 +2287,7 @@ static int __init capidrv_init(void) return -EIO; } - capi20_set_callback(&global.ap, lower_callback); + register_capictr_notifier(&capictr_nb); errcode = capi20_get_profile(0, &profile); if (errcode != CAPI_NOERROR) { @@ -2300,6 +2309,7 @@ static int __init capidrv_init(void) static void __exit capidrv_exit(void) { + unregister_capictr_notifier(&capictr_nb); capi20_release(&global.ap); proc_exit(); diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index 9362a7a66aa..e08914d33be 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -44,12 +44,10 @@ module_param(showcapimsgs, uint, 0); /* ------------------------------------------------------------- */ -struct capi_notifier { +struct capictr_event { struct work_struct work; - unsigned int cmd; + unsigned int type; u32 controller; - u16 applid; - u32 ncci; }; /* ------------------------------------------------------------- */ @@ -71,6 +69,8 @@ struct capi_ctr *capi_controller[CAPI_MAXCONTR]; static int ncontrollers; +static BLOCKING_NOTIFIER_HEAD(ctr_notifier_list); + /* -------- controller ref counting -------------------------------------- */ static inline struct capi_ctr * @@ -165,8 +165,6 @@ static void release_appl(struct capi_ctr *ctr, u16 applid) capi_ctr_put(ctr); } -/* -------- KCI_CONTRUP --------------------------------------- */ - static void notify_up(u32 contr) { struct capi20_appl *ap; @@ -188,16 +186,11 @@ static void notify_up(u32 contr) if (!ap || ap->release_in_progress) continue; register_appl(ctr, applid, &ap->rparam); - if (ap->callback && !ap->release_in_progress) - ap->callback(KCI_CONTRUP, contr, - &ctr->profile); } } else printk(KERN_WARNING "%s: invalid contr %d\n", __func__, contr); } -/* -------- KCI_CONTRDOWN ------------------------------------- */ - static void ctr_down(struct capi_ctr *ctr) { struct capi20_appl *ap; @@ -215,11 +208,8 @@ static void ctr_down(struct capi_ctr *ctr) for (applid = 1; applid <= CAPI_MAXAPPL; applid++) { ap = get_capi_appl_by_nr(applid); - if (ap && !ap->release_in_progress) { - if (ap->callback) - ap->callback(KCI_CONTRDOWN, ctr->cnr, NULL); + if (ap && !ap->release_in_progress) capi_ctr_put(ctr); - } } } @@ -237,45 +227,63 @@ static void notify_down(u32 contr) printk(KERN_WARNING "%s: invalid contr %d\n", __func__, contr); } -static void notify_handler(struct work_struct *work) +static int +notify_handler(struct notifier_block *nb, unsigned long val, void *v) { - struct capi_notifier *np = - container_of(work, struct capi_notifier, work); + u32 contr = (long)v; - switch (np->cmd) { - case KCI_CONTRUP: - notify_up(np->controller); + switch (val) { + case CAPICTR_UP: + notify_up(contr); break; - case KCI_CONTRDOWN: - notify_down(np->controller); + case CAPICTR_DOWN: + notify_down(contr); break; } + return NOTIFY_OK; +} + +static void do_notify_work(struct work_struct *work) +{ + struct capictr_event *event = + container_of(work, struct capictr_event, work); - kfree(np); + blocking_notifier_call_chain(&ctr_notifier_list, event->type, + (void *)(long)event->controller); + kfree(event); } /* * The notifier will result in adding/deleteing of devices. Devices can * only removed in user process, not in bh. */ -static int notify_push(unsigned int cmd, u32 controller, u16 applid, u32 ncci) +static int notify_push(unsigned int event_type, u32 controller) { - struct capi_notifier *np = kmalloc(sizeof(*np), GFP_ATOMIC); + struct capictr_event *event = kmalloc(sizeof(*event), GFP_ATOMIC); - if (!np) + if (!event) return -ENOMEM; - INIT_WORK(&np->work, notify_handler); - np->cmd = cmd; - np->controller = controller; - np->applid = applid; - np->ncci = ncci; + INIT_WORK(&event->work, do_notify_work); + event->type = event_type; + event->controller = controller; - schedule_work(&np->work); + schedule_work(&event->work); return 0; } - +int register_capictr_notifier(struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&ctr_notifier_list, nb); +} +EXPORT_SYMBOL_GPL(register_capictr_notifier); + +int unregister_capictr_notifier(struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&ctr_notifier_list, nb); +} +EXPORT_SYMBOL_GPL(unregister_capictr_notifier); + /* -------- Receiver ------------------------------------------ */ static void recv_handler(struct work_struct *work) @@ -401,7 +409,7 @@ void capi_ctr_ready(struct capi_ctr *ctr) printk(KERN_NOTICE "kcapi: controller [%03d] \"%s\" ready.\n", ctr->cnr, ctr->name); - notify_push(KCI_CONTRUP, ctr->cnr, 0, 0); + notify_push(CAPICTR_UP, ctr->cnr); } EXPORT_SYMBOL(capi_ctr_ready); @@ -418,7 +426,7 @@ void capi_ctr_down(struct capi_ctr *ctr) { printk(KERN_NOTICE "kcapi: controller [%03d] down.\n", ctr->cnr); - notify_push(KCI_CONTRDOWN, ctr->cnr, 0, 0); + notify_push(CAPICTR_DOWN, ctr->cnr); } EXPORT_SYMBOL(capi_ctr_down); @@ -633,7 +641,6 @@ u16 capi20_register(struct capi20_appl *ap) ap->nrecvdatapkt = 0; ap->nsentctlpkt = 0; ap->nsentdatapkt = 0; - ap->callback = NULL; mutex_init(&ap->recv_mtx); skb_queue_head_init(&ap->recv_queue); INIT_WORK(&ap->recv_work, recv_handler); @@ -1137,30 +1144,6 @@ int capi20_manufacturer(unsigned int cmd, void __user *data) EXPORT_SYMBOL(capi20_manufacturer); -/* temporary hack */ - -/** - * capi20_set_callback() - set CAPI application notification callback function - * @ap: CAPI application descriptor structure. - * @callback: callback function (NULL to remove). - * - * If not NULL, the callback function will be called to notify the - * application of the addition or removal of a controller. - * The first argument (cmd) will tell whether the controller was added - * (KCI_CONTRUP) or removed (KCI_CONTRDOWN). - * The second argument (contr) will be the controller number. - * For cmd==KCI_CONTRUP the third argument (data) will be a pointer to the - * new controller's capability profile structure. - */ - -void capi20_set_callback(struct capi20_appl *ap, - void (*callback) (unsigned int cmd, __u32 contr, void *data)) -{ - ap->callback = callback; -} - -EXPORT_SYMBOL(capi20_set_callback); - /* ------------------------------------------------------------- */ /* -------- Init & Cleanup ------------------------------------- */ /* ------------------------------------------------------------- */ @@ -1169,10 +1152,17 @@ EXPORT_SYMBOL(capi20_set_callback); * init / exit functions */ +static struct notifier_block capictr_nb = { + .notifier_call = notify_handler, + .priority = INT_MAX, +}; + static int __init kcapi_init(void) { int err; + register_capictr_notifier(&capictr_nb); + err = cdebug_init(); if (!err) kcapi_proc_init(); diff --git a/include/linux/kernelcapi.h b/include/linux/kernelcapi.h index a53e932f80f..9c2683929fd 100644 --- a/include/linux/kernelcapi.h +++ b/include/linux/kernelcapi.h @@ -48,9 +48,7 @@ typedef struct kcapi_carddef { #include #include #include - -#define KCI_CONTRUP 0 /* arg: struct capi_profile */ -#define KCI_CONTRDOWN 1 /* arg: NULL */ +#include struct capi20_appl { u16 applid; @@ -67,11 +65,6 @@ struct capi20_appl { struct sk_buff_head recv_queue; struct work_struct recv_work; int release_in_progress; - - /* ugly hack to allow for notification of added/removed - * controllers. The Right Way (tm) is known. XXX - */ - void (*callback) (unsigned int cmd, __u32 contr, void *data); }; u16 capi20_isinstalled(void); @@ -84,11 +77,11 @@ u16 capi20_get_serial(u32 contr, u8 serial[CAPI_SERIAL_LEN]); u16 capi20_get_profile(u32 contr, struct capi_profile *profp); int capi20_manufacturer(unsigned int cmd, void __user *data); -/* temporary hack XXX */ -void capi20_set_callback(struct capi20_appl *ap, - void (*callback) (unsigned int cmd, __u32 contr, void *data)); - +#define CAPICTR_UP 0 +#define CAPICTR_DOWN 1 +int register_capictr_notifier(struct notifier_block *nb); +int unregister_capictr_notifier(struct notifier_block *nb); #define CAPI_NOERROR 0x0000 -- cgit v1.2.3-70-g09d2 From 0ca3a017a7373a4545dd7b345a8a0cecc16bc7e2 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 8 Feb 2010 10:12:14 +0000 Subject: CAPI: Rework locking of controller data structures This patch applies the mutex so far only protecting the controller list to (almost) all accesses of controller data structures. It also reworks waiting on state changes in old_capi_manufacturer so that it no longer poll and holds a module reference to the controller owner while waiting (the latter was partly done already). Modification and checking of the blocked state remains racy by design, the caller is responsible for dealing with this. Signed-off-by: Jan Kiszka Signed-off-by: David S. Miller --- drivers/isdn/capi/kcapi.c | 291 +++++++++++++++++++++++++++++------------ drivers/isdn/capi/kcapi.h | 5 +- drivers/isdn/capi/kcapi_proc.c | 5 + include/linux/isdn/capilli.h | 5 +- 4 files changed, 217 insertions(+), 89 deletions(-) (limited to 'include') diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index e08914d33be..a99f7e3f8f5 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -61,11 +61,12 @@ static char capi_manufakturer[64] = "AVM Berlin"; LIST_HEAD(capi_drivers); DEFINE_MUTEX(capi_drivers_lock); +struct capi_ctr *capi_controller[CAPI_MAXCONTR]; +DEFINE_MUTEX(capi_controller_lock); + static DEFINE_RWLOCK(application_lock); -static DEFINE_MUTEX(controller_mutex); struct capi20_appl *capi_applications[CAPI_MAXAPPL]; -struct capi_ctr *capi_controller[CAPI_MAXCONTR]; static int ncontrollers; @@ -171,13 +172,15 @@ static void notify_up(u32 contr) struct capi_ctr *ctr; u16 applid; + mutex_lock(&capi_controller_lock); + if (showcapimsgs & 1) printk(KERN_DEBUG "kcapi: notify up contr %d\n", contr); ctr = get_capi_ctr_by_nr(contr); if (ctr) { if (ctr->state == CAPI_CTR_RUNNING) - return; + goto unlock_out; ctr->state = CAPI_CTR_RUNNING; @@ -187,19 +190,24 @@ static void notify_up(u32 contr) continue; register_appl(ctr, applid, &ap->rparam); } + + wake_up_interruptible_all(&ctr->state_wait_queue); } else printk(KERN_WARNING "%s: invalid contr %d\n", __func__, contr); + +unlock_out: + mutex_unlock(&capi_controller_lock); } -static void ctr_down(struct capi_ctr *ctr) +static void ctr_down(struct capi_ctr *ctr, int new_state) { struct capi20_appl *ap; u16 applid; - if (ctr->state == CAPI_CTR_DETECTED) + if (ctr->state == CAPI_CTR_DETECTED || ctr->state == CAPI_CTR_DETACHED) return; - ctr->state = CAPI_CTR_DETECTED; + ctr->state = new_state; memset(ctr->manu, 0, sizeof(ctr->manu)); memset(&ctr->version, 0, sizeof(ctr->version)); @@ -211,20 +219,26 @@ static void ctr_down(struct capi_ctr *ctr) if (ap && !ap->release_in_progress) capi_ctr_put(ctr); } + + wake_up_interruptible_all(&ctr->state_wait_queue); } static void notify_down(u32 contr) { struct capi_ctr *ctr; + mutex_lock(&capi_controller_lock); + if (showcapimsgs & 1) printk(KERN_DEBUG "kcapi: notify down contr %d\n", contr); ctr = get_capi_ctr_by_nr(contr); if (ctr) - ctr_down(ctr); + ctr_down(ctr, CAPI_CTR_DETECTED); else printk(KERN_WARNING "%s: invalid contr %d\n", __func__, contr); + + mutex_unlock(&capi_controller_lock); } static int @@ -436,6 +450,9 @@ EXPORT_SYMBOL(capi_ctr_down); * @ctr: controller descriptor structure. * * Called by hardware driver to stop data flow. + * + * Note: The caller is responsible for synchronizing concurrent state changes + * as well as invocations of capi_ctr_handle_message. */ void capi_ctr_suspend_output(struct capi_ctr *ctr) @@ -454,6 +471,9 @@ EXPORT_SYMBOL(capi_ctr_suspend_output); * @ctr: controller descriptor structure. * * Called by hardware driver to resume data flow. + * + * Note: The caller is responsible for synchronizing concurrent state changes + * as well as invocations of capi_ctr_handle_message. */ void capi_ctr_resume_output(struct capi_ctr *ctr) @@ -481,21 +501,19 @@ int attach_capi_ctr(struct capi_ctr *ctr) { int i; - mutex_lock(&controller_mutex); + mutex_lock(&capi_controller_lock); for (i = 0; i < CAPI_MAXCONTR; i++) { if (!capi_controller[i]) break; } if (i == CAPI_MAXCONTR) { - mutex_unlock(&controller_mutex); + mutex_unlock(&capi_controller_lock); printk(KERN_ERR "kcapi: out of controller slots\n"); return -EBUSY; } capi_controller[i] = ctr; - mutex_unlock(&controller_mutex); - ctr->nrecvctlpkt = 0; ctr->nrecvdatapkt = 0; ctr->nsentctlpkt = 0; @@ -504,11 +522,15 @@ int attach_capi_ctr(struct capi_ctr *ctr) ctr->state = CAPI_CTR_DETECTED; ctr->blocked = 0; ctr->traceflag = showcapimsgs; + init_waitqueue_head(&ctr->state_wait_queue); sprintf(ctr->procfn, "capi/controllers/%d", ctr->cnr); ctr->procent = proc_create_data(ctr->procfn, 0, NULL, ctr->proc_fops, ctr); ncontrollers++; + + mutex_unlock(&capi_controller_lock); + printk(KERN_NOTICE "kcapi: controller [%03d]: %s attached\n", ctr->cnr, ctr->name); return 0; @@ -527,19 +549,29 @@ EXPORT_SYMBOL(attach_capi_ctr); int detach_capi_ctr(struct capi_ctr *ctr) { - ctr_down(ctr); + int err = 0; - ncontrollers--; + mutex_lock(&capi_controller_lock); - if (ctr->procent) { - remove_proc_entry(ctr->procfn, NULL); - ctr->procent = NULL; + ctr_down(ctr, CAPI_CTR_DETACHED); + + if (capi_controller[ctr->cnr - 1] != ctr) { + err = -EINVAL; + goto unlock_out; } capi_controller[ctr->cnr - 1] = NULL; + ncontrollers--; + + if (ctr->procent) + remove_proc_entry(ctr->procfn, NULL); + printk(KERN_NOTICE "kcapi: controller [%03d]: %s unregistered\n", ctr->cnr, ctr->name); - return 0; +unlock_out: + mutex_unlock(&capi_controller_lock); + + return err; } EXPORT_SYMBOL(detach_capi_ctr); @@ -589,13 +621,21 @@ EXPORT_SYMBOL(unregister_capi_driver); u16 capi20_isinstalled(void) { + u16 ret = CAPI_REGNOTINSTALLED; int i; - for (i = 0; i < CAPI_MAXCONTR; i++) { + + mutex_lock(&capi_controller_lock); + + for (i = 0; i < CAPI_MAXCONTR; i++) if (capi_controller[i] && - capi_controller[i]->state == CAPI_CTR_RUNNING) - return CAPI_NOERROR; - } - return CAPI_REGNOTINSTALLED; + capi_controller[i]->state == CAPI_CTR_RUNNING) { + ret = CAPI_NOERROR; + break; + } + + mutex_unlock(&capi_controller_lock); + + return ret; } EXPORT_SYMBOL(capi20_isinstalled); @@ -648,14 +688,16 @@ u16 capi20_register(struct capi20_appl *ap) write_unlock_irqrestore(&application_lock, flags); - mutex_lock(&controller_mutex); + mutex_lock(&capi_controller_lock); + for (i = 0; i < CAPI_MAXCONTR; i++) { if (!capi_controller[i] || capi_controller[i]->state != CAPI_CTR_RUNNING) continue; register_appl(capi_controller[i], applid, &ap->rparam); } - mutex_unlock(&controller_mutex); + + mutex_unlock(&capi_controller_lock); if (showcapimsgs & 1) { printk(KERN_DEBUG "kcapi: appl %d up\n", applid); @@ -688,14 +730,16 @@ u16 capi20_release(struct capi20_appl *ap) capi_applications[ap->applid - 1] = NULL; write_unlock_irqrestore(&application_lock, flags); - mutex_lock(&controller_mutex); + mutex_lock(&capi_controller_lock); + for (i = 0; i < CAPI_MAXCONTR; i++) { if (!capi_controller[i] || capi_controller[i]->state != CAPI_CTR_RUNNING) continue; release_appl(capi_controller[i], ap->applid); } - mutex_unlock(&controller_mutex); + + mutex_unlock(&capi_controller_lock); flush_scheduled_work(); skb_queue_purge(&ap->recv_queue); @@ -734,6 +778,12 @@ u16 capi20_put_message(struct capi20_appl *ap, struct sk_buff *skb) || !capi_cmd_valid(CAPIMSG_COMMAND(skb->data)) || !capi_subcmd_valid(CAPIMSG_SUBCOMMAND(skb->data))) return CAPI_ILLCMDORSUBCMDORMSGTOSMALL; + + /* + * The controller reference is protected by the existence of the + * application passed to us. We assume that the caller properly + * synchronizes this service with capi20_release. + */ ctr = get_capi_ctr_by_nr(CAPIMSG_CONTROLLER(skb->data)); if (!ctr || ctr->state != CAPI_CTR_RUNNING) { ctr = get_capi_ctr_by_nr(1); /* XXX why? */ @@ -798,16 +848,24 @@ EXPORT_SYMBOL(capi20_put_message); u16 capi20_get_manufacturer(u32 contr, u8 *buf) { struct capi_ctr *ctr; + u16 ret; if (contr == 0) { strlcpy(buf, capi_manufakturer, CAPI_MANUFACTURER_LEN); return CAPI_NOERROR; } + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(contr); - if (!ctr || ctr->state != CAPI_CTR_RUNNING) - return CAPI_REGNOTINSTALLED; - strlcpy(buf, ctr->manu, CAPI_MANUFACTURER_LEN); - return CAPI_NOERROR; + if (ctr && ctr->state == CAPI_CTR_RUNNING) { + strlcpy(buf, ctr->manu, CAPI_MANUFACTURER_LEN); + ret = CAPI_NOERROR; + } else + ret = CAPI_REGNOTINSTALLED; + + mutex_unlock(&capi_controller_lock); + return ret; } EXPORT_SYMBOL(capi20_get_manufacturer); @@ -825,17 +883,24 @@ EXPORT_SYMBOL(capi20_get_manufacturer); u16 capi20_get_version(u32 contr, struct capi_version *verp) { struct capi_ctr *ctr; + u16 ret; if (contr == 0) { *verp = driver_version; return CAPI_NOERROR; } + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(contr); - if (!ctr || ctr->state != CAPI_CTR_RUNNING) - return CAPI_REGNOTINSTALLED; + if (ctr && ctr->state == CAPI_CTR_RUNNING) { + memcpy(verp, &ctr->version, sizeof(capi_version)); + ret = CAPI_NOERROR; + } else + ret = CAPI_REGNOTINSTALLED; - memcpy(verp, &ctr->version, sizeof(capi_version)); - return CAPI_NOERROR; + mutex_unlock(&capi_controller_lock); + return ret; } EXPORT_SYMBOL(capi20_get_version); @@ -853,17 +918,24 @@ EXPORT_SYMBOL(capi20_get_version); u16 capi20_get_serial(u32 contr, u8 *serial) { struct capi_ctr *ctr; + u16 ret; if (contr == 0) { strlcpy(serial, driver_serial, CAPI_SERIAL_LEN); return CAPI_NOERROR; } + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(contr); - if (!ctr || ctr->state != CAPI_CTR_RUNNING) - return CAPI_REGNOTINSTALLED; + if (ctr && ctr->state == CAPI_CTR_RUNNING) { + strlcpy(serial, ctr->serial, CAPI_SERIAL_LEN); + ret = CAPI_NOERROR; + } else + ret = CAPI_REGNOTINSTALLED; - strlcpy(serial, ctr->serial, CAPI_SERIAL_LEN); - return CAPI_NOERROR; + mutex_unlock(&capi_controller_lock); + return ret; } EXPORT_SYMBOL(capi20_get_serial); @@ -881,21 +953,64 @@ EXPORT_SYMBOL(capi20_get_serial); u16 capi20_get_profile(u32 contr, struct capi_profile *profp) { struct capi_ctr *ctr; + u16 ret; if (contr == 0) { profp->ncontroller = ncontrollers; return CAPI_NOERROR; } + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(contr); - if (!ctr || ctr->state != CAPI_CTR_RUNNING) - return CAPI_REGNOTINSTALLED; + if (ctr && ctr->state == CAPI_CTR_RUNNING) { + memcpy(profp, &ctr->profile, sizeof(struct capi_profile)); + ret = CAPI_NOERROR; + } else + ret = CAPI_REGNOTINSTALLED; - memcpy(profp, &ctr->profile, sizeof(struct capi_profile)); - return CAPI_NOERROR; + mutex_unlock(&capi_controller_lock); + return ret; } EXPORT_SYMBOL(capi20_get_profile); +/* Must be called with capi_controller_lock held. */ +static int wait_on_ctr_state(struct capi_ctr *ctr, unsigned int state) +{ + DEFINE_WAIT(wait); + int retval = 0; + + ctr = capi_ctr_get(ctr); + if (!ctr) + return -ESRCH; + + for (;;) { + prepare_to_wait(&ctr->state_wait_queue, &wait, + TASK_INTERRUPTIBLE); + + if (ctr->state == state) + break; + if (ctr->state == CAPI_CTR_DETACHED) { + retval = -ESRCH; + break; + } + if (signal_pending(current)) { + retval = -EINTR; + break; + } + + mutex_unlock(&capi_controller_lock); + schedule(); + mutex_lock(&capi_controller_lock); + } + finish_wait(&ctr->state_wait_queue, &wait); + + capi_ctr_put(ctr); + + return retval; +} + #ifdef AVMB1_COMPAT static int old_capi_manufacturer(unsigned int cmd, void __user *data) { @@ -973,27 +1088,30 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) sizeof(avmb1_loadandconfigdef))) return -EFAULT; } + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(ldef.contr); - if (!ctr) - return -EINVAL; - ctr = capi_ctr_get(ctr); - if (!ctr) - return -ESRCH; + if (!ctr) { + retval = -EINVAL; + goto load_unlock_out; + } + if (ctr->load_firmware == NULL) { printk(KERN_DEBUG "kcapi: load: no load function\n"); - capi_ctr_put(ctr); - return -ESRCH; + retval = -ESRCH; + goto load_unlock_out; } if (ldef.t4file.len <= 0) { printk(KERN_DEBUG "kcapi: load: invalid parameter: length of t4file is %d ?\n", ldef.t4file.len); - capi_ctr_put(ctr); - return -EINVAL; + retval = -EINVAL; + goto load_unlock_out; } if (ldef.t4file.data == NULL) { printk(KERN_DEBUG "kcapi: load: invalid parameter: dataptr is 0\n"); - capi_ctr_put(ctr); - return -EINVAL; + retval = -EINVAL; + goto load_unlock_out; } ldata.firmware.user = 1; @@ -1005,52 +1123,47 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) if (ctr->state != CAPI_CTR_DETECTED) { printk(KERN_INFO "kcapi: load: contr=%d not in detect state\n", ldef.contr); - capi_ctr_put(ctr); - return -EBUSY; + retval = -EBUSY; + goto load_unlock_out; } ctr->state = CAPI_CTR_LOADING; retval = ctr->load_firmware(ctr, &ldata); - if (retval) { ctr->state = CAPI_CTR_DETECTED; - capi_ctr_put(ctr); - return retval; + goto load_unlock_out; } - while (ctr->state != CAPI_CTR_RUNNING) { - - msleep_interruptible(100); /* 0.1 sec */ + retval = wait_on_ctr_state(ctr, CAPI_CTR_RUNNING); - if (signal_pending(current)) { - capi_ctr_put(ctr); - return -EINTR; - } - } - capi_ctr_put(ctr); - return 0; +load_unlock_out: + mutex_unlock(&capi_controller_lock); + return retval; case AVMB1_RESETCARD: if (copy_from_user(&rdef, data, sizeof(avmb1_resetdef))) return -EFAULT; + + retval = 0; + + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(rdef.contr); - if (!ctr) - return -ESRCH; + if (!ctr) { + retval = -ESRCH; + goto reset_unlock_out; + } if (ctr->state == CAPI_CTR_DETECTED) - return 0; + goto reset_unlock_out; ctr->reset_ctr(ctr); - while (ctr->state > CAPI_CTR_DETECTED) { - - msleep_interruptible(100); /* 0.1 sec */ - - if (signal_pending(current)) - return -EINTR; - } - return 0; + retval = wait_on_ctr_state(ctr, CAPI_CTR_DETECTED); +reset_unlock_out: + mutex_unlock(&capi_controller_lock); + return retval; } return -EINVAL; } @@ -1068,6 +1181,7 @@ static int old_capi_manufacturer(unsigned int cmd, void __user *data) int capi20_manufacturer(unsigned int cmd, void __user *data) { struct capi_ctr *ctr; + int retval; switch (cmd) { #ifdef AVMB1_COMPAT @@ -1085,14 +1199,20 @@ int capi20_manufacturer(unsigned int cmd, void __user *data) if (copy_from_user(&fdef, data, sizeof(kcapi_flagdef))) return -EFAULT; + mutex_lock(&capi_controller_lock); + ctr = get_capi_ctr_by_nr(fdef.contr); - if (!ctr) - return -ESRCH; + if (ctr) { + ctr->traceflag = fdef.flag; + printk(KERN_INFO "kcapi: contr [%03d] set trace=%d\n", + ctr->cnr, ctr->traceflag); + retval = 0; + } else + retval = -ESRCH; + + mutex_unlock(&capi_controller_lock); - ctr->traceflag = fdef.flag; - printk(KERN_INFO "kcapi: contr [%03d] set trace=%d\n", - ctr->cnr, ctr->traceflag); - return 0; + return retval; } case KCAPI_CMD_ADDCARD: { @@ -1100,7 +1220,6 @@ int capi20_manufacturer(unsigned int cmd, void __user *data) struct capi_driver *driver = NULL; capicardparams cparams; kcapi_carddef cdef; - int retval; if ((retval = copy_from_user(&cdef, data, sizeof(cdef)))) return retval; diff --git a/drivers/isdn/capi/kcapi.h b/drivers/isdn/capi/kcapi.h index 07c58500fe4..f4620b38ec5 100644 --- a/drivers/isdn/capi/kcapi.h +++ b/drivers/isdn/capi/kcapi.h @@ -24,6 +24,7 @@ printk(KERN_DEBUG "%s: " format "\n" , __func__ , ## arg); \ #endif enum { + CAPI_CTR_DETACHED = 0, CAPI_CTR_DETECTED = 1, CAPI_CTR_LOADING = 2, CAPI_CTR_RUNNING = 3, @@ -32,8 +33,10 @@ enum { extern struct list_head capi_drivers; extern struct mutex capi_drivers_lock; -extern struct capi20_appl *capi_applications[CAPI_MAXAPPL]; extern struct capi_ctr *capi_controller[CAPI_MAXCONTR]; +extern struct mutex capi_controller_lock; + +extern struct capi20_appl *capi_applications[CAPI_MAXAPPL]; #ifdef CONFIG_PROC_FS diff --git a/drivers/isdn/capi/kcapi_proc.c b/drivers/isdn/capi/kcapi_proc.c index 71b07610ff3..3e6e17a2438 100644 --- a/drivers/isdn/capi/kcapi_proc.c +++ b/drivers/isdn/capi/kcapi_proc.c @@ -35,7 +35,10 @@ static char *state2str(unsigned short state) // --------------------------------------------------------------------------- static void *controller_start(struct seq_file *seq, loff_t *pos) + __acquires(capi_controller_lock) { + mutex_lock(&capi_controller_lock); + if (*pos < CAPI_MAXCONTR) return &capi_controller[*pos]; @@ -52,7 +55,9 @@ static void *controller_next(struct seq_file *seq, void *v, loff_t *pos) } static void controller_stop(struct seq_file *seq, void *v) + __releases(capi_controller_lock) { + mutex_unlock(&capi_controller_lock); } static int controller_show(struct seq_file *seq, void *v) diff --git a/include/linux/isdn/capilli.h b/include/linux/isdn/capilli.h index 856f38eddd7..11b57c48585 100644 --- a/include/linux/isdn/capilli.h +++ b/include/linux/isdn/capilli.h @@ -66,9 +66,10 @@ struct capi_ctr { unsigned long nsentdatapkt; int cnr; /* controller number */ - volatile unsigned short state; /* controller state */ - volatile int blocked; /* output blocked */ + unsigned short state; /* controller state */ + int blocked; /* output blocked */ int traceflag; /* capi trace */ + wait_queue_head_t state_wait_queue; struct proc_dir_entry *procent; char procfn[128]; -- cgit v1.2.3-70-g09d2 From 7d720c3e4f0c4fc152a6bf17e24244a3c85412d2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Feb 2010 15:20:26 +0000 Subject: percpu: add __percpu sparse annotations to net Add __percpu sparse annotations to net. These annotations are to make sparse consider percpu variables to be in a different address space and warn if accessed without going through percpu accessors. This patch doesn't affect normal builds. The macro and type tricks around snmp stats make things a bit interesting. DEFINE/DECLARE_SNMP_STAT() macros mark the target field as __percpu and SNMP_UPD_PO_STATS() macro is updated accordingly. All snmp_mib_*() users which used to cast the argument to (void **) are updated to cast it to (void __percpu **). Signed-off-by: Tejun Heo Acked-by: David S. Miller Cc: Patrick McHardy Cc: Arnaldo Carvalho de Melo Cc: Vlad Yasevich Cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- include/net/ip.h | 6 +++--- include/net/ipcomp.h | 2 +- include/net/neighbour.h | 2 +- include/net/netns/conntrack.h | 2 +- include/net/netns/core.h | 2 +- include/net/route.h | 2 +- include/net/snmp.h | 12 +++++++---- include/net/tcp.h | 2 +- net/8021q/vlan.h | 2 +- net/dccp/proto.c | 5 +++-- net/ipv4/af_inet.c | 46 +++++++++++++++++++++---------------------- net/ipv4/proc.c | 28 +++++++++++++------------- net/ipv4/route.c | 2 +- net/ipv4/tcp.c | 21 +++++++++++--------- net/ipv6/addrconf.c | 24 +++++++++++----------- net/ipv6/af_inet6.c | 28 +++++++++++++------------- net/ipv6/proc.c | 23 ++++++++++++---------- net/sctp/proc.c | 2 +- net/sctp/protocol.c | 5 +++-- net/xfrm/xfrm_ipcomp.c | 16 +++++++-------- net/xfrm/xfrm_policy.c | 6 +++--- net/xfrm/xfrm_proc.c | 3 ++- 22 files changed, 127 insertions(+), 114 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index 7bc47873e3f..503994a38ed 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -174,9 +174,9 @@ extern struct ipv4_config ipv4_config; #define NET_ADD_STATS_BH(net, field, adnd) SNMP_ADD_STATS_BH((net)->mib.net_statistics, field, adnd) #define NET_ADD_STATS_USER(net, field, adnd) SNMP_ADD_STATS_USER((net)->mib.net_statistics, field, adnd) -extern unsigned long snmp_fold_field(void *mib[], int offt); -extern int snmp_mib_init(void *ptr[2], size_t mibsize); -extern void snmp_mib_free(void *ptr[2]); +extern unsigned long snmp_fold_field(void __percpu *mib[], int offt); +extern int snmp_mib_init(void __percpu *ptr[2], size_t mibsize); +extern void snmp_mib_free(void __percpu *ptr[2]); extern struct local_ports { seqlock_t lock; diff --git a/include/net/ipcomp.h b/include/net/ipcomp.h index 2a1092abaa0..cc4f30cd731 100644 --- a/include/net/ipcomp.h +++ b/include/net/ipcomp.h @@ -9,7 +9,7 @@ struct crypto_comp; struct ipcomp_data { u16 threshold; - struct crypto_comp **tfms; + struct crypto_comp * __percpu *tfms; }; struct ip_comp_hdr; diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 7834f470d14..da1d58be31b 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -164,7 +164,7 @@ struct neigh_table { rwlock_t lock; unsigned long last_rand; struct kmem_cache *kmem_cachep; - struct neigh_statistics *stats; + struct neigh_statistics __percpu *stats; struct neighbour **hash_buckets; unsigned int hash_mask; __u32 hash_rnd; diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 63d449807d9..d4958d4c657 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -17,7 +17,7 @@ struct netns_ct { struct hlist_head *expect_hash; struct hlist_nulls_head unconfirmed; struct hlist_nulls_head dying; - struct ip_conntrack_stat *stat; + struct ip_conntrack_stat __percpu *stat; int sysctl_events; unsigned int sysctl_events_retry_timeout; int sysctl_acct; diff --git a/include/net/netns/core.h b/include/net/netns/core.h index 24d4be76bbd..78eb1ff7547 100644 --- a/include/net/netns/core.h +++ b/include/net/netns/core.h @@ -10,7 +10,7 @@ struct netns_core { int sysctl_somaxconn; - struct prot_inuse *inuse; + struct prot_inuse __percpu *inuse; }; #endif diff --git a/include/net/route.h b/include/net/route.h index bce6dd68d27..2c9fba7f773 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -101,7 +101,7 @@ struct rt_cache_stat { unsigned int out_hlist_search; }; -extern struct ip_rt_acct *ip_rt_acct; +extern struct ip_rt_acct __percpu *ip_rt_acct; struct in_device; extern int ip_rt_init(void); diff --git a/include/net/snmp.h b/include/net/snmp.h index da02ee027d6..692ee0061dc 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -129,9 +129,9 @@ struct linux_xfrm_mib { * nonlocked_atomic_inc() primitives -AK */ #define DEFINE_SNMP_STAT(type, name) \ - __typeof__(type) *name[2] + __typeof__(type) __percpu *name[2] #define DECLARE_SNMP_STAT(type, name) \ - extern __typeof__(type) *name[2] + extern __typeof__(type) __percpu *name[2] #define SNMP_STAT_BHPTR(name) (name[0]) #define SNMP_STAT_USRPTR(name) (name[1]) @@ -148,9 +148,13 @@ struct linux_xfrm_mib { __this_cpu_add(mib[0]->mibs[field], addend) #define SNMP_ADD_STATS_USER(mib, field, addend) \ this_cpu_add(mib[1]->mibs[field], addend) +/* + * Use "__typeof__(*mib[0]) *ptr" instead of "__typeof__(mib[0]) ptr" + * to make @ptr a non-percpu pointer. + */ #define SNMP_UPD_PO_STATS(mib, basefield, addend) \ do { \ - __typeof__(mib[0]) ptr; \ + __typeof__(*mib[0]) *ptr; \ preempt_disable(); \ ptr = this_cpu_ptr((mib)[!in_softirq()]); \ ptr->mibs[basefield##PKTS]++; \ @@ -159,7 +163,7 @@ struct linux_xfrm_mib { } while (0) #define SNMP_UPD_PO_STATS_BH(mib, basefield, addend) \ do { \ - __typeof__(mib[0]) ptr = \ + __typeof__(*mib[0]) *ptr = \ __this_cpu_ptr((mib)[!in_softirq()]); \ ptr->mibs[basefield##PKTS]++; \ ptr->mibs[basefield##OCTETS] += addend;\ diff --git a/include/net/tcp.h b/include/net/tcp.h index 87d164b9bd8..75a00c80bdd 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1189,7 +1189,7 @@ extern int tcp_v4_md5_do_del(struct sock *sk, #define tcp_twsk_md5_key(twsk) NULL #endif -extern struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(struct sock *); +extern struct tcp_md5sig_pool * __percpu *tcp_alloc_md5sig_pool(struct sock *); extern void tcp_free_md5sig_pool(void); extern struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu); diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h index 5685296017e..6abdcac1b2e 100644 --- a/net/8021q/vlan.h +++ b/net/8021q/vlan.h @@ -61,7 +61,7 @@ struct vlan_dev_info { struct proc_dir_entry *dent; unsigned long cnt_inc_headroom_on_tx; unsigned long cnt_encap_on_xmit; - struct vlan_rx_stats *vlan_rx_stats; + struct vlan_rx_stats __percpu *vlan_rx_stats; }; static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev) diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 85ec1cb7fd4..0ef7061920c 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -1005,12 +1005,13 @@ EXPORT_SYMBOL_GPL(dccp_shutdown); static inline int dccp_mib_init(void) { - return snmp_mib_init((void**)dccp_statistics, sizeof(struct dccp_mib)); + return snmp_mib_init((void __percpu **)dccp_statistics, + sizeof(struct dccp_mib)); } static inline void dccp_mib_exit(void) { - snmp_mib_free((void**)dccp_statistics); + snmp_mib_free((void __percpu **)dccp_statistics); } static int thash_entries; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 7d12c6a9b19..33b7dffa773 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1385,7 +1385,7 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, } EXPORT_SYMBOL_GPL(inet_ctl_sock_create); -unsigned long snmp_fold_field(void *mib[], int offt) +unsigned long snmp_fold_field(void __percpu *mib[], int offt) { unsigned long res = 0; int i; @@ -1398,7 +1398,7 @@ unsigned long snmp_fold_field(void *mib[], int offt) } EXPORT_SYMBOL_GPL(snmp_fold_field); -int snmp_mib_init(void *ptr[2], size_t mibsize) +int snmp_mib_init(void __percpu *ptr[2], size_t mibsize) { BUG_ON(ptr == NULL); ptr[0] = __alloc_percpu(mibsize, __alignof__(unsigned long long)); @@ -1416,7 +1416,7 @@ err0: } EXPORT_SYMBOL_GPL(snmp_mib_init); -void snmp_mib_free(void *ptr[2]) +void snmp_mib_free(void __percpu *ptr[2]) { BUG_ON(ptr == NULL); free_percpu(ptr[0]); @@ -1460,25 +1460,25 @@ static const struct net_protocol icmp_protocol = { static __net_init int ipv4_mib_init_net(struct net *net) { - if (snmp_mib_init((void **)net->mib.tcp_statistics, + if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics, sizeof(struct tcp_mib)) < 0) goto err_tcp_mib; - if (snmp_mib_init((void **)net->mib.ip_statistics, + if (snmp_mib_init((void __percpu **)net->mib.ip_statistics, sizeof(struct ipstats_mib)) < 0) goto err_ip_mib; - if (snmp_mib_init((void **)net->mib.net_statistics, + if (snmp_mib_init((void __percpu **)net->mib.net_statistics, sizeof(struct linux_mib)) < 0) goto err_net_mib; - if (snmp_mib_init((void **)net->mib.udp_statistics, + if (snmp_mib_init((void __percpu **)net->mib.udp_statistics, sizeof(struct udp_mib)) < 0) goto err_udp_mib; - if (snmp_mib_init((void **)net->mib.udplite_statistics, + if (snmp_mib_init((void __percpu **)net->mib.udplite_statistics, sizeof(struct udp_mib)) < 0) goto err_udplite_mib; - if (snmp_mib_init((void **)net->mib.icmp_statistics, + if (snmp_mib_init((void __percpu **)net->mib.icmp_statistics, sizeof(struct icmp_mib)) < 0) goto err_icmp_mib; - if (snmp_mib_init((void **)net->mib.icmpmsg_statistics, + if (snmp_mib_init((void __percpu **)net->mib.icmpmsg_statistics, sizeof(struct icmpmsg_mib)) < 0) goto err_icmpmsg_mib; @@ -1486,30 +1486,30 @@ static __net_init int ipv4_mib_init_net(struct net *net) return 0; err_icmpmsg_mib: - snmp_mib_free((void **)net->mib.icmp_statistics); + snmp_mib_free((void __percpu **)net->mib.icmp_statistics); err_icmp_mib: - snmp_mib_free((void **)net->mib.udplite_statistics); + snmp_mib_free((void __percpu **)net->mib.udplite_statistics); err_udplite_mib: - snmp_mib_free((void **)net->mib.udp_statistics); + snmp_mib_free((void __percpu **)net->mib.udp_statistics); err_udp_mib: - snmp_mib_free((void **)net->mib.net_statistics); + snmp_mib_free((void __percpu **)net->mib.net_statistics); err_net_mib: - snmp_mib_free((void **)net->mib.ip_statistics); + snmp_mib_free((void __percpu **)net->mib.ip_statistics); err_ip_mib: - snmp_mib_free((void **)net->mib.tcp_statistics); + snmp_mib_free((void __percpu **)net->mib.tcp_statistics); err_tcp_mib: return -ENOMEM; } static __net_exit void ipv4_mib_exit_net(struct net *net) { - snmp_mib_free((void **)net->mib.icmpmsg_statistics); - snmp_mib_free((void **)net->mib.icmp_statistics); - snmp_mib_free((void **)net->mib.udplite_statistics); - snmp_mib_free((void **)net->mib.udp_statistics); - snmp_mib_free((void **)net->mib.net_statistics); - snmp_mib_free((void **)net->mib.ip_statistics); - snmp_mib_free((void **)net->mib.tcp_statistics); + snmp_mib_free((void __percpu **)net->mib.icmpmsg_statistics); + snmp_mib_free((void __percpu **)net->mib.icmp_statistics); + snmp_mib_free((void __percpu **)net->mib.udplite_statistics); + snmp_mib_free((void __percpu **)net->mib.udp_statistics); + snmp_mib_free((void __percpu **)net->mib.net_statistics); + snmp_mib_free((void __percpu **)net->mib.ip_statistics); + snmp_mib_free((void __percpu **)net->mib.tcp_statistics); } static __net_initdata struct pernet_operations ipv4_mib_ops = { diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 1b09a6dde7c..242ed230737 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -280,7 +280,7 @@ static void icmpmsg_put(struct seq_file *seq) count = 0; for (i = 0; i < ICMPMSG_MIB_MAX; i++) { - val = snmp_fold_field((void **) net->mib.icmpmsg_statistics, i); + val = snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, i); if (val) { type[count] = i; vals[count++] = val; @@ -307,18 +307,18 @@ static void icmp_put(struct seq_file *seq) for (i=0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " Out%s", icmpmibmap[i].name); seq_printf(seq, "\nIcmp: %lu %lu", - snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_INMSGS), - snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_INERRORS)); + snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INMSGS), + snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INERRORS)); for (i=0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **) net->mib.icmpmsg_statistics, + snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, icmpmibmap[i].index)); seq_printf(seq, " %lu %lu", - snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS), - snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS)); + snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS), + snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS)); for (i=0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **) net->mib.icmpmsg_statistics, + snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, icmpmibmap[i].index | 0x100)); } @@ -341,7 +341,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) for (i = 0; snmp4_ipstats_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.ip_statistics, + snmp_fold_field((void __percpu **)net->mib.ip_statistics, snmp4_ipstats_list[i].entry)); icmp_put(seq); /* RFC 2011 compatibility */ @@ -356,11 +356,11 @@ static int snmp_seq_show(struct seq_file *seq, void *v) /* MaxConn field is signed, RFC 2012 */ if (snmp4_tcp_list[i].entry == TCP_MIB_MAXCONN) seq_printf(seq, " %ld", - snmp_fold_field((void **)net->mib.tcp_statistics, + snmp_fold_field((void __percpu **)net->mib.tcp_statistics, snmp4_tcp_list[i].entry)); else seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.tcp_statistics, + snmp_fold_field((void __percpu **)net->mib.tcp_statistics, snmp4_tcp_list[i].entry)); } @@ -371,7 +371,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nUdp:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.udp_statistics, + snmp_fold_field((void __percpu **)net->mib.udp_statistics, snmp4_udp_list[i].entry)); /* the UDP and UDP-Lite MIBs are the same */ @@ -382,7 +382,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nUdpLite:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.udplite_statistics, + snmp_fold_field((void __percpu **)net->mib.udplite_statistics, snmp4_udp_list[i].entry)); seq_putc(seq, '\n'); @@ -419,7 +419,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nTcpExt:"); for (i = 0; snmp4_net_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.net_statistics, + snmp_fold_field((void __percpu **)net->mib.net_statistics, snmp4_net_list[i].entry)); seq_puts(seq, "\nIpExt:"); @@ -429,7 +429,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nIpExt:"); for (i = 0; snmp4_ipextstats_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void **)net->mib.ip_statistics, + snmp_fold_field((void __percpu **)net->mib.ip_statistics, snmp4_ipextstats_list[i].entry)); seq_putc(seq, '\n'); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index b16dfadbe6d..04762d3bef7 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3334,7 +3334,7 @@ static __net_initdata struct pernet_operations rt_secret_timer_ops = { #ifdef CONFIG_NET_CLS_ROUTE -struct ip_rt_acct *ip_rt_acct __read_mostly; +struct ip_rt_acct __percpu *ip_rt_acct __read_mostly; #endif /* CONFIG_NET_CLS_ROUTE */ static __initdata unsigned long rhash_entries; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index d5d69ea8f24..e471d037fcc 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2788,10 +2788,10 @@ EXPORT_SYMBOL(tcp_gro_complete); #ifdef CONFIG_TCP_MD5SIG static unsigned long tcp_md5sig_users; -static struct tcp_md5sig_pool **tcp_md5sig_pool; +static struct tcp_md5sig_pool * __percpu *tcp_md5sig_pool; static DEFINE_SPINLOCK(tcp_md5sig_pool_lock); -static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool) +static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool * __percpu *pool) { int cpu; for_each_possible_cpu(cpu) { @@ -2808,7 +2808,7 @@ static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool) void tcp_free_md5sig_pool(void) { - struct tcp_md5sig_pool **pool = NULL; + struct tcp_md5sig_pool * __percpu *pool = NULL; spin_lock_bh(&tcp_md5sig_pool_lock); if (--tcp_md5sig_users == 0) { @@ -2822,10 +2822,11 @@ void tcp_free_md5sig_pool(void) EXPORT_SYMBOL(tcp_free_md5sig_pool); -static struct tcp_md5sig_pool **__tcp_alloc_md5sig_pool(struct sock *sk) +static struct tcp_md5sig_pool * __percpu * +__tcp_alloc_md5sig_pool(struct sock *sk) { int cpu; - struct tcp_md5sig_pool **pool; + struct tcp_md5sig_pool * __percpu *pool; pool = alloc_percpu(struct tcp_md5sig_pool *); if (!pool) @@ -2852,9 +2853,9 @@ out_free: return NULL; } -struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(struct sock *sk) +struct tcp_md5sig_pool * __percpu *tcp_alloc_md5sig_pool(struct sock *sk) { - struct tcp_md5sig_pool **pool; + struct tcp_md5sig_pool * __percpu *pool; int alloc = 0; retry: @@ -2873,7 +2874,9 @@ retry: if (alloc) { /* we cannot hold spinlock here because this may sleep. */ - struct tcp_md5sig_pool **p = __tcp_alloc_md5sig_pool(sk); + struct tcp_md5sig_pool * __percpu *p; + + p = __tcp_alloc_md5sig_pool(sk); spin_lock_bh(&tcp_md5sig_pool_lock); if (!p) { tcp_md5sig_users--; @@ -2897,7 +2900,7 @@ EXPORT_SYMBOL(tcp_alloc_md5sig_pool); struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu) { - struct tcp_md5sig_pool **p; + struct tcp_md5sig_pool * __percpu *p; spin_lock_bh(&tcp_md5sig_pool_lock); p = tcp_md5sig_pool; if (p) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index c79cbff5437..b0d4a4b23db 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -278,31 +278,31 @@ static void addrconf_mod_timer(struct inet6_ifaddr *ifp, static int snmp6_alloc_dev(struct inet6_dev *idev) { - if (snmp_mib_init((void **)idev->stats.ipv6, + if (snmp_mib_init((void __percpu **)idev->stats.ipv6, sizeof(struct ipstats_mib)) < 0) goto err_ip; - if (snmp_mib_init((void **)idev->stats.icmpv6, + if (snmp_mib_init((void __percpu **)idev->stats.icmpv6, sizeof(struct icmpv6_mib)) < 0) goto err_icmp; - if (snmp_mib_init((void **)idev->stats.icmpv6msg, + if (snmp_mib_init((void __percpu **)idev->stats.icmpv6msg, sizeof(struct icmpv6msg_mib)) < 0) goto err_icmpmsg; return 0; err_icmpmsg: - snmp_mib_free((void **)idev->stats.icmpv6); + snmp_mib_free((void __percpu **)idev->stats.icmpv6); err_icmp: - snmp_mib_free((void **)idev->stats.ipv6); + snmp_mib_free((void __percpu **)idev->stats.ipv6); err_ip: return -ENOMEM; } static void snmp6_free_dev(struct inet6_dev *idev) { - snmp_mib_free((void **)idev->stats.icmpv6msg); - snmp_mib_free((void **)idev->stats.icmpv6); - snmp_mib_free((void **)idev->stats.ipv6); + snmp_mib_free((void __percpu **)idev->stats.icmpv6msg); + snmp_mib_free((void __percpu **)idev->stats.icmpv6); + snmp_mib_free((void __percpu **)idev->stats.ipv6); } /* Nobody refers to this device, we may destroy it. */ @@ -3766,8 +3766,8 @@ static inline size_t inet6_if_nlmsg_size(void) ); } -static inline void __snmp6_fill_stats(u64 *stats, void **mib, int items, - int bytes) +static inline void __snmp6_fill_stats(u64 *stats, void __percpu **mib, + int items, int bytes) { int i; int pad = bytes - sizeof(u64) * items; @@ -3786,10 +3786,10 @@ static void snmp6_fill_stats(u64 *stats, struct inet6_dev *idev, int attrtype, { switch(attrtype) { case IFLA_INET6_STATS: - __snmp6_fill_stats(stats, (void **)idev->stats.ipv6, IPSTATS_MIB_MAX, bytes); + __snmp6_fill_stats(stats, (void __percpu **)idev->stats.ipv6, IPSTATS_MIB_MAX, bytes); break; case IFLA_INET6_ICMP6STATS: - __snmp6_fill_stats(stats, (void **)idev->stats.icmpv6, ICMP6_MIB_MAX, bytes); + __snmp6_fill_stats(stats, (void __percpu **)idev->stats.icmpv6, ICMP6_MIB_MAX, bytes); break; } } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index e29160ff4a0..37d14e735c2 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -971,41 +971,41 @@ static void ipv6_packet_cleanup(void) static int __net_init ipv6_init_mibs(struct net *net) { - if (snmp_mib_init((void **)net->mib.udp_stats_in6, + if (snmp_mib_init((void __percpu **)net->mib.udp_stats_in6, sizeof (struct udp_mib)) < 0) return -ENOMEM; - if (snmp_mib_init((void **)net->mib.udplite_stats_in6, + if (snmp_mib_init((void __percpu **)net->mib.udplite_stats_in6, sizeof (struct udp_mib)) < 0) goto err_udplite_mib; - if (snmp_mib_init((void **)net->mib.ipv6_statistics, + if (snmp_mib_init((void __percpu **)net->mib.ipv6_statistics, sizeof(struct ipstats_mib)) < 0) goto err_ip_mib; - if (snmp_mib_init((void **)net->mib.icmpv6_statistics, + if (snmp_mib_init((void __percpu **)net->mib.icmpv6_statistics, sizeof(struct icmpv6_mib)) < 0) goto err_icmp_mib; - if (snmp_mib_init((void **)net->mib.icmpv6msg_statistics, + if (snmp_mib_init((void __percpu **)net->mib.icmpv6msg_statistics, sizeof(struct icmpv6msg_mib)) < 0) goto err_icmpmsg_mib; return 0; err_icmpmsg_mib: - snmp_mib_free((void **)net->mib.icmpv6_statistics); + snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics); err_icmp_mib: - snmp_mib_free((void **)net->mib.ipv6_statistics); + snmp_mib_free((void __percpu **)net->mib.ipv6_statistics); err_ip_mib: - snmp_mib_free((void **)net->mib.udplite_stats_in6); + snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6); err_udplite_mib: - snmp_mib_free((void **)net->mib.udp_stats_in6); + snmp_mib_free((void __percpu **)net->mib.udp_stats_in6); return -ENOMEM; } static void ipv6_cleanup_mibs(struct net *net) { - snmp_mib_free((void **)net->mib.udp_stats_in6); - snmp_mib_free((void **)net->mib.udplite_stats_in6); - snmp_mib_free((void **)net->mib.ipv6_statistics); - snmp_mib_free((void **)net->mib.icmpv6_statistics); - snmp_mib_free((void **)net->mib.icmpv6msg_statistics); + snmp_mib_free((void __percpu **)net->mib.udp_stats_in6); + snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6); + snmp_mib_free((void __percpu **)net->mib.ipv6_statistics); + snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics); + snmp_mib_free((void __percpu **)net->mib.icmpv6msg_statistics); } static int __net_init inet6_net_init(struct net *net) diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index bfe2598dd56..58344c0fbd1 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -136,7 +136,7 @@ static const struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_SENTINEL }; -static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) +static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void __percpu **mib) { char name[32]; int i; @@ -170,7 +170,7 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) return; } -static void snmp6_seq_show_item(struct seq_file *seq, void **mib, +static void snmp6_seq_show_item(struct seq_file *seq, void __percpu **mib, const struct snmp_mib *itemlist) { int i; @@ -183,14 +183,15 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) { struct net *net = (struct net *)seq->private; - snmp6_seq_show_item(seq, (void **)net->mib.ipv6_statistics, + snmp6_seq_show_item(seq, (void __percpu **)net->mib.ipv6_statistics, snmp6_ipstats_list); - snmp6_seq_show_item(seq, (void **)net->mib.icmpv6_statistics, + snmp6_seq_show_item(seq, (void __percpu **)net->mib.icmpv6_statistics, snmp6_icmp6_list); - snmp6_seq_show_icmpv6msg(seq, (void **)net->mib.icmpv6msg_statistics); - snmp6_seq_show_item(seq, (void **)net->mib.udp_stats_in6, + snmp6_seq_show_icmpv6msg(seq, + (void __percpu **)net->mib.icmpv6msg_statistics); + snmp6_seq_show_item(seq, (void __percpu **)net->mib.udp_stats_in6, snmp6_udp6_list); - snmp6_seq_show_item(seq, (void **)net->mib.udplite_stats_in6, + snmp6_seq_show_item(seq, (void __percpu **)net->mib.udplite_stats_in6, snmp6_udplite6_list); return 0; } @@ -213,9 +214,11 @@ static int snmp6_dev_seq_show(struct seq_file *seq, void *v) struct inet6_dev *idev = (struct inet6_dev *)seq->private; seq_printf(seq, "%-32s\t%u\n", "ifIndex", idev->dev->ifindex); - snmp6_seq_show_item(seq, (void **)idev->stats.ipv6, snmp6_ipstats_list); - snmp6_seq_show_item(seq, (void **)idev->stats.icmpv6, snmp6_icmp6_list); - snmp6_seq_show_icmpv6msg(seq, (void **)idev->stats.icmpv6msg); + snmp6_seq_show_item(seq, (void __percpu **)idev->stats.ipv6, + snmp6_ipstats_list); + snmp6_seq_show_item(seq, (void __percpu **)idev->stats.icmpv6, + snmp6_icmp6_list); + snmp6_seq_show_icmpv6msg(seq, (void __percpu **)idev->stats.icmpv6msg); return 0; } diff --git a/net/sctp/proc.c b/net/sctp/proc.c index a5ac6e0a8d9..784bcc9a979 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -83,7 +83,7 @@ static int sctp_snmp_seq_show(struct seq_file *seq, void *v) for (i = 0; sctp_snmp_list[i].name != NULL; i++) seq_printf(seq, "%-32s\t%ld\n", sctp_snmp_list[i].name, - snmp_fold_field((void **)sctp_statistics, + snmp_fold_field((void __percpu **)sctp_statistics, sctp_snmp_list[i].entry)); return 0; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index a3c8988758b..9687177b026 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -996,12 +996,13 @@ int sctp_register_pf(struct sctp_pf *pf, sa_family_t family) static inline int init_sctp_mibs(void) { - return snmp_mib_init((void**)sctp_statistics, sizeof(struct sctp_mib)); + return snmp_mib_init((void __percpu **)sctp_statistics, + sizeof(struct sctp_mib)); } static inline void cleanup_sctp_mibs(void) { - snmp_mib_free((void**)sctp_statistics); + snmp_mib_free((void __percpu **)sctp_statistics); } static void sctp_v4_pf_init(void) diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c index 42cd18391f4..0fc5ff66d1f 100644 --- a/net/xfrm/xfrm_ipcomp.c +++ b/net/xfrm/xfrm_ipcomp.c @@ -30,12 +30,12 @@ struct ipcomp_tfms { struct list_head list; - struct crypto_comp **tfms; + struct crypto_comp * __percpu *tfms; int users; }; static DEFINE_MUTEX(ipcomp_resource_mutex); -static void **ipcomp_scratches; +static void * __percpu *ipcomp_scratches; static int ipcomp_scratch_users; static LIST_HEAD(ipcomp_tfms_list); @@ -200,7 +200,7 @@ EXPORT_SYMBOL_GPL(ipcomp_output); static void ipcomp_free_scratches(void) { int i; - void **scratches; + void * __percpu *scratches; if (--ipcomp_scratch_users) return; @@ -215,10 +215,10 @@ static void ipcomp_free_scratches(void) free_percpu(scratches); } -static void **ipcomp_alloc_scratches(void) +static void * __percpu *ipcomp_alloc_scratches(void) { int i; - void **scratches; + void * __percpu *scratches; if (ipcomp_scratch_users++) return ipcomp_scratches; @@ -239,7 +239,7 @@ static void **ipcomp_alloc_scratches(void) return scratches; } -static void ipcomp_free_tfms(struct crypto_comp **tfms) +static void ipcomp_free_tfms(struct crypto_comp * __percpu *tfms) { struct ipcomp_tfms *pos; int cpu; @@ -267,10 +267,10 @@ static void ipcomp_free_tfms(struct crypto_comp **tfms) free_percpu(tfms); } -static struct crypto_comp **ipcomp_alloc_tfms(const char *alg_name) +static struct crypto_comp * __percpu *ipcomp_alloc_tfms(const char *alg_name) { struct ipcomp_tfms *pos; - struct crypto_comp **tfms; + struct crypto_comp * __percpu *tfms; int cpu; /* This can be any valid CPU ID so we don't need locking. */ diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index eb870fcc29c..cfceb6616ec 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2428,19 +2428,19 @@ static int __net_init xfrm_statistics_init(struct net *net) { int rv; - if (snmp_mib_init((void **)net->mib.xfrm_statistics, + if (snmp_mib_init((void __percpu **)net->mib.xfrm_statistics, sizeof(struct linux_xfrm_mib)) < 0) return -ENOMEM; rv = xfrm_proc_init(net); if (rv < 0) - snmp_mib_free((void **)net->mib.xfrm_statistics); + snmp_mib_free((void __percpu **)net->mib.xfrm_statistics); return rv; } static void xfrm_statistics_fini(struct net *net) { xfrm_proc_fini(net); - snmp_mib_free((void **)net->mib.xfrm_statistics); + snmp_mib_free((void __percpu **)net->mib.xfrm_statistics); } #else static int __net_init xfrm_statistics_init(struct net *net) diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index c083a4e4e79..003f2c437ac 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -50,7 +50,8 @@ static int xfrm_statistics_seq_show(struct seq_file *seq, void *v) int i; for (i=0; xfrm_mib_list[i].name; i++) seq_printf(seq, "%-24s\t%lu\n", xfrm_mib_list[i].name, - snmp_fold_field((void **)net->mib.xfrm_statistics, + snmp_fold_field((void __percpu **) + net->mib.xfrm_statistics, xfrm_mib_list[i].entry)); return 0; } -- cgit v1.2.3-70-g09d2 From 47d742752df4c1088589d4424840bc761613ab2a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Feb 2010 15:21:08 +0000 Subject: percpu: add __percpu sparse annotations to net drivers Add __percpu sparse annotations to net drivers. These annotations are to make sparse consider percpu variables to be in a different address space and warn if accessed without going through percpu accessors. This patch doesn't affect normal builds. Signed-off-by: Tejun Heo Acked-by: David S. Miller Cc: Eric Dumazet Cc: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/chelsio/sge.c | 2 +- drivers/net/loopback.c | 16 +++++++++------- drivers/net/veth.c | 4 ++-- include/linux/if_macvlan.h | 2 +- 4 files changed, 13 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/net/chelsio/sge.c b/drivers/net/chelsio/sge.c index 109d2783e4d..8e12505ac3a 100644 --- a/drivers/net/chelsio/sge.c +++ b/drivers/net/chelsio/sge.c @@ -267,7 +267,7 @@ struct sge { struct sk_buff *espibug_skb[MAX_NPORTS]; u32 sge_control; /* shadow value of sge control reg */ struct sge_intr_counts stats; - struct sge_port_stats *port_stats[MAX_NPORTS]; + struct sge_port_stats __percpu *port_stats[MAX_NPORTS]; struct sched *tx_sched; struct cmdQ cmdQ[SGE_CMDQ_N] ____cacheline_aligned_in_smp; }; diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index b9fcc981983..72b7949c91b 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -72,7 +72,8 @@ struct pcpu_lstats { static netdev_tx_t loopback_xmit(struct sk_buff *skb, struct net_device *dev) { - struct pcpu_lstats *pcpu_lstats, *lb_stats; + struct pcpu_lstats __percpu *pcpu_lstats; + struct pcpu_lstats *lb_stats; int len; skb_orphan(skb); @@ -80,7 +81,7 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb, skb->protocol = eth_type_trans(skb, dev); /* it's OK to use per_cpu_ptr() because BHs are off */ - pcpu_lstats = dev->ml_priv; + pcpu_lstats = (void __percpu __force *)dev->ml_priv; lb_stats = this_cpu_ptr(pcpu_lstats); len = skb->len; @@ -95,14 +96,14 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb, static struct net_device_stats *loopback_get_stats(struct net_device *dev) { - const struct pcpu_lstats *pcpu_lstats; + const struct pcpu_lstats __percpu *pcpu_lstats; struct net_device_stats *stats = &dev->stats; unsigned long bytes = 0; unsigned long packets = 0; unsigned long drops = 0; int i; - pcpu_lstats = dev->ml_priv; + pcpu_lstats = (void __percpu __force *)dev->ml_priv; for_each_possible_cpu(i) { const struct pcpu_lstats *lb_stats; @@ -135,19 +136,20 @@ static const struct ethtool_ops loopback_ethtool_ops = { static int loopback_dev_init(struct net_device *dev) { - struct pcpu_lstats *lstats; + struct pcpu_lstats __percpu *lstats; lstats = alloc_percpu(struct pcpu_lstats); if (!lstats) return -ENOMEM; - dev->ml_priv = lstats; + dev->ml_priv = (void __force *)lstats; return 0; } static void loopback_dev_free(struct net_device *dev) { - struct pcpu_lstats *lstats = dev->ml_priv; + struct pcpu_lstats __percpu *lstats = + (void __percpu __force *)dev->ml_priv; free_percpu(lstats); free_netdev(dev); diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 3a15de56df9..35609e64f6f 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -34,7 +34,7 @@ struct veth_net_stats { struct veth_priv { struct net_device *peer; - struct veth_net_stats *stats; + struct veth_net_stats __percpu *stats; unsigned ip_summed; }; @@ -263,7 +263,7 @@ static int veth_change_mtu(struct net_device *dev, int new_mtu) static int veth_dev_init(struct net_device *dev) { - struct veth_net_stats *stats; + struct veth_net_stats __percpu *stats; struct veth_priv *priv; stats = alloc_percpu(struct veth_net_stats); diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index 51f1512045e..f9cb9ba1475 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -30,7 +30,7 @@ struct macvlan_dev { struct hlist_node hlist; struct macvlan_port *port; struct net_device *lowerdev; - struct macvlan_rx_stats *rx_stats; + struct macvlan_rx_stats __percpu *rx_stats; enum macvlan_mode mode; int (*receive)(struct sk_buff *skb); int (*forward)(struct net_device *dev, struct sk_buff *skb); -- cgit v1.2.3-70-g09d2 From 9874c41cd5e70bfc97bcd52a8b6c98c2a6ba7299 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 16 Feb 2010 18:40:04 +0000 Subject: ipv6.h: reassembly: replace calculated magic number with multiplication On Tue, 2010-02-16 at 16:47 +0100, Patrick McHardy wrote: > Joe Perches wrote: > >> @@ -246,6 +246,8 @@ extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb); > >> int ip6_frag_nqueues(struct net *net); > >> int ip6_frag_mem(struct net *net); > >> > >> +#define IPV6_FRAG_HIGH_THRESH 262144 /* == 256*1024 */ > >> +#define IPV6_FRAG_LOW_THRESH 196608 /* == 192*1024 */ > >> #define IPV6_FRAG_TIMEOUT (60*HZ) /* 60 seconds */ > > > > 196608 isn't a number I want to remember. > > Is this better as: > > > > #define IPV6_FRAG_HIGH_THRESH (256 * 1024) /* 262144 */ > > #define IPV6_FRAG_LOW_THRESH (192 * 1024) /* 196608 */ > > Please send a patch, I'll apply it once these patches are in Dave's > tree. Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/net/ipv6.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 639ec53ea08..d067db1f88c 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -246,9 +246,9 @@ extern int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb); int ip6_frag_nqueues(struct net *net); int ip6_frag_mem(struct net *net); -#define IPV6_FRAG_HIGH_THRESH 262144 /* == 256*1024 */ -#define IPV6_FRAG_LOW_THRESH 196608 /* == 192*1024 */ -#define IPV6_FRAG_TIMEOUT (60*HZ) /* 60 seconds */ +#define IPV6_FRAG_HIGH_THRESH (256 * 1024) /* 262144 */ +#define IPV6_FRAG_LOW_THRESH (192 * 1024) /* 196608 */ +#define IPV6_FRAG_TIMEOUT (60 * HZ) /* 60 seconds */ extern int __ipv6_addr_type(const struct in6_addr *addr); static inline int ipv6_addr_type(const struct in6_addr *addr) -- cgit v1.2.3-70-g09d2 From 60b86755929e1a7e9038c8d860a8491cfdf8d93a Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 17 Feb 2010 10:30:23 +0000 Subject: usbnet: Convert dev(dbg|err|warn|info) macros to netdev_ These macros are too similar to the dev_ equivalents but take a usbnet * argument. Convert them to the recently introduced netdev_ macros and remove the old macros. The old macros had "\n" appended to the format string. Add the "\n" to the converted uses. Some existing uses of the dev macros in cdc_eem.c probably mistakenly had trailing "\n". No "\n" added there. Fix net1080 this/other log message inversion. Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- drivers/net/usb/asix.c | 97 ++++++------ drivers/net/usb/cdc_eem.c | 10 +- drivers/net/usb/cdc_ether.c | 18 +-- drivers/net/usb/dm9601.c | 49 +++--- drivers/net/usb/int51x1.c | 15 +- drivers/net/usb/net1080.c | 101 ++++++------ drivers/net/usb/rndis_host.c | 10 +- drivers/net/usb/smsc95xx.c | 188 +++++++++++----------- drivers/net/usb/usbnet.c | 162 +++++++++---------- drivers/net/wireless/rndis_wlan.c | 325 ++++++++++++++++++++------------------ include/linux/usb/usbnet.h | 21 --- 11 files changed, 507 insertions(+), 489 deletions(-) (limited to 'include') diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c index f02551713b1..f605204de3e 100644 --- a/drivers/net/usb/asix.c +++ b/drivers/net/usb/asix.c @@ -184,8 +184,8 @@ static int asix_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index, void *buf; int err = -ENOMEM; - devdbg(dev,"asix_read_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d", - cmd, value, index, size); + netdev_dbg(dev->net, "asix_read_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d\n", + cmd, value, index, size); buf = kmalloc(size, GFP_KERNEL); if (!buf) @@ -217,8 +217,8 @@ static int asix_write_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index, void *buf = NULL; int err = -ENOMEM; - devdbg(dev,"asix_write_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d", - cmd, value, index, size); + netdev_dbg(dev->net, "asix_write_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d\n", + cmd, value, index, size); if (data) { buf = kmalloc(size, GFP_KERNEL); @@ -264,15 +264,15 @@ asix_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index, int status; struct urb *urb; - devdbg(dev,"asix_write_cmd_async() cmd=0x%02x value=0x%04x index=0x%04x size=%d", - cmd, value, index, size); + netdev_dbg(dev->net, "asix_write_cmd_async() cmd=0x%02x value=0x%04x index=0x%04x size=%d\n", + cmd, value, index, size); if ((urb = usb_alloc_urb(0, GFP_ATOMIC)) == NULL) { - deverr(dev, "Error allocating URB in write_cmd_async!"); + netdev_err(dev->net, "Error allocating URB in write_cmd_async!\n"); return; } if ((req = kmalloc(sizeof(struct usb_ctrlrequest), GFP_ATOMIC)) == NULL) { - deverr(dev, "Failed to allocate memory for control request"); + netdev_err(dev->net, "Failed to allocate memory for control request\n"); usb_free_urb(urb); return; } @@ -289,8 +289,8 @@ asix_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index, asix_async_cmd_callback, req); if((status = usb_submit_urb(urb, GFP_ATOMIC)) < 0) { - deverr(dev, "Error submitting the control message: status=%d", - status); + netdev_err(dev->net, "Error submitting the control message: status=%d\n", + status); kfree(req); usb_free_urb(urb); } @@ -314,7 +314,7 @@ static int asix_rx_fixup(struct usbnet *dev, struct sk_buff *skb) while (skb->len > 0) { if ((short)(header & 0x0000ffff) != ~((short)((header & 0xffff0000) >> 16))) { - deverr(dev,"asix_rx_fixup() Bad Header Length"); + netdev_err(dev->net, "asix_rx_fixup() Bad Header Length\n"); } /* get the packet length */ size = (u16) (header & 0x0000ffff); @@ -322,7 +322,8 @@ static int asix_rx_fixup(struct usbnet *dev, struct sk_buff *skb) if ((skb->len) - ((size + 1) & 0xfffe) == 0) return 2; if (size > ETH_FRAME_LEN) { - deverr(dev,"asix_rx_fixup() Bad RX Length %d", size); + netdev_err(dev->net, "asix_rx_fixup() Bad RX Length %d\n", + size); return 0; } ax_skb = skb_clone(skb, GFP_ATOMIC); @@ -348,7 +349,8 @@ static int asix_rx_fixup(struct usbnet *dev, struct sk_buff *skb) } if (skb->len < 0) { - deverr(dev,"asix_rx_fixup() Bad SKB Length %d", skb->len); + netdev_err(dev->net, "asix_rx_fixup() Bad SKB Length %d\n", + skb->len); return 0; } return 1; @@ -409,7 +411,7 @@ static void asix_status(struct usbnet *dev, struct urb *urb) usbnet_defer_kevent (dev, EVENT_LINK_RESET ); } else netif_carrier_off(dev->net); - devdbg(dev, "Link Status is: %d", link); + netdev_dbg(dev->net, "Link Status is: %d\n", link); } } @@ -418,7 +420,7 @@ static inline int asix_set_sw_mii(struct usbnet *dev) int ret; ret = asix_write_cmd(dev, AX_CMD_SET_SW_MII, 0x0000, 0, 0, NULL); if (ret < 0) - deverr(dev, "Failed to enable software MII access"); + netdev_err(dev->net, "Failed to enable software MII access\n"); return ret; } @@ -427,7 +429,7 @@ static inline int asix_set_hw_mii(struct usbnet *dev) int ret; ret = asix_write_cmd(dev, AX_CMD_SET_HW_MII, 0x0000, 0, 0, NULL); if (ret < 0) - deverr(dev, "Failed to enable hardware MII access"); + netdev_err(dev->net, "Failed to enable hardware MII access\n"); return ret; } @@ -436,13 +438,14 @@ static inline int asix_get_phy_addr(struct usbnet *dev) u8 buf[2]; int ret = asix_read_cmd(dev, AX_CMD_READ_PHY_ID, 0, 0, 2, buf); - devdbg(dev, "asix_get_phy_addr()"); + netdev_dbg(dev->net, "asix_get_phy_addr()\n"); if (ret < 0) { - deverr(dev, "Error reading PHYID register: %02x", ret); + netdev_err(dev->net, "Error reading PHYID register: %02x\n", ret); goto out; } - devdbg(dev, "asix_get_phy_addr() returning 0x%04x", *((__le16 *)buf)); + netdev_dbg(dev->net, "asix_get_phy_addr() returning 0x%04x\n", + *((__le16 *)buf)); ret = buf[1]; out: @@ -455,7 +458,7 @@ static int asix_sw_reset(struct usbnet *dev, u8 flags) ret = asix_write_cmd(dev, AX_CMD_SW_RESET, flags, 0, 0, NULL); if (ret < 0) - deverr(dev,"Failed to send software reset: %02x", ret); + netdev_err(dev->net, "Failed to send software reset: %02x\n", ret); return ret; } @@ -466,7 +469,7 @@ static u16 asix_read_rx_ctl(struct usbnet *dev) int ret = asix_read_cmd(dev, AX_CMD_READ_RX_CTL, 0, 0, 2, &v); if (ret < 0) { - deverr(dev, "Error reading RX_CTL register: %02x", ret); + netdev_err(dev->net, "Error reading RX_CTL register: %02x\n", ret); goto out; } ret = le16_to_cpu(v); @@ -478,11 +481,11 @@ static int asix_write_rx_ctl(struct usbnet *dev, u16 mode) { int ret; - devdbg(dev,"asix_write_rx_ctl() - mode = 0x%04x", mode); + netdev_dbg(dev->net, "asix_write_rx_ctl() - mode = 0x%04x\n", mode); ret = asix_write_cmd(dev, AX_CMD_WRITE_RX_CTL, mode, 0, 0, NULL); if (ret < 0) - deverr(dev, "Failed to write RX_CTL mode to 0x%04x: %02x", - mode, ret); + netdev_err(dev->net, "Failed to write RX_CTL mode to 0x%04x: %02x\n", + mode, ret); return ret; } @@ -493,7 +496,8 @@ static u16 asix_read_medium_status(struct usbnet *dev) int ret = asix_read_cmd(dev, AX_CMD_READ_MEDIUM_STATUS, 0, 0, 2, &v); if (ret < 0) { - deverr(dev, "Error reading Medium Status register: %02x", ret); + netdev_err(dev->net, "Error reading Medium Status register: %02x\n", + ret); goto out; } ret = le16_to_cpu(v); @@ -505,11 +509,11 @@ static int asix_write_medium_mode(struct usbnet *dev, u16 mode) { int ret; - devdbg(dev,"asix_write_medium_mode() - mode = 0x%04x", mode); + netdev_dbg(dev->net, "asix_write_medium_mode() - mode = 0x%04x\n", mode); ret = asix_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL); if (ret < 0) - deverr(dev, "Failed to write Medium Mode mode to 0x%04x: %02x", - mode, ret); + netdev_err(dev->net, "Failed to write Medium Mode mode to 0x%04x: %02x\n", + mode, ret); return ret; } @@ -518,11 +522,11 @@ static int asix_write_gpio(struct usbnet *dev, u16 value, int sleep) { int ret; - devdbg(dev,"asix_write_gpio() - value = 0x%04x", value); + netdev_dbg(dev->net, "asix_write_gpio() - value = 0x%04x\n", value); ret = asix_write_cmd(dev, AX_CMD_WRITE_GPIOS, value, 0, 0, NULL); if (ret < 0) - deverr(dev, "Failed to write GPIO value 0x%04x: %02x", - value, ret); + netdev_err(dev->net, "Failed to write GPIO value 0x%04x: %02x\n", + value, ret); if (sleep) msleep(sleep); @@ -588,7 +592,8 @@ static int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) asix_set_hw_mii(dev); mutex_unlock(&dev->phy_mutex); - devdbg(dev, "asix_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x", phy_id, loc, le16_to_cpu(res)); + netdev_dbg(dev->net, "asix_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x\n", + phy_id, loc, le16_to_cpu(res)); return le16_to_cpu(res); } @@ -599,7 +604,8 @@ asix_mdio_write(struct net_device *netdev, int phy_id, int loc, int val) struct usbnet *dev = netdev_priv(netdev); __le16 res = cpu_to_le16(val); - devdbg(dev, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x", phy_id, loc, val); + netdev_dbg(dev->net, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n", + phy_id, loc, val); mutex_lock(&dev->phy_mutex); asix_set_sw_mii(dev); asix_write_cmd(dev, AX_CMD_WRITE_MII_REG, phy_id, (__u16)loc, 2, &res); @@ -800,7 +806,8 @@ static int ax88172_link_reset(struct usbnet *dev) if (ecmd.duplex != DUPLEX_FULL) mode |= ~AX88172_MEDIUM_FD; - devdbg(dev, "ax88172_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode); + netdev_dbg(dev->net, "ax88172_link_reset() speed: %d duplex: %d setting mode to 0x%04x\n", + ecmd.speed, ecmd.duplex, mode); asix_write_medium_mode(dev, mode); @@ -902,7 +909,8 @@ static int ax88772_link_reset(struct usbnet *dev) if (ecmd.duplex != DUPLEX_FULL) mode &= ~AX_MEDIUM_FD; - devdbg(dev, "ax88772_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode); + netdev_dbg(dev->net, "ax88772_link_reset() speed: %d duplex: %d setting mode to 0x%04x\n", + ecmd.speed, ecmd.duplex, mode); asix_write_medium_mode(dev, mode); @@ -1059,10 +1067,10 @@ static int marvell_phy_init(struct usbnet *dev) struct asix_data *data = (struct asix_data *)&dev->data; u16 reg; - devdbg(dev,"marvell_phy_init()"); + netdev_dbg(dev->net, "marvell_phy_init()\n"); reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_MARVELL_STATUS); - devdbg(dev,"MII_MARVELL_STATUS = 0x%04x", reg); + netdev_dbg(dev->net, "MII_MARVELL_STATUS = 0x%04x\n", reg); asix_mdio_write(dev->net, dev->mii.phy_id, MII_MARVELL_CTRL, MARVELL_CTRL_RXDELAY | MARVELL_CTRL_TXDELAY); @@ -1070,7 +1078,7 @@ static int marvell_phy_init(struct usbnet *dev) if (data->ledmode) { reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_MARVELL_LED_CTRL); - devdbg(dev,"MII_MARVELL_LED_CTRL (1) = 0x%04x", reg); + netdev_dbg(dev->net, "MII_MARVELL_LED_CTRL (1) = 0x%04x\n", reg); reg &= 0xf8ff; reg |= (1 + 0x0100); @@ -1079,7 +1087,7 @@ static int marvell_phy_init(struct usbnet *dev) reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_MARVELL_LED_CTRL); - devdbg(dev,"MII_MARVELL_LED_CTRL (2) = 0x%04x", reg); + netdev_dbg(dev->net, "MII_MARVELL_LED_CTRL (2) = 0x%04x\n", reg); reg &= 0xfc0f; } @@ -1090,7 +1098,7 @@ static int marvell_led_status(struct usbnet *dev, u16 speed) { u16 reg = asix_mdio_read(dev->net, dev->mii.phy_id, MARVELL_LED_MANUAL); - devdbg(dev, "marvell_led_status() read 0x%04x", reg); + netdev_dbg(dev->net, "marvell_led_status() read 0x%04x\n", reg); /* Clear out the center LED bits - 0x03F0 */ reg &= 0xfc0f; @@ -1106,7 +1114,7 @@ static int marvell_led_status(struct usbnet *dev, u16 speed) reg |= 0x02f0; } - devdbg(dev, "marvell_led_status() writing 0x%04x", reg); + netdev_dbg(dev->net, "marvell_led_status() writing 0x%04x\n", reg); asix_mdio_write(dev->net, dev->mii.phy_id, MARVELL_LED_MANUAL, reg); return 0; @@ -1118,7 +1126,7 @@ static int ax88178_link_reset(struct usbnet *dev) struct ethtool_cmd ecmd; struct asix_data *data = (struct asix_data *)&dev->data; - devdbg(dev,"ax88178_link_reset()"); + netdev_dbg(dev->net, "ax88178_link_reset()\n"); mii_check_media(&dev->mii, 1, 1); mii_ethtool_gset(&dev->mii, &ecmd); @@ -1138,7 +1146,8 @@ static int ax88178_link_reset(struct usbnet *dev) else mode &= ~AX_MEDIUM_FD; - devdbg(dev, "ax88178_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode); + netdev_dbg(dev->net, "ax88178_link_reset() speed: %d duplex: %d setting mode to 0x%04x\n", + ecmd.speed, ecmd.duplex, mode); asix_write_medium_mode(dev, mode); @@ -1188,7 +1197,7 @@ static int ax88178_change_mtu(struct net_device *net, int new_mtu) struct usbnet *dev = netdev_priv(net); int ll_mtu = new_mtu + net->hard_header_len + 4; - devdbg(dev, "ax88178_change_mtu() new_mtu=%d", new_mtu); + netdev_dbg(dev->net, "ax88178_change_mtu() new_mtu=%d\n", new_mtu); if (new_mtu <= 0 || ll_mtu > 16384) return -EINVAL; diff --git a/drivers/net/usb/cdc_eem.c b/drivers/net/usb/cdc_eem.c index c337ffc3304..a4a85a6ed86 100644 --- a/drivers/net/usb/cdc_eem.c +++ b/drivers/net/usb/cdc_eem.c @@ -73,7 +73,7 @@ static void eem_linkcmd(struct usbnet *dev, struct sk_buff *skb) usb_free_urb(urb); fail: dev_kfree_skb(skb); - devwarn(dev, "link cmd failure\n"); + netdev_warn(dev->net, "link cmd failure\n"); return; } } @@ -212,7 +212,8 @@ static int eem_rx_fixup(struct usbnet *dev, struct sk_buff *skb) * b15: 1 (EEM command) */ if (header & BIT(14)) { - devdbg(dev, "reserved command %04x\n", header); + netdev_dbg(dev->net, "reserved command %04x\n", + header); continue; } @@ -255,8 +256,9 @@ static int eem_rx_fixup(struct usbnet *dev, struct sk_buff *skb) case 1: /* Echo response */ case 5: /* Tickle */ default: /* reserved */ - devwarn(dev, "unexpected link command %d\n", - bmEEMCmd); + netdev_warn(dev->net, + "unexpected link command %d\n", + bmEEMCmd); continue; } diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c index 4f27f022fbf..7e5a75269cc 100644 --- a/drivers/net/usb/cdc_ether.c +++ b/drivers/net/usb/cdc_ether.c @@ -340,9 +340,9 @@ EXPORT_SYMBOL_GPL(usbnet_cdc_unbind); static void dumpspeed(struct usbnet *dev, __le32 *speeds) { if (netif_msg_timer(dev)) - devinfo(dev, "link speeds: %u kbps up, %u kbps down", - __le32_to_cpu(speeds[0]) / 1000, - __le32_to_cpu(speeds[1]) / 1000); + netdev_info(dev->net, "link speeds: %u kbps up, %u kbps down\n", + __le32_to_cpu(speeds[0]) / 1000, + __le32_to_cpu(speeds[1]) / 1000); } static void cdc_status(struct usbnet *dev, struct urb *urb) @@ -362,8 +362,8 @@ static void cdc_status(struct usbnet *dev, struct urb *urb) switch (event->bNotificationType) { case USB_CDC_NOTIFY_NETWORK_CONNECTION: if (netif_msg_timer(dev)) - devdbg(dev, "CDC: carrier %s", - event->wValue ? "on" : "off"); + netdev_dbg(dev->net, "CDC: carrier %s\n", + event->wValue ? "on" : "off"); if (event->wValue) netif_carrier_on(dev->net); else @@ -371,8 +371,8 @@ static void cdc_status(struct usbnet *dev, struct urb *urb) break; case USB_CDC_NOTIFY_SPEED_CHANGE: /* tx/rx rates */ if (netif_msg_timer(dev)) - devdbg(dev, "CDC: speed change (len %d)", - urb->actual_length); + netdev_dbg(dev->net, "CDC: speed change (len %d)\n", + urb->actual_length); if (urb->actual_length != (sizeof *event + 8)) set_bit(EVENT_STS_SPLIT, &dev->flags); else @@ -382,8 +382,8 @@ static void cdc_status(struct usbnet *dev, struct urb *urb) * but there are no standard formats for the response data. */ default: - deverr(dev, "CDC: unexpected notification %02x!", - event->bNotificationType); + netdev_err(dev->net, "CDC: unexpected notification %02x!\n", + event->bNotificationType); break; } } diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c index c820fec6204..269339769f4 100644 --- a/drivers/net/usb/dm9601.c +++ b/drivers/net/usb/dm9601.c @@ -58,7 +58,7 @@ static int dm_read(struct usbnet *dev, u8 reg, u16 length, void *data) void *buf; int err = -ENOMEM; - devdbg(dev, "dm_read() reg=0x%02x length=%d", reg, length); + netdev_dbg(dev->net, "dm_read() reg=0x%02x length=%d\n", reg, length); buf = kmalloc(length, GFP_KERNEL); if (!buf) @@ -89,7 +89,7 @@ static int dm_write(struct usbnet *dev, u8 reg, u16 length, void *data) void *buf = NULL; int err = -ENOMEM; - devdbg(dev, "dm_write() reg=0x%02x, length=%d", reg, length); + netdev_dbg(dev->net, "dm_write() reg=0x%02x, length=%d\n", reg, length); if (data) { buf = kmalloc(length, GFP_KERNEL); @@ -112,7 +112,8 @@ static int dm_write(struct usbnet *dev, u8 reg, u16 length, void *data) static int dm_write_reg(struct usbnet *dev, u8 reg, u8 value) { - devdbg(dev, "dm_write_reg() reg=0x%02x, value=0x%02x", reg, value); + netdev_dbg(dev->net, "dm_write_reg() reg=0x%02x, value=0x%02x\n", + reg, value); return usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), DM_WRITE_REG, @@ -142,13 +143,13 @@ static void dm_write_async_helper(struct usbnet *dev, u8 reg, u8 value, urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { - deverr(dev, "Error allocating URB in dm_write_async_helper!"); + netdev_err(dev->net, "Error allocating URB in dm_write_async_helper!\n"); return; } req = kmalloc(sizeof(struct usb_ctrlrequest), GFP_ATOMIC); if (!req) { - deverr(dev, "Failed to allocate memory for control request"); + netdev_err(dev->net, "Failed to allocate memory for control request\n"); usb_free_urb(urb); return; } @@ -166,8 +167,8 @@ static void dm_write_async_helper(struct usbnet *dev, u8 reg, u8 value, status = usb_submit_urb(urb, GFP_ATOMIC); if (status < 0) { - deverr(dev, "Error submitting the control message: status=%d", - status); + netdev_err(dev->net, "Error submitting the control message: status=%d\n", + status); kfree(req); usb_free_urb(urb); } @@ -175,15 +176,15 @@ static void dm_write_async_helper(struct usbnet *dev, u8 reg, u8 value, static void dm_write_async(struct usbnet *dev, u8 reg, u16 length, void *data) { - devdbg(dev, "dm_write_async() reg=0x%02x length=%d", reg, length); + netdev_dbg(dev->net, "dm_write_async() reg=0x%02x length=%d\n", reg, length); dm_write_async_helper(dev, reg, 0, length, data); } static void dm_write_reg_async(struct usbnet *dev, u8 reg, u8 value) { - devdbg(dev, "dm_write_reg_async() reg=0x%02x value=0x%02x", - reg, value); + netdev_dbg(dev->net, "dm_write_reg_async() reg=0x%02x value=0x%02x\n", + reg, value); dm_write_async_helper(dev, reg, value, 0, NULL); } @@ -211,7 +212,7 @@ static int dm_read_shared_word(struct usbnet *dev, int phy, u8 reg, __le16 *valu } if (i == DM_TIMEOUT) { - deverr(dev, "%s read timed out!", phy ? "phy" : "eeprom"); + netdev_err(dev->net, "%s read timed out!\n", phy ? "phy" : "eeprom"); ret = -EIO; goto out; } @@ -219,8 +220,8 @@ static int dm_read_shared_word(struct usbnet *dev, int phy, u8 reg, __le16 *valu dm_write_reg(dev, DM_SHARED_CTRL, 0x0); ret = dm_read(dev, DM_SHARED_DATA, 2, value); - devdbg(dev, "read shared %d 0x%02x returned 0x%04x, %d", - phy, reg, *value, ret); + netdev_dbg(dev->net, "read shared %d 0x%02x returned 0x%04x, %d\n", + phy, reg, *value, ret); out: mutex_unlock(&dev->phy_mutex); @@ -254,7 +255,7 @@ static int dm_write_shared_word(struct usbnet *dev, int phy, u8 reg, __le16 valu } if (i == DM_TIMEOUT) { - deverr(dev, "%s write timed out!", phy ? "phy" : "eeprom"); + netdev_err(dev->net, "%s write timed out!\n", phy ? "phy" : "eeprom"); ret = -EIO; goto out; } @@ -304,15 +305,15 @@ static int dm9601_mdio_read(struct net_device *netdev, int phy_id, int loc) __le16 res; if (phy_id) { - devdbg(dev, "Only internal phy supported"); + netdev_dbg(dev->net, "Only internal phy supported\n"); return 0; } dm_read_shared_word(dev, 1, loc, &res); - devdbg(dev, - "dm9601_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x", - phy_id, loc, le16_to_cpu(res)); + netdev_dbg(dev->net, + "dm9601_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x\n", + phy_id, loc, le16_to_cpu(res)); return le16_to_cpu(res); } @@ -324,12 +325,12 @@ static void dm9601_mdio_write(struct net_device *netdev, int phy_id, int loc, __le16 res = cpu_to_le16(val); if (phy_id) { - devdbg(dev, "Only internal phy supported"); + netdev_dbg(dev->net, "Only internal phy supported\n"); return; } - devdbg(dev,"dm9601_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x", - phy_id, loc, val); + netdev_dbg(dev->net, "dm9601_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n", + phy_id, loc, val); dm_write_shared_word(dev, 1, loc, res); } @@ -592,7 +593,7 @@ static void dm9601_status(struct usbnet *dev, struct urb *urb) } else netif_carrier_off(dev->net); - devdbg(dev, "Link Status is: %d", link); + netdev_dbg(dev->net, "Link Status is: %d\n", link); } } @@ -603,8 +604,8 @@ static int dm9601_link_reset(struct usbnet *dev) mii_check_media(&dev->mii, 1, 1); mii_ethtool_gset(&dev->mii, &ecmd); - devdbg(dev, "link_reset() speed: %d duplex: %d", - ecmd.speed, ecmd.duplex); + netdev_dbg(dev->net, "link_reset() speed: %d duplex: %d\n", + ecmd.speed, ecmd.duplex); return 0; } diff --git a/drivers/net/usb/int51x1.c b/drivers/net/usb/int51x1.c index 9ab5c1983a7..3c228df5706 100644 --- a/drivers/net/usb/int51x1.c +++ b/drivers/net/usb/int51x1.c @@ -51,7 +51,7 @@ static int int51x1_rx_fixup(struct usbnet *dev, struct sk_buff *skb) int len; if (!(pskb_may_pull(skb, INT51X1_HEADER_SIZE))) { - deverr(dev, "unexpected tiny rx frame"); + netdev_err(dev->net, "unexpected tiny rx frame\n"); return 0; } @@ -138,25 +138,25 @@ static void int51x1_set_multicast(struct net_device *netdev) if (netdev->flags & IFF_PROMISC) { /* do not expect to see traffic of other PLCs */ filter |= PACKET_TYPE_PROMISCUOUS; - devinfo(dev, "promiscuous mode enabled"); + netdev_info(dev->net, "promiscuous mode enabled\n"); } else if (!netdev_mc_empty(netdev) || (netdev->flags & IFF_ALLMULTI)) { filter |= PACKET_TYPE_ALL_MULTICAST; - devdbg(dev, "receive all multicast enabled"); + netdev_dbg(dev->net, "receive all multicast enabled\n"); } else { /* ~PROMISCUOUS, ~MULTICAST */ - devdbg(dev, "receive own packets only"); + netdev_dbg(dev->net, "receive own packets only\n"); } urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { - devwarn(dev, "Error allocating URB"); + netdev_warn(dev->net, "Error allocating URB\n"); return; } req = kmalloc(sizeof(*req), GFP_ATOMIC); if (!req) { - devwarn(dev, "Error allocating control msg"); + netdev_warn(dev->net, "Error allocating control msg\n"); goto out; } @@ -173,7 +173,8 @@ static void int51x1_set_multicast(struct net_device *netdev) status = usb_submit_urb(urb, GFP_ATOMIC); if (status < 0) { - devwarn(dev, "Error submitting control msg, sts=%d", status); + netdev_warn(dev->net, "Error submitting control msg, sts=%d\n", + status); goto out1; } return; diff --git a/drivers/net/usb/net1080.c b/drivers/net/usb/net1080.c index aeb1ab03a9e..f6994053c10 100644 --- a/drivers/net/usb/net1080.c +++ b/drivers/net/usb/net1080.c @@ -205,23 +205,24 @@ static inline void nc_dump_usbctl(struct usbnet *dev, u16 usbctl) { if (!netif_msg_link(dev)) return; - devdbg(dev, "net1080 %s-%s usbctl 0x%x:%s%s%s%s%s;" - " this%s%s;" - " other%s%s; r/o 0x%x", - dev->udev->bus->bus_name, dev->udev->devpath, - usbctl, - (usbctl & USBCTL_ENABLE_LANG) ? " lang" : "", - (usbctl & USBCTL_ENABLE_MFGR) ? " mfgr" : "", - (usbctl & USBCTL_ENABLE_PROD) ? " prod" : "", - (usbctl & USBCTL_ENABLE_SERIAL) ? " serial" : "", - (usbctl & USBCTL_ENABLE_DEFAULTS) ? " defaults" : "", - - (usbctl & USBCTL_FLUSH_OTHER) ? " FLUSH" : "", - (usbctl & USBCTL_DISCONN_OTHER) ? " DIS" : "", - (usbctl & USBCTL_FLUSH_THIS) ? " FLUSH" : "", - (usbctl & USBCTL_DISCONN_THIS) ? " DIS" : "", - usbctl & ~USBCTL_WRITABLE_MASK - ); + netdev_dbg(dev->net, "net1080 %s-%s usbctl 0x%x:%s%s%s%s%s;" + " this%s%s;" + " other%s%s; r/o 0x%x\n", + dev->udev->bus->bus_name, dev->udev->devpath, + usbctl, + (usbctl & USBCTL_ENABLE_LANG) ? " lang" : "", + (usbctl & USBCTL_ENABLE_MFGR) ? " mfgr" : "", + (usbctl & USBCTL_ENABLE_PROD) ? " prod" : "", + (usbctl & USBCTL_ENABLE_SERIAL) ? " serial" : "", + (usbctl & USBCTL_ENABLE_DEFAULTS) ? " defaults" : "", + + (usbctl & USBCTL_FLUSH_THIS) ? " FLUSH" : "", + (usbctl & USBCTL_DISCONN_THIS) ? " DIS" : "", + + (usbctl & USBCTL_FLUSH_OTHER) ? " FLUSH" : "", + (usbctl & USBCTL_DISCONN_OTHER) ? " DIS" : "", + + usbctl & ~USBCTL_WRITABLE_MASK); } /*-------------------------------------------------------------------------*/ @@ -250,28 +251,25 @@ static inline void nc_dump_status(struct usbnet *dev, u16 status) { if (!netif_msg_link(dev)) return; - devdbg(dev, "net1080 %s-%s status 0x%x:" - " this (%c) PKT=%d%s%s%s;" - " other PKT=%d%s%s%s; unspec 0x%x", - dev->udev->bus->bus_name, dev->udev->devpath, - status, - - // XXX the packet counts don't seem right - // (1 at reset, not 0); maybe UNSPEC too - - (status & STATUS_PORT_A) ? 'A' : 'B', - STATUS_PACKETS_THIS(status), - (status & STATUS_CONN_THIS) ? " CON" : "", - (status & STATUS_SUSPEND_THIS) ? " SUS" : "", - (status & STATUS_MAILBOX_THIS) ? " MBOX" : "", - - STATUS_PACKETS_OTHER(status), - (status & STATUS_CONN_OTHER) ? " CON" : "", - (status & STATUS_SUSPEND_OTHER) ? " SUS" : "", - (status & STATUS_MAILBOX_OTHER) ? " MBOX" : "", - - status & STATUS_UNSPEC_MASK - ); + netdev_dbg(dev->net, "net1080 %s-%s status 0x%x: this (%c) PKT=%d%s%s%s; other PKT=%d%s%s%s; unspec 0x%x\n", + dev->udev->bus->bus_name, dev->udev->devpath, + status, + + // XXX the packet counts don't seem right + // (1 at reset, not 0); maybe UNSPEC too + + (status & STATUS_PORT_A) ? 'A' : 'B', + STATUS_PACKETS_THIS(status), + (status & STATUS_CONN_THIS) ? " CON" : "", + (status & STATUS_SUSPEND_THIS) ? " SUS" : "", + (status & STATUS_MAILBOX_THIS) ? " MBOX" : "", + + STATUS_PACKETS_OTHER(status), + (status & STATUS_CONN_OTHER) ? " CON" : "", + (status & STATUS_SUSPEND_OTHER) ? " SUS" : "", + (status & STATUS_MAILBOX_OTHER) ? " MBOX" : "", + + status & STATUS_UNSPEC_MASK); } /*-------------------------------------------------------------------------*/ @@ -287,9 +285,9 @@ static inline void nc_dump_status(struct usbnet *dev, u16 status) static inline void nc_dump_ttl(struct usbnet *dev, u16 ttl) { if (netif_msg_link(dev)) - devdbg(dev, "net1080 %s-%s ttl 0x%x this = %d, other = %d", - dev->udev->bus->bus_name, dev->udev->devpath, - ttl, TTL_THIS(ttl), TTL_OTHER(ttl)); + netdev_dbg(dev->net, "net1080 %s-%s ttl 0x%x this = %d, other = %d\n", + dev->udev->bus->bus_name, dev->udev->devpath, + ttl, TTL_THIS(ttl), TTL_OTHER(ttl)); } /*-------------------------------------------------------------------------*/ @@ -335,10 +333,9 @@ static int net1080_reset(struct usbnet *dev) dbg("%s: assigned TTL, %d ms", dev->net->name, NC_READ_TTL_MS); if (netif_msg_link(dev)) - devinfo(dev, "port %c, peer %sconnected", - (status & STATUS_PORT_A) ? 'A' : 'B', - (status & STATUS_CONN_OTHER) ? "" : "dis" - ); + netdev_info(dev->net, "port %c, peer %sconnected\n", + (status & STATUS_PORT_A) ? 'A' : 'B', + (status & STATUS_CONN_OTHER) ? "" : "dis"); retval = 0; done: @@ -416,7 +413,7 @@ static void nc_ensure_sync(struct usbnet *dev) } if (netif_msg_rx_err(dev)) - devdbg(dev, "flush net1080; too many framing errors"); + netdev_dbg(dev->net, "flush net1080; too many framing errors\n"); dev->frame_errors = 0; } } @@ -486,8 +483,8 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb) return 0; } #if 0 - devdbg(dev, "frame hdr_len, - header->packet_len, header->packet_id); + netdev_dbg(dev->net, "frame hdr_len, + header->packet_len, header->packet_id); #endif dev->frame_errors = 0; return 1; @@ -547,9 +544,9 @@ encapsulate: trailer = (struct nc_trailer *) skb_put(skb, sizeof *trailer); put_unaligned(header->packet_id, &trailer->packet_id); #if 0 - devdbg(dev, "frame >tx h %d p %d id %d", - header->hdr_len, header->packet_len, - header->packet_id); + netdev_dbg(dev->net, "frame >tx h %d p %d id %d\n", + header->hdr_len, header->packet_len, + header->packet_id); #endif return skb; } diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c index 490fa8f5542..f01f0240185 100644 --- a/drivers/net/usb/rndis_host.c +++ b/drivers/net/usb/rndis_host.c @@ -57,8 +57,8 @@ */ void rndis_status(struct usbnet *dev, struct urb *urb) { - devdbg(dev, "rndis status urb, len %d stat %d", - urb->actual_length, urb->status); + netdev_dbg(dev->net, "rndis status urb, len %d stat %d\n", + urb->actual_length, urb->status); // FIXME for keepalives, respond immediately (asynchronously) // if not an RNDIS status, do like cdc_status(dev,urb) does } @@ -497,9 +497,9 @@ int rndis_rx_fixup(struct usbnet *dev, struct sk_buff *skb) skb->len < msg_len || (data_offset + data_len + 8) > msg_len)) { dev->net->stats.rx_frame_errors++; - devdbg(dev, "bad rndis message %d/%d/%d/%d, len %d", - le32_to_cpu(hdr->msg_type), - msg_len, data_offset, data_len, skb->len); + netdev_dbg(dev->net, "bad rndis message %d/%d/%d/%d, len %d\n", + le32_to_cpu(hdr->msg_type), + msg_len, data_offset, data_len, skb->len); return 0; } skb_pull(skb, 8 + data_offset); diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c index 48555d0e374..d4cbcefbff3 100644 --- a/drivers/net/usb/smsc95xx.c +++ b/drivers/net/usb/smsc95xx.c @@ -78,7 +78,7 @@ static int smsc95xx_read_reg(struct usbnet *dev, u32 index, u32 *data) 00, index, buf, 4, USB_CTRL_GET_TIMEOUT); if (unlikely(ret < 0)) - devwarn(dev, "Failed to read register index 0x%08x", index); + netdev_warn(dev->net, "Failed to read register index 0x%08x\n", index); le32_to_cpus(buf); *data = *buf; @@ -106,7 +106,7 @@ static int smsc95xx_write_reg(struct usbnet *dev, u32 index, u32 data) 00, index, buf, 4, USB_CTRL_SET_TIMEOUT); if (unlikely(ret < 0)) - devwarn(dev, "Failed to write register index 0x%08x", index); + netdev_warn(dev->net, "Failed to write register index 0x%08x\n", index); kfree(buf); @@ -138,7 +138,7 @@ static int smsc95xx_mdio_read(struct net_device *netdev, int phy_id, int idx) /* confirm MII not busy */ if (smsc95xx_phy_wait_not_busy(dev)) { - devwarn(dev, "MII is busy in smsc95xx_mdio_read"); + netdev_warn(dev->net, "MII is busy in smsc95xx_mdio_read\n"); mutex_unlock(&dev->phy_mutex); return -EIO; } @@ -150,7 +150,7 @@ static int smsc95xx_mdio_read(struct net_device *netdev, int phy_id, int idx) smsc95xx_write_reg(dev, MII_ADDR, addr); if (smsc95xx_phy_wait_not_busy(dev)) { - devwarn(dev, "Timed out reading MII reg %02X", idx); + netdev_warn(dev->net, "Timed out reading MII reg %02X\n", idx); mutex_unlock(&dev->phy_mutex); return -EIO; } @@ -172,7 +172,7 @@ static void smsc95xx_mdio_write(struct net_device *netdev, int phy_id, int idx, /* confirm MII not busy */ if (smsc95xx_phy_wait_not_busy(dev)) { - devwarn(dev, "MII is busy in smsc95xx_mdio_write"); + netdev_warn(dev->net, "MII is busy in smsc95xx_mdio_write\n"); mutex_unlock(&dev->phy_mutex); return; } @@ -187,7 +187,7 @@ static void smsc95xx_mdio_write(struct net_device *netdev, int phy_id, int idx, smsc95xx_write_reg(dev, MII_ADDR, addr); if (smsc95xx_phy_wait_not_busy(dev)) - devwarn(dev, "Timed out writing MII reg %02X", idx); + netdev_warn(dev->net, "Timed out writing MII reg %02X\n", idx); mutex_unlock(&dev->phy_mutex); } @@ -205,7 +205,7 @@ static int smsc95xx_wait_eeprom(struct usbnet *dev) } while (!time_after(jiffies, start_time + HZ)); if (val & (E2P_CMD_TIMEOUT_ | E2P_CMD_BUSY_)) { - devwarn(dev, "EEPROM read operation timeout"); + netdev_warn(dev->net, "EEPROM read operation timeout\n"); return -EIO; } @@ -226,7 +226,7 @@ static int smsc95xx_eeprom_confirm_not_busy(struct usbnet *dev) udelay(40); } while (!time_after(jiffies, start_time + HZ)); - devwarn(dev, "EEPROM is busy"); + netdev_warn(dev->net, "EEPROM is busy\n"); return -EIO; } @@ -308,7 +308,7 @@ static void smsc95xx_async_cmd_callback(struct urb *urb) int status = urb->status; if (status < 0) - devwarn(dev, "async callback failed with %d", status); + netdev_warn(dev->net, "async callback failed with %d\n", status); kfree(usb_context); usb_free_urb(urb); @@ -323,13 +323,13 @@ static int smsc95xx_write_reg_async(struct usbnet *dev, u16 index, u32 *data) urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { - devwarn(dev, "Error allocating URB"); + netdev_warn(dev->net, "Error allocating URB\n"); return -ENOMEM; } usb_context = kmalloc(sizeof(struct usb_context), GFP_ATOMIC); if (usb_context == NULL) { - devwarn(dev, "Error allocating control msg"); + netdev_warn(dev->net, "Error allocating control msg\n"); usb_free_urb(urb); return -ENOMEM; } @@ -348,7 +348,8 @@ static int smsc95xx_write_reg_async(struct usbnet *dev, u16 index, u32 *data) status = usb_submit_urb(urb, GFP_ATOMIC); if (status < 0) { - devwarn(dev, "Error submitting control msg, sts=%d", status); + netdev_warn(dev->net, "Error submitting control msg, sts=%d\n", + status); kfree(usb_context); usb_free_urb(urb); } @@ -376,12 +377,12 @@ static void smsc95xx_set_multicast(struct net_device *netdev) if (dev->net->flags & IFF_PROMISC) { if (netif_msg_drv(dev)) - devdbg(dev, "promiscuous mode enabled"); + netdev_dbg(dev->net, "promiscuous mode enabled\n"); pdata->mac_cr |= MAC_CR_PRMS_; pdata->mac_cr &= ~(MAC_CR_MCPAS_ | MAC_CR_HPFILT_); } else if (dev->net->flags & IFF_ALLMULTI) { if (netif_msg_drv(dev)) - devdbg(dev, "receive all multicast enabled"); + netdev_dbg(dev->net, "receive all multicast enabled\n"); pdata->mac_cr |= MAC_CR_MCPAS_; pdata->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_HPFILT_); } else if (!netdev_mc_empty(dev->net)) { @@ -401,20 +402,20 @@ static void smsc95xx_set_multicast(struct net_device *netdev) else hash_lo |= mask; } else { - devwarn(dev, "dmi_addrlen != 6"); + netdev_warn(dev->net, "dmi_addrlen != 6\n"); } mc_list = mc_list->next; } if (count != ((u32) netdev_mc_count(dev->net))) - devwarn(dev, "mc_count != dev->mc_count"); + netdev_warn(dev->net, "mc_count != dev->mc_count\n"); if (netif_msg_drv(dev)) - devdbg(dev, "HASHH=0x%08X, HASHL=0x%08X", hash_hi, - hash_lo); + netdev_dbg(dev->net, "HASHH=0x%08X, HASHL=0x%08X\n", + hash_hi, hash_lo); } else { if (netif_msg_drv(dev)) - devdbg(dev, "receive own packets only"); + netdev_dbg(dev->net, "receive own packets only\n"); pdata->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_ | MAC_CR_HPFILT_); } @@ -434,7 +435,7 @@ static void smsc95xx_phy_update_flowcontrol(struct usbnet *dev, u8 duplex, int ret = smsc95xx_read_reg(dev, AFC_CFG, &afc_cfg); if (ret < 0) { - devwarn(dev, "error reading AFC_CFG"); + netdev_warn(dev->net, "error reading AFC_CFG\n"); return; } @@ -452,12 +453,12 @@ static void smsc95xx_phy_update_flowcontrol(struct usbnet *dev, u8 duplex, afc_cfg &= ~0xF; if (netif_msg_link(dev)) - devdbg(dev, "rx pause %s, tx pause %s", - (cap & FLOW_CTRL_RX ? "enabled" : "disabled"), - (cap & FLOW_CTRL_TX ? "enabled" : "disabled")); + netdev_dbg(dev->net, "rx pause %s, tx pause %s\n", + cap & FLOW_CTRL_RX ? "enabled" : "disabled", + cap & FLOW_CTRL_TX ? "enabled" : "disabled"); } else { if (netif_msg_link(dev)) - devdbg(dev, "half duplex"); + netdev_dbg(dev->net, "half duplex\n"); flow = 0; afc_cfg |= 0xF; } @@ -486,8 +487,8 @@ static int smsc95xx_link_reset(struct usbnet *dev) rmtadv = smsc95xx_mdio_read(dev->net, mii->phy_id, MII_LPA); if (netif_msg_link(dev)) - devdbg(dev, "speed: %d duplex: %d lcladv: %04x rmtadv: %04x", - ecmd.speed, ecmd.duplex, lcladv, rmtadv); + netdev_dbg(dev->net, "speed: %d duplex: %d lcladv: %04x rmtadv: %04x\n", + ecmd.speed, ecmd.duplex, lcladv, rmtadv); spin_lock_irqsave(&pdata->mac_cr_lock, flags); if (ecmd.duplex != DUPLEX_FULL) { @@ -511,7 +512,8 @@ static void smsc95xx_status(struct usbnet *dev, struct urb *urb) u32 intdata; if (urb->actual_length != 4) { - devwarn(dev, "unexpected urb length %d", urb->actual_length); + netdev_warn(dev->net, "unexpected urb length %d\n", + urb->actual_length); return; } @@ -519,12 +521,13 @@ static void smsc95xx_status(struct usbnet *dev, struct urb *urb) le32_to_cpus(&intdata); if (netif_msg_link(dev)) - devdbg(dev, "intdata: 0x%08X", intdata); + netdev_dbg(dev->net, "intdata: 0x%08X\n", intdata); if (intdata & INT_ENP_PHY_INT_) usbnet_defer_kevent(dev, EVENT_LINK_RESET); else - devwarn(dev, "unexpected interrupt, intdata=0x%08X", intdata); + netdev_warn(dev->net, "unexpected interrupt, intdata=0x%08X\n", + intdata); } /* Enable or disable Tx & Rx checksum offload engines */ @@ -534,7 +537,7 @@ static int smsc95xx_set_csums(struct usbnet *dev) u32 read_buf; int ret = smsc95xx_read_reg(dev, COE_CR, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read COE_CR: %d", ret); + netdev_warn(dev->net, "Failed to read COE_CR: %d\n", ret); return ret; } @@ -550,12 +553,12 @@ static int smsc95xx_set_csums(struct usbnet *dev) ret = smsc95xx_write_reg(dev, COE_CR, read_buf); if (ret < 0) { - devwarn(dev, "Failed to write COE_CR: %d", ret); + netdev_warn(dev->net, "Failed to write COE_CR: %d\n", ret); return ret; } if (netif_msg_hw(dev)) - devdbg(dev, "COE_CR = 0x%08x", read_buf); + netdev_dbg(dev->net, "COE_CR = 0x%08x\n", read_buf); return 0; } @@ -580,8 +583,8 @@ static int smsc95xx_ethtool_set_eeprom(struct net_device *netdev, struct usbnet *dev = netdev_priv(netdev); if (ee->magic != LAN95XX_EEPROM_MAGIC) { - devwarn(dev, "EEPROM: magic value mismatch, magic = 0x%x", - ee->magic); + netdev_warn(dev->net, "EEPROM: magic value mismatch, magic = 0x%x\n", + ee->magic); return -EINVAL; } @@ -660,7 +663,7 @@ static void smsc95xx_init_mac_address(struct usbnet *dev) if (is_valid_ether_addr(dev->net->dev_addr)) { /* eeprom values are valid so use them */ if (netif_msg_ifup(dev)) - devdbg(dev, "MAC address read from EEPROM"); + netdev_dbg(dev->net, "MAC address read from EEPROM\n"); return; } } @@ -668,7 +671,7 @@ static void smsc95xx_init_mac_address(struct usbnet *dev) /* no eeprom, or eeprom values are invalid. generate random MAC */ random_ether_addr(dev->net->dev_addr); if (netif_msg_ifup(dev)) - devdbg(dev, "MAC address set to random_ether_addr"); + netdev_dbg(dev->net, "MAC address set to random_ether_addr\n"); } static int smsc95xx_set_mac_address(struct usbnet *dev) @@ -680,13 +683,13 @@ static int smsc95xx_set_mac_address(struct usbnet *dev) ret = smsc95xx_write_reg(dev, ADDRL, addr_lo); if (ret < 0) { - devwarn(dev, "Failed to write ADDRL: %d", ret); + netdev_warn(dev->net, "Failed to write ADDRL: %d\n", ret); return ret; } ret = smsc95xx_write_reg(dev, ADDRH, addr_hi); if (ret < 0) { - devwarn(dev, "Failed to write ADDRH: %d", ret); + netdev_warn(dev->net, "Failed to write ADDRH: %d\n", ret); return ret; } @@ -748,7 +751,7 @@ static int smsc95xx_phy_initialize(struct usbnet *dev) mii_nway_restart(&dev->mii); if (netif_msg_ifup(dev)) - devdbg(dev, "phy initialised successfully"); + netdev_dbg(dev->net, "phy initialised successfully\n"); return 0; } @@ -760,13 +763,13 @@ static int smsc95xx_reset(struct usbnet *dev) int ret = 0, timeout; if (netif_msg_ifup(dev)) - devdbg(dev, "entering smsc95xx_reset"); + netdev_dbg(dev->net, "entering smsc95xx_reset\n"); write_buf = HW_CFG_LRST_; ret = smsc95xx_write_reg(dev, HW_CFG, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write HW_CFG_LRST_ bit in HW_CFG " - "register, ret = %d", ret); + netdev_warn(dev->net, "Failed to write HW_CFG_LRST_ bit in HW_CFG register, ret = %d\n", + ret); return ret; } @@ -774,7 +777,7 @@ static int smsc95xx_reset(struct usbnet *dev) do { ret = smsc95xx_read_reg(dev, HW_CFG, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read HW_CFG: %d", ret); + netdev_warn(dev->net, "Failed to read HW_CFG: %d\n", ret); return ret; } msleep(10); @@ -782,14 +785,14 @@ static int smsc95xx_reset(struct usbnet *dev) } while ((read_buf & HW_CFG_LRST_) && (timeout < 100)); if (timeout >= 100) { - devwarn(dev, "timeout waiting for completion of Lite Reset"); + netdev_warn(dev->net, "timeout waiting for completion of Lite Reset\n"); return ret; } write_buf = PM_CTL_PHY_RST_; ret = smsc95xx_write_reg(dev, PM_CTRL, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write PM_CTRL: %d", ret); + netdev_warn(dev->net, "Failed to write PM_CTRL: %d\n", ret); return ret; } @@ -797,7 +800,7 @@ static int smsc95xx_reset(struct usbnet *dev) do { ret = smsc95xx_read_reg(dev, PM_CTRL, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read PM_CTRL: %d", ret); + netdev_warn(dev->net, "Failed to read PM_CTRL: %d\n", ret); return ret; } msleep(10); @@ -805,7 +808,7 @@ static int smsc95xx_reset(struct usbnet *dev) } while ((read_buf & PM_CTL_PHY_RST_) && (timeout < 100)); if (timeout >= 100) { - devwarn(dev, "timeout waiting for PHY Reset"); + netdev_warn(dev->net, "timeout waiting for PHY Reset\n"); return ret; } @@ -816,34 +819,34 @@ static int smsc95xx_reset(struct usbnet *dev) return ret; if (netif_msg_ifup(dev)) - devdbg(dev, "MAC Address: %pM", dev->net->dev_addr); + netdev_dbg(dev->net, "MAC Address: %pM\n", dev->net->dev_addr); ret = smsc95xx_read_reg(dev, HW_CFG, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read HW_CFG: %d", ret); + netdev_warn(dev->net, "Failed to read HW_CFG: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from HW_CFG : 0x%08x", read_buf); + netdev_dbg(dev->net, "Read Value from HW_CFG : 0x%08x\n", read_buf); read_buf |= HW_CFG_BIR_; ret = smsc95xx_write_reg(dev, HW_CFG, read_buf); if (ret < 0) { - devwarn(dev, "Failed to write HW_CFG_BIR_ bit in HW_CFG " - "register, ret = %d", ret); + netdev_warn(dev->net, "Failed to write HW_CFG_BIR_ bit in HW_CFG register, ret = %d\n", + ret); return ret; } ret = smsc95xx_read_reg(dev, HW_CFG, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read HW_CFG: %d", ret); + netdev_warn(dev->net, "Failed to read HW_CFG: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from HW_CFG after writing " - "HW_CFG_BIR_: 0x%08x", read_buf); + netdev_dbg(dev->net, "Read Value from HW_CFG after writing HW_CFG_BIR_: 0x%08x\n", + read_buf); if (!turbo_mode) { burst_cap = 0; @@ -857,46 +860,46 @@ static int smsc95xx_reset(struct usbnet *dev) } if (netif_msg_ifup(dev)) - devdbg(dev, "rx_urb_size=%ld", (ulong)dev->rx_urb_size); + netdev_dbg(dev->net, "rx_urb_size=%ld\n", (ulong)dev->rx_urb_size); ret = smsc95xx_write_reg(dev, BURST_CAP, burst_cap); if (ret < 0) { - devwarn(dev, "Failed to write BURST_CAP: %d", ret); + netdev_warn(dev->net, "Failed to write BURST_CAP: %d\n", ret); return ret; } ret = smsc95xx_read_reg(dev, BURST_CAP, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read BURST_CAP: %d", ret); + netdev_warn(dev->net, "Failed to read BURST_CAP: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from BURST_CAP after writing: 0x%08x", - read_buf); + netdev_dbg(dev->net, "Read Value from BURST_CAP after writing: 0x%08x\n", + read_buf); read_buf = DEFAULT_BULK_IN_DELAY; ret = smsc95xx_write_reg(dev, BULK_IN_DLY, read_buf); if (ret < 0) { - devwarn(dev, "ret = %d", ret); + netdev_warn(dev->net, "ret = %d\n", ret); return ret; } ret = smsc95xx_read_reg(dev, BULK_IN_DLY, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read BULK_IN_DLY: %d", ret); + netdev_warn(dev->net, "Failed to read BULK_IN_DLY: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from BULK_IN_DLY after writing: " - "0x%08x", read_buf); + netdev_dbg(dev->net, "Read Value from BULK_IN_DLY after writing: 0x%08x\n", + read_buf); ret = smsc95xx_read_reg(dev, HW_CFG, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read HW_CFG: %d", ret); + netdev_warn(dev->net, "Failed to read HW_CFG: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from HW_CFG: 0x%08x", read_buf); + netdev_dbg(dev->net, "Read Value from HW_CFG: 0x%08x\n", read_buf); if (turbo_mode) read_buf |= (HW_CFG_MEF_ | HW_CFG_BCE_); @@ -908,41 +911,43 @@ static int smsc95xx_reset(struct usbnet *dev) ret = smsc95xx_write_reg(dev, HW_CFG, read_buf); if (ret < 0) { - devwarn(dev, "Failed to write HW_CFG register, ret=%d", ret); + netdev_warn(dev->net, "Failed to write HW_CFG register, ret=%d\n", + ret); return ret; } ret = smsc95xx_read_reg(dev, HW_CFG, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read HW_CFG: %d", ret); + netdev_warn(dev->net, "Failed to read HW_CFG: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "Read Value from HW_CFG after writing: 0x%08x", - read_buf); + netdev_dbg(dev->net, "Read Value from HW_CFG after writing: 0x%08x\n", + read_buf); write_buf = 0xFFFFFFFF; ret = smsc95xx_write_reg(dev, INT_STS, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write INT_STS register, ret=%d", ret); + netdev_warn(dev->net, "Failed to write INT_STS register, ret=%d\n", + ret); return ret; } ret = smsc95xx_read_reg(dev, ID_REV, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read ID_REV: %d", ret); + netdev_warn(dev->net, "Failed to read ID_REV: %d\n", ret); return ret; } if (netif_msg_ifup(dev)) - devdbg(dev, "ID_REV = 0x%08x", read_buf); + netdev_dbg(dev->net, "ID_REV = 0x%08x\n", read_buf); /* Configure GPIO pins as LED outputs */ write_buf = LED_GPIO_CFG_SPD_LED | LED_GPIO_CFG_LNK_LED | LED_GPIO_CFG_FDX_LED; ret = smsc95xx_write_reg(dev, LED_GPIO_CFG, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write LED_GPIO_CFG register, ret=%d", - ret); + netdev_warn(dev->net, "Failed to write LED_GPIO_CFG register, ret=%d\n", + ret); return ret; } @@ -950,21 +955,21 @@ static int smsc95xx_reset(struct usbnet *dev) write_buf = 0; ret = smsc95xx_write_reg(dev, FLOW, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write FLOW: %d", ret); + netdev_warn(dev->net, "Failed to write FLOW: %d\n", ret); return ret; } read_buf = AFC_CFG_DEFAULT; ret = smsc95xx_write_reg(dev, AFC_CFG, read_buf); if (ret < 0) { - devwarn(dev, "Failed to write AFC_CFG: %d", ret); + netdev_warn(dev->net, "Failed to write AFC_CFG: %d\n", ret); return ret; } /* Don't need mac_cr_lock during initialisation */ ret = smsc95xx_read_reg(dev, MAC_CR, &pdata->mac_cr); if (ret < 0) { - devwarn(dev, "Failed to read MAC_CR: %d", ret); + netdev_warn(dev->net, "Failed to read MAC_CR: %d\n", ret); return ret; } @@ -973,7 +978,7 @@ static int smsc95xx_reset(struct usbnet *dev) write_buf = (u32)ETH_P_8021Q; ret = smsc95xx_write_reg(dev, VLAN1, write_buf); if (ret < 0) { - devwarn(dev, "Failed to write VAN1: %d", ret); + netdev_warn(dev->net, "Failed to write VAN1: %d\n", ret); return ret; } @@ -981,7 +986,7 @@ static int smsc95xx_reset(struct usbnet *dev) ethtool_op_set_tx_hw_csum(netdev, pdata->use_tx_csum); ret = smsc95xx_set_csums(dev); if (ret < 0) { - devwarn(dev, "Failed to set csum offload: %d", ret); + netdev_warn(dev->net, "Failed to set csum offload: %d\n", ret); return ret; } @@ -992,7 +997,7 @@ static int smsc95xx_reset(struct usbnet *dev) ret = smsc95xx_read_reg(dev, INT_EP_CTL, &read_buf); if (ret < 0) { - devwarn(dev, "Failed to read INT_EP_CTL: %d", ret); + netdev_warn(dev->net, "Failed to read INT_EP_CTL: %d\n", ret); return ret; } @@ -1001,7 +1006,7 @@ static int smsc95xx_reset(struct usbnet *dev) ret = smsc95xx_write_reg(dev, INT_EP_CTL, read_buf); if (ret < 0) { - devwarn(dev, "Failed to write INT_EP_CTL: %d", ret); + netdev_warn(dev->net, "Failed to write INT_EP_CTL: %d\n", ret); return ret; } @@ -1009,7 +1014,7 @@ static int smsc95xx_reset(struct usbnet *dev) smsc95xx_start_rx_path(dev); if (netif_msg_ifup(dev)) - devdbg(dev, "smsc95xx_reset, return 0"); + netdev_dbg(dev->net, "smsc95xx_reset, return 0\n"); return 0; } @@ -1034,7 +1039,7 @@ static int smsc95xx_bind(struct usbnet *dev, struct usb_interface *intf) ret = usbnet_get_endpoints(dev, intf); if (ret < 0) { - devwarn(dev, "usbnet_get_endpoints failed: %d", ret); + netdev_warn(dev->net, "usbnet_get_endpoints failed: %d\n", ret); return ret; } @@ -1043,7 +1048,7 @@ static int smsc95xx_bind(struct usbnet *dev, struct usb_interface *intf) pdata = (struct smsc95xx_priv *)(dev->data[0]); if (!pdata) { - devwarn(dev, "Unable to allocate struct smsc95xx_priv"); + netdev_warn(dev->net, "Unable to allocate struct smsc95xx_priv\n"); return -ENOMEM; } @@ -1067,7 +1072,7 @@ static void smsc95xx_unbind(struct usbnet *dev, struct usb_interface *intf) struct smsc95xx_priv *pdata = (struct smsc95xx_priv *)(dev->data[0]); if (pdata) { if (netif_msg_ifdown(dev)) - devdbg(dev, "free pdata"); + netdev_dbg(dev->net, "free pdata\n"); kfree(pdata); pdata = NULL; dev->data[0] = 0; @@ -1102,7 +1107,8 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb) if (unlikely(header & RX_STS_ES_)) { if (netif_msg_rx_err(dev)) - devdbg(dev, "Error header=0x%08x", header); + netdev_dbg(dev->net, "Error header=0x%08x\n", + header); dev->net->stats.rx_errors++; dev->net->stats.rx_dropped++; @@ -1120,8 +1126,8 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb) /* ETH_FRAME_LEN + 4(CRC) + 2(COE) + 4(Vlan) */ if (unlikely(size > (ETH_FRAME_LEN + 12))) { if (netif_msg_rx_err(dev)) - devdbg(dev, "size err header=0x%08x", - header); + netdev_dbg(dev->net, "size err header=0x%08x\n", + header); return 0; } @@ -1137,7 +1143,7 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb) ax_skb = skb_clone(skb, GFP_ATOMIC); if (unlikely(!ax_skb)) { - devwarn(dev, "Error allocating skb"); + netdev_warn(dev->net, "Error allocating skb\n"); return 0; } @@ -1161,7 +1167,7 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb) } if (unlikely(skb->len < 0)) { - devwarn(dev, "invalid rx length<0 %d", skb->len); + netdev_warn(dev->net, "invalid rx length<0 %d\n", skb->len); return 0; } diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 035fab04c0a..8e732930d24 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -243,12 +243,12 @@ void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb) dev->net->stats.rx_bytes += skb->len; if (netif_msg_rx_status (dev)) - devdbg (dev, "< rx, len %zu, type 0x%x", - skb->len + sizeof (struct ethhdr), skb->protocol); + netdev_dbg(dev->net, "< rx, len %zu, type 0x%x\n", + skb->len + sizeof (struct ethhdr), skb->protocol); memset (skb->cb, 0, sizeof (struct skb_data)); status = netif_rx (skb); if (status != NET_RX_SUCCESS && netif_msg_rx_err (dev)) - devdbg (dev, "netif_rx status %d", status); + netdev_dbg(dev->net, "netif_rx status %d\n", status); } EXPORT_SYMBOL_GPL(usbnet_skb_return); @@ -313,9 +313,9 @@ void usbnet_defer_kevent (struct usbnet *dev, int work) { set_bit (work, &dev->flags); if (!schedule_work (&dev->kevent)) - deverr (dev, "kevent %d may have been dropped", work); + netdev_err(dev->net, "kevent %d may have been dropped\n", work); else - devdbg (dev, "kevent %d scheduled", work); + netdev_dbg(dev->net, "kevent %d scheduled\n", work); } EXPORT_SYMBOL_GPL(usbnet_defer_kevent); @@ -333,7 +333,7 @@ static void rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags) if ((skb = alloc_skb (size + NET_IP_ALIGN, flags)) == NULL) { if (netif_msg_rx_err (dev)) - devdbg (dev, "no rx skb"); + netdev_dbg(dev->net, "no rx skb\n"); usbnet_defer_kevent (dev, EVENT_RX_MEMORY); usb_free_urb (urb); return; @@ -364,12 +364,12 @@ static void rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags) break; case -ENODEV: if (netif_msg_ifdown (dev)) - devdbg (dev, "device gone"); + netdev_dbg(dev->net, "device gone\n"); netif_device_detach (dev->net); break; default: if (netif_msg_rx_err (dev)) - devdbg (dev, "rx submit, %d", retval); + netdev_dbg(dev->net, "rx submit, %d\n", retval); tasklet_schedule (&dev->bh); break; case 0: @@ -377,7 +377,7 @@ static void rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags) } } else { if (netif_msg_ifdown (dev)) - devdbg (dev, "rx: stopped"); + netdev_dbg(dev->net, "rx: stopped\n"); retval = -ENOLINK; } spin_unlock_irqrestore (&dev->rxq.lock, lockflags); @@ -401,7 +401,7 @@ static inline void rx_process (struct usbnet *dev, struct sk_buff *skb) usbnet_skb_return (dev, skb); else { if (netif_msg_rx_err (dev)) - devdbg (dev, "drop"); + netdev_dbg(dev->net, "drop\n"); error: dev->net->stats.rx_errors++; skb_queue_tail (&dev->done, skb); @@ -429,7 +429,7 @@ static void rx_complete (struct urb *urb) dev->net->stats.rx_errors++; dev->net->stats.rx_length_errors++; if (netif_msg_rx_err (dev)) - devdbg (dev, "rx length %d", skb->len); + netdev_dbg(dev->net, "rx length %d\n", skb->len); } break; @@ -447,7 +447,7 @@ static void rx_complete (struct urb *urb) case -ECONNRESET: /* async unlink */ case -ESHUTDOWN: /* hardware gone */ if (netif_msg_ifdown (dev)) - devdbg (dev, "rx shutdown, code %d", urb_status); + netdev_dbg(dev->net, "rx shutdown, code %d\n", urb_status); goto block; /* we get controller i/o faults during khubd disconnect() delays. @@ -461,7 +461,7 @@ static void rx_complete (struct urb *urb) if (!timer_pending (&dev->delay)) { mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES); if (netif_msg_link (dev)) - devdbg (dev, "rx throttle %d", urb_status); + netdev_dbg(dev->net, "rx throttle %d\n", urb_status); } block: entry->state = rx_cleanup; @@ -478,7 +478,7 @@ block: entry->state = rx_cleanup; dev->net->stats.rx_errors++; if (netif_msg_rx_err (dev)) - devdbg (dev, "rx status %d", urb_status); + netdev_dbg(dev->net, "rx status %d\n", urb_status); break; } @@ -493,7 +493,7 @@ block: usb_free_urb (urb); } if (netif_msg_rx_err (dev)) - devdbg (dev, "no read resubmitted"); + netdev_dbg(dev->net, "no read resubmitted\n"); } static void intr_complete (struct urb *urb) @@ -511,14 +511,14 @@ static void intr_complete (struct urb *urb) case -ENOENT: /* urb killed */ case -ESHUTDOWN: /* hardware gone */ if (netif_msg_ifdown (dev)) - devdbg (dev, "intr shutdown, code %d", status); + netdev_dbg(dev->net, "intr shutdown, code %d\n", status); return; /* NOTE: not throttling like RX/TX, since this endpoint * already polls infrequently */ default: - devdbg (dev, "intr status %d", status); + netdev_dbg(dev->net, "intr status %d\n", status); break; } @@ -528,7 +528,7 @@ static void intr_complete (struct urb *urb) memset(urb->transfer_buffer, 0, urb->transfer_buffer_length); status = usb_submit_urb (urb, GFP_ATOMIC); if (status != 0 && netif_msg_timer (dev)) - deverr(dev, "intr resubmit --> %d", status); + netdev_err(dev->net, "intr resubmit --> %d\n", status); } /*-------------------------------------------------------------------------*/ @@ -537,7 +537,7 @@ void usbnet_pause_rx(struct usbnet *dev) set_bit(EVENT_RX_PAUSED, &dev->flags); if (netif_msg_rx_status(dev)) - devdbg(dev, "paused rx queue enabled"); + netdev_dbg(dev->net, "paused rx queue enabled\n"); } EXPORT_SYMBOL_GPL(usbnet_pause_rx); @@ -556,7 +556,8 @@ void usbnet_resume_rx(struct usbnet *dev) tasklet_schedule(&dev->bh); if (netif_msg_rx_status(dev)) - devdbg(dev, "paused rx queue disabled, %d skbs requeued", num); + netdev_dbg(dev->net, "paused rx queue disabled, %d skbs requeued\n", + num); } EXPORT_SYMBOL_GPL(usbnet_resume_rx); @@ -589,7 +590,7 @@ static int unlink_urbs (struct usbnet *dev, struct sk_buff_head *q) // these (async) unlinks complete immediately retval = usb_unlink_urb (urb); if (retval != -EINPROGRESS && retval != 0) - devdbg (dev, "unlink urb err, %d", retval); + netdev_dbg(dev->net, "unlink urb err, %d\n", retval); else count++; } @@ -632,8 +633,8 @@ static void usbnet_terminate_urbs(struct usbnet *dev) schedule_timeout(UNLINK_TIMEOUT_MS); set_current_state(TASK_UNINTERRUPTIBLE); if (netif_msg_ifdown(dev)) - devdbg(dev, "waited for %d urb completions", - temp); + netdev_dbg(dev->net, "waited for %d urb completions\n", + temp); } set_current_state(TASK_RUNNING); dev->wait = NULL; @@ -649,21 +650,20 @@ int usbnet_stop (struct net_device *net) netif_stop_queue (net); if (netif_msg_ifdown (dev)) - devinfo (dev, "stop stats: rx/tx %ld/%ld, errs %ld/%ld", - net->stats.rx_packets, net->stats.tx_packets, - net->stats.rx_errors, net->stats.tx_errors - ); + netdev_info(dev->net, "stop stats: rx/tx %ld/%ld, errs %ld/%ld\n", + net->stats.rx_packets, net->stats.tx_packets, + net->stats.rx_errors, net->stats.tx_errors); /* allow minidriver to stop correctly (wireless devices to turn off * radio etc) */ if (info->stop) { retval = info->stop(dev); if (retval < 0 && netif_msg_ifdown(dev)) - devinfo(dev, - "stop fail (%d) usbnet usb-%s-%s, %s", - retval, - dev->udev->bus->bus_name, dev->udev->devpath, - info->description); + netdev_info(dev->net, + "stop fail (%d) usbnet usb-%s-%s, %s\n", + retval, + dev->udev->bus->bus_name, dev->udev->devpath, + info->description); } if (!(info->flags & FLAG_AVOID_UNLINK_URBS)) @@ -703,29 +703,31 @@ int usbnet_open (struct net_device *net) if ((retval = usb_autopm_get_interface(dev->intf)) < 0) { if (netif_msg_ifup (dev)) - devinfo (dev, - "resumption fail (%d) usbnet usb-%s-%s, %s", - retval, - dev->udev->bus->bus_name, dev->udev->devpath, - info->description); + netdev_info(dev->net, + "resumption fail (%d) usbnet usb-%s-%s, %s\n", + retval, + dev->udev->bus->bus_name, + dev->udev->devpath, + info->description); goto done_nopm; } // put into "known safe" state if (info->reset && (retval = info->reset (dev)) < 0) { if (netif_msg_ifup (dev)) - devinfo (dev, - "open reset fail (%d) usbnet usb-%s-%s, %s", - retval, - dev->udev->bus->bus_name, dev->udev->devpath, - info->description); + netdev_info(dev->net, + "open reset fail (%d) usbnet usb-%s-%s, %s\n", + retval, + dev->udev->bus->bus_name, + dev->udev->devpath, + info->description); goto done; } // insist peer be connected if (info->check_connect && (retval = info->check_connect (dev)) < 0) { if (netif_msg_ifup (dev)) - devdbg (dev, "can't open; %d", retval); + netdev_dbg(dev->net, "can't open; %d\n", retval); goto done; } @@ -734,7 +736,7 @@ int usbnet_open (struct net_device *net) retval = usb_submit_urb (dev->interrupt, GFP_KERNEL); if (retval < 0) { if (netif_msg_ifup (dev)) - deverr (dev, "intr submit %d", retval); + netdev_err(dev->net, "intr submit %d\n", retval); goto done; } } @@ -756,10 +758,9 @@ int usbnet_open (struct net_device *net) else framing = "simple"; - devinfo (dev, "open: enable queueing " - "(rx %d, tx %d) mtu %d %s framing", - (int)RX_QLEN (dev), (int)TX_QLEN (dev), dev->net->mtu, - framing); + netdev_info(dev->net, "open: enable queueing (rx %d, tx %d) mtu %d %s framing\n", + (int)RX_QLEN(dev), (int)TX_QLEN(dev), + dev->net->mtu, framing); } // delay posting reads until we're fully open @@ -908,8 +909,8 @@ kevent (struct work_struct *work) status != -ESHUTDOWN) { if (netif_msg_tx_err (dev)) fail_pipe: - deverr (dev, "can't clear tx halt, status %d", - status); + netdev_err(dev->net, "can't clear tx halt, status %d\n", + status); } else { clear_bit (EVENT_TX_HALT, &dev->flags); if (status != -ESHUTDOWN) @@ -928,8 +929,8 @@ fail_pipe: status != -ESHUTDOWN) { if (netif_msg_rx_err (dev)) fail_halt: - deverr (dev, "can't clear rx halt, status %d", - status); + netdev_err(dev->net, "can't clear rx halt, status %d\n", + status); } else { clear_bit (EVENT_RX_HALT, &dev->flags); tasklet_schedule (&dev->bh); @@ -967,18 +968,18 @@ fail_lowmem: if(info->link_reset && (retval = info->link_reset(dev)) < 0) { usb_autopm_put_interface(dev->intf); skip_reset: - devinfo(dev, "link reset failed (%d) usbnet usb-%s-%s, %s", - retval, - dev->udev->bus->bus_name, dev->udev->devpath, - info->description); + netdev_info(dev->net, "link reset failed (%d) usbnet usb-%s-%s, %s\n", + retval, + dev->udev->bus->bus_name, + dev->udev->devpath, + info->description); } else { usb_autopm_put_interface(dev->intf); } } if (dev->flags) - devdbg (dev, "kevent done, flags = 0x%lx", - dev->flags); + netdev_dbg(dev->net, "kevent done, flags = 0x%lx\n", dev->flags); } /*-------------------------------------------------------------------------*/ @@ -1015,14 +1016,15 @@ static void tx_complete (struct urb *urb) mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES); if (netif_msg_link (dev)) - devdbg (dev, "tx throttle %d", - urb->status); + netdev_dbg(dev->net, "tx throttle %d\n", + urb->status); } netif_stop_queue (dev->net); break; default: if (netif_msg_tx_err (dev)) - devdbg (dev, "tx err %d", entry->urb->status); + netdev_dbg(dev->net, "tx err %d\n", + entry->urb->status); break; } } @@ -1065,7 +1067,7 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, skb = info->tx_fixup (dev, skb, GFP_ATOMIC); if (!skb) { if (netif_msg_tx_err (dev)) - devdbg (dev, "can't tx_fixup skb"); + netdev_dbg(dev->net, "can't tx_fixup skb\n"); goto drop; } } @@ -1073,7 +1075,7 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, if (!(urb = usb_alloc_urb (0, GFP_ATOMIC))) { if (netif_msg_tx_err (dev)) - devdbg (dev, "no urb"); + netdev_dbg(dev->net, "no urb\n"); goto drop; } @@ -1113,7 +1115,7 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, /* no use to process more packets */ netif_stop_queue(net); spin_unlock_irqrestore(&dev->txq.lock, flags); - devdbg(dev, "Delaying transmission for resumption"); + netdev_dbg(dev->net, "Delaying transmission for resumption\n"); goto deferred; } #endif @@ -1127,7 +1129,7 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, default: usb_autopm_put_interface_async(dev->intf); if (netif_msg_tx_err (dev)) - devdbg (dev, "tx: submit urb err %d", retval); + netdev_dbg(dev->net, "tx: submit urb err %d\n", retval); break; case 0: net->trans_start = jiffies; @@ -1139,15 +1141,15 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, if (retval) { if (netif_msg_tx_err (dev)) - devdbg (dev, "drop, code %d", retval); + netdev_dbg(dev->net, "drop, code %d\n", retval); drop: dev->net->stats.tx_dropped++; if (skb) dev_kfree_skb_any (skb); usb_free_urb (urb); } else if (netif_msg_tx_queued (dev)) { - devdbg (dev, "> tx, len %d, type 0x%x", - length, skb->protocol); + netdev_dbg(dev->net, "> tx, len %d, type 0x%x\n", + length, skb->protocol); } #ifdef CONFIG_PM deferred: @@ -1179,7 +1181,7 @@ static void usbnet_bh (unsigned long param) dev_kfree_skb (skb); continue; default: - devdbg (dev, "bogus skb state %d", entry->state); + netdev_dbg(dev->net, "bogus skb state %d\n", entry->state); } } @@ -1208,8 +1210,8 @@ static void usbnet_bh (unsigned long param) rx_submit (dev, urb, GFP_ATOMIC); } if (temp != dev->rxq.qlen && netif_msg_link (dev)) - devdbg (dev, "rxqlen %d --> %d", - temp, dev->rxq.qlen); + netdev_dbg(dev->net, "rxqlen %d --> %d\n", + temp, dev->rxq.qlen); if (dev->rxq.qlen < qlen) tasklet_schedule (&dev->bh); } @@ -1241,10 +1243,10 @@ void usbnet_disconnect (struct usb_interface *intf) xdev = interface_to_usbdev (intf); if (netif_msg_probe (dev)) - devinfo (dev, "unregister '%s' usb-%s-%s, %s", - intf->dev.driver->name, - xdev->bus->bus_name, xdev->devpath, - dev->driver_info->description); + netdev_info(dev->net, "unregister '%s' usb-%s-%s, %s\n", + intf->dev.driver->name, + xdev->bus->bus_name, xdev->devpath, + dev->driver_info->description); net = dev->net; unregister_netdev (net); @@ -1408,11 +1410,11 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) if (status) goto out3; if (netif_msg_probe (dev)) - devinfo (dev, "register '%s' at usb-%s-%s, %s, %pM", - udev->dev.driver->name, - xdev->bus->bus_name, xdev->devpath, - dev->driver_info->description, - net->dev_addr); + netdev_info(dev->net, "register '%s' at usb-%s-%s, %s, %pM\n", + udev->dev.driver->name, + xdev->bus->bus_name, xdev->devpath, + dev->driver_info->description, + net->dev_addr); // ok, it's ready to go. usb_set_intfdata (udev, dev); diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 14692bc51b5..65cbd060cc3 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -728,9 +728,9 @@ static int rndis_query_oid(struct usbnet *dev, __le32 oid, void *data, int *len) ret = rndis_command(dev, u.header, buflen); priv->current_command_oid = 0; if (ret < 0) - devdbg(dev, "rndis_query_oid(%s): rndis_command() failed, %d " - "(%08x)", oid_to_string(oid), ret, - le32_to_cpu(u.get_c->status)); + netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n", + __func__, oid_to_string(oid), ret, + le32_to_cpu(u.get_c->status)); if (ret == 0) { memcpy(data, u.buf + le32_to_cpu(u.get_c->offset) + 8, *len); @@ -741,9 +741,9 @@ static int rndis_query_oid(struct usbnet *dev, __le32 oid, void *data, int *len) ret = rndis_error_status(u.get_c->status); if (ret < 0) - devdbg(dev, "rndis_query_oid(%s): device returned " - "error, 0x%08x (%d)", oid_to_string(oid), - le32_to_cpu(u.get_c->status), ret); + netdev_dbg(dev->net, "%s(%s): device returned error, 0x%08x (%d)\n", + __func__, oid_to_string(oid), + le32_to_cpu(u.get_c->status), ret); } mutex_unlock(&priv->command_lock); @@ -791,17 +791,17 @@ static int rndis_set_oid(struct usbnet *dev, __le32 oid, void *data, int len) ret = rndis_command(dev, u.header, buflen); priv->current_command_oid = 0; if (ret < 0) - devdbg(dev, "rndis_set_oid(%s): rndis_command() failed, %d " - "(%08x)", oid_to_string(oid), ret, - le32_to_cpu(u.set_c->status)); + netdev_dbg(dev->net, "%s(%s): rndis_command() failed, %d (%08x)\n", + __func__, oid_to_string(oid), ret, + le32_to_cpu(u.set_c->status)); if (ret == 0) { ret = rndis_error_status(u.set_c->status); if (ret < 0) - devdbg(dev, "rndis_set_oid(%s): device returned error, " - "0x%08x (%d)", oid_to_string(oid), - le32_to_cpu(u.set_c->status), ret); + netdev_dbg(dev->net, "%s(%s): device returned error, 0x%08x (%d)\n", + __func__, oid_to_string(oid), + le32_to_cpu(u.set_c->status), ret); } mutex_unlock(&priv->command_lock); @@ -870,11 +870,11 @@ static int rndis_set_config_parameter(struct usbnet *dev, char *param, #endif if (value_type == 2) - devdbg(dev, "setting config parameter: %s, value: %s", - param, (u8 *)value); + netdev_dbg(dev->net, "setting config parameter: %s, value: %s\n", + param, (u8 *)value); else - devdbg(dev, "setting config parameter: %s, value: %d", - param, *(u32 *)value); + netdev_dbg(dev->net, "setting config parameter: %s, value: %d\n", + param, *(u32 *)value); infobuf->name_offs = cpu_to_le32(sizeof(*infobuf)); infobuf->name_length = cpu_to_le32(param_len); @@ -897,20 +897,21 @@ static int rndis_set_config_parameter(struct usbnet *dev, char *param, } #ifdef DEBUG - devdbg(dev, "info buffer (len: %d):", info_len); + netdev_dbg(dev->net, "info buffer (len: %d)\n", info_len); for (i = 0; i < info_len; i += 12) { u32 *tmp = (u32 *)((u8 *)infobuf + i); - devdbg(dev, "%08X:%08X:%08X", - cpu_to_be32(tmp[0]), - cpu_to_be32(tmp[1]), - cpu_to_be32(tmp[2])); + netdev_dbg(dev->net, "%08X:%08X:%08X\n", + cpu_to_be32(tmp[0]), + cpu_to_be32(tmp[1]), + cpu_to_be32(tmp[2])); } #endif ret = rndis_set_oid(dev, OID_GEN_RNDIS_CONFIG_PARAMETER, infobuf, info_len); if (ret != 0) - devdbg(dev, "setting rndis config parameter failed, %d.", ret); + netdev_dbg(dev->net, "setting rndis config parameter failed, %d\n", + ret); kfree(infobuf); return ret; @@ -945,13 +946,13 @@ static int set_essid(struct usbnet *usbdev, struct ndis_80211_ssid *ssid) ret = rndis_set_oid(usbdev, OID_802_11_SSID, ssid, sizeof(*ssid)); if (ret < 0) { - devwarn(usbdev, "setting SSID failed (%08X)", ret); + netdev_warn(usbdev->net, "setting SSID failed (%08X)\n", ret); return ret; } if (ret == 0) { memcpy(&priv->essid, ssid, sizeof(priv->essid)); priv->radio_on = true; - devdbg(usbdev, "set_essid: radio_on = true"); + netdev_dbg(usbdev->net, "%s(): radio_on = true\n", __func__); } return ret; @@ -963,7 +964,8 @@ static int set_bssid(struct usbnet *usbdev, u8 bssid[ETH_ALEN]) ret = rndis_set_oid(usbdev, OID_802_11_BSSID, bssid, ETH_ALEN); if (ret < 0) { - devwarn(usbdev, "setting BSSID[%pM] failed (%08X)", bssid, ret); + netdev_warn(usbdev->net, "setting BSSID[%pM] failed (%08X)\n", + bssid, ret); return ret; } @@ -1021,7 +1023,8 @@ static int disassociate(struct usbnet *usbdev, bool reset_ssid) ret = rndis_set_oid(usbdev, OID_802_11_DISASSOCIATE, NULL, 0); if (ret == 0) { priv->radio_on = false; - devdbg(usbdev, "disassociate: radio_on = false"); + netdev_dbg(usbdev->net, "%s(): radio_on = false\n", + __func__); if (reset_ssid) msleep(100); @@ -1054,8 +1057,8 @@ static int set_auth_mode(struct usbnet *usbdev, u32 wpa_version, __le32 tmp; int auth_mode, ret; - devdbg(usbdev, "set_auth_mode: wpa_version=0x%x authalg=0x%x " - "keymgmt=0x%x", wpa_version, auth_type, keymgmt); + netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x authalg=0x%x keymgmt=0x%x\n", + __func__, wpa_version, auth_type, keymgmt); if (wpa_version & NL80211_WPA_VERSION_2) { if (keymgmt & RNDIS_WLAN_KEY_MGMT_802_1X) @@ -1082,7 +1085,8 @@ static int set_auth_mode(struct usbnet *usbdev, u32 wpa_version, ret = rndis_set_oid(usbdev, OID_802_11_AUTHENTICATION_MODE, &tmp, sizeof(tmp)); if (ret != 0) { - devwarn(usbdev, "setting auth mode failed (%08X)", ret); + netdev_warn(usbdev->net, "setting auth mode failed (%08X)\n", + ret); return ret; } @@ -1098,7 +1102,8 @@ static int set_priv_filter(struct usbnet *usbdev) struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev); __le32 tmp; - devdbg(usbdev, "set_priv_filter: wpa_version=0x%x", priv->wpa_version); + netdev_dbg(usbdev->net, "%s(): wpa_version=0x%x\n", + __func__, priv->wpa_version); if (priv->wpa_version & NL80211_WPA_VERSION_2 || priv->wpa_version & NL80211_WPA_VERSION_1) @@ -1116,8 +1121,8 @@ static int set_encr_mode(struct usbnet *usbdev, int pairwise, int groupwise) __le32 tmp; int encr_mode, ret; - devdbg(usbdev, "set_encr_mode: cipher_pair=0x%x cipher_group=0x%x", - pairwise, groupwise); + netdev_dbg(usbdev->net, "%s(): cipher_pair=0x%x cipher_group=0x%x\n", + __func__, pairwise, groupwise); if (pairwise & RNDIS_WLAN_ALG_CCMP) encr_mode = NDIS_80211_ENCR_CCMP_ENABLED; @@ -1136,7 +1141,8 @@ static int set_encr_mode(struct usbnet *usbdev, int pairwise, int groupwise) ret = rndis_set_oid(usbdev, OID_802_11_ENCRYPTION_STATUS, &tmp, sizeof(tmp)); if (ret != 0) { - devwarn(usbdev, "setting encr mode failed (%08X)", ret); + netdev_warn(usbdev->net, "setting encr mode failed (%08X)\n", + ret); return ret; } @@ -1151,13 +1157,15 @@ static int set_infra_mode(struct usbnet *usbdev, int mode) __le32 tmp; int ret; - devdbg(usbdev, "set_infra_mode: infra_mode=0x%x", priv->infra_mode); + netdev_dbg(usbdev->net, "%s(): infra_mode=0x%x\n", + __func__, priv->infra_mode); tmp = cpu_to_le32(mode); ret = rndis_set_oid(usbdev, OID_802_11_INFRASTRUCTURE_MODE, &tmp, sizeof(tmp)); if (ret != 0) { - devwarn(usbdev, "setting infra mode failed (%08X)", ret); + netdev_warn(usbdev->net, "setting infra mode failed (%08X)\n", + ret); return ret; } @@ -1174,7 +1182,7 @@ static int set_rts_threshold(struct usbnet *usbdev, u32 rts_threshold) { __le32 tmp; - devdbg(usbdev, "set_rts_threshold %i", rts_threshold); + netdev_dbg(usbdev->net, "%s(): %i\n", __func__, rts_threshold); if (rts_threshold < 0 || rts_threshold > 2347) rts_threshold = 2347; @@ -1188,7 +1196,7 @@ static int set_frag_threshold(struct usbnet *usbdev, u32 frag_threshold) { __le32 tmp; - devdbg(usbdev, "set_frag_threshold %i", frag_threshold); + netdev_dbg(usbdev->net, "%s(): %i\n", __func__, frag_threshold); if (frag_threshold < 256 || frag_threshold > 2346) frag_threshold = 2346; @@ -1222,7 +1230,7 @@ static int set_channel(struct usbnet *usbdev, int channel) unsigned int dsconfig; int len, ret; - devdbg(usbdev, "set_channel(%d)", channel); + netdev_dbg(usbdev->net, "%s(%d)\n", __func__, channel); /* this OID is valid only when not associated */ if (is_associated(usbdev)) @@ -1233,7 +1241,8 @@ static int set_channel(struct usbnet *usbdev, int channel) len = sizeof(config); ret = rndis_query_oid(usbdev, OID_802_11_CONFIGURATION, &config, &len); if (ret < 0) { - devdbg(usbdev, "set_channel: querying configuration failed"); + netdev_dbg(usbdev->net, "%s(): querying configuration failed\n", + __func__); return ret; } @@ -1241,7 +1250,7 @@ static int set_channel(struct usbnet *usbdev, int channel) ret = rndis_set_oid(usbdev, OID_802_11_CONFIGURATION, &config, sizeof(config)); - devdbg(usbdev, "set_channel: %d -> %d", channel, ret); + netdev_dbg(usbdev->net, "%s(): %d -> %d\n", __func__, channel, ret); return ret; } @@ -1255,7 +1264,8 @@ static int add_wep_key(struct usbnet *usbdev, const u8 *key, int key_len, u32 cipher; int ret; - devdbg(usbdev, "add_wep_key(idx: %d, len: %d)", index, key_len); + netdev_dbg(usbdev->net, "%s(idx: %d, len: %d)\n", + __func__, index, key_len); if ((key_len != 5 && key_len != 13) || index < 0 || index > 3) return -EINVAL; @@ -1277,15 +1287,15 @@ static int add_wep_key(struct usbnet *usbdev, const u8 *key, int key_len, ret = set_encr_mode(usbdev, RNDIS_WLAN_ALG_WEP, RNDIS_WLAN_ALG_NONE); if (ret) - devwarn(usbdev, "encryption couldn't be enabled (%08X)", - ret); + netdev_warn(usbdev->net, "encryption couldn't be enabled (%08X)\n", + ret); } ret = rndis_set_oid(usbdev, OID_802_11_ADD_WEP, &ndis_key, sizeof(ndis_key)); if (ret != 0) { - devwarn(usbdev, "adding encryption key %d failed (%08X)", - index+1, ret); + netdev_warn(usbdev->net, "adding encryption key %d failed (%08X)\n", + index + 1, ret); return ret; } @@ -1307,22 +1317,23 @@ static int add_wpa_key(struct usbnet *usbdev, const u8 *key, int key_len, int ret; if (index < 0 || index >= 4) { - devdbg(usbdev, "add_wpa_key: index out of range (%i)", index); + netdev_dbg(usbdev->net, "%s(): index out of range (%i)\n", + __func__, index); return -EINVAL; } if (key_len > sizeof(ndis_key.material) || key_len < 0) { - devdbg(usbdev, "add_wpa_key: key length out of range (%i)", - key_len); + netdev_dbg(usbdev->net, "%s(): key length out of range (%i)\n", + __func__, key_len); return -EINVAL; } if (flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ) { if (!rx_seq || seq_len <= 0) { - devdbg(usbdev, "add_wpa_key: recv seq flag without" - "buffer"); + netdev_dbg(usbdev->net, "%s(): recv seq flag without buffer\n", + __func__); return -EINVAL; } if (rx_seq && seq_len > sizeof(ndis_key.rsc)) { - devdbg(usbdev, "add_wpa_key: too big recv seq buffer"); + netdev_dbg(usbdev->net, "%s(): too big recv seq buffer\n", __func__); return -EINVAL; } } @@ -1330,15 +1341,16 @@ static int add_wpa_key(struct usbnet *usbdev, const u8 *key, int key_len, is_addr_ok = addr && !is_zero_ether_addr(addr) && !is_broadcast_ether_addr(addr); if ((flags & NDIS_80211_ADDKEY_PAIRWISE_KEY) && !is_addr_ok) { - devdbg(usbdev, "add_wpa_key: pairwise but bssid invalid (%pM)", - addr); + netdev_dbg(usbdev->net, "%s(): pairwise but bssid invalid (%pM)\n", + __func__, addr); return -EINVAL; } - devdbg(usbdev, "add_wpa_key(%i): flags:%i%i%i", index, - !!(flags & NDIS_80211_ADDKEY_TRANSMIT_KEY), - !!(flags & NDIS_80211_ADDKEY_PAIRWISE_KEY), - !!(flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ)); + netdev_dbg(usbdev->net, "%s(%i): flags:%i%i%i\n", + __func__, index, + !!(flags & NDIS_80211_ADDKEY_TRANSMIT_KEY), + !!(flags & NDIS_80211_ADDKEY_PAIRWISE_KEY), + !!(flags & NDIS_80211_ADDKEY_SET_INIT_RECV_SEQ)); memset(&ndis_key, 0, sizeof(ndis_key)); @@ -1372,7 +1384,8 @@ static int add_wpa_key(struct usbnet *usbdev, const u8 *key, int key_len, ret = rndis_set_oid(usbdev, OID_802_11_ADD_KEY, &ndis_key, le32_to_cpu(ndis_key.size)); - devdbg(usbdev, "add_wpa_key: OID_802_11_ADD_KEY -> %08X", ret); + netdev_dbg(usbdev->net, "%s(): OID_802_11_ADD_KEY -> %08X\n", + __func__, ret); if (ret != 0) return ret; @@ -1401,7 +1414,7 @@ static int restore_key(struct usbnet *usbdev, int key_idx) key = priv->encr_keys[key_idx]; - devdbg(usbdev, "restore_key: %i:%i", key_idx, key.len); + netdev_dbg(usbdev->net, "%s(): %i:%i\n", __func__, key_idx, key.len); if (key.len == 0) return 0; @@ -1436,8 +1449,9 @@ static int remove_key(struct usbnet *usbdev, int index, const u8 *bssid) is_wpa = is_wpa_key(priv, index); - devdbg(usbdev, "remove_key: %i:%s:%i", index, is_wpa ? "wpa" : "wep", - priv->encr_keys[index].len); + netdev_dbg(usbdev->net, "%s(): %i:%s:%i\n", + __func__, index, is_wpa ? "wpa" : "wep", + priv->encr_keys[index].len); clear_key(priv, index); @@ -1464,9 +1478,9 @@ static int remove_key(struct usbnet *usbdev, int index, const u8 *bssid) ret = rndis_set_oid(usbdev, OID_802_11_REMOVE_WEP, &keyindex, sizeof(keyindex)); if (ret != 0) { - devwarn(usbdev, - "removing encryption key %d failed (%08X)", - index, ret); + netdev_warn(usbdev->net, + "removing encryption key %d failed (%08X)\n", + index, ret); return ret; } } @@ -1498,9 +1512,9 @@ static void set_multicast_list(struct usbnet *usbdev) size = min(priv->multicast_size, netdev_mc_count(usbdev->net)); buf = kmalloc(size * ETH_ALEN, GFP_KERNEL); if (!buf) { - devwarn(usbdev, - "couldn't alloc %d bytes of memory", - size * ETH_ALEN); + netdev_warn(usbdev->net, + "couldn't alloc %d bytes of memory\n", + size * ETH_ALEN); return; } @@ -1520,8 +1534,8 @@ static void set_multicast_list(struct usbnet *usbdev) else filter |= RNDIS_PACKET_TYPE_ALL_MULTICAST; - devdbg(usbdev, "OID_802_3_MULTICAST_LIST(%d, max: %d) -> %d", - i, priv->multicast_size, ret); + netdev_dbg(usbdev->net, "OID_802_3_MULTICAST_LIST(%d, max: %d) -> %d\n", + i, priv->multicast_size, ret); kfree(buf); } @@ -1529,12 +1543,12 @@ static void set_multicast_list(struct usbnet *usbdev) ret = rndis_set_oid(usbdev, OID_GEN_CURRENT_PACKET_FILTER, &filter, sizeof(filter)); if (ret < 0) { - devwarn(usbdev, "couldn't set packet filter: %08x", - le32_to_cpu(filter)); + netdev_warn(usbdev->net, "couldn't set packet filter: %08x\n", + le32_to_cpu(filter)); } - devdbg(usbdev, "OID_GEN_CURRENT_PACKET_FILTER(%08x) -> %d", - le32_to_cpu(filter), ret); + netdev_dbg(usbdev->net, "OID_GEN_CURRENT_PACKET_FILTER(%08x) -> %d\n", + le32_to_cpu(filter), ret); } /* @@ -1592,7 +1606,8 @@ static int rndis_set_tx_power(struct wiphy *wiphy, enum tx_power_setting type, struct rndis_wlan_private *priv = wiphy_priv(wiphy); struct usbnet *usbdev = priv->usbdev; - devdbg(usbdev, "rndis_set_tx_power type:0x%x dbm:%i", type, dbm); + netdev_dbg(usbdev->net, "%s(): type:0x%x dbm:%i\n", + __func__, type, dbm); /* Device doesn't support changing txpower after initialization, only * turn off/on radio. Support 'auto' mode and setting same dBm that is @@ -1615,7 +1630,7 @@ static int rndis_get_tx_power(struct wiphy *wiphy, int *dbm) *dbm = get_bcm4320_power_dbm(priv); - devdbg(usbdev, "rndis_get_tx_power dbm:%i", *dbm); + netdev_dbg(usbdev->net, "%s(): dbm:%i\n", __func__, *dbm); return 0; } @@ -1629,7 +1644,7 @@ static int rndis_scan(struct wiphy *wiphy, struct net_device *dev, int ret; __le32 tmp; - devdbg(usbdev, "cfg80211.scan"); + netdev_dbg(usbdev->net, "cfg80211.scan\n"); /* Get current bssid list from device before new scan, as new scan * clears internal bssid list. @@ -1669,8 +1684,8 @@ static struct cfg80211_bss *rndis_bss_info_update(struct usbnet *usbdev, int ie_len, bssid_len; u8 *ie; - devdbg(usbdev, " found bssid: '%.32s' [%pM]", bssid->ssid.essid, - bssid->mac); + netdev_dbg(usbdev->net, " found bssid: '%.32s' [%pM]\n", + bssid->ssid.essid, bssid->mac); /* parse bssid structure */ bssid_len = le32_to_cpu(bssid->length); @@ -1712,7 +1727,7 @@ static int rndis_check_bssid_list(struct usbnet *usbdev) int ret = -EINVAL, len, count, bssid_len; bool resized = false; - devdbg(usbdev, "check_bssid_list"); + netdev_dbg(usbdev->net, "check_bssid_list\n"); len = CONTROL_BUFFER_SIZE; resize_buf: @@ -1736,8 +1751,8 @@ resize_buf: bssid = bssid_list->bssid; bssid_len = le32_to_cpu(bssid->length); count = le32_to_cpu(bssid_list->num_items); - devdbg(usbdev, "check_bssid_list: %d BSSIDs found (buflen: %d)", count, - len); + netdev_dbg(usbdev->net, "check_bssid_list: %d BSSIDs found (buflen: %d)\n", + count, len); while (count && ((void *)bssid + bssid_len) <= (buf + len)) { rndis_bss_info_update(usbdev, bssid); @@ -1759,7 +1774,7 @@ static void rndis_get_scan_results(struct work_struct *work) struct usbnet *usbdev = priv->usbdev; int ret; - devdbg(usbdev, "get_scan_results"); + netdev_dbg(usbdev->net, "get_scan_results\n"); if (!priv->scan_request) return; @@ -1793,7 +1808,7 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, if (sme->crypto.n_ciphers_pairwise > 0 && pairwise == RNDIS_WLAN_ALG_NONE) { - deverr(usbdev, "Unsupported pairwise cipher"); + netdev_err(usbdev->net, "Unsupported pairwise cipher\n"); return -ENOTSUPP; } @@ -1803,28 +1818,30 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, if (sme->crypto.n_akm_suites > 0 && keymgmt == RNDIS_WLAN_KEY_MGMT_NONE) { - deverr(usbdev, "Invalid keymgmt"); + netdev_err(usbdev->net, "Invalid keymgmt\n"); return -ENOTSUPP; } - devdbg(usbdev, "cfg80211.connect('%.32s':[%pM]:%d:[%d,0x%x:0x%x]:[0x%x:" - "0x%x]:0x%x)", sme->ssid, sme->bssid, chan, - sme->privacy, sme->crypto.wpa_versions, sme->auth_type, - groupwise, pairwise, keymgmt); + netdev_dbg(usbdev->net, "cfg80211.connect('%.32s':[%pM]:%d:[%d,0x%x:0x%x]:[0x%x:0x%x]:0x%x)\n", + sme->ssid, sme->bssid, chan, + sme->privacy, sme->crypto.wpa_versions, sme->auth_type, + groupwise, pairwise, keymgmt); if (is_associated(usbdev)) disassociate(usbdev, false); ret = set_infra_mode(usbdev, NDIS_80211_INFRA_INFRA); if (ret < 0) { - devdbg(usbdev, "connect: set_infra_mode failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_infra_mode failed, %d\n", + ret); goto err_turn_radio_on; } ret = set_auth_mode(usbdev, sme->crypto.wpa_versions, sme->auth_type, keymgmt); if (ret < 0) { - devdbg(usbdev, "connect: set_auth_mode failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_auth_mode failed, %d\n", + ret); goto err_turn_radio_on; } @@ -1832,14 +1849,16 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, ret = set_encr_mode(usbdev, pairwise, groupwise); if (ret < 0) { - devdbg(usbdev, "connect: set_encr_mode failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_encr_mode failed, %d\n", + ret); goto err_turn_radio_on; } if (channel) { ret = set_channel(usbdev, chan); if (ret < 0) { - devdbg(usbdev, "connect: set_channel failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_channel failed, %d\n", + ret); goto err_turn_radio_on; } } @@ -1848,8 +1867,8 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, priv->encr_tx_key_index = sme->key_idx; ret = add_wep_key(usbdev, sme->key, sme->key_len, sme->key_idx); if (ret < 0) { - devdbg(usbdev, "connect: add_wep_key failed, %d " - "(%d, %d)", ret, sme->key_len, sme->key_idx); + netdev_dbg(usbdev->net, "connect: add_wep_key failed, %d (%d, %d)\n", + ret, sme->key_len, sme->key_idx); goto err_turn_radio_on; } } @@ -1858,7 +1877,8 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, !is_broadcast_ether_addr(sme->bssid)) { ret = set_bssid(usbdev, sme->bssid); if (ret < 0) { - devdbg(usbdev, "connect: set_bssid failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_bssid failed, %d\n", + ret); goto err_turn_radio_on; } } else @@ -1880,7 +1900,7 @@ static int rndis_connect(struct wiphy *wiphy, struct net_device *dev, ret = set_essid(usbdev, &ssid); if (ret < 0) - devdbg(usbdev, "connect: set_essid failed, %d", ret); + netdev_dbg(usbdev->net, "connect: set_essid failed, %d\n", ret); return ret; err_turn_radio_on: @@ -1895,7 +1915,7 @@ static int rndis_disconnect(struct wiphy *wiphy, struct net_device *dev, struct rndis_wlan_private *priv = wiphy_priv(wiphy); struct usbnet *usbdev = priv->usbdev; - devdbg(usbdev, "cfg80211.disconnect(%d)", reason_code); + netdev_dbg(usbdev->net, "cfg80211.disconnect(%d)\n", reason_code); priv->connected = false; memset(priv->bssid, 0, ETH_ALEN); @@ -1929,21 +1949,23 @@ static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev, alg = RNDIS_WLAN_ALG_NONE; } - devdbg(usbdev, "cfg80211.join_ibss('%.32s':[%pM]:%d:%d)", params->ssid, - params->bssid, chan, params->privacy); + netdev_dbg(usbdev->net, "cfg80211.join_ibss('%.32s':[%pM]:%d:%d)\n", + params->ssid, params->bssid, chan, params->privacy); if (is_associated(usbdev)) disassociate(usbdev, false); ret = set_infra_mode(usbdev, NDIS_80211_INFRA_ADHOC); if (ret < 0) { - devdbg(usbdev, "join_ibss: set_infra_mode failed, %d", ret); + netdev_dbg(usbdev->net, "join_ibss: set_infra_mode failed, %d\n", + ret); goto err_turn_radio_on; } ret = set_auth_mode(usbdev, 0, auth_type, RNDIS_WLAN_KEY_MGMT_NONE); if (ret < 0) { - devdbg(usbdev, "join_ibss: set_auth_mode failed, %d", ret); + netdev_dbg(usbdev->net, "join_ibss: set_auth_mode failed, %d\n", + ret); goto err_turn_radio_on; } @@ -1951,15 +1973,16 @@ static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev, ret = set_encr_mode(usbdev, alg, RNDIS_WLAN_ALG_NONE); if (ret < 0) { - devdbg(usbdev, "join_ibss: set_encr_mode failed, %d", ret); + netdev_dbg(usbdev->net, "join_ibss: set_encr_mode failed, %d\n", + ret); goto err_turn_radio_on; } if (channel) { ret = set_channel(usbdev, chan); if (ret < 0) { - devdbg(usbdev, "join_ibss: set_channel failed, %d", - ret); + netdev_dbg(usbdev->net, "join_ibss: set_channel failed, %d\n", + ret); goto err_turn_radio_on; } } @@ -1968,7 +1991,8 @@ static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev, !is_broadcast_ether_addr(params->bssid)) { ret = set_bssid(usbdev, params->bssid); if (ret < 0) { - devdbg(usbdev, "join_ibss: set_bssid failed, %d", ret); + netdev_dbg(usbdev->net, "join_ibss: set_bssid failed, %d\n", + ret); goto err_turn_radio_on; } } else @@ -1988,7 +2012,8 @@ static int rndis_join_ibss(struct wiphy *wiphy, struct net_device *dev, ret = set_essid(usbdev, &ssid); if (ret < 0) - devdbg(usbdev, "join_ibss: set_essid failed, %d", ret); + netdev_dbg(usbdev->net, "join_ibss: set_essid failed, %d\n", + ret); return ret; err_turn_radio_on: @@ -2002,7 +2027,7 @@ static int rndis_leave_ibss(struct wiphy *wiphy, struct net_device *dev) struct rndis_wlan_private *priv = wiphy_priv(wiphy); struct usbnet *usbdev = priv->usbdev; - devdbg(usbdev, "cfg80211.leave_ibss()"); + netdev_dbg(usbdev->net, "cfg80211.leave_ibss()\n"); priv->connected = false; memset(priv->bssid, 0, ETH_ALEN); @@ -2028,8 +2053,8 @@ static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev, struct usbnet *usbdev = priv->usbdev; __le32 flags; - devdbg(usbdev, "rndis_add_key(%i, %pM, %08x)", key_index, mac_addr, - params->cipher); + netdev_dbg(usbdev->net, "%s(%i, %pM, %08x)\n", + __func__, key_index, mac_addr, params->cipher); switch (params->cipher) { case WLAN_CIPHER_SUITE_WEP40: @@ -2050,8 +2075,8 @@ static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev, key_index, mac_addr, params->seq, params->seq_len, params->cipher, flags); default: - devdbg(usbdev, "rndis_add_key: unsupported cipher %08x", - params->cipher); + netdev_dbg(usbdev->net, "%s(): unsupported cipher %08x\n", + __func__, params->cipher); return -ENOTSUPP; } } @@ -2062,7 +2087,7 @@ static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev, struct rndis_wlan_private *priv = wiphy_priv(wiphy); struct usbnet *usbdev = priv->usbdev; - devdbg(usbdev, "rndis_del_key(%i, %pM)", key_index, mac_addr); + netdev_dbg(usbdev->net, "%s(%i, %pM)\n", __func__, key_index, mac_addr); return remove_key(usbdev, key_index, mac_addr); } @@ -2074,7 +2099,7 @@ static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev, struct usbnet *usbdev = priv->usbdev; struct rndis_wlan_encr_key key; - devdbg(usbdev, "rndis_set_default_key(%i)", key_index); + netdev_dbg(usbdev->net, "%s(%i)\n", __func__, key_index); priv->encr_tx_key_index = key_index; @@ -2188,7 +2213,8 @@ static void rndis_wlan_do_link_up_work(struct usbnet *usbdev) if (ret < 0) memset(bssid, 0, sizeof(bssid)); - devdbg(usbdev, "link up work: [%pM] %s", bssid, roamed ? "roamed" : ""); + netdev_dbg(usbdev->net, "link up work: [%pM]%s\n", + bssid, roamed ? " roamed" : ""); /* Internal bss list in device always contains at least the currently * connected bss and we can get it to cfg80211 with @@ -2270,8 +2296,8 @@ static void rndis_wlan_auth_indication(struct usbnet *usbdev, /* must have at least one array entry */ if (len < offsetof(struct ndis_80211_status_indication, u) + sizeof(struct ndis_80211_auth_request)) { - devinfo(usbdev, "authentication indication: " - "too short message (%i)", len); + netdev_info(usbdev->net, "authentication indication: too short message (%i)\n", + len); return; } @@ -2298,8 +2324,8 @@ static void rndis_wlan_auth_indication(struct usbnet *usbdev, type = "group_error"; } - devinfo(usbdev, "authentication indication: %s (0x%08x)", type, - le32_to_cpu(auth_req->flags)); + netdev_info(usbdev->net, "authentication indication: %s (0x%08x)\n", + type, le32_to_cpu(auth_req->flags)); if (pairwise_error) { key_type = NL80211_KEYTYPE_PAIRWISE; @@ -2335,8 +2361,8 @@ static void rndis_wlan_pmkid_cand_list_indication(struct usbnet *usbdev, if (len < offsetof(struct ndis_80211_status_indication, u) + sizeof(struct ndis_80211_pmkid_cand_list)) { - devinfo(usbdev, "pmkid candidate list indication: " - "too short message (%i)", len); + netdev_info(usbdev->net, "pmkid candidate list indication: too short message (%i)\n", + len); return; } @@ -2346,18 +2372,16 @@ static void rndis_wlan_pmkid_cand_list_indication(struct usbnet *usbdev, offsetof(struct ndis_80211_status_indication, u); if (len < expected_len) { - devinfo(usbdev, "pmkid candidate list indication: " - "list larger than buffer (%i < %i)", - len, expected_len); + netdev_info(usbdev->net, "pmkid candidate list indication: list larger than buffer (%i < %i)\n", + len, expected_len); return; } cand_list = &indication->u.cand_list; - devinfo(usbdev, "pmkid candidate list indication: " - "version %i, candidates %i", - le32_to_cpu(cand_list->version), - le32_to_cpu(cand_list->num_candidates)); + netdev_info(usbdev->net, "pmkid candidate list indication: version %i, candidates %i\n", + le32_to_cpu(cand_list->version), + le32_to_cpu(cand_list->num_candidates)); if (le32_to_cpu(cand_list->version) != 1) return; @@ -2366,8 +2390,8 @@ static void rndis_wlan_pmkid_cand_list_indication(struct usbnet *usbdev, struct ndis_80211_pmkid_candidate *cand = &cand_list->candidate_list[i]; - devdbg(usbdev, "cand[%i]: flags: 0x%08x, bssid: %pM", - i, le32_to_cpu(cand->flags), cand->bssid); + netdev_dbg(usbdev->net, "cand[%i]: flags: 0x%08x, bssid: %pM\n", + i, le32_to_cpu(cand->flags), cand->bssid); #if 0 struct iw_pmkid_cand pcand; @@ -2398,15 +2422,14 @@ static void rndis_wlan_media_specific_indication(struct usbnet *usbdev, len = le32_to_cpu(msg->length); if (len < 8) { - devinfo(usbdev, "media specific indication, " - "ignore too short message (%i < 8)", len); + netdev_info(usbdev->net, "media specific indication, ignore too short message (%i < 8)\n", + len); return; } if (offset + len > buflen) { - devinfo(usbdev, "media specific indication, " - "too large to fit to buffer (%i > %i)", - offset + len, buflen); + netdev_info(usbdev->net, "media specific indication, too large to fit to buffer (%i > %i)\n", + offset + len, buflen); return; } @@ -2414,13 +2437,13 @@ static void rndis_wlan_media_specific_indication(struct usbnet *usbdev, switch (le32_to_cpu(indication->status_type)) { case NDIS_80211_STATUSTYPE_RADIOSTATE: - devinfo(usbdev, "radio state indication: %i", - le32_to_cpu(indication->u.radio_status)); + netdev_info(usbdev->net, "radio state indication: %i\n", + le32_to_cpu(indication->u.radio_status)); return; case NDIS_80211_STATUSTYPE_MEDIASTREAMMODE: - devinfo(usbdev, "media stream mode indication: %i", - le32_to_cpu(indication->u.media_stream_mode)); + netdev_info(usbdev->net, "media stream mode indication: %i\n", + le32_to_cpu(indication->u.media_stream_mode)); return; case NDIS_80211_STATUSTYPE_AUTHENTICATION: @@ -2432,9 +2455,8 @@ static void rndis_wlan_media_specific_indication(struct usbnet *usbdev, return; default: - devinfo(usbdev, "media specific indication: " - "unknown status type 0x%08x", - le32_to_cpu(indication->status_type)); + netdev_info(usbdev->net, "media specific indication: unknown status type 0x%08x\n", + le32_to_cpu(indication->status_type)); } } @@ -2451,14 +2473,13 @@ static void rndis_wlan_indication(struct usbnet *usbdev, void *ind, int buflen) * and userspace to think that device is * roaming/reassociating when it isn't. */ - devdbg(usbdev, "ignored OID_802_11_ADD_KEY triggered " - "'media connect'"); + netdev_dbg(usbdev->net, "ignored OID_802_11_ADD_KEY triggered 'media connect'\n"); return; } usbnet_pause_rx(usbdev); - devinfo(usbdev, "media connect"); + netdev_info(usbdev->net, "media connect\n"); /* queue work to avoid recursive calls into rndis_command */ set_bit(WORK_LINK_UP, &priv->work_pending); @@ -2466,7 +2487,7 @@ static void rndis_wlan_indication(struct usbnet *usbdev, void *ind, int buflen) break; case RNDIS_STATUS_MEDIA_DISCONNECT: - devinfo(usbdev, "media disconnect"); + netdev_info(usbdev->net, "media disconnect\n"); /* queue work to avoid recursive calls into rndis_command */ set_bit(WORK_LINK_DOWN, &priv->work_pending); @@ -2478,8 +2499,8 @@ static void rndis_wlan_indication(struct usbnet *usbdev, void *ind, int buflen) break; default: - devinfo(usbdev, "indication: 0x%08x", - le32_to_cpu(msg->status)); + netdev_info(usbdev->net, "indication: 0x%08x\n", + le32_to_cpu(msg->status)); break; } } @@ -2544,8 +2565,8 @@ static void rndis_device_poller(struct work_struct *work) if (ret == 0) priv->last_qual = level_to_qual(le32_to_cpu(rssi)); - devdbg(usbdev, "dev-poller: OID_802_11_RSSI -> %d, rssi:%d, qual: %d", - ret, le32_to_cpu(rssi), level_to_qual(le32_to_cpu(rssi))); + netdev_dbg(usbdev->net, "dev-poller: OID_802_11_RSSI -> %d, rssi:%d, qual: %d\n", + ret, le32_to_cpu(rssi), level_to_qual(le32_to_cpu(rssi))); /* Workaround transfer stalls on poor quality links. * TODO: find right way to fix these stalls (as stalls do not happen @@ -2838,11 +2859,11 @@ static int rndis_wlan_reset(struct usbnet *usbdev) struct rndis_wlan_private *priv = get_rndis_wlan_priv(usbdev); int retval; - devdbg(usbdev, "rndis_wlan_reset"); + netdev_dbg(usbdev->net, "%s()\n", __func__); retval = rndis_reset(usbdev); if (retval) - devwarn(usbdev, "rndis_reset() failed: %d", retval); + netdev_warn(usbdev->net, "rndis_reset failed: %d\n", retval); /* rndis_reset cleared multicast list, so restore here. (set_multicast_list() also turns on current packet filter) */ @@ -2860,7 +2881,7 @@ static int rndis_wlan_stop(struct usbnet *usbdev) int retval; __le32 filter; - devdbg(usbdev, "rndis_wlan_stop"); + netdev_dbg(usbdev->net, "%s()\n", __func__); retval = disassociate(usbdev, false); diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h index 8ce61359bf7..df1e83dd9a5 100644 --- a/include/linux/usb/usbnet.h +++ b/include/linux/usb/usbnet.h @@ -214,25 +214,4 @@ extern void usbnet_set_msglevel (struct net_device *, u32); extern void usbnet_get_drvinfo (struct net_device *, struct ethtool_drvinfo *); extern int usbnet_nway_reset(struct net_device *net); -/* messaging support includes the interface name, so it must not be - * used before it has one ... notably, in minidriver bind() calls. - */ -#ifdef DEBUG -#define devdbg(usbnet, fmt, arg...) \ - printk(KERN_DEBUG "%s: " fmt "\n" , (usbnet)->net->name , ## arg) -#else -#define devdbg(usbnet, fmt, arg...) \ - ({ if (0) printk(KERN_DEBUG "%s: " fmt "\n" , (usbnet)->net->name , \ - ## arg); 0; }) -#endif - -#define deverr(usbnet, fmt, arg...) \ - printk(KERN_ERR "%s: " fmt "\n" , (usbnet)->net->name , ## arg) -#define devwarn(usbnet, fmt, arg...) \ - printk(KERN_WARNING "%s: " fmt "\n" , (usbnet)->net->name , ## arg) - -#define devinfo(usbnet, fmt, arg...) \ - printk(KERN_INFO "%s: " fmt "\n" , (usbnet)->net->name , ## arg); \ - - #endif /* __LINUX_USB_USBNET_H */ -- cgit v1.2.3-70-g09d2 From 99405162598176e830d17ae6d4f3d9e070ad900c Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Sun, 14 Feb 2010 01:01:10 +0000 Subject: tun: socket filter support This patch adds Linux Socket Filter support to tun driver. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller --- drivers/net/tun.c | 26 ++++++++++++++++++++++++++ include/linux/if_tun.h | 3 +++ 2 files changed, 29 insertions(+) (limited to 'include') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 5adb3d15055..ce1efa4c0b0 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -61,6 +61,7 @@ #include #include #include +#include #include #include #include @@ -366,6 +367,10 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev) if (!check_filter(&tun->txflt, skb)) goto drop; + if (tun->socket.sk->sk_filter && + sk_filter(tun->socket.sk, skb)) + goto drop; + if (skb_queue_len(&tun->socket.sk->sk_receive_queue) >= dev->tx_queue_len) { if (!(tun->flags & TUN_ONE_QUEUE)) { /* Normal queueing mode. */ @@ -1162,6 +1167,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd, struct tun_file *tfile = file->private_data; struct tun_struct *tun; void __user* argp = (void __user*)arg; + struct sock_fprog fprog; struct ifreq ifr; int sndbuf; int ret; @@ -1309,6 +1315,26 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd, tun->socket.sk->sk_sndbuf = sndbuf; break; + case TUNATTACHFILTER: + /* Can be set only for TAPs */ + ret = -EINVAL; + if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV) + break; + ret = -EFAULT; + if (copy_from_user(&fprog, argp, sizeof(fprog))) + break; + + ret = sk_attach_filter(&fprog, tun->socket.sk); + break; + + case TUNDETACHFILTER: + /* Can be set only for TAPs */ + ret = -EINVAL; + if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV) + break; + ret = sk_detach_filter(tun->socket.sk); + break; + default: ret = -EINVAL; break; diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h index 404abe00162..1350a246893 100644 --- a/include/linux/if_tun.h +++ b/include/linux/if_tun.h @@ -18,6 +18,7 @@ #include #include +#include /* Read queue size */ #define TUN_READQ_SIZE 500 @@ -48,6 +49,8 @@ #define TUNGETIFF _IOR('T', 210, unsigned int) #define TUNGETSNDBUF _IOR('T', 211, int) #define TUNSETSNDBUF _IOW('T', 212, int) +#define TUNATTACHFILTER _IOW('T', 213, struct sock_fprog) +#define TUNDETACHFILTER _IOW('T', 214, struct sock_fprog) /* TUNSETIFF ifr flags */ #define IFF_TUN 0x0001 -- cgit v1.2.3-70-g09d2 From 6a443a0f72ad7706345412dbd2e4d4981fdfce39 Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Wed, 17 Feb 2010 15:17:04 +0000 Subject: tg3: Push phylib definitions to phylib This patch pushes phylib definitions out to phylib headers. For phy IDs, this removes some code duplication. Signed-off-by: Matt Carlson Reviewed-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/phy/broadcom.c | 5 +---- drivers/net/tg3.c | 32 ++++++++++++++++---------------- drivers/net/tg3.h | 14 +++----------- include/linux/brcmphy.h | 11 +++++++++++ 4 files changed, 31 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index 33c4b12a63b..f482fc4f8cf 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -18,9 +18,6 @@ #include #include -#define PHY_ID_BCM50610 0x0143bd60 -#define PHY_ID_BCM50610M 0x0143bd70 -#define PHY_ID_BCM57780 0x03625d90 #define BRCM_PHY_MODEL(phydev) \ ((phydev)->drv->phy_id & (phydev)->drv->phy_id_mask) @@ -823,7 +820,7 @@ static struct phy_driver bcm57780_driver = { }; static struct phy_driver bcmac131_driver = { - .phy_id = 0x0143bc70, + .phy_id = PHY_ID_BCMAC131, .phy_id_mask = 0xfffffff0, .name = "Broadcom BCMAC131", .features = PHY_BASIC_FEATURES | diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 965efa92b02..994bac0b018 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -955,17 +955,17 @@ static void tg3_mdio_config_5785(struct tg3 *tp) phydev = tp->mdio_bus->phy_map[TG3_PHY_MII_ADDR]; switch (phydev->drv->phy_id & phydev->drv->phy_id_mask) { - case TG3_PHY_ID_BCM50610: - case TG3_PHY_ID_BCM50610M: + case PHY_ID_BCM50610: + case PHY_ID_BCM50610M: val = MAC_PHYCFG2_50610_LED_MODES; break; - case TG3_PHY_ID_BCMAC131: + case PHY_ID_BCMAC131: val = MAC_PHYCFG2_AC131_LED_MODES; break; - case TG3_PHY_ID_RTL8211C: + case PHY_ID_RTL8211C: val = MAC_PHYCFG2_RTL8211C_LED_MODES; break; - case TG3_PHY_ID_RTL8201E: + case PHY_ID_RTL8201E: val = MAC_PHYCFG2_RTL8201E_LED_MODES; break; default: @@ -1115,12 +1115,12 @@ static int tg3_mdio_init(struct tg3 *tp) } switch (phydev->drv->phy_id & phydev->drv->phy_id_mask) { - case TG3_PHY_ID_BCM57780: + case PHY_ID_BCM57780: phydev->interface = PHY_INTERFACE_MODE_GMII; phydev->dev_flags |= PHY_BRCM_AUTO_PWRDWN_ENABLE; break; - case TG3_PHY_ID_BCM50610: - case TG3_PHY_ID_BCM50610M: + case PHY_ID_BCM50610: + case PHY_ID_BCM50610M: phydev->dev_flags |= PHY_BRCM_CLEAR_RGMII_MODE | PHY_BRCM_RX_REFCLK_UNUSED | PHY_BRCM_DIS_TXCRXC_NOENRGY | @@ -1132,11 +1132,11 @@ static int tg3_mdio_init(struct tg3 *tp) if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN) phydev->dev_flags |= PHY_BRCM_EXT_IBND_TX_ENABLE; /* fallthru */ - case TG3_PHY_ID_RTL8211C: + case PHY_ID_RTL8211C: phydev->interface = PHY_INTERFACE_MODE_RGMII; break; - case TG3_PHY_ID_RTL8201E: - case TG3_PHY_ID_BCMAC131: + case PHY_ID_RTL8201E: + case PHY_ID_BCMAC131: phydev->interface = PHY_INTERFACE_MODE_MII; phydev->dev_flags |= PHY_BRCM_AUTO_PWRDWN_ENABLE; tp->tg3_flags3 |= TG3_FLG3_PHY_IS_FET; @@ -2563,11 +2563,11 @@ static int tg3_set_power_state(struct tg3 *tp, pci_power_t state) phy_start_aneg(phydev); phyid = phydev->drv->phy_id & phydev->drv->phy_id_mask; - if (phyid != TG3_PHY_ID_BCMAC131) { - phyid &= TG3_PHY_OUI_MASK; - if (phyid == TG3_PHY_OUI_1 || - phyid == TG3_PHY_OUI_2 || - phyid == TG3_PHY_OUI_3) + if (phyid != PHY_ID_BCMAC131) { + phyid &= PHY_BCM_OUI_MASK; + if (phyid == PHY_BCM_OUI_1 || + phyid == PHY_BCM_OUI_2 || + phyid == PHY_BCM_OUI_3) do_low_power = true; } } diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h index 800dec463e9..574a1cc4d35 100644 --- a/drivers/net/tg3.h +++ b/drivers/net/tg3.h @@ -2943,22 +2943,14 @@ struct tg3 { #define TG3_PHY_ID_BCM57765 0x5c0d8a40 #define TG3_PHY_ID_BCM5906 0xdc00ac40 #define TG3_PHY_ID_BCM8002 0x60010140 -#define TG3_PHY_ID_BCM50610 0x0143bd60 -#define TG3_PHY_ID_BCM50610M 0x0143bd70 -#define TG3_PHY_ID_BCMAC131 0x0143bc70 -#define TG3_PHY_ID_RTL8211C 0x001cc910 -#define TG3_PHY_ID_RTL8201E 0x00008200 -#define TG3_PHY_ID_BCM57780 0x03625d90 #define TG3_PHY_ID_INVALID 0xffffffff +#define PHY_ID_RTL8211C 0x001cc910 +#define PHY_ID_RTL8201E 0x00008200 + #define TG3_PHY_ID_REV_MASK 0x0000000f #define TG3_PHY_REV_BCM5401_B0 0x1 -#define TG3_PHY_OUI_MASK 0xfffffc00 -#define TG3_PHY_OUI_1 0x00206000 -#define TG3_PHY_OUI_2 0x0143bc00 -#define TG3_PHY_OUI_3 0x03625c00 - /* This macro assumes the passed PHY ID is * already masked with TG3_PHY_ID_MASK. */ diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h index 2b31b91f587..7f437ca1ed4 100644 --- a/include/linux/brcmphy.h +++ b/include/linux/brcmphy.h @@ -1,3 +1,14 @@ +#define PHY_ID_BCM50610 0x0143bd60 +#define PHY_ID_BCM50610M 0x0143bd70 +#define PHY_ID_BCMAC131 0x0143bc70 +#define PHY_ID_BCM57780 0x03625d90 + +#define PHY_BCM_OUI_MASK 0xfffffc00 +#define PHY_BCM_OUI_1 0x00206000 +#define PHY_BCM_OUI_2 0x0143bc00 +#define PHY_BCM_OUI_3 0x03625c00 + + #define PHY_BCM_FLAGS_MODE_COPPER 0x00000001 #define PHY_BCM_FLAGS_MODE_1000BX 0x00000002 #define PHY_BCM_FLAGS_INTF_SGMII 0x00000010 -- cgit v1.2.3-70-g09d2 From 6457d26bd40077238799e31df2b800bcf4ef9177 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 17 Feb 2010 18:48:44 -0800 Subject: IPv6: convert mc_lock to spinlock Only used for writing, so convert to spinlock Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/net/if_inet6.h | 2 +- net/ipv6/mcast.c | 19 ++++++++++--------- 2 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index e9d69d19849..545d8b059be 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -157,7 +157,7 @@ struct inet6_dev { struct ifmcaddr6 *mc_list; struct ifmcaddr6 *mc_tomb; - rwlock_t mc_lock; + spinlock_t mc_lock; unsigned char mc_qrv; unsigned char mc_gq_running; unsigned char mc_ifc_count; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 25f6cca79e6..bcd97191596 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -793,10 +793,10 @@ static void mld_add_delrec(struct inet6_dev *idev, struct ifmcaddr6 *im) } spin_unlock_bh(&im->mca_lock); - write_lock_bh(&idev->mc_lock); + spin_lock_bh(&idev->mc_lock); pmc->next = idev->mc_tomb; idev->mc_tomb = pmc; - write_unlock_bh(&idev->mc_lock); + spin_unlock_bh(&idev->mc_lock); } static void mld_del_delrec(struct inet6_dev *idev, struct in6_addr *pmca) @@ -804,7 +804,7 @@ static void mld_del_delrec(struct inet6_dev *idev, struct in6_addr *pmca) struct ifmcaddr6 *pmc, *pmc_prev; struct ip6_sf_list *psf, *psf_next; - write_lock_bh(&idev->mc_lock); + spin_lock_bh(&idev->mc_lock); pmc_prev = NULL; for (pmc=idev->mc_tomb; pmc; pmc=pmc->next) { if (ipv6_addr_equal(&pmc->mca_addr, pmca)) @@ -817,7 +817,8 @@ static void mld_del_delrec(struct inet6_dev *idev, struct in6_addr *pmca) else idev->mc_tomb = pmc->next; } - write_unlock_bh(&idev->mc_lock); + spin_unlock_bh(&idev->mc_lock); + if (pmc) { for (psf=pmc->mca_tomb; psf; psf=psf_next) { psf_next = psf->sf_next; @@ -832,10 +833,10 @@ static void mld_clear_delrec(struct inet6_dev *idev) { struct ifmcaddr6 *pmc, *nextpmc; - write_lock_bh(&idev->mc_lock); + spin_lock_bh(&idev->mc_lock); pmc = idev->mc_tomb; idev->mc_tomb = NULL; - write_unlock_bh(&idev->mc_lock); + spin_unlock_bh(&idev->mc_lock); for (; pmc; pmc = nextpmc) { nextpmc = pmc->next; @@ -1696,7 +1697,7 @@ static void mld_send_cr(struct inet6_dev *idev) int type, dtype; read_lock_bh(&idev->lock); - write_lock_bh(&idev->mc_lock); + spin_lock(&idev->mc_lock); /* deleted MCA's */ pmc_prev = NULL; @@ -1730,7 +1731,7 @@ static void mld_send_cr(struct inet6_dev *idev) } else pmc_prev = pmc; } - write_unlock_bh(&idev->mc_lock); + spin_unlock(&idev->mc_lock); /* change recs */ for (pmc=idev->mc_list; pmc; pmc=pmc->next) { @@ -2311,7 +2312,7 @@ void ipv6_mc_up(struct inet6_dev *idev) void ipv6_mc_init_dev(struct inet6_dev *idev) { write_lock_bh(&idev->lock); - rwlock_init(&idev->mc_lock); + spin_lock_init(&idev->mc_lock); idev->mc_gq_running = 0; setup_timer(&idev->mc_gq_timer, mld_gq_timer_expire, (unsigned long)idev); -- cgit v1.2.3-70-g09d2 From 2906f66a5682e5670a5eefe991843689b8d8563f Mon Sep 17 00:00:00 2001 From: Venkata Mohan Reddy Date: Thu, 18 Feb 2010 12:31:05 +0100 Subject: ipvs: SCTP Trasport Loadbalancing Support Enhance IPVS to load balance SCTP transport protocol packets. This is done based on the SCTP rfc 4960. All possible control chunks have been taken care. The state machine used in this code looks some what lengthy. I tried to make the state machine easy to understand. Signed-off-by: Venkata Mohan Reddy Koppula Signed-off-by: Simon Horman Signed-off-by: Patrick McHardy --- include/net/ip_vs.h | 22 +- net/netfilter/ipvs/Kconfig | 7 + net/netfilter/ipvs/Makefile | 1 + net/netfilter/ipvs/ip_vs_core.c | 62 +- net/netfilter/ipvs/ip_vs_ctl.c | 5 +- net/netfilter/ipvs/ip_vs_proto.c | 3 + net/netfilter/ipvs/ip_vs_proto_sctp.c | 1183 +++++++++++++++++++++++++++++++++ net/netfilter/ipvs/ip_vs_sync.c | 14 + 8 files changed, 1285 insertions(+), 12 deletions(-) create mode 100644 net/netfilter/ipvs/ip_vs_proto_sctp.c (limited to 'include') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index a816c37417b..fe82b1e10a2 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -224,6 +224,26 @@ enum { IP_VS_ICMP_S_LAST, }; +/* + * SCTP State Values + */ +enum ip_vs_sctp_states { + IP_VS_SCTP_S_NONE, + IP_VS_SCTP_S_INIT_CLI, + IP_VS_SCTP_S_INIT_SER, + IP_VS_SCTP_S_INIT_ACK_CLI, + IP_VS_SCTP_S_INIT_ACK_SER, + IP_VS_SCTP_S_ECHO_CLI, + IP_VS_SCTP_S_ECHO_SER, + IP_VS_SCTP_S_ESTABLISHED, + IP_VS_SCTP_S_SHUT_CLI, + IP_VS_SCTP_S_SHUT_SER, + IP_VS_SCTP_S_SHUT_ACK_CLI, + IP_VS_SCTP_S_SHUT_ACK_SER, + IP_VS_SCTP_S_CLOSED, + IP_VS_SCTP_S_LAST +}; + /* * Delta sequence info structure * Each ip_vs_conn has 2 (output AND input seq. changes). @@ -741,7 +761,7 @@ extern struct ip_vs_protocol ip_vs_protocol_udp; extern struct ip_vs_protocol ip_vs_protocol_icmp; extern struct ip_vs_protocol ip_vs_protocol_esp; extern struct ip_vs_protocol ip_vs_protocol_ah; - +extern struct ip_vs_protocol ip_vs_protocol_sctp; /* * Registering/unregistering scheduler functions diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig index 817a8898203..712ccad1334 100644 --- a/net/netfilter/ipvs/Kconfig +++ b/net/netfilter/ipvs/Kconfig @@ -104,6 +104,13 @@ config IP_VS_PROTO_AH This option enables support for load balancing AH (Authentication Header) transport protocol. Say Y if unsure. +config IP_VS_PROTO_SCTP + bool "SCTP load balancing support" + select LIBCRC32C + ---help--- + This option enables support for load balancing SCTP transport + protocol. Say Y if unsure. + comment "IPVS scheduler" config IP_VS_RR diff --git a/net/netfilter/ipvs/Makefile b/net/netfilter/ipvs/Makefile index 73a46fe1fe4..e3baefd7066 100644 --- a/net/netfilter/ipvs/Makefile +++ b/net/netfilter/ipvs/Makefile @@ -7,6 +7,7 @@ ip_vs_proto-objs-y := ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_TCP) += ip_vs_proto_tcp.o ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_UDP) += ip_vs_proto_udp.o ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH_ESP) += ip_vs_proto_ah_esp.o +ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_SCTP) += ip_vs_proto_sctp.o ip_vs-objs := ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o \ ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o \ diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 847ffca4018..72e96d823eb 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include @@ -81,6 +82,8 @@ const char *ip_vs_proto_name(unsigned proto) return "UDP"; case IPPROTO_TCP: return "TCP"; + case IPPROTO_SCTP: + return "SCTP"; case IPPROTO_ICMP: return "ICMP"; #ifdef CONFIG_IP_VS_IPV6 @@ -589,8 +592,9 @@ void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, ip_send_check(ciph); } - /* the TCP/UDP port */ - if (IPPROTO_TCP == ciph->protocol || IPPROTO_UDP == ciph->protocol) { + /* the TCP/UDP/SCTP port */ + if (IPPROTO_TCP == ciph->protocol || IPPROTO_UDP == ciph->protocol || + IPPROTO_SCTP == ciph->protocol) { __be16 *ports = (void *)ciph + ciph->ihl*4; if (inout) @@ -630,8 +634,9 @@ void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp, ciph->saddr = cp->daddr.in6; } - /* the TCP/UDP port */ - if (IPPROTO_TCP == ciph->nexthdr || IPPROTO_UDP == ciph->nexthdr) { + /* the TCP/UDP/SCTP port */ + if (IPPROTO_TCP == ciph->nexthdr || IPPROTO_UDP == ciph->nexthdr || + IPPROTO_SCTP == ciph->nexthdr) { __be16 *ports = (void *)ciph + sizeof(struct ipv6hdr); if (inout) @@ -679,7 +684,8 @@ static int handle_response_icmp(int af, struct sk_buff *skb, goto out; } - if (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol) + if (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol || + IPPROTO_SCTP == protocol) offset += 2 * sizeof(__u16); if (!skb_make_writable(skb, offset)) goto out; @@ -857,6 +863,21 @@ static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related) } #endif +/* + * Check if sctp chunc is ABORT chunk + */ +static inline int is_sctp_abort(const struct sk_buff *skb, int nh_len) +{ + sctp_chunkhdr_t *sch, schunk; + sch = skb_header_pointer(skb, nh_len + sizeof(sctp_sctphdr_t), + sizeof(schunk), &schunk); + if (sch == NULL) + return 0; + if (sch->type == SCTP_CID_ABORT) + return 1; + return 0; +} + static inline int is_tcp_reset(const struct sk_buff *skb, int nh_len) { struct tcphdr _tcph, *th; @@ -999,7 +1020,8 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, if (unlikely(!cp)) { if (sysctl_ip_vs_nat_icmp_send && (pp->protocol == IPPROTO_TCP || - pp->protocol == IPPROTO_UDP)) { + pp->protocol == IPPROTO_UDP || + pp->protocol == IPPROTO_SCTP)) { __be16 _ports[2], *pptr; pptr = skb_header_pointer(skb, iph.len, @@ -1014,8 +1036,13 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, * existing entry if it is not RST * packet or not TCP packet. */ - if (iph.protocol != IPPROTO_TCP - || !is_tcp_reset(skb, iph.len)) { + if ((iph.protocol != IPPROTO_TCP && + iph.protocol != IPPROTO_SCTP) + || ((iph.protocol == IPPROTO_TCP + && !is_tcp_reset(skb, iph.len)) + || (iph.protocol == IPPROTO_SCTP + && !is_sctp_abort(skb, + iph.len)))) { #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) icmpv6_send(skb, @@ -1235,7 +1262,8 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) /* do the statistics and put it back */ ip_vs_in_stats(cp, skb); - if (IPPROTO_TCP == cih->nexthdr || IPPROTO_UDP == cih->nexthdr) + if (IPPROTO_TCP == cih->nexthdr || IPPROTO_UDP == cih->nexthdr || + IPPROTO_SCTP == cih->nexthdr) offset += 2 * sizeof(__u16); verdict = ip_vs_icmp_xmit_v6(skb, cp, pp, offset); /* do not touch skb anymore */ @@ -1358,6 +1386,21 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, * encorage the standby servers to update the connections timeout */ pkts = atomic_add_return(1, &cp->in_pkts); + if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) && + cp->protocol == IPPROTO_SCTP) { + if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && + (atomic_read(&cp->in_pkts) % + sysctl_ip_vs_sync_threshold[1] + == sysctl_ip_vs_sync_threshold[0])) || + (cp->old_state != cp->state && + ((cp->state == IP_VS_SCTP_S_CLOSED) || + (cp->state == IP_VS_SCTP_S_SHUT_ACK_CLI) || + (cp->state == IP_VS_SCTP_S_SHUT_ACK_SER)))) { + ip_vs_sync_conn(cp); + goto out; + } + } + if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || @@ -1370,6 +1413,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, (cp->state == IP_VS_TCP_S_CLOSE_WAIT) || (cp->state == IP_VS_TCP_S_TIME_WAIT))))) ip_vs_sync_conn(cp); +out: cp->old_state = cp->state; ip_vs_conn_put(cp); diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 00d0b152db3..7ee9c3426f4 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2132,8 +2132,9 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) } } - /* Check for valid protocol: TCP or UDP, even for fwmark!=0 */ - if (usvc.protocol != IPPROTO_TCP && usvc.protocol != IPPROTO_UDP) { + /* Check for valid protocol: TCP or UDP or SCTP, even for fwmark!=0 */ + if (usvc.protocol != IPPROTO_TCP && usvc.protocol != IPPROTO_UDP && + usvc.protocol != IPPROTO_SCTP) { pr_err("set_ctl: invalid protocol: %d %pI4:%d %s\n", usvc.protocol, &usvc.addr.ip, ntohs(usvc.port), usvc.sched_name); diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 3e767167454..0e584553819 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -257,6 +257,9 @@ int __init ip_vs_protocol_init(void) #ifdef CONFIG_IP_VS_PROTO_UDP REGISTER_PROTOCOL(&ip_vs_protocol_udp); #endif +#ifdef CONFIG_IP_VS_PROTO_SCTP + REGISTER_PROTOCOL(&ip_vs_protocol_sctp); +#endif #ifdef CONFIG_IP_VS_PROTO_AH REGISTER_PROTOCOL(&ip_vs_protocol_ah); #endif diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c new file mode 100644 index 00000000000..c9a3f7a21d5 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -0,0 +1,1183 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +static struct ip_vs_conn * +sctp_conn_in_get(int af, + const struct sk_buff *skb, + struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, + unsigned int proto_off, + int inverse) +{ + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) + return ip_vs_conn_in_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + else + return ip_vs_conn_in_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); +} + +static struct ip_vs_conn * +sctp_conn_out_get(int af, + const struct sk_buff *skb, + struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, + unsigned int proto_off, + int inverse) +{ + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) + return ip_vs_conn_out_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + else + return ip_vs_conn_out_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); +} + +static int +sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, + int *verdict, struct ip_vs_conn **cpp) +{ + struct ip_vs_service *svc; + sctp_chunkhdr_t _schunkh, *sch; + sctp_sctphdr_t *sh, _sctph; + struct ip_vs_iphdr iph; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + + sh = skb_header_pointer(skb, iph.len, sizeof(_sctph), &_sctph); + if (sh == NULL) + return 0; + + sch = skb_header_pointer(skb, iph.len + sizeof(sctp_sctphdr_t), + sizeof(_schunkh), &_schunkh); + if (sch == NULL) + return 0; + + if ((sch->type == SCTP_CID_INIT) && + (svc = ip_vs_service_get(af, skb->mark, iph.protocol, + &iph.daddr, sh->dest))) { + if (ip_vs_todrop()) { + /* + * It seems that we are very loaded. + * We have to drop this packet :( + */ + ip_vs_service_put(svc); + *verdict = NF_DROP; + return 0; + } + /* + * Let the virtual server select a real server for the + * incoming connection, and create a connection entry. + */ + *cpp = ip_vs_schedule(svc, skb); + if (!*cpp) { + *verdict = ip_vs_leave(svc, skb, pp); + return 0; + } + ip_vs_service_put(svc); + } + + return 1; +} + +static int +sctp_snat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + sctp_sctphdr_t *sctph; + unsigned int sctphoff; + __be32 crc32; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + sctphoff = sizeof(struct ipv6hdr); + else +#endif + sctphoff = ip_hdrlen(skb); + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, sctphoff + sizeof(*sctph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* Call application helper if needed */ + if (!ip_vs_app_pkt_out(cp, skb)) + return 0; + } + + sctph = (void *) skb_network_header(skb) + sctphoff; + sctph->source = cp->vport; + + /* Calculate the checksum */ + crc32 = sctp_start_cksum((u8 *) sctph, skb_headlen(skb) - sctphoff); + for (skb = skb_shinfo(skb)->frag_list; skb; skb = skb->next) + crc32 = sctp_update_cksum((u8 *) skb->data, skb_headlen(skb), + crc32); + crc32 = sctp_end_cksum(crc32); + sctph->checksum = crc32; + + return 1; +} + +static int +sctp_dnat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + + sctp_sctphdr_t *sctph; + unsigned int sctphoff; + __be32 crc32; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + sctphoff = sizeof(struct ipv6hdr); + else +#endif + sctphoff = ip_hdrlen(skb); + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, sctphoff + sizeof(*sctph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* Call application helper if needed */ + if (!ip_vs_app_pkt_out(cp, skb)) + return 0; + } + + sctph = (void *) skb_network_header(skb) + sctphoff; + sctph->dest = cp->dport; + + /* Calculate the checksum */ + crc32 = sctp_start_cksum((u8 *) sctph, skb_headlen(skb) - sctphoff); + for (skb = skb_shinfo(skb)->frag_list; skb; skb = skb->next) + crc32 = sctp_update_cksum((u8 *) skb->data, skb_headlen(skb), + crc32); + crc32 = sctp_end_cksum(crc32); + sctph->checksum = crc32; + + return 1; +} + +static int +sctp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) +{ + struct sk_buff *list = skb_shinfo(skb)->frag_list; + unsigned int sctphoff; + struct sctphdr *sh, _sctph; + __le32 cmp; + __le32 val; + __u32 tmp; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + sctphoff = sizeof(struct ipv6hdr); + else +#endif + sctphoff = ip_hdrlen(skb); + + sh = skb_header_pointer(skb, sctphoff, sizeof(_sctph), &_sctph); + if (sh == NULL) + return 0; + + cmp = sh->checksum; + + tmp = sctp_start_cksum((__u8 *) sh, skb_headlen(skb)); + for (; list; list = list->next) + tmp = sctp_update_cksum((__u8 *) list->data, + skb_headlen(list), tmp); + + val = sctp_end_cksum(tmp); + + if (val != cmp) { + /* CRC failure, dump it. */ + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "Failed checksum for"); + return 0; + } + return 1; +} + +struct ipvs_sctp_nextstate { + int next_state; +}; +enum ipvs_sctp_event_t { + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_DATA_SER, + IP_VS_SCTP_EVE_INIT_CLI, + IP_VS_SCTP_EVE_INIT_SER, + IP_VS_SCTP_EVE_INIT_ACK_CLI, + IP_VS_SCTP_EVE_INIT_ACK_SER, + IP_VS_SCTP_EVE_COOKIE_ECHO_CLI, + IP_VS_SCTP_EVE_COOKIE_ECHO_SER, + IP_VS_SCTP_EVE_COOKIE_ACK_CLI, + IP_VS_SCTP_EVE_COOKIE_ACK_SER, + IP_VS_SCTP_EVE_ABORT_CLI, + IP_VS_SCTP_EVE__ABORT_SER, + IP_VS_SCTP_EVE_SHUT_CLI, + IP_VS_SCTP_EVE_SHUT_SER, + IP_VS_SCTP_EVE_SHUT_ACK_CLI, + IP_VS_SCTP_EVE_SHUT_ACK_SER, + IP_VS_SCTP_EVE_SHUT_COM_CLI, + IP_VS_SCTP_EVE_SHUT_COM_SER, + IP_VS_SCTP_EVE_LAST +}; + +static enum ipvs_sctp_event_t sctp_events[255] = { + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_INIT_CLI, + IP_VS_SCTP_EVE_INIT_ACK_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_ABORT_CLI, + IP_VS_SCTP_EVE_SHUT_CLI, + IP_VS_SCTP_EVE_SHUT_ACK_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_COOKIE_ECHO_CLI, + IP_VS_SCTP_EVE_COOKIE_ACK_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_DATA_CLI, + IP_VS_SCTP_EVE_SHUT_COM_CLI, +}; + +static struct ipvs_sctp_nextstate + sctp_states_table[IP_VS_SCTP_S_LAST][IP_VS_SCTP_EVE_LAST] = { + /* + * STATE : IP_VS_SCTP_S_NONE + */ + /*next state *//*event */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ }, + }, + /* + * STATE : IP_VS_SCTP_S_INIT_CLI + * Cient sent INIT and is waiting for reply from server(In ECHO_WAIT) + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_INIT_ACK_SER /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ECHO_CLI */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_ECHO_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_INIT_SER + * Server sent INIT and waiting for INIT ACK from the client + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + {IP_VS_SCTP_S_INIT_ACK_CLI /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_INIT_ACK_CLI + * Client sent INIT ACK and waiting for ECHO from the server + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK has been resent by the client, let us stay is in + * the same state + */ + {IP_VS_SCTP_S_INIT_ACK_CLI /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + /* + * INIT_ACK sent by the server, close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * ECHO by client, it should not happen, close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + /* + * ECHO by server, this is what we are expecting, move to ECHO_SER + */ + {IP_VS_SCTP_S_ECHO_SER /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, it should not happen, close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + /* + * Unexpected COOKIE ACK from server, staty in the same state + */ + {IP_VS_SCTP_S_INIT_ACK_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_INIT_ACK_SER + * Server sent INIT ACK and waiting for ECHO from the client + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * Unexpected INIT_ACK by the client, let us close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + /* + * INIT_ACK resent by the server, let us move to same state + */ + {IP_VS_SCTP_S_INIT_ACK_SER /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client send the ECHO, this is what we are expecting, + * move to ECHO_CLI + */ + {IP_VS_SCTP_S_ECHO_CLI /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + /* + * ECHO received from the server, Not sure what to do, + * let us close it + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, let us stay in the same state + */ + {IP_VS_SCTP_S_INIT_ACK_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + /* + * COOKIE ACK from server, hmm... this should not happen, lets close + * the connection. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_ECHO_CLI + * Cient sent ECHO and waiting COOKEI ACK from the Server + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK has been by the client, let us close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_ECHO_CLI /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client resent the ECHO, let us stay in the same state + */ + {IP_VS_SCTP_S_ECHO_CLI /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + /* + * ECHO received from the server, Not sure what to do, + * let us close it + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, this shoud not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + /* + * COOKIE ACK from server, this is what we are awaiting,lets move to + * ESTABLISHED. + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_ECHO_SER + * Server sent ECHO and waiting COOKEI ACK from the client + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_ECHO_SER /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + /* + * INIT_ACK has been by the server, let us close the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent the ECHO, not sure what to do, let's close the + * connection. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + /* + * ECHO resent by the server, stay in the same state + */ + {IP_VS_SCTP_S_ECHO_SER /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, this is what we are expecting, let's move + * to ESTABLISHED. + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + /* + * COOKIE ACK from server, this should not happen, lets close the + * connection. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_ESTABLISHED + * Association established + */ + {{IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent ECHO, Spec(sec 5.2.4) says it may be handled by the + * peer and peer shall move to the ESTABISHED. if it doesn't handle + * it will send ERROR chunk. So, stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, not sure what to do stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + /* + * SHUTDOWN from the client, move to SHUDDOWN_CLI + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + /* + * SHUTDOWN from the server, move to SHUTDOWN_SER + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_SHUT_SER */ }, + /* + * client sent SHUDTDOWN_ACK, this should not happen, let's close + * the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_SHUT_CLI + * SHUTDOWN sent from the client, waitinf for SHUT ACK from the server + */ + /* + * We recieved the data chuck, keep the state unchanged. I assume + * that still data chuncks can be received by both the peers in + * SHUDOWN state + */ + + {{IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent ECHO, Spec(sec 5.2.4) says it may be handled by the + * peer and peer shall move to the ESTABISHED. if it doesn't handle + * it will send ERROR chunk. So, stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, not sure what to do stay in the same state + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + /* + * SHUTDOWN resent from the client, move to SHUDDOWN_CLI + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + /* + * SHUTDOWN from the server, move to SHUTDOWN_SER + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_SHUT_SER */ }, + /* + * client sent SHUDTDOWN_ACK, this should not happen, let's close + * the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + /* + * Server sent SHUTDOWN ACK, this is what we are expecting, let's move + * to SHUDOWN_ACK_SER + */ + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + /* + * SHUTDOWN COM from client, this should not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_SHUT_SER + * SHUTDOWN sent from the server, waitinf for SHUTDOWN ACK from client + */ + /* + * We recieved the data chuck, keep the state unchanged. I assume + * that still data chuncks can be received by both the peers in + * SHUDOWN state + */ + + {{IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent ECHO, Spec(sec 5.2.4) says it may be handled by the + * peer and peer shall move to the ESTABISHED. if it doesn't handle + * it will send ERROR chunk. So, stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, not sure what to do stay in the same state + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + /* + * SHUTDOWN resent from the client, move to SHUDDOWN_CLI + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + /* + * SHUTDOWN resent from the server, move to SHUTDOWN_SER + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_SHUT_SER */ }, + /* + * client sent SHUDTDOWN_ACK, this is what we are expecting, let's + * move to SHUT_ACK_CLI + */ + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + /* + * Server sent SHUTDOWN ACK, this should not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + /* + * SHUTDOWN COM from client, this should not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + + /* + * State : IP_VS_SCTP_S_SHUT_ACK_CLI + * SHUTDOWN ACK from the client, awaiting for SHUTDOWN COM from server + */ + /* + * We recieved the data chuck, keep the state unchanged. I assume + * that still data chuncks can be received by both the peers in + * SHUDOWN state + */ + + {{IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent ECHO, Spec(sec 5.2.4) says it may be handled by the + * peer and peer shall move to the ESTABISHED. if it doesn't handle + * it will send ERROR chunk. So, stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, not sure what to do stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + /* + * SHUTDOWN sent from the client, move to SHUDDOWN_CLI + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + /* + * SHUTDOWN sent from the server, move to SHUTDOWN_SER + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_SHUT_SER */ }, + /* + * client resent SHUDTDOWN_ACK, let's stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_CLI /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + /* + * Server sent SHUTDOWN ACK, this should not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + /* + * SHUTDOWN COM from client, this should not happen, let's close the + * connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + /* + * SHUTDOWN COMPLETE from server this is what we are expecting. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + + /* + * State : IP_VS_SCTP_S_SHUT_ACK_SER + * SHUTDOWN ACK from the server, awaiting for SHUTDOWN COM from client + */ + /* + * We recieved the data chuck, keep the state unchanged. I assume + * that still data chuncks can be received by both the peers in + * SHUDOWN state + */ + + {{IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_DATA_SER */ }, + /* + * We have got an INIT from client. From the spec.“Upon receipt of + * an INIT in the COOKIE-WAIT state, an endpoint MUST respond with + * an INIT ACK using the same parameters it sent in its original + * INIT chunk (including its Initiate Tag, unchanged”). + */ + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + /* + * INIT_ACK sent by the server, Unexpected INIT ACK, spec says, + * “If an INIT ACK is received by an endpoint in any state other + * than the COOKIE-WAIT state, the endpoint should discard the + * INIT ACK chunk”. Stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + /* + * Client sent ECHO, Spec(sec 5.2.4) says it may be handled by the + * peer and peer shall move to the ESTABISHED. if it doesn't handle + * it will send ERROR chunk. So, stay in the same state + */ + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_ESTABLISHED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + /* + * COOKIE ACK from client, not sure what to do stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + /* + * SHUTDOWN sent from the client, move to SHUDDOWN_CLI + */ + {IP_VS_SCTP_S_SHUT_CLI /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + /* + * SHUTDOWN sent from the server, move to SHUTDOWN_SER + */ + {IP_VS_SCTP_S_SHUT_SER /* IP_VS_SCTP_EVE_SHUT_SER */ }, + /* + * client sent SHUDTDOWN_ACK, this should not happen let's close + * the connection. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + /* + * Server resent SHUTDOWN ACK, stay in the same state + */ + {IP_VS_SCTP_S_SHUT_ACK_SER /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + /* + * SHUTDOWN COM from client, this what we are expecting, let's close + * the connection + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + /* + * SHUTDOWN COMPLETE from server this should not happen. + */ + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + }, + /* + * State : IP_VS_SCTP_S_CLOSED + */ + {{IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_DATA_SER */ }, + {IP_VS_SCTP_S_INIT_CLI /* IP_VS_SCTP_EVE_INIT_CLI */ }, + {IP_VS_SCTP_S_INIT_SER /* IP_VS_SCTP_EVE_INIT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_INIT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ECHO_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_COOKIE_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_ABORT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_ACK_SER */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_CLI */ }, + {IP_VS_SCTP_S_CLOSED /* IP_VS_SCTP_EVE_SHUT_COM_SER */ } + } +}; + +/* + * Timeout table[state] + */ +static int sctp_timeouts[IP_VS_SCTP_S_LAST + 1] = { + [IP_VS_SCTP_S_NONE] = 2 * HZ, + [IP_VS_SCTP_S_INIT_CLI] = 1 * 60 * HZ, + [IP_VS_SCTP_S_INIT_SER] = 1 * 60 * HZ, + [IP_VS_SCTP_S_INIT_ACK_CLI] = 1 * 60 * HZ, + [IP_VS_SCTP_S_INIT_ACK_SER] = 1 * 60 * HZ, + [IP_VS_SCTP_S_ECHO_CLI] = 1 * 60 * HZ, + [IP_VS_SCTP_S_ECHO_SER] = 1 * 60 * HZ, + [IP_VS_SCTP_S_ESTABLISHED] = 15 * 60 * HZ, + [IP_VS_SCTP_S_SHUT_CLI] = 1 * 60 * HZ, + [IP_VS_SCTP_S_SHUT_SER] = 1 * 60 * HZ, + [IP_VS_SCTP_S_SHUT_ACK_CLI] = 1 * 60 * HZ, + [IP_VS_SCTP_S_SHUT_ACK_SER] = 1 * 60 * HZ, + [IP_VS_SCTP_S_CLOSED] = 10 * HZ, + [IP_VS_SCTP_S_LAST] = 2 * HZ, +}; + +static const char *sctp_state_name_table[IP_VS_SCTP_S_LAST + 1] = { + [IP_VS_SCTP_S_NONE] = "NONE", + [IP_VS_SCTP_S_INIT_CLI] = "INIT_CLI", + [IP_VS_SCTP_S_INIT_SER] = "INIT_SER", + [IP_VS_SCTP_S_INIT_ACK_CLI] = "INIT_ACK_CLI", + [IP_VS_SCTP_S_INIT_ACK_SER] = "INIT_ACK_SER", + [IP_VS_SCTP_S_ECHO_CLI] = "COOKIE_ECHO_CLI", + [IP_VS_SCTP_S_ECHO_SER] = "COOKIE_ECHO_SER", + [IP_VS_SCTP_S_ESTABLISHED] = "ESTABISHED", + [IP_VS_SCTP_S_SHUT_CLI] = "SHUTDOWN_CLI", + [IP_VS_SCTP_S_SHUT_SER] = "SHUTDOWN_SER", + [IP_VS_SCTP_S_SHUT_ACK_CLI] = "SHUTDOWN_ACK_CLI", + [IP_VS_SCTP_S_SHUT_ACK_SER] = "SHUTDOWN_ACK_SER", + [IP_VS_SCTP_S_CLOSED] = "CLOSED", + [IP_VS_SCTP_S_LAST] = "BUG!" +}; + + +static const char *sctp_state_name(int state) +{ + if (state >= IP_VS_SCTP_S_LAST) + return "ERR!"; + if (sctp_state_name_table[state]) + return sctp_state_name_table[state]; + return "?"; +} + +static void sctp_timeout_change(struct ip_vs_protocol *pp, int flags) +{ +} + +static int +sctp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) +{ + +return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_SCTP_S_LAST, + sctp_state_name_table, sname, to); +} + +static inline int +set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, + int direction, const struct sk_buff *skb) +{ + sctp_chunkhdr_t _sctpch, *sch; + unsigned char chunk_type; + int event, next_state; + int ihl; + +#ifdef CONFIG_IP_VS_IPV6 + ihl = cp->af == AF_INET ? ip_hdrlen(skb) : sizeof(struct ipv6hdr); +#else + ihl = ip_hdrlen(skb); +#endif + + sch = skb_header_pointer(skb, ihl + sizeof(sctp_sctphdr_t), + sizeof(_sctpch), &_sctpch); + if (sch == NULL) + return 0; + + chunk_type = sch->type; + /* + * Section 3: Multiple chunks can be bundled into one SCTP packet + * up to the MTU size, except for the INIT, INIT ACK, and + * SHUTDOWN COMPLETE chunks. These chunks MUST NOT be bundled with + * any other chunk in a packet. + * + * Section 3.3.7: DATA chunks MUST NOT be bundled with ABORT. Control + * chunks (except for INIT, INIT ACK, and SHUTDOWN COMPLETE) MAY be + * bundled with an ABORT, but they MUST be placed before the ABORT + * in the SCTP packet or they will be ignored by the receiver. + */ + if ((sch->type == SCTP_CID_COOKIE_ECHO) || + (sch->type == SCTP_CID_COOKIE_ACK)) { + sch = skb_header_pointer(skb, (ihl + sizeof(sctp_sctphdr_t) + + sch->length), sizeof(_sctpch), &_sctpch); + if (sch) { + if (sch->type == SCTP_CID_ABORT) + chunk_type = sch->type; + } + } + + event = sctp_events[chunk_type]; + + /* + * If the direction is IP_VS_DIR_OUTPUT, this event is from server + */ + if (direction == IP_VS_DIR_OUTPUT) + event++; + /* + * get next state + */ + next_state = sctp_states_table[cp->state][event].next_state; + + if (next_state != cp->state) { + struct ip_vs_dest *dest = cp->dest; + + IP_VS_DBG_BUF(8, "%s %s %s:%d->" + "%s:%d state: %s->%s conn->refcnt:%d\n", + pp->name, + ((direction == IP_VS_DIR_OUTPUT) ? + "output " : "input "), + IP_VS_DBG_ADDR(cp->af, &cp->daddr), + ntohs(cp->dport), + IP_VS_DBG_ADDR(cp->af, &cp->caddr), + ntohs(cp->cport), + sctp_state_name(cp->state), + sctp_state_name(next_state), + atomic_read(&cp->refcnt)); + if (dest) { + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (next_state != IP_VS_SCTP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags |= IP_VS_CONN_F_INACTIVE; + } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && + (next_state == IP_VS_SCTP_S_ESTABLISHED)) { + atomic_inc(&dest->activeconns); + atomic_dec(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } + } + + cp->timeout = pp->timeout_table[cp->state = next_state]; + + return 1; +} + +static int +sctp_state_transition(struct ip_vs_conn *cp, int direction, + const struct sk_buff *skb, struct ip_vs_protocol *pp) +{ + int ret = 0; + + spin_lock(&cp->lock); + ret = set_sctp_state(pp, cp, direction, skb); + spin_unlock(&cp->lock); + + return ret; +} + +/* + * Hash table for SCTP application incarnations + */ +#define SCTP_APP_TAB_BITS 4 +#define SCTP_APP_TAB_SIZE (1 << SCTP_APP_TAB_BITS) +#define SCTP_APP_TAB_MASK (SCTP_APP_TAB_SIZE - 1) + +static struct list_head sctp_apps[SCTP_APP_TAB_SIZE]; +static DEFINE_SPINLOCK(sctp_app_lock); + +static inline __u16 sctp_app_hashkey(__be16 port) +{ + return (((__force u16)port >> SCTP_APP_TAB_BITS) ^ (__force u16)port) + & SCTP_APP_TAB_MASK; +} + +static int sctp_register_app(struct ip_vs_app *inc) +{ + struct ip_vs_app *i; + __u16 hash; + __be16 port = inc->port; + int ret = 0; + + hash = sctp_app_hashkey(port); + + spin_lock_bh(&sctp_app_lock); + list_for_each_entry(i, &sctp_apps[hash], p_list) { + if (i->port == port) { + ret = -EEXIST; + goto out; + } + } + list_add(&inc->p_list, &sctp_apps[hash]); + atomic_inc(&ip_vs_protocol_sctp.appcnt); +out: + spin_unlock_bh(&sctp_app_lock); + + return ret; +} + +static void sctp_unregister_app(struct ip_vs_app *inc) +{ + spin_lock_bh(&sctp_app_lock); + atomic_dec(&ip_vs_protocol_sctp.appcnt); + list_del(&inc->p_list); + spin_unlock_bh(&sctp_app_lock); +} + +static int sctp_app_conn_bind(struct ip_vs_conn *cp) +{ + int hash; + struct ip_vs_app *inc; + int result = 0; + + /* Default binding: bind app only for NAT */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) + return 0; + /* Lookup application incarnations and bind the right one */ + hash = sctp_app_hashkey(cp->vport); + + spin_lock(&sctp_app_lock); + list_for_each_entry(inc, &sctp_apps[hash], p_list) { + if (inc->port == cp->vport) { + if (unlikely(!ip_vs_app_inc_get(inc))) + break; + spin_unlock(&sctp_app_lock); + + IP_VS_DBG_BUF(9, "%s: Binding conn %s:%u->" + "%s:%u to app %s on port %u\n", + __func__, + IP_VS_DBG_ADDR(cp->af, &cp->caddr), + ntohs(cp->cport), + IP_VS_DBG_ADDR(cp->af, &cp->vaddr), + ntohs(cp->vport), + inc->name, ntohs(inc->port)); + cp->app = inc; + if (inc->init_conn) + result = inc->init_conn(inc, cp); + goto out; + } + } + spin_unlock(&sctp_app_lock); +out: + return result; +} + +static void ip_vs_sctp_init(struct ip_vs_protocol *pp) +{ + IP_VS_INIT_HASH_TABLE(sctp_apps); + pp->timeout_table = sctp_timeouts; +} + + +static void ip_vs_sctp_exit(struct ip_vs_protocol *pp) +{ + +} + +struct ip_vs_protocol ip_vs_protocol_sctp = { + .name = "SCTP", + .protocol = IPPROTO_SCTP, + .num_states = IP_VS_SCTP_S_LAST, + .dont_defrag = 0, + .appcnt = ATOMIC_INIT(0), + .init = ip_vs_sctp_init, + .exit = ip_vs_sctp_exit, + .register_app = sctp_register_app, + .unregister_app = sctp_unregister_app, + .conn_schedule = sctp_conn_schedule, + .conn_in_get = sctp_conn_in_get, + .conn_out_get = sctp_conn_out_get, + .snat_handler = sctp_snat_handler, + .dnat_handler = sctp_dnat_handler, + .csum_check = sctp_csum_check, + .state_name = sctp_state_name, + .state_transition = sctp_state_transition, + .app_conn_bind = sctp_app_conn_bind, + .debug_packet = ip_vs_tcpudp_debug_packet, + .timeout_change = sctp_timeout_change, + .set_state_timeout = sctp_set_state_timeout, +}; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index e177f0dc208..8fb0ae61676 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -400,6 +400,11 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) flags |= IP_VS_CONN_F_INACTIVE; else flags &= ~IP_VS_CONN_F_INACTIVE; + } else if (s->protocol == IPPROTO_SCTP) { + if (state != IP_VS_SCTP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; } cp = ip_vs_conn_new(AF_INET, s->protocol, (union nf_inet_addr *)&s->caddr, @@ -434,6 +439,15 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) atomic_dec(&dest->inactconns); cp->flags &= ~IP_VS_CONN_F_INACTIVE; } + } else if ((cp->dest) && (cp->protocol == IPPROTO_SCTP) && + (cp->state != state)) { + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_SCTP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } } if (opt) -- cgit v1.2.3-70-g09d2 From 37ee3d5b3e979a168536e7e2f15bd1e769cb4122 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 18 Feb 2010 19:04:44 +0100 Subject: netfilter: nf_defrag_ipv4: fix compilation error with NF_CONNTRACK=n As reported by Randy Dunlap , compilation of nf_defrag_ipv4 fails with: include/net/netfilter/nf_conntrack.h:94: error: field 'ct_general' has incomplete type include/net/netfilter/nf_conntrack.h:178: error: 'const struct sk_buff' has no member named 'nfct' include/net/netfilter/nf_conntrack.h:185: error: implicit declaration of function 'nf_conntrack_put' include/net/netfilter/nf_conntrack.h:294: error: 'const struct sk_buff' has no member named 'nfct' net/ipv4/netfilter/nf_defrag_ipv4.c:45: error: 'struct sk_buff' has no member named 'nfct' net/ipv4/netfilter/nf_defrag_ipv4.c:46: error: 'struct sk_buff' has no member named 'nfct' net/nf_conntrack.h must not be included with NF_CONNTRACK=n, add a few #ifdefs. Long term the header file should be fixed to be usable even with NF_CONNTRACK=n. Tested-by: Randy Dunlap Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_zones.h | 6 ++++-- net/ipv4/netfilter/nf_defrag_ipv4.c | 6 +++++- 2 files changed, 9 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_zones.h b/include/net/netfilter/nf_conntrack_zones.h index 0bbb2bd51e8..034efe8d45a 100644 --- a/include/net/netfilter/nf_conntrack_zones.h +++ b/include/net/netfilter/nf_conntrack_zones.h @@ -1,10 +1,11 @@ #ifndef _NF_CONNTRACK_ZONES_H #define _NF_CONNTRACK_ZONES_H -#include - #define NF_CT_DEFAULT_ZONE 0 +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) +#include + struct nf_conntrack_zone { u16 id; }; @@ -20,4 +21,5 @@ static inline u16 nf_ct_zone(const struct nf_conn *ct) return NF_CT_DEFAULT_ZONE; } +#endif /* CONFIG_NF_CONNTRACK || CONFIG_NF_CONNTRACK_MODULE */ #endif /* _NF_CONNTRACK_ZONES_H */ diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index d498a704d45..cb763ae9ed9 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -16,9 +16,11 @@ #include #include -#include #include +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) #include +#endif +#include /* Returns new sk_buff, or NULL */ static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) @@ -42,8 +44,10 @@ static enum ip_defrag_users nf_ct_defrag_user(unsigned int hooknum, { u16 zone = NF_CT_DEFAULT_ZONE; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) if (skb->nfct) zone = nf_ct_zone((struct nf_conn *)skb->nfct); +#endif #ifdef CONFIG_BRIDGE_NETFILTER if (skb->nf_bridge && -- cgit v1.2.3-70-g09d2 From 501c774cb13c3ef8fb7fc5f08fa19473f7d9a0db Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 18 Feb 2010 05:46:50 +0000 Subject: net/macvtap: add vhost support This adds support for passing a macvtap file descriptor into vhost-net, much like we already do for tun/tap. Most of the new code is taken from the respective patch in the tun driver and may get consolidated in the future. Signed-off-by: Arnd Bergmann Acked-by: Sridhar Samudrala Signed-off-by: David S. Miller --- drivers/net/macvtap.c | 98 ++++++++++++++++++++++++++++++++++++---------- drivers/vhost/Kconfig | 2 +- drivers/vhost/net.c | 8 +++- include/linux/if_macvlan.h | 13 ++++++ 4 files changed, 97 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 70509974976..e354501ab29 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -58,6 +58,8 @@ static unsigned int macvtap_major; static struct class *macvtap_class; static struct cdev macvtap_cdev; +static const struct proto_ops macvtap_socket_ops; + /* * RCU usage: * The macvtap_queue and the macvlan_dev are loosely coupled, the @@ -176,7 +178,7 @@ static int macvtap_forward(struct net_device *dev, struct sk_buff *skb) return -ENOLINK; skb_queue_tail(&q->sk.sk_receive_queue, skb); - wake_up(q->sk.sk_sleep); + wake_up_interruptible_poll(q->sk.sk_sleep, POLLIN | POLLRDNORM | POLLRDBAND); return 0; } @@ -242,7 +244,7 @@ static void macvtap_sock_write_space(struct sock *sk) return; if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) - wake_up_interruptible_sync(sk->sk_sleep); + wake_up_interruptible_poll(sk->sk_sleep, POLLOUT | POLLWRNORM | POLLWRBAND); } static int macvtap_open(struct inode *inode, struct file *file) @@ -270,6 +272,8 @@ static int macvtap_open(struct inode *inode, struct file *file) init_waitqueue_head(&q->sock.wait); q->sock.type = SOCK_RAW; q->sock.state = SS_CONNECTED; + q->sock.file = file; + q->sock.ops = &macvtap_socket_ops; sock_init_data(&q->sock, &q->sk); q->sk.sk_write_space = macvtap_sock_write_space; @@ -387,32 +391,20 @@ static ssize_t macvtap_put_user(struct macvtap_queue *q, rcu_read_lock_bh(); vlan = rcu_dereference(q->vlan); - macvlan_count_rx(vlan, len, ret == 0, 0); + if (vlan) + macvlan_count_rx(vlan, len, ret == 0, 0); rcu_read_unlock_bh(); return ret ? ret : len; } -static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv, - unsigned long count, loff_t pos) +static ssize_t macvtap_do_read(struct macvtap_queue *q, struct kiocb *iocb, + const struct iovec *iv, unsigned long len, + int noblock) { - struct file *file = iocb->ki_filp; - struct macvtap_queue *q = file->private_data; - DECLARE_WAITQUEUE(wait, current); struct sk_buff *skb; - ssize_t len, ret = 0; - - if (!q) { - ret = -ENOLINK; - goto out; - } - - len = iov_length(iv, count); - if (len < 0) { - ret = -EINVAL; - goto out; - } + ssize_t ret = 0; add_wait_queue(q->sk.sk_sleep, &wait); while (len) { @@ -421,7 +413,7 @@ static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv, /* Read frames from the queue */ skb = skb_dequeue(&q->sk.sk_receive_queue); if (!skb) { - if (file->f_flags & O_NONBLOCK) { + if (noblock) { ret = -EAGAIN; break; } @@ -440,7 +432,24 @@ static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv, current->state = TASK_RUNNING; remove_wait_queue(q->sk.sk_sleep, &wait); + return ret; +} + +static ssize_t macvtap_aio_read(struct kiocb *iocb, const struct iovec *iv, + unsigned long count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct macvtap_queue *q = file->private_data; + ssize_t len, ret = 0; + len = iov_length(iv, count); + if (len < 0) { + ret = -EINVAL; + goto out; + } + + ret = macvtap_do_read(q, iocb, iv, len, file->f_flags & O_NONBLOCK); + ret = min_t(ssize_t, ret, len); /* XXX copied from tun.c. Why? */ out: return ret; } @@ -538,6 +547,53 @@ static const struct file_operations macvtap_fops = { #endif }; +static int macvtap_sendmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct macvtap_queue *q = container_of(sock, struct macvtap_queue, sock); + return macvtap_get_user(q, m->msg_iov, total_len, + m->msg_flags & MSG_DONTWAIT); +} + +static int macvtap_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len, + int flags) +{ + struct macvtap_queue *q = container_of(sock, struct macvtap_queue, sock); + int ret; + if (flags & ~(MSG_DONTWAIT|MSG_TRUNC)) + return -EINVAL; + ret = macvtap_do_read(q, iocb, m->msg_iov, total_len, + flags & MSG_DONTWAIT); + if (ret > total_len) { + m->msg_flags |= MSG_TRUNC; + ret = flags & MSG_TRUNC ? ret : total_len; + } + return ret; +} + +/* Ops structure to mimic raw sockets with tun */ +static const struct proto_ops macvtap_socket_ops = { + .sendmsg = macvtap_sendmsg, + .recvmsg = macvtap_recvmsg, +}; + +/* Get an underlying socket object from tun file. Returns error unless file is + * attached to a device. The returned object works like a packet socket, it + * can be used for sock_sendmsg/sock_recvmsg. The caller is responsible for + * holding a reference to the file for as long as the socket is in use. */ +struct socket *macvtap_get_socket(struct file *file) +{ + struct macvtap_queue *q; + if (file->f_op != &macvtap_fops) + return ERR_PTR(-EINVAL); + q = file->private_data; + if (!q) + return ERR_PTR(-EBADFD); + return &q->sock; +} +EXPORT_SYMBOL_GPL(macvtap_get_socket); + static int macvtap_init(void) { int err; diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig index 9e9355367bb..e4e2fd1b510 100644 --- a/drivers/vhost/Kconfig +++ b/drivers/vhost/Kconfig @@ -1,6 +1,6 @@ config VHOST_NET tristate "Host kernel accelerator for virtio net (EXPERIMENTAL)" - depends on NET && EVENTFD && (TUN || !TUN) && EXPERIMENTAL + depends on NET && EVENTFD && (TUN || !TUN) && (MACVTAP || !MACVTAP) && EXPERIMENTAL ---help--- This kernel module can be loaded in host kernel to accelerate guest networking with virtio_net. Not to be confused with virtio_net diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 4c8928319e1..91a324cc229 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -22,6 +22,7 @@ #include #include #include +#include #include @@ -452,13 +453,16 @@ err: return ERR_PTR(r); } -static struct socket *get_tun_socket(int fd) +static struct socket *get_tap_socket(int fd) { struct file *file = fget(fd); struct socket *sock; if (!file) return ERR_PTR(-EBADF); sock = tun_get_socket(file); + if (!IS_ERR(sock)) + return sock; + sock = macvtap_get_socket(file); if (IS_ERR(sock)) fput(file); return sock; @@ -473,7 +477,7 @@ static struct socket *get_socket(int fd) sock = get_raw_socket(fd); if (!IS_ERR(sock)) return sock; - sock = get_tun_socket(fd); + sock = get_tap_socket(fd); if (!IS_ERR(sock)) return sock; return ERR_PTR(-ENOTSOCK); diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index f9cb9ba1475..b78a712247d 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -7,6 +7,19 @@ #include #include +#if defined(CONFIG_MACVTAP) || defined(CONFIG_MACVTAP_MODULE) +struct socket *macvtap_get_socket(struct file *); +#else +#include +#include +struct file; +struct socket; +static inline struct socket *macvtap_get_socket(struct file *f) +{ + return ERR_PTR(-EINVAL); +} +#endif /* CONFIG_MACVTAP */ + struct macvlan_port; struct macvtap_queue; -- cgit v1.2.3-70-g09d2 From bbef49daca35d4fd21bf606a10b6980f17d9df5d Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 18 Feb 2010 08:13:30 +0000 Subject: ipv6: use standard lists for FIB walks Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 2 +- net/ipv6/ip6_fib.c | 17 ++++------------- 2 files changed, 5 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 8f279ddb359..86f46c49e31 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -124,7 +124,7 @@ static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) } struct fib6_walker_t { - struct fib6_walker_t *prev, *next; + struct list_head lh; struct fib6_node *root, *node; struct rt6_info *leaf; unsigned char state; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 77e122f53ea..2f9847924fa 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -93,29 +93,20 @@ static __u32 rt_sernum; static void fib6_gc_timer_cb(unsigned long arg); -static struct fib6_walker_t fib6_walker_list = { - .prev = &fib6_walker_list, - .next = &fib6_walker_list, -}; - -#define FOR_WALKERS(w) for ((w)=fib6_walker_list.next; (w) != &fib6_walker_list; (w)=(w)->next) +static LIST_HEAD(fib6_walkers); +#define FOR_WALKERS(w) list_for_each_entry(w, &fib6_walkers, lh) static inline void fib6_walker_link(struct fib6_walker_t *w) { write_lock_bh(&fib6_walker_lock); - w->next = fib6_walker_list.next; - w->prev = &fib6_walker_list; - w->next->prev = w; - w->prev->next = w; + list_add(&w->lh, &fib6_walkers); write_unlock_bh(&fib6_walker_lock); } static inline void fib6_walker_unlink(struct fib6_walker_t *w) { write_lock_bh(&fib6_walker_lock); - w->next->prev = w->prev; - w->prev->next = w->next; - w->prev = w->next = w; + list_del(&w->lh); write_unlock_bh(&fib6_walker_lock); } static __inline__ u32 fib6_new_sernum(void) -- cgit v1.2.3-70-g09d2 From 3ffe533c87281b68d469b279ff3a5056f9c75862 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 18 Feb 2010 08:25:24 +0000 Subject: ipv6: drop unused "dev" arg of icmpv6_send() Dunno, what was the idea, it wasn't used for a long time. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- drivers/infiniband/ulp/ipoib/ipoib_cm.c | 2 +- include/linux/icmpv6.h | 3 +-- net/ipv4/ip_gre.c | 2 +- net/ipv6/exthdrs.c | 2 +- net/ipv6/icmp.c | 5 ++--- net/ipv6/ip6_input.c | 3 +-- net/ipv6/ip6_output.c | 11 +++++------ net/ipv6/ip6_tunnel.c | 6 +++--- net/ipv6/mip6.c | 2 +- net/ipv6/netfilter/ip6t_REJECT.c | 2 +- net/ipv6/reassembly.c | 2 +- net/ipv6/route.c | 4 ++-- net/ipv6/sit.c | 2 +- net/ipv6/tunnel6.c | 4 ++-- net/ipv6/udp.c | 5 ++--- net/ipv6/xfrm6_output.c | 2 +- net/netfilter/ipvs/ip_vs_core.c | 5 ++--- net/netfilter/ipvs/ip_vs_xmit.c | 10 +++++----- 18 files changed, 33 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 30bdf427ee6..83a7751c38d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -1374,7 +1374,7 @@ static void ipoib_cm_skb_reap(struct work_struct *work) icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) else if (skb->protocol == htons(ETH_P_IPV6)) - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, priv->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); #endif dev_kfree_skb_any(skb); diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index c0d8357917e..4c4c74ec598 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -174,8 +174,7 @@ struct icmp6_filter { extern void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, - __u32 info, - struct net_device *dev); + __u32 info); extern int icmpv6_init(void); extern int icmpv6_err_convert(u8 type, u8 code, diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index a2a5983dbf0..c0c5274d027 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -793,7 +793,7 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev } if (mtu >= IPV6_MIN_MTU && mtu < skb->len - tunnel->hlen + gre_hlen) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); ip_rt_put(rt); goto tx_error; } diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 4bac362b133..074f2c084f9 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -481,7 +481,7 @@ looped_back: IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS); icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, - 0, skb->dev); + 0); kfree_skb(skb); return -1; } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 9a37379d741..eb9abe24bdf 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -114,7 +114,7 @@ static __inline__ void icmpv6_xmit_unlock(struct sock *sk) */ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) { - icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev); + icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos); kfree_skb(skb); } @@ -300,8 +300,7 @@ static inline void mip6_addr_swap(struct sk_buff *skb) {} /* * Send an ICMP message in response to a packet in error */ -void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, - struct net_device *dev) +void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) { struct net *net = dev_net(skb->dev); struct inet6_dev *idev = NULL; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 237e2dba6e9..e28f9203dec 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -216,8 +216,7 @@ resubmit: IP6_INC_STATS_BH(net, idev, IPSTATS_MIB_INUNKNOWNPROTOS); icmpv6_send(skb, ICMPV6_PARAMPROB, - ICMPV6_UNK_NEXTHDR, nhoff, - skb->dev); + ICMPV6_UNK_NEXTHDR, nhoff); } } else IP6_INC_STATS_BH(net, idev, IPSTATS_MIB_INDELIVERS); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index eb6d0972863..1a5fe9ad194 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -267,7 +267,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (net_ratelimit()) printk(KERN_DEBUG "IPv6: sending pkt_too_big to self\n"); skb->dev = dst->dev; - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); return -EMSGSIZE; @@ -441,8 +441,7 @@ int ip6_forward(struct sk_buff *skb) if (hdr->hop_limit <= 1) { /* Force OUTPUT device used as source address */ skb->dev = dst->dev; - icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, - 0, skb->dev); + icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0); IP6_INC_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS); @@ -504,7 +503,7 @@ int ip6_forward(struct sk_buff *skb) goto error; if (addrtype & IPV6_ADDR_LINKLOCAL) { icmpv6_send(skb, ICMPV6_DEST_UNREACH, - ICMPV6_NOT_NEIGHBOUR, 0, skb->dev); + ICMPV6_NOT_NEIGHBOUR, 0); goto error; } } @@ -512,7 +511,7 @@ int ip6_forward(struct sk_buff *skb) if (skb->len > dst_mtu(dst)) { /* Again, force OUTPUT device used as source address */ skb->dev = dst->dev; - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, dst_mtu(dst), skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, dst_mtu(dst)); IP6_INC_STATS_BH(net, ip6_dst_idev(dst), IPSTATS_MIB_INTOOBIGERRORS); IP6_INC_STATS_BH(net, @@ -627,7 +626,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) */ if (!skb->local_df) { skb->dev = skb_dst(skb)->dev; - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 9b02492d870..138980eec21 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -622,7 +622,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (rt && rt->rt6i_dev) skb2->dev = rt->rt6i_dev; - icmpv6_send(skb2, rel_type, rel_code, rel_info, skb2->dev); + icmpv6_send(skb2, rel_type, rel_code, rel_info); if (rt) dst_release(&rt->u.dst); @@ -1014,7 +1014,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) tel = (struct ipv6_tlv_tnl_enc_lim *)&skb_network_header(skb)[offset]; if (tel->encap_limit == 0) { icmpv6_send(skb, ICMPV6_PARAMPROB, - ICMPV6_HDR_FIELD, offset + 2, skb->dev); + ICMPV6_HDR_FIELD, offset + 2); return -1; } encap_limit = tel->encap_limit - 1; @@ -1033,7 +1033,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) err = ip6_tnl_xmit2(skb, dev, dsfield, &fl, encap_limit, &mtu); if (err != 0) { if (err == -EMSGSIZE) - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); return -1; } diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c index f797e8c6f3b..2794b600283 100644 --- a/net/ipv6/mip6.c +++ b/net/ipv6/mip6.c @@ -56,7 +56,7 @@ static inline void *mip6_padn(__u8 *data, __u8 padlen) static inline void mip6_param_prob(struct sk_buff *skb, u8 code, int pos) { - icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev); + icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos); } static int mip6_mh_len(int type) diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index 8311ca31816..dd8afbaf00a 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -169,7 +169,7 @@ send_unreach(struct net *net, struct sk_buff *skb_in, unsigned char code, if (hooknum == NF_INET_LOCAL_OUT && skb_in->dev == NULL) skb_in->dev = net->loopback_dev; - icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0, NULL); + icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0); } static unsigned int diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index b2847ed6a7d..a555156e977 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -228,7 +228,7 @@ static void ip6_frag_expire(unsigned long data) pointer directly, device might already disappeared. */ fq->q.fragments->dev = dev; - icmpv6_send(fq->q.fragments, ICMPV6_TIME_EXCEED, ICMPV6_EXC_FRAGTIME, 0, dev); + icmpv6_send(fq->q.fragments, ICMPV6_TIME_EXCEED, ICMPV6_EXC_FRAGTIME, 0); out_rcu_unlock: rcu_read_unlock(); out: diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8500156f263..88c0a5c49ae 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -909,7 +909,7 @@ static void ip6_link_failure(struct sk_buff *skb) { struct rt6_info *rt; - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0, skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); rt = (struct rt6_info *) skb_dst(skb); if (rt) { @@ -1884,7 +1884,7 @@ static int ip6_pkt_drop(struct sk_buff *skb, u8 code, int ipstats_mib_noroutes) ipstats_mib_noroutes); break; } - icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0, skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0); kfree_skb(skb); return 0; } diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 96eb2d4641c..b1eea811be4 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -743,7 +743,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); ip_rt_put(rt); goto tx_error; } diff --git a/net/ipv6/tunnel6.c b/net/ipv6/tunnel6.c index 51e2832d13a..e17bc1dfc1a 100644 --- a/net/ipv6/tunnel6.c +++ b/net/ipv6/tunnel6.c @@ -98,7 +98,7 @@ static int tunnel6_rcv(struct sk_buff *skb) if (!handler->handler(skb)) return 0; - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); drop: kfree_skb(skb); @@ -116,7 +116,7 @@ static int tunnel46_rcv(struct sk_buff *skb) if (!handler->handler(skb)) return 0; - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); drop: kfree_skb(skb); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index a7af9d68cd6..52b8347ae3b 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -680,12 +680,11 @@ static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, int proto) { + struct net *net = dev_net(skb->dev); struct sock *sk; struct udphdr *uh; - struct net_device *dev = skb->dev; struct in6_addr *saddr, *daddr; u32 ulen = 0; - struct net *net = dev_net(skb->dev); if (!pskb_may_pull(skb, sizeof(struct udphdr))) goto short_packet; @@ -744,7 +743,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, UDP6_INC_STATS_BH(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); kfree_skb(skb); return 0; diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index c4f4eef032a..0c92112dcba 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -38,7 +38,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb) if (!skb->local_df && skb->len > mtu) { skb->dev = dst->dev; - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); ret = -EMSGSIZE; } diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 72e96d823eb..44590887a92 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -515,8 +515,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, */ #ifdef CONFIG_IP_VS_IPV6 if (svc->af == AF_INET6) - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, - skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); else #endif icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); @@ -1048,7 +1047,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, - 0, skb->dev); + 0); else #endif icmp_send(skb, diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 30b3189bd29..223b5018c7d 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -311,7 +311,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, mtu = dst_mtu(&rt->u.dst); if (skb->len > mtu) { dst_release(&rt->u.dst); - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error; } @@ -454,7 +454,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, mtu = dst_mtu(&rt->u.dst); if (skb->len > mtu) { dst_release(&rt->u.dst); - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP_VS_DBG_RL_PKT(0, pp, skb, 0, "ip_vs_nat_xmit_v6(): frag needed for"); goto tx_error; @@ -672,7 +672,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr)) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); dst_release(&rt->u.dst); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error; @@ -814,7 +814,7 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->u.dst); if (skb->len > mtu) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); dst_release(&rt->u.dst); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error; @@ -965,7 +965,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, mtu = dst_mtu(&rt->u.dst); if (skb->len > mtu) { dst_release(&rt->u.dst); - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error; } -- cgit v1.2.3-70-g09d2 From b54452b07a7b1b8cc1385edba3ef2ef6d4679d5a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 18 Feb 2010 08:14:31 +0000 Subject: const: struct nla_policy Make remaining netlink policies as const. Fixup coding style where needed. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/netlink.h | 2 +- kernel/taskstats.c | 6 ++---- net/dcb/dcbnl.c | 16 ++++++++-------- net/irda/irnetlink.c | 2 +- net/wimax/op-msg.c | 3 +-- net/wimax/op-reset.c | 3 +-- net/wimax/op-rfkill.c | 3 +-- net/wimax/op-state-get.c | 3 +-- net/wimax/stack.c | 3 +-- net/wireless/nl80211.c | 14 +++++--------- 10 files changed, 22 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/net/netlink.h b/include/net/netlink.h index a63b2192ac1..f82e463c875 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -196,7 +196,7 @@ enum { * All other Exact length of attribute payload * * Example: - * static struct nla_policy my_policy[ATTR_MAX+1] __read_mostly = { + * static const struct nla_policy my_policy[ATTR_MAX+1] = { * [ATTR_FOO] = { .type = NLA_U16 }, * [ATTR_BAR] = { .type = NLA_STRING, .len = BARSIZ }, * [ATTR_BAZ] = { .len = sizeof(struct mystruct) }, diff --git a/kernel/taskstats.c b/kernel/taskstats.c index ea8384d3caa..899ca51be5e 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -46,15 +46,13 @@ static struct genl_family family = { .maxattr = TASKSTATS_CMD_ATTR_MAX, }; -static struct nla_policy taskstats_cmd_get_policy[TASKSTATS_CMD_ATTR_MAX+1] -__read_mostly = { +static const struct nla_policy taskstats_cmd_get_policy[TASKSTATS_CMD_ATTR_MAX+1] = { [TASKSTATS_CMD_ATTR_PID] = { .type = NLA_U32 }, [TASKSTATS_CMD_ATTR_TGID] = { .type = NLA_U32 }, [TASKSTATS_CMD_ATTR_REGISTER_CPUMASK] = { .type = NLA_STRING }, [TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK] = { .type = NLA_STRING },}; -static struct nla_policy -cgroupstats_cmd_get_policy[CGROUPSTATS_CMD_ATTR_MAX+1] __read_mostly = { +static const struct nla_policy cgroupstats_cmd_get_policy[CGROUPSTATS_CMD_ATTR_MAX+1] = { [CGROUPSTATS_CMD_ATTR_FD] = { .type = NLA_U32 }, }; diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c index db9f5b39388..813e399220a 100644 --- a/net/dcb/dcbnl.c +++ b/net/dcb/dcbnl.c @@ -54,7 +54,7 @@ MODULE_LICENSE("GPL"); /**************** DCB attribute policies *************************************/ /* DCB netlink attributes policy */ -static struct nla_policy dcbnl_rtnl_policy[DCB_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_rtnl_policy[DCB_ATTR_MAX + 1] = { [DCB_ATTR_IFNAME] = {.type = NLA_NUL_STRING, .len = IFNAMSIZ - 1}, [DCB_ATTR_STATE] = {.type = NLA_U8}, [DCB_ATTR_PFC_CFG] = {.type = NLA_NESTED}, @@ -68,7 +68,7 @@ static struct nla_policy dcbnl_rtnl_policy[DCB_ATTR_MAX + 1] = { }; /* DCB priority flow control to User Priority nested attributes */ -static struct nla_policy dcbnl_pfc_up_nest[DCB_PFC_UP_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_pfc_up_nest[DCB_PFC_UP_ATTR_MAX + 1] = { [DCB_PFC_UP_ATTR_0] = {.type = NLA_U8}, [DCB_PFC_UP_ATTR_1] = {.type = NLA_U8}, [DCB_PFC_UP_ATTR_2] = {.type = NLA_U8}, @@ -81,7 +81,7 @@ static struct nla_policy dcbnl_pfc_up_nest[DCB_PFC_UP_ATTR_MAX + 1] = { }; /* DCB priority grouping nested attributes */ -static struct nla_policy dcbnl_pg_nest[DCB_PG_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_pg_nest[DCB_PG_ATTR_MAX + 1] = { [DCB_PG_ATTR_TC_0] = {.type = NLA_NESTED}, [DCB_PG_ATTR_TC_1] = {.type = NLA_NESTED}, [DCB_PG_ATTR_TC_2] = {.type = NLA_NESTED}, @@ -103,7 +103,7 @@ static struct nla_policy dcbnl_pg_nest[DCB_PG_ATTR_MAX + 1] = { }; /* DCB traffic class nested attributes. */ -static struct nla_policy dcbnl_tc_param_nest[DCB_TC_ATTR_PARAM_MAX + 1] = { +static const struct nla_policy dcbnl_tc_param_nest[DCB_TC_ATTR_PARAM_MAX + 1] = { [DCB_TC_ATTR_PARAM_PGID] = {.type = NLA_U8}, [DCB_TC_ATTR_PARAM_UP_MAPPING] = {.type = NLA_U8}, [DCB_TC_ATTR_PARAM_STRICT_PRIO] = {.type = NLA_U8}, @@ -112,7 +112,7 @@ static struct nla_policy dcbnl_tc_param_nest[DCB_TC_ATTR_PARAM_MAX + 1] = { }; /* DCB capabilities nested attributes. */ -static struct nla_policy dcbnl_cap_nest[DCB_CAP_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_cap_nest[DCB_CAP_ATTR_MAX + 1] = { [DCB_CAP_ATTR_ALL] = {.type = NLA_FLAG}, [DCB_CAP_ATTR_PG] = {.type = NLA_U8}, [DCB_CAP_ATTR_PFC] = {.type = NLA_U8}, @@ -124,14 +124,14 @@ static struct nla_policy dcbnl_cap_nest[DCB_CAP_ATTR_MAX + 1] = { }; /* DCB capabilities nested attributes. */ -static struct nla_policy dcbnl_numtcs_nest[DCB_NUMTCS_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_numtcs_nest[DCB_NUMTCS_ATTR_MAX + 1] = { [DCB_NUMTCS_ATTR_ALL] = {.type = NLA_FLAG}, [DCB_NUMTCS_ATTR_PG] = {.type = NLA_U8}, [DCB_NUMTCS_ATTR_PFC] = {.type = NLA_U8}, }; /* DCB BCN nested attributes. */ -static struct nla_policy dcbnl_bcn_nest[DCB_BCN_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_bcn_nest[DCB_BCN_ATTR_MAX + 1] = { [DCB_BCN_ATTR_RP_0] = {.type = NLA_U8}, [DCB_BCN_ATTR_RP_1] = {.type = NLA_U8}, [DCB_BCN_ATTR_RP_2] = {.type = NLA_U8}, @@ -160,7 +160,7 @@ static struct nla_policy dcbnl_bcn_nest[DCB_BCN_ATTR_MAX + 1] = { }; /* DCB APP nested attributes. */ -static struct nla_policy dcbnl_app_nest[DCB_APP_ATTR_MAX + 1] = { +static const struct nla_policy dcbnl_app_nest[DCB_APP_ATTR_MAX + 1] = { [DCB_APP_ATTR_IDTYPE] = {.type = NLA_U8}, [DCB_APP_ATTR_ID] = {.type = NLA_U16}, [DCB_APP_ATTR_PRIORITY] = {.type = NLA_U8}, diff --git a/net/irda/irnetlink.c b/net/irda/irnetlink.c index 476b307bd80..69b5b75f543 100644 --- a/net/irda/irnetlink.c +++ b/net/irda/irnetlink.c @@ -124,7 +124,7 @@ static int irda_nl_get_mode(struct sk_buff *skb, struct genl_info *info) return ret; } -static struct nla_policy irda_nl_policy[IRDA_NL_ATTR_MAX + 1] = { +static const struct nla_policy irda_nl_policy[IRDA_NL_ATTR_MAX + 1] = { [IRDA_NL_ATTR_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ-1 }, [IRDA_NL_ATTR_MODE] = { .type = NLA_U32 }, diff --git a/net/wimax/op-msg.c b/net/wimax/op-msg.c index d3bfb6ef13a..7718657e93d 100644 --- a/net/wimax/op-msg.c +++ b/net/wimax/op-msg.c @@ -320,8 +320,7 @@ int wimax_msg(struct wimax_dev *wimax_dev, const char *pipe_name, EXPORT_SYMBOL_GPL(wimax_msg); -static const -struct nla_policy wimax_gnl_msg_policy[WIMAX_GNL_ATTR_MAX + 1] = { +static const struct nla_policy wimax_gnl_msg_policy[WIMAX_GNL_ATTR_MAX + 1] = { [WIMAX_GNL_MSG_IFIDX] = { .type = NLA_U32, }, diff --git a/net/wimax/op-reset.c b/net/wimax/op-reset.c index 35f370091f4..4dc82a54ba3 100644 --- a/net/wimax/op-reset.c +++ b/net/wimax/op-reset.c @@ -91,8 +91,7 @@ int wimax_reset(struct wimax_dev *wimax_dev) EXPORT_SYMBOL(wimax_reset); -static const -struct nla_policy wimax_gnl_reset_policy[WIMAX_GNL_ATTR_MAX + 1] = { +static const struct nla_policy wimax_gnl_reset_policy[WIMAX_GNL_ATTR_MAX + 1] = { [WIMAX_GNL_RESET_IFIDX] = { .type = NLA_U32, }, diff --git a/net/wimax/op-rfkill.c b/net/wimax/op-rfkill.c index ae752a64d92..e978c7136c9 100644 --- a/net/wimax/op-rfkill.c +++ b/net/wimax/op-rfkill.c @@ -410,8 +410,7 @@ void wimax_rfkill_rm(struct wimax_dev *wimax_dev) * just query). */ -static const -struct nla_policy wimax_gnl_rfkill_policy[WIMAX_GNL_ATTR_MAX + 1] = { +static const struct nla_policy wimax_gnl_rfkill_policy[WIMAX_GNL_ATTR_MAX + 1] = { [WIMAX_GNL_RFKILL_IFIDX] = { .type = NLA_U32, }, diff --git a/net/wimax/op-state-get.c b/net/wimax/op-state-get.c index a76b8fcb056..11ad3356eb5 100644 --- a/net/wimax/op-state-get.c +++ b/net/wimax/op-state-get.c @@ -33,8 +33,7 @@ #include "debug-levels.h" -static const -struct nla_policy wimax_gnl_state_get_policy[WIMAX_GNL_ATTR_MAX + 1] = { +static const struct nla_policy wimax_gnl_state_get_policy[WIMAX_GNL_ATTR_MAX + 1] = { [WIMAX_GNL_STGET_IFIDX] = { .type = NLA_U32, }, diff --git a/net/wimax/stack.c b/net/wimax/stack.c index c8866412f83..813e1eaea29 100644 --- a/net/wimax/stack.c +++ b/net/wimax/stack.c @@ -75,8 +75,7 @@ MODULE_PARM_DESC(debug, * close to where the data is generated. */ /* -static const -struct nla_policy wimax_gnl_re_status_change[WIMAX_GNL_ATTR_MAX + 1] = { +static const struct nla_policy wimax_gnl_re_status_change[WIMAX_GNL_ATTR_MAX + 1] = { [WIMAX_GNL_STCH_STATE_OLD] = { .type = NLA_U8 }, [WIMAX_GNL_STCH_STATE_NEW] = { .type = NLA_U8 }, }; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 5b79ecf17be..a001ea32cb7 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -58,7 +58,7 @@ static int get_rdev_dev_by_info_ifindex(struct genl_info *info, } /* policy for the attributes */ -static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { +static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_WIPHY] = { .type = NLA_U32 }, [NL80211_ATTR_WIPHY_NAME] = { .type = NLA_NUL_STRING, .len = 20-1 }, @@ -148,8 +148,7 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { }; /* policy for the attributes */ -static struct nla_policy -nl80211_key_policy[NL80211_KEY_MAX + 1] __read_mostly = { +static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = { [NL80211_KEY_DATA] = { .type = NLA_BINARY, .len = WLAN_MAX_KEY_LEN }, [NL80211_KEY_IDX] = { .type = NLA_U8 }, [NL80211_KEY_CIPHER] = { .type = NLA_U32 }, @@ -2501,8 +2500,7 @@ static int nl80211_set_bss(struct sk_buff *skb, struct genl_info *info) return err; } -static const struct nla_policy - reg_rule_policy[NL80211_REG_RULE_ATTR_MAX + 1] = { +static const struct nla_policy reg_rule_policy[NL80211_REG_RULE_ATTR_MAX + 1] = { [NL80211_ATTR_REG_RULE_FLAGS] = { .type = NLA_U32 }, [NL80211_ATTR_FREQ_RANGE_START] = { .type = NLA_U32 }, [NL80211_ATTR_FREQ_RANGE_END] = { .type = NLA_U32 }, @@ -2671,8 +2669,7 @@ do {\ } \ } while (0);\ -static struct nla_policy -nl80211_meshconf_params_policy[NL80211_MESHCONF_ATTR_MAX+1] __read_mostly = { +static const struct nla_policy nl80211_meshconf_params_policy[NL80211_MESHCONF_ATTR_MAX+1] = { [NL80211_MESHCONF_RETRY_TIMEOUT] = { .type = NLA_U16 }, [NL80211_MESHCONF_CONFIRM_TIMEOUT] = { .type = NLA_U16 }, [NL80211_MESHCONF_HOLDING_TIMEOUT] = { .type = NLA_U16 }, @@ -4470,8 +4467,7 @@ static u32 rateset_to_mask(struct ieee80211_supported_band *sband, return mask; } -static struct nla_policy -nl80211_txattr_policy[NL80211_TXRATE_MAX + 1] __read_mostly = { +static const struct nla_policy nl80211_txattr_policy[NL80211_TXRATE_MAX + 1] = { [NL80211_TXRATE_LEGACY] = { .type = NLA_BINARY, .len = NL80211_MAX_SUPP_RATES }, }; -- cgit v1.2.3-70-g09d2 From 5aa4b32fc86408705337e941ed716880c63d1590 Mon Sep 17 00:00:00 2001 From: Andreas Petlund Date: Thu, 18 Feb 2010 02:45:45 +0000 Subject: net: TCP thin-stream detection Inline function to dynamically detect thin streams based on the number of packets in flight. Used to dynamically trigger thin-stream mechanisms if enabled by ioctl or sysctl. Signed-off-by: Andreas Petlund Signed-off-by: David S. Miller --- Documentation/networking/tcp-thin.txt | 47 +++++++++++++++++++++++++++++++++++ include/net/tcp.h | 8 ++++++ 2 files changed, 55 insertions(+) create mode 100644 Documentation/networking/tcp-thin.txt (limited to 'include') diff --git a/Documentation/networking/tcp-thin.txt b/Documentation/networking/tcp-thin.txt new file mode 100644 index 00000000000..151e229980f --- /dev/null +++ b/Documentation/networking/tcp-thin.txt @@ -0,0 +1,47 @@ +Thin-streams and TCP +==================== +A wide range of Internet-based services that use reliable transport +protocols display what we call thin-stream properties. This means +that the application sends data with such a low rate that the +retransmission mechanisms of the transport protocol are not fully +effective. In time-dependent scenarios (like online games, control +systems, stock trading etc.) where the user experience depends +on the data delivery latency, packet loss can be devastating for +the service quality. Extreme latencies are caused by TCP's +dependency on the arrival of new data from the application to trigger +retransmissions effectively through fast retransmit instead of +waiting for long timeouts. + +After analysing a large number of time-dependent interactive +applications, we have seen that they often produce thin streams +and also stay with this traffic pattern throughout its entire +lifespan. The combination of time-dependency and the fact that the +streams provoke high latencies when using TCP is unfortunate. + +In order to reduce application-layer latency when packets are lost, +a set of mechanisms has been made, which address these latency issues +for thin streams. In short, if the kernel detects a thin stream, +the retransmission mechanisms are modified in the following manner: + +1) If the stream is thin, fast retransmit on the first dupACK. +2) If the stream is thin, do not apply exponential backoff. + +These enhancements are applied only if the stream is detected as +thin. This is accomplished by defining a threshold for the number +of packets in flight. If there are less than 4 packets in flight, +fast retransmissions can not be triggered, and the stream is prone +to experience high retransmission latencies. + +Since these mechanisms are targeted at time-dependent applications, +they must be specifically activated by the application using the +TCP_THIN_LINEAR_TIMEOUTS and TCP_THIN_DUPACK IOCTLS or the +tcp_thin_linear_timeouts and tcp_thin_dupack sysctls. Both +modifications are turned off by default. + +References +========== +More information on the modifications, as well as a wide range of +experimental data can be found here: +"Improving latency for interactive, thin-stream applications over +reliable transport" +http://simula.no/research/nd/publications/Simula.nd.477/simula_pdf_file diff --git a/include/net/tcp.h b/include/net/tcp.h index 75a00c80bdd..0bdc3f64024 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1386,6 +1386,14 @@ static inline void tcp_highest_sack_combine(struct sock *sk, tcp_sk(sk)->highest_sack = new; } +/* Determines whether this is a thin stream (which may suffer from + * increased latency). Used to trigger latency-reducing mechanisms. + */ +static inline unsigned int tcp_stream_is_thin(struct tcp_sock *tp) +{ + return tp->packets_out < 4 && !tcp_in_initial_slowstart(tp); +} + /* /proc */ enum tcp_seq_states { TCP_SEQ_STATE_LISTENING, -- cgit v1.2.3-70-g09d2 From 36e31b0af58728071e8023cf8e20c5166b700717 Mon Sep 17 00:00:00 2001 From: Andreas Petlund Date: Thu, 18 Feb 2010 02:47:01 +0000 Subject: net: TCP thin linear timeouts This patch will make TCP use only linear timeouts if the stream is thin. This will help to avoid the very high latencies that thin stream suffer because of exponential backoff. This mechanism is only active if enabled by iocontrol or syscontrol and the stream is identified as thin. A maximum of 6 linear timeouts is tried before exponential backoff is resumed. Signed-off-by: Andreas Petlund Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 12 ++++++++++++ include/linux/tcp.h | 5 ++++- include/net/tcp.h | 4 ++++ net/ipv4/sysctl_net_ipv4.c | 7 +++++++ net/ipv4/tcp.c | 7 +++++++ net/ipv4/tcp_timer.c | 21 ++++++++++++++++++++- 6 files changed, 54 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 2dc7a1d9768..f147310d9af 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -487,6 +487,18 @@ tcp_dma_copybreak - INTEGER and CONFIG_NET_DMA is enabled. Default: 4096 +tcp_thin_linear_timeouts - BOOLEAN + Enable dynamic triggering of linear timeouts for thin streams. + If set, a check is performed upon retransmission by timeout to + determine if the stream is thin (less than 4 packets in flight). + As long as the stream is found to be thin, up to 6 linear + timeouts may be performed before exponential backoff mode is + initiated. This improves retransmission latency for + non-aggressive thin streams, often found to be time-dependent. + For more information on thin streams, see + Documentation/networking/tcp-thin.txt + Default: 0 + UDP variables: udp_mem - vector of 3 INTEGERs: min, pressure, max diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 7fee8a4df93..3ba8b074612 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -103,6 +103,7 @@ enum { #define TCP_CONGESTION 13 /* Congestion control algorithm */ #define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ #define TCP_COOKIE_TRANSACTIONS 15 /* TCP Cookie Transactions */ +#define TCP_THIN_LINEAR_TIMEOUTS 16 /* Use linear timeouts for thin streams*/ /* for TCP_INFO socket option */ #define TCPI_OPT_TIMESTAMPS 1 @@ -340,7 +341,9 @@ struct tcp_sock { u32 frto_highmark; /* snd_nxt when RTO occurred */ u16 advmss; /* Advertised MSS */ u8 frto_counter; /* Number of new acks after RTO */ - u8 nonagle; /* Disable Nagle algorithm? */ + u8 nonagle : 4,/* Disable Nagle algorithm? */ + thin_lto : 1,/* Use linear timeouts for thin streams */ + unused : 3; /* RTT measurement */ u32 srtt; /* smoothed round trip time << 3 */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 0bdc3f64024..6278fc734ab 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -196,6 +196,9 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCP_NAGLE_CORK 2 /* Socket is corked */ #define TCP_NAGLE_PUSH 4 /* Cork is overridden for already queued data */ +/* TCP thin-stream limits */ +#define TCP_THIN_LINEAR_RETRIES 6 /* After 6 linear retries, do exp. backoff */ + extern struct inet_timewait_death_row tcp_death_row; /* sysctl variables for tcp */ @@ -241,6 +244,7 @@ extern int sysctl_tcp_workaround_signed_windows; extern int sysctl_tcp_slow_start_after_idle; extern int sysctl_tcp_max_ssthresh; extern int sysctl_tcp_cookie_size; +extern int sysctl_tcp_thin_linear_timeouts; extern atomic_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 7e3712ce399..e6a2460587d 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -575,6 +575,13 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_thin_linear_timeouts", + .data = &sysctl_tcp_thin_linear_timeouts, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { .procname = "udp_mem", .data = &sysctl_udp_mem, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e471d037fcc..21bae9afefe 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2229,6 +2229,13 @@ static int do_tcp_setsockopt(struct sock *sk, int level, } break; + case TCP_THIN_LINEAR_TIMEOUTS: + if (val < 0 || val > 1) + err = -EINVAL; + else + tp->thin_lto = val; + break; + case TCP_CORK: /* When set indicates to always queue non-full frames. * Later the user clears this option and we transmit diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index de7d1bf9114..a17629b8912 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -29,6 +29,7 @@ int sysctl_tcp_keepalive_intvl __read_mostly = TCP_KEEPALIVE_INTVL; int sysctl_tcp_retries1 __read_mostly = TCP_RETR1; int sysctl_tcp_retries2 __read_mostly = TCP_RETR2; int sysctl_tcp_orphan_retries __read_mostly; +int sysctl_tcp_thin_linear_timeouts __read_mostly; static void tcp_write_timer(unsigned long); static void tcp_delack_timer(unsigned long); @@ -415,7 +416,25 @@ void tcp_retransmit_timer(struct sock *sk) icsk->icsk_retransmits++; out_reset_timer: - icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX); + /* If stream is thin, use linear timeouts. Since 'icsk_backoff' is + * used to reset timer, set to 0. Recalculate 'icsk_rto' as this + * might be increased if the stream oscillates between thin and thick, + * thus the old value might already be too high compared to the value + * set by 'tcp_set_rto' in tcp_input.c which resets the rto without + * backoff. Limit to TCP_THIN_LINEAR_RETRIES before initiating + * exponential backoff behaviour to avoid continue hammering + * linear-timeout retransmissions into a black hole + */ + if (sk->sk_state == TCP_ESTABLISHED && + (tp->thin_lto || sysctl_tcp_thin_linear_timeouts) && + tcp_stream_is_thin(tp) && + icsk->icsk_retransmits <= TCP_THIN_LINEAR_RETRIES) { + icsk->icsk_backoff = 0; + icsk->icsk_rto = min(__tcp_set_rto(tp), TCP_RTO_MAX); + } else { + /* Use normal (exponential) backoff */ + icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX); + } inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX); if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1)) __sk_dst_reset(sk); -- cgit v1.2.3-70-g09d2 From 7e38017557bc0b87434d184f8804cadb102bb903 Mon Sep 17 00:00:00 2001 From: Andreas Petlund Date: Thu, 18 Feb 2010 04:48:19 +0000 Subject: net: TCP thin dupack This patch enables fast retransmissions after one dupACK for TCP if the stream is identified as thin. This will reduce latencies for thin streams that are not able to trigger fast retransmissions due to high packet interarrival time. This mechanism is only active if enabled by iocontrol or syscontrol and the stream is identified as thin. Signed-off-by: Andreas Petlund Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 12 ++++++++++++ include/linux/tcp.h | 4 +++- include/net/tcp.h | 1 + net/ipv4/sysctl_net_ipv4.c | 7 +++++++ net/ipv4/tcp.c | 7 +++++++ net/ipv4/tcp_input.c | 12 ++++++++++++ 6 files changed, 42 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index f147310d9af..2571a62d923 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -499,6 +499,18 @@ tcp_thin_linear_timeouts - BOOLEAN Documentation/networking/tcp-thin.txt Default: 0 +tcp_thin_dupack - BOOLEAN + Enable dynamic triggering of retransmissions after one dupACK + for thin streams. If set, a check is performed upon reception + of a dupACK to determine if the stream is thin (less than 4 + packets in flight). As long as the stream is found to be thin, + data is retransmitted on the first received dupACK. This + improves retransmission latency for non-aggressive thin + streams, often found to be time-dependent. + For more information on thin streams, see + Documentation/networking/tcp-thin.txt + Default: 0 + UDP variables: udp_mem - vector of 3 INTEGERs: min, pressure, max diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 3ba8b074612..a778ee02459 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -104,6 +104,7 @@ enum { #define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ #define TCP_COOKIE_TRANSACTIONS 15 /* TCP Cookie Transactions */ #define TCP_THIN_LINEAR_TIMEOUTS 16 /* Use linear timeouts for thin streams*/ +#define TCP_THIN_DUPACK 17 /* Fast retrans. after 1 dupack */ /* for TCP_INFO socket option */ #define TCPI_OPT_TIMESTAMPS 1 @@ -343,7 +344,8 @@ struct tcp_sock { u8 frto_counter; /* Number of new acks after RTO */ u8 nonagle : 4,/* Disable Nagle algorithm? */ thin_lto : 1,/* Use linear timeouts for thin streams */ - unused : 3; + thin_dupack : 1,/* Fast retransmit on first dupack */ + unused : 2; /* RTT measurement */ u32 srtt; /* smoothed round trip time << 3 */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 6278fc734ab..56f0aec40ed 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -245,6 +245,7 @@ extern int sysctl_tcp_slow_start_after_idle; extern int sysctl_tcp_max_ssthresh; extern int sysctl_tcp_cookie_size; extern int sysctl_tcp_thin_linear_timeouts; +extern int sysctl_tcp_thin_dupack; extern atomic_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e6a2460587d..c1bc074f61b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -582,6 +582,13 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_thin_dupack", + .data = &sysctl_tcp_thin_dupack, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { .procname = "udp_mem", .data = &sysctl_udp_mem, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 21bae9afefe..5901010fad5 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2236,6 +2236,13 @@ static int do_tcp_setsockopt(struct sock *sk, int level, tp->thin_lto = val; break; + case TCP_THIN_DUPACK: + if (val < 0 || val > 1) + err = -EINVAL; + else + tp->thin_dupack = val; + break; + case TCP_CORK: /* When set indicates to always queue non-full frames. * Later the user clears this option and we transmit diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 3fddc69cccc..788851ca8c5 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -89,6 +89,8 @@ int sysctl_tcp_frto __read_mostly = 2; int sysctl_tcp_frto_response __read_mostly; int sysctl_tcp_nometrics_save __read_mostly; +int sysctl_tcp_thin_dupack __read_mostly; + int sysctl_tcp_moderate_rcvbuf __read_mostly = 1; int sysctl_tcp_abc __read_mostly; @@ -2447,6 +2449,16 @@ static int tcp_time_to_recover(struct sock *sk) return 1; } + /* If a thin stream is detected, retransmit after first + * received dupack. Employ only if SACK is supported in order + * to avoid possible corner-case series of spurious retransmissions + * Use only if there are no unsent data. + */ + if ((tp->thin_dupack || sysctl_tcp_thin_dupack) && + tcp_stream_is_thin(tp) && tcp_dupack_heuristics(tp) > 1 && + tcp_is_sack(tp) && !tcp_send_head(sk)) + return 1; + return 0; } -- cgit v1.2.3-70-g09d2 From 72032fdbcde8b333e65b3430e1bcb4358e2d6716 Mon Sep 17 00:00:00 2001 From: jamal Date: Thu, 18 Feb 2010 03:35:07 +0000 Subject: xfrm: Introduce LINUX_MIB_XFRMFWDHDRERROR XFRMINHDRERROR counter is ambigous when validating forwarding path. It makes it tricky to debug when you have both in and fwd validation. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/linux/snmp.h | 1 + net/xfrm/xfrm_policy.c | 3 +-- net/xfrm/xfrm_proc.c | 1 + 3 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/snmp.h b/include/linux/snmp.h index 0f953fe4041..e28f5a0182e 100644 --- a/include/linux/snmp.h +++ b/include/linux/snmp.h @@ -257,6 +257,7 @@ enum LINUX_MIB_XFRMOUTPOLBLOCK, /* XfrmOutPolBlock */ LINUX_MIB_XFRMOUTPOLDEAD, /* XfrmOutPolDead */ LINUX_MIB_XFRMOUTPOLERROR, /* XfrmOutPolError */ + LINUX_MIB_XFRMFWDHDRERROR, /* XfrmFwdHdrError*/ __LINUX_MIB_XFRMMAX }; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 2c5d93181f1..4368e7b8846 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2045,8 +2045,7 @@ int __xfrm_route_forward(struct sk_buff *skb, unsigned short family) int res; if (xfrm_decode_session(skb, &fl, family) < 0) { - /* XXX: we should have something like FWDHDRERROR here. */ - XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR); + XFRM_INC_STATS(net, LINUX_MIB_XFRMFWDHDRERROR); return 0; } diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index 003f2c437ac..58d9ae00559 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -41,6 +41,7 @@ static const struct snmp_mib xfrm_mib_list[] = { SNMP_MIB_ITEM("XfrmOutPolBlock", LINUX_MIB_XFRMOUTPOLBLOCK), SNMP_MIB_ITEM("XfrmOutPolDead", LINUX_MIB_XFRMOUTPOLDEAD), SNMP_MIB_ITEM("XfrmOutPolError", LINUX_MIB_XFRMOUTPOLERROR), + SNMP_MIB_ITEM("XfrmFwdHdrError", LINUX_MIB_XFRMFWDHDRERROR), SNMP_MIB_SENTINEL }; -- cgit v1.2.3-70-g09d2 From 4bac6b180771f7ef5275b1a6d88e630ca3a3d6f0 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 19 Feb 2010 08:03:28 +0100 Subject: netfilter: restore POST_ROUTING hook in NF_HOOK_COND Commit 2249065 ("netfilter: get rid of the grossness in netfilter.h") inverted the logic for conditional hook invocation, breaking the POST_ROUTING hook invoked by ip_output(). Correct the logic and remove an unnecessary initialization. Reported-by: Stephen Hemminger Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 70079454ffd..89341c32631 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -212,8 +212,9 @@ NF_HOOK_COND(uint8_t pf, unsigned int hook, struct sk_buff *skb, struct net_device *in, struct net_device *out, int (*okfn)(struct sk_buff *), bool cond) { - int ret = 1; - if (cond || + int ret; + + if (!cond || (ret = nf_hook_thresh(pf, hook, skb, in, out, okfn, INT_MIN) == 1)) ret = okfn(skb); return ret; -- cgit v1.2.3-70-g09d2 From fbc87d67af5ccd733f894273b215564c67e3a749 Mon Sep 17 00:00:00 2001 From: Pavel Roskin Date: Thu, 11 Feb 2010 17:56:06 -0500 Subject: Introduce PCMCIA_DEVICE_PROD_ID3 ...and use it in hostap_cs and orinoco_cs. Another PCMCIA device with Intersil Prism chipset has been reported: Socket 0: product info: "Gigabyte", "GN-WLM01_P25L_ADAPTER", "ISL37300P", "Eval-RevA" manfid: 0x02e0, 0x1011 function: 6 (network) As it's the case with some other Prism based devices, the third ID string contains a design name that should be sufficient to identify the card as having Intersil Prism chipset and thus compatible with both orinoco_cs and hostap_cs. Introduce PCMCIA_DEVICE_PROD_ID3 that matches the third ID string only. Use it in orinoco_cs and hostap_cs to match cards with the third ID string indicating Prism chipset. Remove corresponding entries that use PCMCIA_DEVICE_PROD_ID123. Reported-by: Ozzy Signed-off-by: Pavel Roskin Signed-off-by: John W. Linville --- drivers/net/wireless/hostap/hostap_cs.c | 17 ++++------------- drivers/net/wireless/orinoco/orinoco_cs.c | 9 ++++----- include/pcmcia/device_id.h | 5 +++++ 3 files changed, 13 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/hostap/hostap_cs.c b/drivers/net/wireless/hostap/hostap_cs.c index c9640a3e02c..d19748d90aa 100644 --- a/drivers/net/wireless/hostap/hostap_cs.c +++ b/drivers/net/wireless/hostap/hostap_cs.c @@ -793,13 +793,6 @@ static struct pcmcia_device_id hostap_cs_ids[] = { 0x4b801a17), PCMCIA_MFC_DEVICE_PROD_ID12(0, "SanDisk", "ConnectPlus", 0x7a954bd9, 0x74be00c6), - PCMCIA_DEVICE_PROD_ID123( - "Intersil", "PRISM 2_5 PCMCIA ADAPTER", "ISL37300P", - 0x4b801a17, 0x6345a0bf, 0xc9049a39), - /* D-Link DWL-650 Rev. P1; manfid 0x000b, 0x7110 */ - PCMCIA_DEVICE_PROD_ID123( - "D-Link", "DWL-650 Wireless PC Card RevP", "ISL37101P-10", - 0x1a424a1c, 0x6ea57632, 0xdd97a26b), PCMCIA_DEVICE_PROD_ID123( "Addtron", "AWP-100 Wireless PCMCIA", "Version 01.02", 0xe6ec52ce, 0x08649af2, 0x4b74baa0), @@ -833,15 +826,13 @@ static struct pcmcia_device_id hostap_cs_ids[] = { "Allied Telesyn", "AT-WCL452 Wireless PCMCIA Radio", "Ver. 1.00", 0x5cd01705, 0x4271660f, 0x9d08ee12), - PCMCIA_DEVICE_PROD_ID123( - "corega", "WL PCCL-11", "ISL37300P", - 0xa21501a, 0x59868926, 0xc9049a39), - PCMCIA_DEVICE_PROD_ID123( - "The Linksys Group, Inc.", "Wireless Network CF Card", "ISL37300P", - 0xa5f472c2, 0x9c05598d, 0xc9049a39), PCMCIA_DEVICE_PROD_ID123( "Wireless LAN" , "11Mbps PC Card", "Version 01.02", 0x4b8870ff, 0x70e946d1, 0x4b74baa0), + PCMCIA_DEVICE_PROD_ID3("HFA3863", 0x355cb092), + PCMCIA_DEVICE_PROD_ID3("ISL37100P", 0x630d52b2), + PCMCIA_DEVICE_PROD_ID3("ISL37101P-10", 0xdd97a26b), + PCMCIA_DEVICE_PROD_ID3("ISL37300P", 0xc9049a39), PCMCIA_DEVICE_NULL }; MODULE_DEVICE_TABLE(pcmcia, hostap_cs_ids); diff --git a/drivers/net/wireless/orinoco/orinoco_cs.c b/drivers/net/wireless/orinoco/orinoco_cs.c index f27bb8367c9..1d4ada188ed 100644 --- a/drivers/net/wireless/orinoco/orinoco_cs.c +++ b/drivers/net/wireless/orinoco/orinoco_cs.c @@ -407,7 +407,6 @@ static struct pcmcia_device_id orinoco_cs_ids[] = { PCMCIA_DEVICE_PROD_ID12("3Com", "3CRWE737A AirConnect Wireless LAN PC Card", 0x41240e5b, 0x56010af3), PCMCIA_DEVICE_PROD_ID12("ACTIONTEC", "PRISM Wireless LAN PC Card", 0x393089da, 0xa71e69d5), PCMCIA_DEVICE_PROD_ID12("Addtron", "AWP-100 Wireless PCMCIA", 0xe6ec52ce, 0x08649af2), - PCMCIA_DEVICE_PROD_ID123("AIRVAST", "IEEE 802.11b Wireless PCMCIA Card", "HFA3863", 0xea569531, 0x4bcb9645, 0x355cb092), PCMCIA_DEVICE_PROD_ID12("Allied Telesyn", "AT-WCL452 Wireless PCMCIA Radio", 0x5cd01705, 0x4271660f), PCMCIA_DEVICE_PROD_ID12("ASUS", "802_11b_PC_CARD_25", 0x78fc06ee, 0xdb9aa842), PCMCIA_DEVICE_PROD_ID12("ASUS", "802_11B_CF_CARD_25", 0x78fc06ee, 0x45a50c1e), @@ -417,7 +416,6 @@ static struct pcmcia_device_id orinoco_cs_ids[] = { PCMCIA_DEVICE_PROD_ID12("BUFFALO", "WLI-CF-S11G", 0x2decece3, 0x82067c18), PCMCIA_DEVICE_PROD_ID12("Cabletron", "RoamAbout 802.11 DS", 0x32d445f5, 0xedeffd90), PCMCIA_DEVICE_PROD_ID12("Compaq", "WL200_11Mbps_Wireless_PCI_Card", 0x54f7c49c, 0x15a75e5b), - PCMCIA_DEVICE_PROD_ID123("corega", "WL PCCL-11", "ISL37300P", 0x0a21501a, 0x59868926, 0xc9049a39), PCMCIA_DEVICE_PROD_ID12("corega K.K.", "Wireless LAN PCC-11", 0x5261440f, 0xa6405584), PCMCIA_DEVICE_PROD_ID12("corega K.K.", "Wireless LAN PCCA-11", 0x5261440f, 0xdf6115f9), PCMCIA_DEVICE_PROD_ID12("corega_K.K.", "Wireless_LAN_PCCB-11", 0x29e33311, 0xee7a27ae), @@ -432,7 +430,6 @@ static struct pcmcia_device_id orinoco_cs_ids[] = { PCMCIA_DEVICE_PROD_ID12("INTERSIL", "HFA384x/IEEE", 0x74c5e40d, 0xdb472a18), PCMCIA_DEVICE_PROD_ID12("INTERSIL", "I-GATE 11M PC Card / PC Card plus", 0x74c5e40d, 0x8304ff77), PCMCIA_DEVICE_PROD_ID12("Intersil", "PRISM 2_5 PCMCIA ADAPTER", 0x4b801a17, 0x6345a0bf), - PCMCIA_DEVICE_PROD_ID123("Intersil", "PRISM Freedom PCMCIA Adapter", "ISL37100P", 0x4b801a17, 0xf222ec2d, 0x630d52b2), PCMCIA_DEVICE_PROD_ID12("LeArtery", "SYNCBYAIR 11Mbps Wireless LAN PC Card", 0x7e3b326a, 0x49893e92), PCMCIA_DEVICE_PROD_ID12("Linksys", "Wireless CompactFlash Card", 0x0733cc81, 0x0c52f395), PCMCIA_DEVICE_PROD_ID12("Lucent Technologies", "WaveLAN/IEEE", 0x23eb9949, 0xc562e72a), @@ -445,7 +442,6 @@ static struct pcmcia_device_id orinoco_cs_ids[] = { PCMCIA_DEVICE_PROD_ID12("Nortel Networks", "emobility 802.11 Wireless LAN PC Card", 0x2d617ea0, 0x88cd5767), PCMCIA_DEVICE_PROD_ID12("OEM", "PRISM2 IEEE 802.11 PC-Card", 0xfea54c90, 0x48f2bdd6), PCMCIA_DEVICE_PROD_ID12("OTC", "Wireless AirEZY 2411-PCC WLAN Card", 0x4ac44287, 0x235a6bed), - PCMCIA_DEVICE_PROD_ID123("PCMCIA", "11M WLAN Card v2.5", "ISL37300P", 0x281f1c5d, 0x6e440487, 0xc9049a39), PCMCIA_DEVICE_PROD_ID12("PLANEX", "GeoWave/GW-CF110", 0x209f40ab, 0xd9715264), PCMCIA_DEVICE_PROD_ID12("PLANEX", "GeoWave/GW-NS110", 0x209f40ab, 0x46263178), PCMCIA_DEVICE_PROD_ID12("PROXIM", "LAN PC CARD HARMONY 80211B", 0xc6536a5e, 0x090c3cd9), @@ -454,8 +450,11 @@ static struct pcmcia_device_id orinoco_cs_ids[] = { PCMCIA_DEVICE_PROD_ID12("SMC", "SMC2532W-B EliteConnect Wireless Adapter", 0xc4f8b18b, 0x196bd757), PCMCIA_DEVICE_PROD_ID12("SMC", "SMC2632W", 0xc4f8b18b, 0x474a1f2a), PCMCIA_DEVICE_PROD_ID12("Symbol Technologies", "LA4111 Spectrum24 Wireless LAN PC Card", 0x3f02b4d6, 0x3663cb0e), - PCMCIA_DEVICE_PROD_ID123("The Linksys Group, Inc.", "Instant Wireless Network PC Card", "ISL37300P", 0xa5f472c2, 0x590eb502, 0xc9049a39), PCMCIA_DEVICE_PROD_ID12("ZoomAir 11Mbps High", "Rate wireless Networking", 0x273fe3db, 0x32a1eaee), + PCMCIA_DEVICE_PROD_ID3("HFA3863", 0x355cb092), + PCMCIA_DEVICE_PROD_ID3("ISL37100P", 0x630d52b2), + PCMCIA_DEVICE_PROD_ID3("ISL37101P-10", 0xdd97a26b), + PCMCIA_DEVICE_PROD_ID3("ISL37300P", 0xc9049a39), PCMCIA_DEVICE_NULL, }; MODULE_DEVICE_TABLE(pcmcia, orinoco_cs_ids); diff --git a/include/pcmcia/device_id.h b/include/pcmcia/device_id.h index c33ea08352b..63e5b8f6b7d 100644 --- a/include/pcmcia/device_id.h +++ b/include/pcmcia/device_id.h @@ -34,6 +34,11 @@ .prod_id = { NULL, (v2), NULL, NULL }, \ .prod_id_hash = { 0, (vh2), 0, 0 }, } +#define PCMCIA_DEVICE_PROD_ID3(v3, vh3) { \ + .match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID3, \ + .prod_id = { NULL, NULL, (v3), NULL }, \ + .prod_id_hash = { 0, 0, (vh3), 0 }, } + #define PCMCIA_DEVICE_PROD_ID12(v1, v2, vh1, vh2) { \ .match_flags = PCMCIA_DEV_ID_MATCH_PROD_ID1| \ PCMCIA_DEV_ID_MATCH_PROD_ID2, \ -- cgit v1.2.3-70-g09d2 From ffb9eb3d8b450c22bbbc688c6b630141ac476fd9 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Wed, 17 Feb 2010 17:58:10 +0200 Subject: nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo Signed-off-by: John W. Linville --- include/linux/nl80211.h | 10 ++++ include/net/cfg80211.h | 7 +-- net/wireless/core.c | 16 +++--- net/wireless/nl80211.c | 131 +++++++++++++++++++++++++++++++++++++++++++++ net/wireless/wext-compat.c | 10 ++-- 5 files changed, 159 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 8e6384f8fda..28ba20fda3e 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -416,6 +416,9 @@ enum nl80211_commands { NL80211_CMD_ACTION, NL80211_CMD_ACTION_TX_STATUS, + NL80211_CMD_SET_POWER_SAVE, + NL80211_CMD_GET_POWER_SAVE, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -837,6 +840,8 @@ enum nl80211_attrs { NL80211_ATTR_ACK, + NL80211_ATTR_PS_STATE, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -1573,4 +1578,9 @@ enum nl80211_band { NL80211_BAND_5GHZ, }; +enum nl80211_ps_state { + NL80211_PS_DISABLED, + NL80211_PS_ENABLED, +}; + #endif /* __LINUX_NL80211_H */ diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 7188934b64d..3d134a1fb96 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1150,7 +1150,6 @@ struct cfg80211_ops { enum nl80211_channel_type channel_type, const u8 *buf, size_t len, u64 *cookie); - /* some temporary stuff to finish wext */ int (*set_power_mgmt)(struct wiphy *wiphy, struct net_device *dev, bool enabled, int timeout); }; @@ -1489,6 +1488,9 @@ struct wireless_dev { struct cfg80211_internal_bss *auth_bsses[MAX_AUTH_BSSES]; struct cfg80211_internal_bss *current_bss; /* associated / joined */ + bool ps; + int ps_timeout; + #ifdef CONFIG_CFG80211_WEXT /* wext data */ struct { @@ -1500,8 +1502,7 @@ struct wireless_dev { u8 bssid[ETH_ALEN], prev_bssid[ETH_ALEN]; u8 ssid[IEEE80211_MAX_SSID_LEN]; s8 default_key, default_mgmt_key; - bool ps, prev_bssid_valid; - int ps_timeout; + bool prev_bssid_valid; } wext; #endif }; diff --git a/net/wireless/core.c b/net/wireless/core.c index 51908dc2ea0..7fdb9409ad2 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -698,19 +698,21 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, wdev->wext.default_key = -1; wdev->wext.default_mgmt_key = -1; wdev->wext.connect.auth_type = NL80211_AUTHTYPE_AUTOMATIC; +#endif + if (wdev->wiphy->flags & WIPHY_FLAG_PS_ON_BY_DEFAULT) - wdev->wext.ps = true; + wdev->ps = true; else - wdev->wext.ps = false; - wdev->wext.ps_timeout = 100; + wdev->ps = false; + wdev->ps_timeout = 100; if (rdev->ops->set_power_mgmt) if (rdev->ops->set_power_mgmt(wdev->wiphy, dev, - wdev->wext.ps, - wdev->wext.ps_timeout)) { + wdev->ps, + wdev->ps_timeout)) { /* assume this means it's off */ - wdev->wext.ps = false; + wdev->ps = false; } -#endif + if (!dev->ethtool_ops) dev->ethtool_ops = &cfg80211_ethtool_ops; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 32811208135..b0495a1da22 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -148,6 +148,7 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { [NL80211_ATTR_FRAME] = { .type = NLA_BINARY, .len = IEEE80211_MAX_DATA_LEN }, [NL80211_ATTR_FRAME_MATCH] = { .type = NLA_BINARY, }, + [NL80211_ATTR_PS_STATE] = { .type = NLA_U32 }, }; /* policy for the attributes */ @@ -4663,6 +4664,124 @@ unlock_rtnl: return err; } +static int nl80211_set_power_save(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + struct wireless_dev *wdev; + struct net_device *dev; + u8 ps_state; + bool state; + int err; + + if (!info->attrs[NL80211_ATTR_PS_STATE]) { + err = -EINVAL; + goto out; + } + + ps_state = nla_get_u32(info->attrs[NL80211_ATTR_PS_STATE]); + + if (ps_state != NL80211_PS_DISABLED && ps_state != NL80211_PS_ENABLED) { + err = -EINVAL; + goto out; + } + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rdev; + + wdev = dev->ieee80211_ptr; + + if (!rdev->ops->set_power_mgmt) { + err = -EOPNOTSUPP; + goto unlock_rdev; + } + + state = (ps_state == NL80211_PS_ENABLED) ? true : false; + + if (state == wdev->ps) + goto unlock_rdev; + + wdev->ps = state; + + if (rdev->ops->set_power_mgmt(wdev->wiphy, dev, wdev->ps, + wdev->ps_timeout)) + /* assume this means it's off */ + wdev->ps = false; + +unlock_rdev: + cfg80211_unlock_rdev(rdev); + dev_put(dev); + rtnl_unlock(); + +out: + return err; +} + +static int nl80211_get_power_save(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev; + enum nl80211_ps_state ps_state; + struct wireless_dev *wdev; + struct net_device *dev; + struct sk_buff *msg; + void *hdr; + int err; + + rtnl_lock(); + + err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev); + if (err) + goto unlock_rtnl; + + wdev = dev->ieee80211_ptr; + + if (!rdev->ops->set_power_mgmt) { + err = -EOPNOTSUPP; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + err = -ENOMEM; + goto out; + } + + hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0, + NL80211_CMD_GET_POWER_SAVE); + if (!hdr) { + err = -ENOMEM; + goto free_msg; + } + + if (wdev->ps) + ps_state = NL80211_PS_ENABLED; + else + ps_state = NL80211_PS_DISABLED; + + NLA_PUT_U32(msg, NL80211_ATTR_PS_STATE, ps_state); + + genlmsg_end(msg, hdr); + err = genlmsg_reply(msg, info); + goto out; + +nla_put_failure: + err = -ENOBUFS; + +free_msg: + nlmsg_free(msg); + +out: + cfg80211_unlock_rdev(rdev); + dev_put(dev); + +unlock_rtnl: + rtnl_unlock(); + + return err; +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -4955,6 +5074,18 @@ static struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, + { + .cmd = NL80211_CMD_SET_POWER_SAVE, + .doit = nl80211_set_power_save, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_GET_POWER_SAVE, + .doit = nl80211_get_power_save, + .policy = nl80211_policy, + /* can be retrieved by unprivileged users */ + }, }; static struct genl_multicast_group nl80211_mlme_mcgrp = { diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index b17eeae448d..9ab51838849 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -1099,8 +1099,8 @@ int cfg80211_wext_siwpower(struct net_device *dev, { struct wireless_dev *wdev = dev->ieee80211_ptr; struct cfg80211_registered_device *rdev = wiphy_to_dev(wdev->wiphy); - bool ps = wdev->wext.ps; - int timeout = wdev->wext.ps_timeout; + bool ps = wdev->ps; + int timeout = wdev->ps_timeout; int err; if (wdev->iftype != NL80211_IFTYPE_STATION) @@ -1133,8 +1133,8 @@ int cfg80211_wext_siwpower(struct net_device *dev, if (err) return err; - wdev->wext.ps = ps; - wdev->wext.ps_timeout = timeout; + wdev->ps = ps; + wdev->ps_timeout = timeout; return 0; @@ -1147,7 +1147,7 @@ int cfg80211_wext_giwpower(struct net_device *dev, { struct wireless_dev *wdev = dev->ieee80211_ptr; - wrq->disabled = !wdev->wext.ps; + wrq->disabled = !wdev->ps; return 0; } -- cgit v1.2.3-70-g09d2 From 1cc523271ef0b6305c565a143e3d48f6fff826dd Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 22 Feb 2010 07:57:17 +0000 Subject: seq_file: add RCU versions of new hlist/list iterators (v3) Many usages of seq_file use RCU protected lists, so non RCU iterators will not work safely. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- fs/seq_file.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/rculist.h | 5 ++++ include/linux/seq_file.h | 15 +++++++--- 3 files changed, 87 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/fs/seq_file.c b/fs/seq_file.c index f65b16f02da..5afd554efad 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -750,3 +750,74 @@ struct hlist_node *seq_hlist_next(void *v, struct hlist_head *head, return node->next; } EXPORT_SYMBOL(seq_hlist_next); + +/** + * seq_hlist_start_rcu - start an iteration of a hlist protected by RCU + * @head: the head of the hlist + * @pos: the start position of the sequence + * + * Called at seq_file->op->start(). + * + * This list-traversal primitive may safely run concurrently with + * the _rcu list-mutation primitives such as hlist_add_head_rcu() + * as long as the traversal is guarded by rcu_read_lock(). + */ +struct hlist_node *seq_hlist_start_rcu(struct hlist_head *head, + loff_t pos) +{ + struct hlist_node *node; + + __hlist_for_each_rcu(node, head) + if (pos-- == 0) + return node; + return NULL; +} +EXPORT_SYMBOL(seq_hlist_start_rcu); + +/** + * seq_hlist_start_head_rcu - start an iteration of a hlist protected by RCU + * @head: the head of the hlist + * @pos: the start position of the sequence + * + * Called at seq_file->op->start(). Call this function if you want to + * print a header at the top of the output. + * + * This list-traversal primitive may safely run concurrently with + * the _rcu list-mutation primitives such as hlist_add_head_rcu() + * as long as the traversal is guarded by rcu_read_lock(). + */ +struct hlist_node *seq_hlist_start_head_rcu(struct hlist_head *head, + loff_t pos) +{ + if (!pos) + return SEQ_START_TOKEN; + + return seq_hlist_start_rcu(head, pos - 1); +} +EXPORT_SYMBOL(seq_hlist_start_head_rcu); + +/** + * seq_hlist_next_rcu - move to the next position of the hlist protected by RCU + * @v: the current iterator + * @head: the head of the hlist + * @pos: the current posision + * + * Called at seq_file->op->next(). + * + * This list-traversal primitive may safely run concurrently with + * the _rcu list-mutation primitives such as hlist_add_head_rcu() + * as long as the traversal is guarded by rcu_read_lock(). + */ +struct hlist_node *seq_hlist_next_rcu(void *v, + struct hlist_head *head, + loff_t *ppos) +{ + struct hlist_node *node = v; + + ++*ppos; + if (v == SEQ_START_TOKEN) + return rcu_dereference(head->first); + else + return rcu_dereference(node->next); +} +EXPORT_SYMBOL(seq_hlist_next_rcu); diff --git a/include/linux/rculist.h b/include/linux/rculist.h index 1bf0f708c4f..701fe9cb552 100644 --- a/include/linux/rculist.h +++ b/include/linux/rculist.h @@ -406,6 +406,11 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev, n->next->pprev = &n->next; } +#define __hlist_for_each_rcu(pos, head) \ + for (pos = rcu_dereference((head)->first); \ + pos && ({ prefetch(pos->next); 1; }); \ + pos = rcu_dereference(pos->next)) + /** * hlist_for_each_entry_rcu - iterate over rcu list of given type * @tpos: the type * to use as a loop cursor. diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index c95bcdc18f4..03c0232b416 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -140,10 +140,17 @@ extern struct list_head *seq_list_next(void *v, struct list_head *head, */ extern struct hlist_node *seq_hlist_start(struct hlist_head *head, - loff_t pos); + loff_t pos); extern struct hlist_node *seq_hlist_start_head(struct hlist_head *head, - loff_t pos); + loff_t pos); extern struct hlist_node *seq_hlist_next(void *v, struct hlist_head *head, - loff_t *ppos); - + loff_t *ppos); + +extern struct hlist_node *seq_hlist_start_rcu(struct hlist_head *head, + loff_t pos); +extern struct hlist_node *seq_hlist_start_head_rcu(struct hlist_head *head, + loff_t pos); +extern struct hlist_node *seq_hlist_next_rcu(void *v, + struct hlist_head *head, + loff_t *ppos); #endif -- cgit v1.2.3-70-g09d2 From 808f5114a9206fee855117d416440e1071ab375c Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 22 Feb 2010 07:57:18 +0000 Subject: packet: convert socket list to RCU (v3) Convert AF_PACKET to use RCU, eliminating one more reader/writer lock. There is no need for a real sk_del_node_init_rcu(), because sk_del_node_init is doing the equivalent thing to hlst_del_init_rcu already; but added some comments to try and make that obvious. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/net/netns/packet.h | 4 +-- include/net/sock.h | 10 ++++++++ net/packet/af_packet.c | 62 +++++++++++++++++++++++----------------------- 3 files changed, 43 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/net/netns/packet.h b/include/net/netns/packet.h index 637daf69888..cb4e894c0f8 100644 --- a/include/net/netns/packet.h +++ b/include/net/netns/packet.h @@ -4,11 +4,11 @@ #ifndef __NETNS_PACKET_H__ #define __NETNS_PACKET_H__ -#include +#include #include struct netns_packet { - rwlock_t sklist_lock; + spinlock_t sklist_lock; struct hlist_head sklist; }; diff --git a/include/net/sock.h b/include/net/sock.h index 580d51fa28e..6cb1676e409 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -381,6 +381,7 @@ static __inline__ void __sk_del_node(struct sock *sk) __hlist_del(&sk->sk_node); } +/* NB: equivalent to hlist_del_init_rcu */ static __inline__ int __sk_del_node_init(struct sock *sk) { if (sk_hashed(sk)) { @@ -421,6 +422,7 @@ static __inline__ int sk_del_node_init(struct sock *sk) } return rc; } +#define sk_del_node_init_rcu(sk) sk_del_node_init(sk) static __inline__ int __sk_nulls_del_node_init_rcu(struct sock *sk) { @@ -454,6 +456,12 @@ static __inline__ void sk_add_node(struct sock *sk, struct hlist_head *list) __sk_add_node(sk, list); } +static __inline__ void sk_add_node_rcu(struct sock *sk, struct hlist_head *list) +{ + sock_hold(sk); + hlist_add_head_rcu(&sk->sk_node, list); +} + static __inline__ void __sk_nulls_add_node_rcu(struct sock *sk, struct hlist_nulls_head *list) { hlist_nulls_add_head_rcu(&sk->sk_nulls_node, list); @@ -478,6 +486,8 @@ static __inline__ void sk_add_bind_node(struct sock *sk, #define sk_for_each(__sk, node, list) \ hlist_for_each_entry(__sk, node, list, sk_node) +#define sk_for_each_rcu(__sk, node, list) \ + hlist_for_each_entry_rcu(__sk, node, list, sk_node) #define sk_nulls_for_each(__sk, node, list) \ hlist_nulls_for_each_entry(__sk, node, list, sk_nulls_node) #define sk_nulls_for_each_rcu(__sk, node, list) \ diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 10f7295bcef..2f0369367ee 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1262,24 +1262,22 @@ static int packet_release(struct socket *sock) net = sock_net(sk); po = pkt_sk(sk); - write_lock_bh(&net->packet.sklist_lock); - sk_del_node_init(sk); + spin_lock_bh(&net->packet.sklist_lock); + sk_del_node_init_rcu(sk); sock_prot_inuse_add(net, sk->sk_prot, -1); - write_unlock_bh(&net->packet.sklist_lock); - - /* - * Unhook packet receive handler. - */ + spin_unlock_bh(&net->packet.sklist_lock); + spin_lock(&po->bind_lock); if (po->running) { /* - * Remove the protocol hook + * Remove from protocol table */ - dev_remove_pack(&po->prot_hook); po->running = 0; po->num = 0; + __dev_remove_pack(&po->prot_hook); __sock_put(sk); } + spin_unlock(&po->bind_lock); packet_flush_mclist(sk); @@ -1291,10 +1289,10 @@ static int packet_release(struct socket *sock) if (po->tx_ring.pg_vec) packet_set_ring(sk, &req, 1, 1); + synchronize_net(); /* * Now the socket is dead. No more input will appear. */ - sock_orphan(sk); sock->sk = NULL; @@ -1478,10 +1476,11 @@ static int packet_create(struct net *net, struct socket *sock, int protocol, po->running = 1; } - write_lock_bh(&net->packet.sklist_lock); - sk_add_node(sk, &net->packet.sklist); + spin_lock_bh(&net->packet.sklist_lock); + sk_add_node_rcu(sk, &net->packet.sklist); sock_prot_inuse_add(net, &packet_proto, 1); - write_unlock_bh(&net->packet.sklist_lock); + spin_unlock_bh(&net->packet.sklist_lock); + return 0; out: return err; @@ -2075,8 +2074,8 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void struct net_device *dev = data; struct net *net = dev_net(dev); - read_lock(&net->packet.sklist_lock); - sk_for_each(sk, node, &net->packet.sklist) { + rcu_read_lock(); + sk_for_each_rcu(sk, node, &net->packet.sklist) { struct packet_sock *po = pkt_sk(sk); switch (msg) { @@ -2104,18 +2103,19 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void } break; case NETDEV_UP: - spin_lock(&po->bind_lock); - if (dev->ifindex == po->ifindex && po->num && - !po->running) { - dev_add_pack(&po->prot_hook); - sock_hold(sk); - po->running = 1; + if (dev->ifindex == po->ifindex) { + spin_lock(&po->bind_lock); + if (po->num && !po->running) { + dev_add_pack(&po->prot_hook); + sock_hold(sk); + po->running = 1; + } + spin_unlock(&po->bind_lock); } - spin_unlock(&po->bind_lock); break; } } - read_unlock(&net->packet.sklist_lock); + rcu_read_unlock(); return NOTIFY_DONE; } @@ -2512,24 +2512,24 @@ static struct notifier_block packet_netdev_notifier = { #ifdef CONFIG_PROC_FS static void *packet_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(seq_file_net(seq)->packet.sklist_lock) + __acquires(RCU) { struct net *net = seq_file_net(seq); - read_lock(&net->packet.sklist_lock); - return seq_hlist_start_head(&net->packet.sklist, *pos); + + rcu_read_lock(); + return seq_hlist_start_head_rcu(&net->packet.sklist, *pos); } static void *packet_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct net *net = seq_file_net(seq); - return seq_hlist_next(v, &net->packet.sklist, pos); + return seq_hlist_next_rcu(v, &net->packet.sklist, pos); } static void packet_seq_stop(struct seq_file *seq, void *v) - __releases(seq_file_net(seq)->packet.sklist_lock) + __releases(RCU) { - struct net *net = seq_file_net(seq); - read_unlock(&net->packet.sklist_lock); + rcu_read_unlock(); } static int packet_seq_show(struct seq_file *seq, void *v) @@ -2581,7 +2581,7 @@ static const struct file_operations packet_seq_fops = { static int __net_init packet_net_init(struct net *net) { - rwlock_init(&net->packet.sklist_lock); + spin_lock_init(&net->packet.sklist_lock); INIT_HLIST_HEAD(&net->packet.sklist); if (!proc_net_fops_create(net, "packet", 0, &packet_seq_fops)) -- cgit v1.2.3-70-g09d2 From bf825f81b454fae2ffe1b675f3a549656726440e Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 22 Feb 2010 11:32:54 +0000 Subject: xfrm: introduce basic mark infrastructure Add basic structuring and accessors for xfrm mark Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/linux/xfrm.h | 12 +++++++++--- include/net/xfrm.h | 22 ++++++++++++++++++++++ 2 files changed, 31 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 29e04beb1fc..b971e384849 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -267,8 +267,8 @@ enum xfrm_attr_type_t { XFRMA_ALG_COMP, /* struct xfrm_algo */ XFRMA_ENCAP, /* struct xfrm_algo + struct xfrm_encap_tmpl */ XFRMA_TMPL, /* 1 or more struct xfrm_user_tmpl */ - XFRMA_SA, - XFRMA_POLICY, + XFRMA_SA, /* struct xfrm_usersa_info */ + XFRMA_POLICY, /*struct xfrm_userpolicy_info */ XFRMA_SEC_CTX, /* struct xfrm_sec_ctx */ XFRMA_LTIME_VAL, XFRMA_REPLAY_VAL, @@ -276,17 +276,23 @@ enum xfrm_attr_type_t { XFRMA_ETIMER_THRESH, XFRMA_SRCADDR, /* xfrm_address_t */ XFRMA_COADDR, /* xfrm_address_t */ - XFRMA_LASTUSED, + XFRMA_LASTUSED, /* unsigned long */ XFRMA_POLICY_TYPE, /* struct xfrm_userpolicy_type */ XFRMA_MIGRATE, XFRMA_ALG_AEAD, /* struct xfrm_algo_aead */ XFRMA_KMADDRESS, /* struct xfrm_user_kmaddress */ XFRMA_ALG_AUTH_TRUNC, /* struct xfrm_algo_auth */ + XFRMA_MARK, /* struct xfrm_mark */ __XFRMA_MAX #define XFRMA_MAX (__XFRMA_MAX - 1) }; +struct xfrm_mark { + __u32 v; /* value */ + __u32 m; /* mask */ +}; + enum xfrm_sadattr_type_t { XFRMA_SAD_UNSPEC, XFRMA_SAD_CNT, diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 0beb413c01c..39f151c7f25 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -140,6 +140,7 @@ struct xfrm_state { struct xfrm_id id; struct xfrm_selector sel; + struct xfrm_mark mark; u32 genid; @@ -481,6 +482,7 @@ struct xfrm_policy { u32 priority; u32 index; + struct xfrm_mark mark; struct xfrm_selector selector; struct xfrm_lifetime_cfg lft; struct xfrm_lifetime_cur curlft; @@ -1570,4 +1572,24 @@ static inline struct xfrm_state *xfrm_input_state(struct sk_buff *skb) } #endif +static inline int xfrm_mark_get(struct nlattr **attrs, struct xfrm_mark *m) +{ + if (attrs[XFRMA_MARK]) + memcpy(m, nla_data(attrs[XFRMA_MARK]), sizeof(m)); + else + m->v = m->m = 0; + + return m->v & m->m; +} + +static inline int xfrm_mark_put(struct sk_buff *skb, struct xfrm_mark *m) +{ + if (m->m | m->v) + NLA_PUT(skb, XFRMA_MARK, sizeof(struct xfrm_mark), m); + return 0; + +nla_put_failure: + return -1; +} + #endif /* _NET_XFRM_H */ -- cgit v1.2.3-70-g09d2 From bd55775c8dd656fc69b3a42a1c4ab32abb7e8af9 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 22 Feb 2010 16:20:22 -0800 Subject: xfrm: SA lookups signature with mark pass mark to all SA lookups to prepare them for when we add code to have them search. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/xfrm.h | 23 ++++++++++++++------ net/core/pktgen.c | 3 ++- net/ipv4/ah4.c | 2 +- net/ipv4/esp4.c | 2 +- net/ipv4/ipcomp.c | 6 ++++-- net/ipv6/ah6.c | 2 +- net/ipv6/esp6.c | 2 +- net/ipv6/ipcomp6.c | 6 ++++-- net/ipv6/xfrm6_input.c | 2 +- net/key/af_key.c | 14 ++++++------ net/xfrm/xfrm_input.c | 2 +- net/xfrm/xfrm_state.c | 58 +++++++++++++++++++++++++++++--------------------- net/xfrm/xfrm_user.c | 17 +++++++++------ 13 files changed, 84 insertions(+), 55 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 39f151c7f25..693523c870b 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1319,7 +1319,7 @@ extern struct xfrm_state *xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t struct flowi *fl, struct xfrm_tmpl *tmpl, struct xfrm_policy *pol, int *err, unsigned short family); -extern struct xfrm_state * xfrm_stateonly_find(struct net *net, +extern struct xfrm_state *xfrm_stateonly_find(struct net *net, u32 mark, xfrm_address_t *daddr, xfrm_address_t *saddr, unsigned short family, @@ -1328,8 +1328,14 @@ extern int xfrm_state_check_expire(struct xfrm_state *x); extern void xfrm_state_insert(struct xfrm_state *x); extern int xfrm_state_add(struct xfrm_state *x); extern int xfrm_state_update(struct xfrm_state *x); -extern struct xfrm_state *xfrm_state_lookup(struct net *net, xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family); -extern struct xfrm_state *xfrm_state_lookup_byaddr(struct net *net, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family); +extern struct xfrm_state *xfrm_state_lookup(struct net *net, u32 mark, + xfrm_address_t *daddr, __be32 spi, + u8 proto, unsigned short family); +extern struct xfrm_state *xfrm_state_lookup_byaddr(struct net *net, u32 mark, + xfrm_address_t *daddr, + xfrm_address_t *saddr, + u8 proto, + unsigned short family); #ifdef CONFIG_XFRM_SUB_POLICY extern int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n, unsigned short family); @@ -1366,7 +1372,8 @@ struct xfrmk_spdinfo { u32 spdhmcnt; }; -extern struct xfrm_state *xfrm_find_acq_byseq(struct net *net, u32 seq); +extern struct xfrm_state *xfrm_find_acq_byseq(struct net *net, u32 mark, + u32 seq); extern int xfrm_state_delete(struct xfrm_state *x); extern int xfrm_state_flush(struct net *net, u8 proto, struct xfrm_audit *audit_info); extern void xfrm_sad_getinfo(struct net *net, struct xfrmk_sadinfo *si); @@ -1451,9 +1458,11 @@ struct xfrm_policy *xfrm_policy_byid(struct net *net, u8, int dir, u32 id, int d int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info); u32 xfrm_get_acqseq(void); extern int xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi); -struct xfrm_state * xfrm_find_acq(struct net *net, u8 mode, u32 reqid, u8 proto, - xfrm_address_t *daddr, xfrm_address_t *saddr, - int create, unsigned short family); +struct xfrm_state *xfrm_find_acq(struct net *net, struct xfrm_mark *mark, + u8 mode, u32 reqid, u8 proto, + xfrm_address_t *daddr, + xfrm_address_t *saddr, int create, + unsigned short family); extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol); extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst, struct flowi *fl, int family, int strict); diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 2e692afdc55..43923811bd6 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2188,12 +2188,13 @@ static inline int f_pick(struct pktgen_dev *pkt_dev) /* If there was already an IPSEC SA, we keep it as is, else * we go look for it ... */ +#define DUMMY_MARK 0 static void get_ipsec_sa(struct pktgen_dev *pkt_dev, int flow) { struct xfrm_state *x = pkt_dev->flows[flow].x; if (!x) { /*slow path: we dont already have xfrm_state*/ - x = xfrm_stateonly_find(&init_net, + x = xfrm_stateonly_find(&init_net, DUMMY_MARK, (xfrm_address_t *)&pkt_dev->cur_daddr, (xfrm_address_t *)&pkt_dev->cur_saddr, AF_INET, diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 7ed3e4ae93a..987b47dc69a 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -393,7 +393,7 @@ static void ah4_err(struct sk_buff *skb, u32 info) icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return; - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, ah->spi, IPPROTO_AH, AF_INET); + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, ah->spi, IPPROTO_AH, AF_INET); if (!x) return; printk(KERN_DEBUG "pmtu discovery on SA AH/%08x/%08x\n", diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 1948895beb6..14ca1f1c3fb 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -422,7 +422,7 @@ static void esp4_err(struct sk_buff *skb, u32 info) icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return; - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, esph->spi, IPPROTO_ESP, AF_INET); + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, esph->spi, IPPROTO_ESP, AF_INET); if (!x) return; NETDEBUG(KERN_DEBUG "pmtu discovery on SA ESP/%08x/%08x\n", diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c index 83ed7150089..629067571f0 100644 --- a/net/ipv4/ipcomp.c +++ b/net/ipv4/ipcomp.c @@ -36,7 +36,7 @@ static void ipcomp4_err(struct sk_buff *skb, u32 info) return; spi = htonl(ntohs(ipch->cpi)); - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, spi, IPPROTO_COMP, AF_INET); if (!x) return; @@ -63,6 +63,7 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x) t->props.mode = x->props.mode; t->props.saddr.a4 = x->props.saddr.a4; t->props.flags = x->props.flags; + memcpy(&t->mark, &x->mark, sizeof(t->mark)); if (xfrm_init_state(t)) goto error; @@ -87,8 +88,9 @@ static int ipcomp_tunnel_attach(struct xfrm_state *x) struct net *net = xs_net(x); int err = 0; struct xfrm_state *t; + u32 mark = x->mark.v & x->mark.m; - t = xfrm_state_lookup(net, (xfrm_address_t *)&x->id.daddr.a4, + t = xfrm_state_lookup(net, mark, (xfrm_address_t *)&x->id.daddr.a4, x->props.saddr.a4, IPPROTO_IPIP, AF_INET); if (!t) { t = ipcomp_tunnel_create(x); diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index c2f300c314b..5ac89025f9d 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -614,7 +614,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, type != ICMPV6_PKT_TOOBIG) return; - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, ah->spi, IPPROTO_AH, AF_INET6); + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, ah->spi, IPPROTO_AH, AF_INET6); if (!x) return; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 668a46b655e..ee9b93bdd6a 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -365,7 +365,7 @@ static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, type != ICMPV6_PKT_TOOBIG) return; - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, esph->spi, IPPROTO_ESP, AF_INET6); + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, esph->spi, IPPROTO_ESP, AF_INET6); if (!x) return; printk(KERN_DEBUG "pmtu discovery on SA ESP/%08x/%pI6\n", diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index bb42f39c1db..85cccd6ed0b 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -64,7 +64,7 @@ static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, return; spi = htonl(ntohs(ipcomph->cpi)); - x = xfrm_state_lookup(net, (xfrm_address_t *)&iph->daddr, spi, IPPROTO_COMP, AF_INET6); + x = xfrm_state_lookup(net, skb->mark, (xfrm_address_t *)&iph->daddr, spi, IPPROTO_COMP, AF_INET6); if (!x) return; @@ -92,6 +92,7 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x) t->props.family = AF_INET6; t->props.mode = x->props.mode; memcpy(t->props.saddr.a6, x->props.saddr.a6, sizeof(struct in6_addr)); + memcpy(&t->mark, &x->mark, sizeof(t->mark)); if (xfrm_init_state(t)) goto error; @@ -114,10 +115,11 @@ static int ipcomp6_tunnel_attach(struct xfrm_state *x) int err = 0; struct xfrm_state *t = NULL; __be32 spi; + u32 mark = x->mark.m & x->mark.v; spi = xfrm6_tunnel_spi_lookup(net, (xfrm_address_t *)&x->props.saddr); if (spi) - t = xfrm_state_lookup(net, (xfrm_address_t *)&x->id.daddr, + t = xfrm_state_lookup(net, mark, (xfrm_address_t *)&x->id.daddr, spi, IPPROTO_IPV6, AF_INET6); if (!t) { t = ipcomp6_tunnel_create(x); diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index 9084582d236..2bc98ede123 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -101,7 +101,7 @@ int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr, break; } - x = xfrm_state_lookup_byaddr(net, dst, src, proto, AF_INET6); + x = xfrm_state_lookup_byaddr(net, skb->mark, dst, src, proto, AF_INET6); if (!x) continue; diff --git a/net/key/af_key.c b/net/key/af_key.c index da2fe5f5761..aae3cd86ccd 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -43,6 +43,8 @@ struct netns_pfkey { }; static DEFINE_MUTEX(pfkey_mutex); +#define DUMMY_MARK 0 +static struct xfrm_mark dummy_mark = {0, 0}; struct pfkey_sock { /* struct sock must be the first member of struct pfkey_sock */ struct sock sk; @@ -647,7 +649,7 @@ static struct xfrm_state *pfkey_xfrm_state_lookup(struct net *net, struct sadb_ if (!xaddr) return NULL; - return xfrm_state_lookup(net, xaddr, sa->sadb_sa_spi, proto, family); + return xfrm_state_lookup(net, DUMMY_MARK, xaddr, sa->sadb_sa_spi, proto, family); } #define PFKEY_ALIGN8(a) (1 + (((a) - 1) | (8 - 1))) @@ -1316,7 +1318,7 @@ static int pfkey_getspi(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h } if (hdr->sadb_msg_seq) { - x = xfrm_find_acq_byseq(net, hdr->sadb_msg_seq); + x = xfrm_find_acq_byseq(net, DUMMY_MARK, hdr->sadb_msg_seq); if (x && xfrm_addr_cmp(&x->id.daddr, xdaddr, family)) { xfrm_state_put(x); x = NULL; @@ -1324,7 +1326,7 @@ static int pfkey_getspi(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h } if (!x) - x = xfrm_find_acq(net, mode, reqid, proto, xdaddr, xsaddr, 1, family); + x = xfrm_find_acq(net, &dummy_mark, mode, reqid, proto, xdaddr, xsaddr, 1, family); if (x == NULL) return -ENOENT; @@ -1373,7 +1375,7 @@ static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg * if (hdr->sadb_msg_seq == 0 || hdr->sadb_msg_errno == 0) return 0; - x = xfrm_find_acq_byseq(net, hdr->sadb_msg_seq); + x = xfrm_find_acq_byseq(net, DUMMY_MARK, hdr->sadb_msg_seq); if (x == NULL) return 0; @@ -2572,8 +2574,8 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h return -EINVAL; delete = (hdr->sadb_msg_type == SADB_X_SPDDELETE2); - xp = xfrm_policy_byid(net, XFRM_POLICY_TYPE_MAIN, dir, - pol->sadb_x_policy_id, delete, &err); + xp = xfrm_policy_byid(net, XFRM_POLICY_TYPE_MAIN, + dir, pol->sadb_x_policy_id, delete, &err); if (xp == NULL) return -ENOENT; diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index e0009c17d80..45f1c98d4fc 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -152,7 +152,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop; } - x = xfrm_state_lookup(net, daddr, spi, nexthdr, family); + x = xfrm_state_lookup(net, skb->mark, daddr, spi, nexthdr, family); if (x == NULL) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOSTATES); xfrm_audit_state_notfound(skb, family, spi, seq); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 9fa3322b2a7..9f8530356b8 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -669,7 +669,7 @@ xfrm_init_tempsel(struct xfrm_state *x, struct flowi *fl, return 0; } -static struct xfrm_state *__xfrm_state_lookup(struct net *net, xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family) +static struct xfrm_state *__xfrm_state_lookup(struct net *net, u32 mark, xfrm_address_t *daddr, __be32 spi, u8 proto, unsigned short family) { unsigned int h = xfrm_spi_hash(net, daddr, spi, proto, family); struct xfrm_state *x; @@ -689,7 +689,7 @@ static struct xfrm_state *__xfrm_state_lookup(struct net *net, xfrm_address_t *d return NULL; } -static struct xfrm_state *__xfrm_state_lookup_byaddr(struct net *net, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family) +static struct xfrm_state *__xfrm_state_lookup_byaddr(struct net *net, u32 mark, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family) { unsigned int h = xfrm_src_hash(net, daddr, saddr, family); struct xfrm_state *x; @@ -713,12 +713,14 @@ static inline struct xfrm_state * __xfrm_state_locate(struct xfrm_state *x, int use_spi, int family) { struct net *net = xs_net(x); + u32 mark = x->mark.v & x->mark.m; if (use_spi) - return __xfrm_state_lookup(net, &x->id.daddr, x->id.spi, - x->id.proto, family); + return __xfrm_state_lookup(net, mark, &x->id.daddr, + x->id.spi, x->id.proto, family); else - return __xfrm_state_lookup_byaddr(net, &x->id.daddr, + return __xfrm_state_lookup_byaddr(net, mark, + &x->id.daddr, &x->props.saddr, x->id.proto, family); } @@ -783,6 +785,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, int acquire_in_progress = 0; int error = 0; struct xfrm_state *best = NULL; + u32 mark = pol->mark.v & pol->mark.m; to_put = NULL; @@ -819,7 +822,7 @@ found: x = best; if (!x && !error && !acquire_in_progress) { if (tmpl->id.spi && - (x0 = __xfrm_state_lookup(net, daddr, tmpl->id.spi, + (x0 = __xfrm_state_lookup(net, mark, daddr, tmpl->id.spi, tmpl->id.proto, family)) != NULL) { to_put = x0; error = -EEXIST; @@ -833,6 +836,7 @@ found: /* Initialize temporary selector matching only * to current session. */ xfrm_init_tempsel(x, fl, tmpl, daddr, saddr, family); + memcpy(&x->mark, &pol->mark, sizeof(x->mark)); error = security_xfrm_state_alloc_acquire(x, pol->security, fl->secid); if (error) { @@ -875,7 +879,7 @@ out: } struct xfrm_state * -xfrm_stateonly_find(struct net *net, +xfrm_stateonly_find(struct net *net, u32 mark, xfrm_address_t *daddr, xfrm_address_t *saddr, unsigned short family, u8 mode, u8 proto, u32 reqid) { @@ -971,7 +975,7 @@ void xfrm_state_insert(struct xfrm_state *x) EXPORT_SYMBOL(xfrm_state_insert); /* xfrm_state_lock is held */ -static struct xfrm_state *__find_acq_core(struct net *net, unsigned short family, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create) +static struct xfrm_state *__find_acq_core(struct net *net, struct xfrm_mark *m, unsigned short family, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create) { unsigned int h = xfrm_dst_hash(net, daddr, saddr, reqid, family); struct hlist_node *entry; @@ -1026,6 +1030,8 @@ static struct xfrm_state *__find_acq_core(struct net *net, unsigned short family x->props.family = family; x->props.mode = mode; x->props.reqid = reqid; + x->mark.v = m->v; + x->mark.m = m->m; x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires; xfrm_state_hold(x); tasklet_hrtimer_start(&x->mtimer, ktime_set(net->xfrm.sysctl_acq_expires, 0), HRTIMER_MODE_REL); @@ -1042,7 +1048,7 @@ static struct xfrm_state *__find_acq_core(struct net *net, unsigned short family return x; } -static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 seq); +static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 mark, u32 seq); int xfrm_state_add(struct xfrm_state *x) { @@ -1050,6 +1056,7 @@ int xfrm_state_add(struct xfrm_state *x) struct xfrm_state *x1, *to_put; int family; int err; + u32 mark = x->mark.v & x->mark.m; int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY); family = x->props.family; @@ -1067,7 +1074,7 @@ int xfrm_state_add(struct xfrm_state *x) } if (use_spi && x->km.seq) { - x1 = __xfrm_find_acq_byseq(net, x->km.seq); + x1 = __xfrm_find_acq_byseq(net, mark, x->km.seq); if (x1 && ((x1->id.proto != x->id.proto) || xfrm_addr_cmp(&x1->id.daddr, &x->id.daddr, family))) { to_put = x1; @@ -1076,8 +1083,8 @@ int xfrm_state_add(struct xfrm_state *x) } if (use_spi && !x1) - x1 = __find_acq_core(net, family, x->props.mode, x->props.reqid, - x->id.proto, + x1 = __find_acq_core(net, &x->mark, family, x->props.mode, + x->props.reqid, x->id.proto, &x->id.daddr, &x->props.saddr, 0); __xfrm_state_bump_genids(x); @@ -1151,6 +1158,8 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig, int *errp) goto error; } + memcpy(&x->mark, &orig->mark, sizeof(x->mark)); + err = xfrm_init_state(x); if (err) goto error; @@ -1342,41 +1351,41 @@ int xfrm_state_check_expire(struct xfrm_state *x) EXPORT_SYMBOL(xfrm_state_check_expire); struct xfrm_state * -xfrm_state_lookup(struct net *net, xfrm_address_t *daddr, __be32 spi, u8 proto, - unsigned short family) +xfrm_state_lookup(struct net *net, u32 mark, xfrm_address_t *daddr, __be32 spi, + u8 proto, unsigned short family) { struct xfrm_state *x; spin_lock_bh(&xfrm_state_lock); - x = __xfrm_state_lookup(net, daddr, spi, proto, family); + x = __xfrm_state_lookup(net, mark, daddr, spi, proto, family); spin_unlock_bh(&xfrm_state_lock); return x; } EXPORT_SYMBOL(xfrm_state_lookup); struct xfrm_state * -xfrm_state_lookup_byaddr(struct net *net, +xfrm_state_lookup_byaddr(struct net *net, u32 mark, xfrm_address_t *daddr, xfrm_address_t *saddr, u8 proto, unsigned short family) { struct xfrm_state *x; spin_lock_bh(&xfrm_state_lock); - x = __xfrm_state_lookup_byaddr(net, daddr, saddr, proto, family); + x = __xfrm_state_lookup_byaddr(net, mark, daddr, saddr, proto, family); spin_unlock_bh(&xfrm_state_lock); return x; } EXPORT_SYMBOL(xfrm_state_lookup_byaddr); struct xfrm_state * -xfrm_find_acq(struct net *net, u8 mode, u32 reqid, u8 proto, +xfrm_find_acq(struct net *net, struct xfrm_mark *mark, u8 mode, u32 reqid, u8 proto, xfrm_address_t *daddr, xfrm_address_t *saddr, int create, unsigned short family) { struct xfrm_state *x; spin_lock_bh(&xfrm_state_lock); - x = __find_acq_core(net, family, mode, reqid, proto, daddr, saddr, create); + x = __find_acq_core(net, mark, family, mode, reqid, proto, daddr, saddr, create); spin_unlock_bh(&xfrm_state_lock); return x; @@ -1423,7 +1432,7 @@ EXPORT_SYMBOL(xfrm_state_sort); /* Silly enough, but I'm lazy to build resolution list */ -static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 seq) +static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 mark, u32 seq) { int i; @@ -1442,12 +1451,12 @@ static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 seq) return NULL; } -struct xfrm_state *xfrm_find_acq_byseq(struct net *net, u32 seq) +struct xfrm_state *xfrm_find_acq_byseq(struct net *net, u32 mark, u32 seq) { struct xfrm_state *x; spin_lock_bh(&xfrm_state_lock); - x = __xfrm_find_acq_byseq(net, seq); + x = __xfrm_find_acq_byseq(net, mark, seq); spin_unlock_bh(&xfrm_state_lock); return x; } @@ -1474,6 +1483,7 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high) int err = -ENOENT; __be32 minspi = htonl(low); __be32 maxspi = htonl(high); + u32 mark = x->mark.v & x->mark.m; spin_lock_bh(&x->lock); if (x->km.state == XFRM_STATE_DEAD) @@ -1486,7 +1496,7 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high) err = -ENOENT; if (minspi == maxspi) { - x0 = xfrm_state_lookup(net, &x->id.daddr, minspi, x->id.proto, x->props.family); + x0 = xfrm_state_lookup(net, mark, &x->id.daddr, minspi, x->id.proto, x->props.family); if (x0) { xfrm_state_put(x0); goto unlock; @@ -1496,7 +1506,7 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high) u32 spi = 0; for (h=0; hid.daddr, htonl(spi), x->id.proto, x->props.family); + x0 = xfrm_state_lookup(net, mark, &x->id.daddr, htonl(spi), x->id.proto, x->props.family); if (x0 == NULL) { x->id.spi = htonl(spi); break; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index ee04e6bf0e5..331ae731080 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -31,6 +31,9 @@ #include #endif +#define DUMMY_MARK 0 +static struct xfrm_mark dummy_mark = {0, 0}; + static inline int aead_len(struct xfrm_algo_aead *alg) { return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); @@ -530,7 +533,7 @@ static struct xfrm_state *xfrm_user_state_lookup(struct net *net, if (xfrm_id_proto_match(p->proto, IPSEC_PROTO_ANY)) { err = -ESRCH; - x = xfrm_state_lookup(net, &p->daddr, p->spi, p->proto, p->family); + x = xfrm_state_lookup(net, DUMMY_MARK, &p->daddr, p->spi, p->proto, p->family); } else { xfrm_address_t *saddr = NULL; @@ -541,7 +544,7 @@ static struct xfrm_state *xfrm_user_state_lookup(struct net *net, } err = -ESRCH; - x = xfrm_state_lookup_byaddr(net, &p->daddr, saddr, + x = xfrm_state_lookup_byaddr(net, DUMMY_MARK, &p->daddr, saddr, p->proto, p->family); } @@ -958,7 +961,7 @@ static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, x = NULL; if (p->info.seq) { - x = xfrm_find_acq_byseq(net, p->info.seq); + x = xfrm_find_acq_byseq(net, DUMMY_MARK, p->info.seq); if (x && xfrm_addr_cmp(&x->id.daddr, daddr, family)) { xfrm_state_put(x); x = NULL; @@ -966,7 +969,7 @@ static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, } if (!x) - x = xfrm_find_acq(net, p->info.mode, p->info.reqid, + x = xfrm_find_acq(net, &dummy_mark, p->info.mode, p->info.reqid, p->info.id.proto, daddr, &p->info.saddr, 1, family); @@ -1598,7 +1601,7 @@ static int xfrm_get_ae(struct sk_buff *skb, struct nlmsghdr *nlh, if (r_skb == NULL) return -ENOMEM; - x = xfrm_state_lookup(net, &id->daddr, id->spi, id->proto, id->family); + x = xfrm_state_lookup(net, DUMMY_MARK, &id->daddr, id->spi, id->proto, id->family); if (x == NULL) { kfree_skb(r_skb); return -ESRCH; @@ -1640,7 +1643,7 @@ static int xfrm_new_ae(struct sk_buff *skb, struct nlmsghdr *nlh, if (!(nlh->nlmsg_flags&NLM_F_REPLACE)) return err; - x = xfrm_state_lookup(net, &p->sa_id.daddr, p->sa_id.spi, p->sa_id.proto, p->sa_id.family); + x = xfrm_state_lookup(net, DUMMY_MARK, &p->sa_id.daddr, p->sa_id.spi, p->sa_id.proto, p->sa_id.family); if (x == NULL) return -ESRCH; @@ -1767,7 +1770,7 @@ static int xfrm_add_sa_expire(struct sk_buff *skb, struct nlmsghdr *nlh, struct xfrm_user_expire *ue = nlmsg_data(nlh); struct xfrm_usersa_info *p = &ue->state; - x = xfrm_state_lookup(net, &p->id.daddr, p->id.spi, p->id.proto, p->family); + x = xfrm_state_lookup(net, DUMMY_MARK, &p->id.daddr, p->id.spi, p->id.proto, p->family); err = -ENOENT; if (x == NULL) -- cgit v1.2.3-70-g09d2 From 8ca2e93b557f2a0b35f7769038abf600177e1122 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 22 Feb 2010 11:32:57 +0000 Subject: xfrm: SP lookups signature with mark pass mark to all SP lookups to prepare them for when we add code to have them search. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/xfrm.h | 5 +++-- net/key/af_key.c | 4 ++-- net/xfrm/xfrm_policy.c | 8 ++++---- net/xfrm/xfrm_user.c | 10 +++++----- 4 files changed, 14 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 693523c870b..a7df3275b86 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1450,11 +1450,12 @@ extern int xfrm_policy_walk(struct net *net, struct xfrm_policy_walk *walk, int (*func)(struct xfrm_policy *, int, int, void*), void *); extern void xfrm_policy_walk_done(struct xfrm_policy_walk *walk); int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl); -struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir, +struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, + u8 type, int dir, struct xfrm_selector *sel, struct xfrm_sec_ctx *ctx, int delete, int *err); -struct xfrm_policy *xfrm_policy_byid(struct net *net, u8, int dir, u32 id, int delete, int *err); +struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u8, int dir, u32 id, int delete, int *err); int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info); u32 xfrm_get_acqseq(void); extern int xfrm_alloc_spi(struct xfrm_state *x, u32 minspi, u32 maxspi); diff --git a/net/key/af_key.c b/net/key/af_key.c index aae3cd86ccd..36870788264 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -2326,7 +2326,7 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg return err; } - xp = xfrm_policy_bysel_ctx(net, XFRM_POLICY_TYPE_MAIN, + xp = xfrm_policy_bysel_ctx(net, DUMMY_MARK, XFRM_POLICY_TYPE_MAIN, pol->sadb_x_policy_dir - 1, &sel, pol_ctx, 1, &err); security_xfrm_policy_free(pol_ctx); @@ -2574,7 +2574,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h return -EINVAL; delete = (hdr->sadb_msg_type == SADB_X_SPDDELETE2); - xp = xfrm_policy_byid(net, XFRM_POLICY_TYPE_MAIN, + xp = xfrm_policy_byid(net, DUMMY_MARK, XFRM_POLICY_TYPE_MAIN, dir, pol->sadb_x_policy_id, delete, &err); if (xp == NULL) return -ENOENT; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index d6eb16d7524..e67d3ca6e65 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -635,8 +635,8 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl) } EXPORT_SYMBOL(xfrm_policy_insert); -struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir, - struct xfrm_selector *sel, +struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u8 type, + int dir, struct xfrm_selector *sel, struct xfrm_sec_ctx *ctx, int delete, int *err) { @@ -676,8 +676,8 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir, } EXPORT_SYMBOL(xfrm_policy_bysel_ctx); -struct xfrm_policy *xfrm_policy_byid(struct net *net, u8 type, int dir, u32 id, - int delete, int *err) +struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u8 type, + int dir, u32 id, int delete, int *err) { struct xfrm_policy *pol, *ret; struct hlist_head *chain; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 331ae731080..02a67b4a64d 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1457,7 +1457,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, return err; if (p->index) - xp = xfrm_policy_byid(net, type, p->dir, p->index, delete, &err); + xp = xfrm_policy_byid(net, DUMMY_MARK, type, p->dir, p->index, delete, &err); else { struct nlattr *rt = attrs[XFRMA_SEC_CTX]; struct xfrm_sec_ctx *ctx; @@ -1474,8 +1474,8 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, if (err) return err; } - xp = xfrm_policy_bysel_ctx(net, type, p->dir, &p->sel, ctx, - delete, &err); + xp = xfrm_policy_bysel_ctx(net, DUMMY_MARK, type, p->dir, + &p->sel, ctx, delete, &err); security_xfrm_policy_free(ctx); } if (xp == NULL) @@ -1712,7 +1712,7 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, return err; if (p->index) - xp = xfrm_policy_byid(net, type, p->dir, p->index, 0, &err); + xp = xfrm_policy_byid(net, DUMMY_MARK, type, p->dir, p->index, 0, &err); else { struct nlattr *rt = attrs[XFRMA_SEC_CTX]; struct xfrm_sec_ctx *ctx; @@ -1729,7 +1729,7 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, if (err) return err; } - xp = xfrm_policy_bysel_ctx(net, type, p->dir, &p->sel, ctx, 0, &err); + xp = xfrm_policy_bysel_ctx(net, DUMMY_MARK, type, p->dir, &p->sel, ctx, 0, &err); security_xfrm_policy_free(ctx); } if (xp == NULL) -- cgit v1.2.3-70-g09d2 From 72b2b1dd77e8feb0b7c0b26dee58f2a1e2c9828c Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 24 Feb 2010 18:32:59 +0100 Subject: netfilter: xtables: replace XT_ENTRY_ITERATE macro The macro is replaced by a list.h-like foreach loop. This makes the code much more inspectable. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 9 ++ include/linux/netfilter_arp/arp_tables.h | 10 +- include/linux/netfilter_ipv4/ip_tables.h | 11 +- include/linux/netfilter_ipv6/ip6_tables.h | 10 +- net/ipv4/netfilter/arp_tables.c | 151 ++++++++++++++++++---------- net/ipv4/netfilter/ip_tables.c | 160 +++++++++++++++++++----------- net/ipv6/netfilter/ip6_tables.c | 160 +++++++++++++++++++----------- 7 files changed, 321 insertions(+), 190 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index a18119fb88f..9df3f5a8f9f 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -139,6 +139,7 @@ struct xt_counters_info { __ret; \ }) +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define XT_ENTRY_ITERATE_CONTINUE(type, entries, size, n, fn, args...) \ ({ \ @@ -163,6 +164,14 @@ struct xt_counters_info { #define XT_ENTRY_ITERATE(type, entries, size, fn, args...) \ XT_ENTRY_ITERATE_CONTINUE(type, entries, size, 0, fn, args) +#endif /* !__KERNEL__ */ + +/* pos is normally a struct ipt_entry/ip6t_entry/etc. */ +#define xt_entry_foreach(pos, ehead, esize) \ + for ((pos) = (typeof(pos))(ehead); \ + (pos) < (typeof(pos))((char *)(ehead) + (esize)); \ + (pos) = (typeof(pos))((char *)(pos) + (pos)->next_offset)) + #ifdef __KERNEL__ #include diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index 0b33980611b..e9948c0560f 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -211,9 +211,11 @@ static __inline__ struct arpt_entry_target *arpt_get_target(struct arpt_entry *e return (void *)e + e->target_offset; } +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define ARPT_ENTRY_ITERATE(entries, size, fn, args...) \ XT_ENTRY_ITERATE(struct arpt_entry, entries, size, fn, ## args) +#endif /* * Main firewall chains definitions and global var's definitions. @@ -291,14 +293,6 @@ compat_arpt_get_target(struct compat_arpt_entry *e) #define COMPAT_ARPT_ALIGN(s) COMPAT_XT_ALIGN(s) -/* fn returns 0 to continue iteration */ -#define COMPAT_ARPT_ENTRY_ITERATE(entries, size, fn, args...) \ - XT_ENTRY_ITERATE(struct compat_arpt_entry, entries, size, fn, ## args) - -#define COMPAT_ARPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \ - XT_ENTRY_ITERATE_CONTINUE(struct compat_arpt_entry, entries, size, n, \ - fn, ## args) - #endif /* CONFIG_COMPAT */ #endif /*__KERNEL__*/ #endif /* _ARPTABLES_H */ diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 364973b4213..5b20ae724b4 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -227,9 +227,11 @@ ipt_get_target(struct ipt_entry *e) #define IPT_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct ipt_entry, e, fn, ## args) +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IPT_ENTRY_ITERATE(entries, size, fn, args...) \ XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args) +#endif /* * Main firewall chains definitions and global var's definitions. @@ -317,15 +319,6 @@ compat_ipt_get_target(struct compat_ipt_entry *e) #define COMPAT_IPT_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct compat_ipt_entry, e, fn, ## args) -/* fn returns 0 to continue iteration */ -#define COMPAT_IPT_ENTRY_ITERATE(entries, size, fn, args...) \ - XT_ENTRY_ITERATE(struct compat_ipt_entry, entries, size, fn, ## args) - -/* fn returns 0 to continue iteration */ -#define COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \ - XT_ENTRY_ITERATE_CONTINUE(struct compat_ipt_entry, entries, size, n, \ - fn, ## args) - #endif /* CONFIG_COMPAT */ #endif /*__KERNEL__*/ #endif /* _IPTABLES_H */ diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index 8031eb486a1..8bb3f5ba5ff 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -284,9 +284,11 @@ ip6t_get_target(struct ip6t_entry *e) #define IP6T_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct ip6t_entry, e, fn, ## args) +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IP6T_ENTRY_ITERATE(entries, size, fn, args...) \ XT_ENTRY_ITERATE(struct ip6t_entry, entries, size, fn, ## args) +#endif /* * Main firewall chains definitions and global var's definitions. @@ -345,14 +347,6 @@ compat_ip6t_get_target(struct compat_ip6t_entry *e) #define COMPAT_IP6T_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct compat_ip6t_entry, e, fn, ## args) -/* fn returns 0 to continue iteration */ -#define COMPAT_IP6T_ENTRY_ITERATE(entries, size, fn, args...) \ - XT_ENTRY_ITERATE(struct compat_ip6t_entry, entries, size, fn, ## args) - -#define COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entries, size, n, fn, args...) \ - XT_ENTRY_ITERATE_CONTINUE(struct compat_ip6t_entry, entries, size, n, \ - fn, ## args) - #endif /* CONFIG_COMPAT */ #endif /*__KERNEL__*/ #endif /* _IP6_TABLES_H */ diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 4db5c1ece0f..f7338869fc4 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -641,8 +641,9 @@ static int translate_table(const char *name, const unsigned int *hook_entries, const unsigned int *underflows) { + struct arpt_entry *iter; unsigned int i; - int ret; + int ret = 0; newinfo->size = size; newinfo->number = number; @@ -657,12 +658,13 @@ static int translate_table(const char *name, i = 0; /* Walk through entries, checking offsets. */ - ret = ARPT_ENTRY_ITERATE(entry0, newinfo->size, - check_entry_size_and_hooks, - newinfo, - entry0, - entry0 + size, - hook_entries, underflows, valid_hooks, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = check_entry_size_and_hooks(iter, newinfo, entry0, + entry0 + size, hook_entries, underflows, + valid_hooks, &i); + if (ret != 0) + break; + } duprintf("translate_table: ARPT_ENTRY_ITERATE gives %d\n", ret); if (ret != 0) return ret; @@ -697,12 +699,16 @@ static int translate_table(const char *name, /* Finally, each sanity check must pass */ i = 0; - ret = ARPT_ENTRY_ITERATE(entry0, newinfo->size, - find_check_entry, name, size, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = find_check_entry(iter, name, size, &i); + if (ret != 0) + break; + } if (ret != 0) { - ARPT_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); + xt_entry_foreach(iter, entry0, newinfo->size) + if (cleanup_entry(iter, &i) != 0) + break; return ret; } @@ -739,6 +745,7 @@ static inline int set_entry_to_counter(const struct arpt_entry *e, static void get_counters(const struct xt_table_info *t, struct xt_counters counters[]) { + struct arpt_entry *iter; unsigned int cpu; unsigned int i; unsigned int curcpu; @@ -754,22 +761,18 @@ static void get_counters(const struct xt_table_info *t, curcpu = smp_processor_id(); i = 0; - ARPT_ENTRY_ITERATE(t->entries[curcpu], - t->size, - set_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[curcpu], t->size) + if (set_entry_to_counter(iter, counters, &i) != 0) + break; for_each_possible_cpu(cpu) { if (cpu == curcpu) continue; i = 0; xt_info_wrlock(cpu); - ARPT_ENTRY_ITERATE(t->entries[cpu], - t->size, - add_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[cpu], t->size) + if (add_entry_to_counter(iter, counters, &i) != 0) + break; xt_info_wrunlock(cpu); } local_bh_enable(); @@ -899,7 +902,9 @@ static int compat_calc_entry(const struct arpt_entry *e, static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { + struct arpt_entry *iter; void *loc_cpu_entry; + int ret = 0; if (!newinfo || !info) return -EINVAL; @@ -908,9 +913,12 @@ static int compat_table_info(const struct xt_table_info *info, memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; loc_cpu_entry = info->entries[raw_smp_processor_id()]; - return ARPT_ENTRY_ITERATE(loc_cpu_entry, info->size, - compat_calc_entry, info, loc_cpu_entry, - newinfo); + xt_entry_foreach(iter, loc_cpu_entry, info->size) { + ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); + if (ret != 0) + break; + } + return ret; } #endif @@ -1025,6 +1033,7 @@ static int __do_replace(struct net *net, const char *name, struct xt_table_info *oldinfo; struct xt_counters *counters; void *loc_cpu_old_entry; + struct arpt_entry *iter; ret = 0; counters = vmalloc_node(num_counters * sizeof(struct xt_counters), @@ -1068,8 +1077,9 @@ static int __do_replace(struct net *net, const char *name, /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; - ARPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry, - NULL); + xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) + if (cleanup_entry(iter, NULL) != 0) + break; xt_free_table_info(oldinfo); if (copy_to_user(counters_ptr, counters, @@ -1095,6 +1105,7 @@ static int do_replace(struct net *net, const void __user *user, struct arpt_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct arpt_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1130,7 +1141,9 @@ static int do_replace(struct net *net, const void __user *user, return 0; free_newinfo_untrans: - ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1163,6 +1176,7 @@ static int do_add_counters(struct net *net, const void __user *user, const struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; + struct arpt_entry *iter; #ifdef CONFIG_COMPAT struct compat_xt_counters_info compat_tmp; @@ -1220,11 +1234,9 @@ static int do_add_counters(struct net *net, const void __user *user, curcpu = smp_processor_id(); loc_cpu_entry = private->entries[curcpu]; xt_info_wrlock(curcpu); - ARPT_ENTRY_ITERATE(loc_cpu_entry, - private->size, - add_counter_to_entry, - paddc, - &i); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (add_counter_to_entry(iter, paddc, &i) != 0) + break; xt_info_wrunlock(curcpu); unlock_up_free: local_bh_enable(); @@ -1388,8 +1400,10 @@ static int translate_compat_table(const char *name, unsigned int i, j; struct xt_table_info *newinfo, *info; void *pos, *entry0, *entry1; + struct compat_arpt_entry *iter0; + struct arpt_entry *iter1; unsigned int size; - int ret; + int ret = 0; info = *pinfo; entry0 = *pentry0; @@ -1406,11 +1420,13 @@ static int translate_compat_table(const char *name, j = 0; xt_compat_lock(NFPROTO_ARP); /* Walk through entries, checking offsets. */ - ret = COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size, - check_compat_entry_size_and_hooks, - info, &size, entry0, - entry0 + total_size, - hook_entries, underflows, &j, name); + xt_entry_foreach(iter0, entry0, total_size) { + ret = check_compat_entry_size_and_hooks(iter0, info, &size, + entry0, entry0 + total_size, hook_entries, underflows, + &j, name); + if (ret != 0) + break; + } if (ret != 0) goto out_unlock; @@ -1451,9 +1467,12 @@ static int translate_compat_table(const char *name, entry1 = newinfo->entries[raw_smp_processor_id()]; pos = entry1; size = total_size; - ret = COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size, - compat_copy_entry_from_user, - &pos, &size, name, newinfo, entry1); + xt_entry_foreach(iter0, entry0, total_size) { + ret = compat_copy_entry_from_user(iter0, &pos, + &size, name, newinfo, entry1); + if (ret != 0) + break; + } xt_compat_flush_offsets(NFPROTO_ARP); xt_compat_unlock(NFPROTO_ARP); if (ret) @@ -1464,13 +1483,28 @@ static int translate_compat_table(const char *name, goto free_newinfo; i = 0; - ret = ARPT_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry, - name, &i); + xt_entry_foreach(iter1, entry1, newinfo->size) { + ret = compat_check_entry(iter1, name, &i); + if (ret != 0) + break; + } if (ret) { + /* + * The first i matches need cleanup_entry (calls ->destroy) + * because they had called ->check already. The other j-i + * entries need only release. + */ + int skip = i; j -= i; - COMPAT_ARPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, - compat_release_entry, &j); - ARPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, &i); + xt_entry_foreach(iter0, entry0, newinfo->size) { + if (skip-- > 0) + continue; + if (compat_release_entry(iter0, &j) != 0) + break; + } + xt_entry_foreach(iter1, entry1, newinfo->size) + if (cleanup_entry(iter1, &i) != 0) + break; xt_free_table_info(newinfo); return ret; } @@ -1488,7 +1522,9 @@ static int translate_compat_table(const char *name, free_newinfo: xt_free_table_info(newinfo); out: - COMPAT_ARPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j); + xt_entry_foreach(iter0, entry0, total_size) + if (compat_release_entry(iter0, &j) != 0) + break; return ret; out_unlock: xt_compat_flush_offsets(NFPROTO_ARP); @@ -1515,6 +1551,7 @@ static int compat_do_replace(struct net *net, void __user *user, struct compat_arpt_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct arpt_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1552,7 +1589,9 @@ static int compat_do_replace(struct net *net, void __user *user, return 0; free_newinfo_untrans: - ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1636,6 +1675,7 @@ static int compat_copy_entries_to_user(unsigned int total_size, int ret = 0; void *loc_cpu_entry; unsigned int i = 0; + struct arpt_entry *iter; counters = alloc_counters(table); if (IS_ERR(counters)) @@ -1645,9 +1685,12 @@ static int compat_copy_entries_to_user(unsigned int total_size, loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - ret = ARPT_ENTRY_ITERATE(loc_cpu_entry, total_size, - compat_copy_entry_to_user, - &pos, &size, counters, &i); + xt_entry_foreach(iter, loc_cpu_entry, total_size) { + ret = compat_copy_entry_to_user(iter, &pos, + &size, counters, &i); + if (ret != 0) + break; + } vfree(counters); return ret; } @@ -1843,13 +1886,15 @@ void arpt_unregister_table(struct xt_table *table) struct xt_table_info *private; void *loc_cpu_entry; struct module *table_owner = table->me; + struct arpt_entry *iter; private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; - ARPT_ENTRY_ITERATE(loc_cpu_entry, private->size, - cleanup_entry, NULL); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (cleanup_entry(iter, NULL) != 0) + break; if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index e94c18bdfc6..b43280aad8a 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -288,6 +288,7 @@ static void trace_packet(const struct sk_buff *skb, const void *table_base; const struct ipt_entry *root; const char *hookname, *chainname, *comment; + const struct ipt_entry *iter; unsigned int rulenum = 0; table_base = private->entries[smp_processor_id()]; @@ -296,10 +297,10 @@ static void trace_packet(const struct sk_buff *skb, hookname = chainname = hooknames[hook]; comment = comments[NF_IP_TRACE_COMMENT_RULE]; - IPT_ENTRY_ITERATE(root, - private->size - private->hook_entry[hook], - get_chainname_rulenum, - e, hookname, &chainname, &comment, &rulenum); + xt_entry_foreach(iter, root, private->size - private->hook_entry[hook]) + if (get_chainname_rulenum(iter, e, hookname, + &chainname, &comment, &rulenum) != 0) + break; nf_log_packet(AF_INET, hook, skb, in, out, &trace_loginfo, "TRACE: %s:%s:%s:%u ", @@ -826,8 +827,9 @@ translate_table(struct net *net, const unsigned int *hook_entries, const unsigned int *underflows) { + struct ipt_entry *iter; unsigned int i; - int ret; + int ret = 0; newinfo->size = size; newinfo->number = number; @@ -841,12 +843,13 @@ translate_table(struct net *net, duprintf("translate_table: size %u\n", newinfo->size); i = 0; /* Walk through entries, checking offsets. */ - ret = IPT_ENTRY_ITERATE(entry0, newinfo->size, - check_entry_size_and_hooks, - newinfo, - entry0, - entry0 + size, - hook_entries, underflows, valid_hooks, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = check_entry_size_and_hooks(iter, newinfo, entry0, + entry0 + size, hook_entries, underflows, + valid_hooks, &i); + if (ret != 0) + break; + } if (ret != 0) return ret; @@ -878,12 +881,16 @@ translate_table(struct net *net, /* Finally, each sanity check must pass */ i = 0; - ret = IPT_ENTRY_ITERATE(entry0, newinfo->size, - find_check_entry, net, name, size, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = find_check_entry(iter, net, name, size, &i); + if (ret != 0) + break; + } if (ret != 0) { - IPT_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, net, &i); + xt_entry_foreach(iter, entry0, newinfo->size) + if (cleanup_entry(iter, net, &i) != 0) + break; return ret; } @@ -923,6 +930,7 @@ static void get_counters(const struct xt_table_info *t, struct xt_counters counters[]) { + struct ipt_entry *iter; unsigned int cpu; unsigned int i; unsigned int curcpu; @@ -938,22 +946,18 @@ get_counters(const struct xt_table_info *t, curcpu = smp_processor_id(); i = 0; - IPT_ENTRY_ITERATE(t->entries[curcpu], - t->size, - set_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[curcpu], t->size) + if (set_entry_to_counter(iter, counters, &i) != 0) + break; for_each_possible_cpu(cpu) { if (cpu == curcpu) continue; i = 0; xt_info_wrlock(cpu); - IPT_ENTRY_ITERATE(t->entries[cpu], - t->size, - add_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[cpu], t->size) + if (add_entry_to_counter(iter, counters, &i) != 0) + break; xt_info_wrunlock(cpu); } local_bh_enable(); @@ -1111,7 +1115,9 @@ static int compat_calc_entry(const struct ipt_entry *e, static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { + struct ipt_entry *iter; void *loc_cpu_entry; + int ret = 0; if (!newinfo || !info) return -EINVAL; @@ -1120,9 +1126,12 @@ static int compat_table_info(const struct xt_table_info *info, memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; loc_cpu_entry = info->entries[raw_smp_processor_id()]; - return IPT_ENTRY_ITERATE(loc_cpu_entry, info->size, - compat_calc_entry, info, loc_cpu_entry, - newinfo); + xt_entry_foreach(iter, loc_cpu_entry, info->size) { + ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); + if (ret != 0) + break; + } + return ret; } #endif @@ -1236,6 +1245,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, struct xt_table_info *oldinfo; struct xt_counters *counters; void *loc_cpu_old_entry; + struct ipt_entry *iter; ret = 0; counters = vmalloc(num_counters * sizeof(struct xt_counters)); @@ -1278,8 +1288,10 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; - IPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry, - net, NULL); + xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; + xt_free_table_info(oldinfo); if (copy_to_user(counters_ptr, counters, sizeof(struct xt_counters) * num_counters) != 0) @@ -1304,6 +1316,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) struct ipt_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct ipt_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1339,7 +1352,9 @@ do_replace(struct net *net, const void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1373,6 +1388,7 @@ do_add_counters(struct net *net, const void __user *user, const struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; + struct ipt_entry *iter; #ifdef CONFIG_COMPAT struct compat_xt_counters_info compat_tmp; @@ -1430,11 +1446,9 @@ do_add_counters(struct net *net, const void __user *user, curcpu = smp_processor_id(); loc_cpu_entry = private->entries[curcpu]; xt_info_wrlock(curcpu); - IPT_ENTRY_ITERATE(loc_cpu_entry, - private->size, - add_counter_to_entry, - paddc, - &i); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (add_counter_to_entry(iter, paddc, &i) != 0) + break; xt_info_wrunlock(curcpu); unlock_up_free: local_bh_enable(); @@ -1720,8 +1734,10 @@ translate_compat_table(struct net *net, unsigned int i, j; struct xt_table_info *newinfo, *info; void *pos, *entry0, *entry1; + struct compat_ipt_entry *iter0; + struct ipt_entry *iter1; unsigned int size; - int ret; + int ret = 0; info = *pinfo; entry0 = *pentry0; @@ -1738,11 +1754,13 @@ translate_compat_table(struct net *net, j = 0; xt_compat_lock(AF_INET); /* Walk through entries, checking offsets. */ - ret = COMPAT_IPT_ENTRY_ITERATE(entry0, total_size, - check_compat_entry_size_and_hooks, - info, &size, entry0, - entry0 + total_size, - hook_entries, underflows, &j, name); + xt_entry_foreach(iter0, entry0, total_size) { + ret = check_compat_entry_size_and_hooks(iter0, info, &size, + entry0, entry0 + total_size, hook_entries, underflows, + &j, name); + if (ret != 0) + break; + } if (ret != 0) goto out_unlock; @@ -1783,9 +1801,12 @@ translate_compat_table(struct net *net, entry1 = newinfo->entries[raw_smp_processor_id()]; pos = entry1; size = total_size; - ret = COMPAT_IPT_ENTRY_ITERATE(entry0, total_size, - compat_copy_entry_from_user, - &pos, &size, name, newinfo, entry1); + xt_entry_foreach(iter0, entry0, total_size) { + ret = compat_copy_entry_from_user(iter0, &pos, + &size, name, newinfo, entry1); + if (ret != 0) + break; + } xt_compat_flush_offsets(AF_INET); xt_compat_unlock(AF_INET); if (ret) @@ -1796,13 +1817,28 @@ translate_compat_table(struct net *net, goto free_newinfo; i = 0; - ret = IPT_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry, - net, name, &i); + xt_entry_foreach(iter1, entry1, newinfo->size) { + ret = compat_check_entry(iter1, net, name, &i); + if (ret != 0) + break; + } if (ret) { + /* + * The first i matches need cleanup_entry (calls ->destroy) + * because they had called ->check already. The other j-i + * entries need only release. + */ + int skip = i; j -= i; - COMPAT_IPT_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, - compat_release_entry, &j); - IPT_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, net, &i); + xt_entry_foreach(iter0, entry0, newinfo->size) { + if (skip-- > 0) + continue; + if (compat_release_entry(iter0, &i) != 0) + break; + } + xt_entry_foreach(iter1, entry1, newinfo->size) + if (cleanup_entry(iter1, net, &i) != 0) + break; xt_free_table_info(newinfo); return ret; } @@ -1820,7 +1856,9 @@ translate_compat_table(struct net *net, free_newinfo: xt_free_table_info(newinfo); out: - COMPAT_IPT_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j); + xt_entry_foreach(iter0, entry0, total_size) + if (compat_release_entry(iter0, &j) != 0) + break; return ret; out_unlock: xt_compat_flush_offsets(AF_INET); @@ -1835,6 +1873,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) struct compat_ipt_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct ipt_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1873,7 +1912,9 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1922,6 +1963,7 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, int ret = 0; const void *loc_cpu_entry; unsigned int i = 0; + struct ipt_entry *iter; counters = alloc_counters(table); if (IS_ERR(counters)) @@ -1934,9 +1976,12 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - ret = IPT_ENTRY_ITERATE(loc_cpu_entry, total_size, - compat_copy_entry_to_user, - &pos, &size, counters, &i); + xt_entry_foreach(iter, loc_cpu_entry, total_size) { + ret = compat_copy_entry_to_user(iter, &pos, + &size, counters, &i); + if (ret != 0) + break; + } vfree(counters); return ret; @@ -2137,12 +2182,15 @@ void ipt_unregister_table(struct net *net, struct xt_table *table) struct xt_table_info *private; void *loc_cpu_entry; struct module *table_owner = table->me; + struct ipt_entry *iter; private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; - IPT_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 4185099c294..23926e38d36 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -318,6 +318,7 @@ static void trace_packet(const struct sk_buff *skb, const void *table_base; const struct ip6t_entry *root; const char *hookname, *chainname, *comment; + const struct ip6t_entry *iter; unsigned int rulenum = 0; table_base = private->entries[smp_processor_id()]; @@ -326,10 +327,10 @@ static void trace_packet(const struct sk_buff *skb, hookname = chainname = hooknames[hook]; comment = comments[NF_IP6_TRACE_COMMENT_RULE]; - IP6T_ENTRY_ITERATE(root, - private->size - private->hook_entry[hook], - get_chainname_rulenum, - e, hookname, &chainname, &comment, &rulenum); + xt_entry_foreach(iter, root, private->size - private->hook_entry[hook]) + if (get_chainname_rulenum(iter, e, hookname, + &chainname, &comment, &rulenum) != 0) + break; nf_log_packet(AF_INET6, hook, skb, in, out, &trace_loginfo, "TRACE: %s:%s:%s:%u ", @@ -857,8 +858,9 @@ translate_table(struct net *net, const unsigned int *hook_entries, const unsigned int *underflows) { + struct ip6t_entry *iter; unsigned int i; - int ret; + int ret = 0; newinfo->size = size; newinfo->number = number; @@ -872,12 +874,13 @@ translate_table(struct net *net, duprintf("translate_table: size %u\n", newinfo->size); i = 0; /* Walk through entries, checking offsets. */ - ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size, - check_entry_size_and_hooks, - newinfo, - entry0, - entry0 + size, - hook_entries, underflows, valid_hooks, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = check_entry_size_and_hooks(iter, newinfo, entry0, + entry0 + size, hook_entries, underflows, + valid_hooks, &i); + if (ret != 0) + break; + } if (ret != 0) return ret; @@ -909,12 +912,16 @@ translate_table(struct net *net, /* Finally, each sanity check must pass */ i = 0; - ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size, - find_check_entry, net, name, size, &i); + xt_entry_foreach(iter, entry0, newinfo->size) { + ret = find_check_entry(iter, net, name, size, &i); + if (ret != 0) + break; + } if (ret != 0) { - IP6T_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, net, &i); + xt_entry_foreach(iter, entry0, newinfo->size) + if (cleanup_entry(iter, net, &i) != 0) + break; return ret; } @@ -954,6 +961,7 @@ static void get_counters(const struct xt_table_info *t, struct xt_counters counters[]) { + struct ip6t_entry *iter; unsigned int cpu; unsigned int i; unsigned int curcpu; @@ -969,22 +977,18 @@ get_counters(const struct xt_table_info *t, curcpu = smp_processor_id(); i = 0; - IP6T_ENTRY_ITERATE(t->entries[curcpu], - t->size, - set_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[curcpu], t->size) + if (set_entry_to_counter(iter, counters, &i) != 0) + break; for_each_possible_cpu(cpu) { if (cpu == curcpu) continue; i = 0; xt_info_wrlock(cpu); - IP6T_ENTRY_ITERATE(t->entries[cpu], - t->size, - add_entry_to_counter, - counters, - &i); + xt_entry_foreach(iter, t->entries[cpu], t->size) + if (add_entry_to_counter(iter, counters, &i) != 0) + break; xt_info_wrunlock(cpu); } local_bh_enable(); @@ -1142,7 +1146,9 @@ static int compat_calc_entry(const struct ip6t_entry *e, static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { + struct ip6t_entry *iter; void *loc_cpu_entry; + int ret = 0; if (!newinfo || !info) return -EINVAL; @@ -1151,9 +1157,12 @@ static int compat_table_info(const struct xt_table_info *info, memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; loc_cpu_entry = info->entries[raw_smp_processor_id()]; - return IP6T_ENTRY_ITERATE(loc_cpu_entry, info->size, - compat_calc_entry, info, loc_cpu_entry, - newinfo); + xt_entry_foreach(iter, loc_cpu_entry, info->size) { + ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); + if (ret != 0) + break; + } + return ret; } #endif @@ -1267,6 +1276,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, struct xt_table_info *oldinfo; struct xt_counters *counters; const void *loc_cpu_old_entry; + struct ip6t_entry *iter; ret = 0; counters = vmalloc_node(num_counters * sizeof(struct xt_counters), @@ -1310,8 +1320,10 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; - IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry, - net, NULL); + xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; + xt_free_table_info(oldinfo); if (copy_to_user(counters_ptr, counters, sizeof(struct xt_counters) * num_counters) != 0) @@ -1336,6 +1348,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) struct ip6t_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct ip6t_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1371,7 +1384,9 @@ do_replace(struct net *net, const void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1405,6 +1420,7 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, const struct xt_table_info *private; int ret = 0; const void *loc_cpu_entry; + struct ip6t_entry *iter; #ifdef CONFIG_COMPAT struct compat_xt_counters_info compat_tmp; @@ -1463,11 +1479,9 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, curcpu = smp_processor_id(); xt_info_wrlock(curcpu); loc_cpu_entry = private->entries[curcpu]; - IP6T_ENTRY_ITERATE(loc_cpu_entry, - private->size, - add_counter_to_entry, - paddc, - &i); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (add_counter_to_entry(iter, paddc, &i) != 0) + break; xt_info_wrunlock(curcpu); unlock_up_free: @@ -1753,8 +1767,10 @@ translate_compat_table(struct net *net, unsigned int i, j; struct xt_table_info *newinfo, *info; void *pos, *entry0, *entry1; + struct compat_ip6t_entry *iter0; + struct ip6t_entry *iter1; unsigned int size; - int ret; + int ret = 0; info = *pinfo; entry0 = *pentry0; @@ -1771,11 +1787,13 @@ translate_compat_table(struct net *net, j = 0; xt_compat_lock(AF_INET6); /* Walk through entries, checking offsets. */ - ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size, - check_compat_entry_size_and_hooks, - info, &size, entry0, - entry0 + total_size, - hook_entries, underflows, &j, name); + xt_entry_foreach(iter0, entry0, total_size) { + ret = check_compat_entry_size_and_hooks(iter0, info, &size, + entry0, entry0 + total_size, hook_entries, underflows, + &j, name); + if (ret != 0) + break; + } if (ret != 0) goto out_unlock; @@ -1816,9 +1834,12 @@ translate_compat_table(struct net *net, entry1 = newinfo->entries[raw_smp_processor_id()]; pos = entry1; size = total_size; - ret = COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size, - compat_copy_entry_from_user, - &pos, &size, name, newinfo, entry1); + xt_entry_foreach(iter0, entry0, total_size) { + ret = compat_copy_entry_from_user(iter0, &pos, + &size, name, newinfo, entry1); + if (ret != 0) + break; + } xt_compat_flush_offsets(AF_INET6); xt_compat_unlock(AF_INET6); if (ret) @@ -1829,13 +1850,28 @@ translate_compat_table(struct net *net, goto free_newinfo; i = 0; - ret = IP6T_ENTRY_ITERATE(entry1, newinfo->size, compat_check_entry, - net, name, &i); + xt_entry_foreach(iter1, entry1, newinfo->size) { + ret = compat_check_entry(iter1, net, name, &i); + if (ret != 0) + break; + } if (ret) { + /* + * The first i matches need cleanup_entry (calls ->destroy) + * because they had called ->check already. The other j-i + * entries need only release. + */ + int skip = i; j -= i; - COMPAT_IP6T_ENTRY_ITERATE_CONTINUE(entry0, newinfo->size, i, - compat_release_entry, &j); - IP6T_ENTRY_ITERATE(entry1, newinfo->size, cleanup_entry, net, &i); + xt_entry_foreach(iter0, entry0, newinfo->size) { + if (skip-- > 0) + continue; + if (compat_release_entry(iter0, &j) != 0) + break; + } + xt_entry_foreach(iter1, entry1, newinfo->size) + if (cleanup_entry(iter1, net, &i) != 0) + break; xt_free_table_info(newinfo); return ret; } @@ -1853,7 +1889,9 @@ translate_compat_table(struct net *net, free_newinfo: xt_free_table_info(newinfo); out: - COMPAT_IP6T_ENTRY_ITERATE(entry0, total_size, compat_release_entry, &j); + xt_entry_foreach(iter0, entry0, total_size) + if (compat_release_entry(iter0, &j) != 0) + break; return ret; out_unlock: xt_compat_flush_offsets(AF_INET6); @@ -1868,6 +1906,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) struct compat_ip6t_replace tmp; struct xt_table_info *newinfo; void *loc_cpu_entry; + struct ip6t_entry *iter; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1906,7 +1945,9 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return 0; free_newinfo_untrans: - IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, newinfo->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; free_newinfo: xt_free_table_info(newinfo); return ret; @@ -1955,6 +1996,7 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, int ret = 0; const void *loc_cpu_entry; unsigned int i = 0; + struct ip6t_entry *iter; counters = alloc_counters(table); if (IS_ERR(counters)) @@ -1967,9 +2009,12 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - ret = IP6T_ENTRY_ITERATE(loc_cpu_entry, total_size, - compat_copy_entry_to_user, - &pos, &size, counters, &i); + xt_entry_foreach(iter, loc_cpu_entry, total_size) { + ret = compat_copy_entry_to_user(iter, &pos, + &size, counters, &i); + if (ret != 0) + break; + } vfree(counters); return ret; @@ -2169,12 +2214,15 @@ void ip6t_unregister_table(struct net *net, struct xt_table *table) struct xt_table_info *private; void *loc_cpu_entry; struct module *table_owner = table->me; + struct ip6t_entry *iter; private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; - IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, net, NULL); + xt_entry_foreach(iter, loc_cpu_entry, private->size) + if (cleanup_entry(iter, net, NULL) != 0) + break; if (private->number > private->initial_entries) module_put(table_owner); xt_free_table_info(private); -- cgit v1.2.3-70-g09d2 From dcea992aca82cb08b4674c4c783e325835408d1e Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 24 Feb 2010 18:34:48 +0100 Subject: netfilter: xtables: replace XT_MATCH_ITERATE macro The macro is replaced by a list.h-like foreach loop. This makes the code more inspectable. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 10 +++- include/linux/netfilter_ipv4/ip_tables.h | 6 +-- include/linux/netfilter_ipv6/ip6_tables.h | 6 +-- net/ipv4/netfilter/ip_tables.c | 78 ++++++++++++++++++++++++------- net/ipv6/netfilter/ip6_tables.c | 78 ++++++++++++++++++++++++------- net/netfilter/xt_TCPMSS.c | 12 +++-- 6 files changed, 141 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 9df3f5a8f9f..84c7c928e9e 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -120,6 +120,7 @@ struct xt_counters_info { #define XT_INV_PROTO 0x40 /* Invert the sense of PROTO. */ +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define XT_MATCH_ITERATE(type, e, fn, args...) \ ({ \ @@ -139,7 +140,6 @@ struct xt_counters_info { __ret; \ }) -#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define XT_ENTRY_ITERATE_CONTINUE(type, entries, size, n, fn, args...) \ ({ \ @@ -172,6 +172,14 @@ struct xt_counters_info { (pos) < (typeof(pos))((char *)(ehead) + (esize)); \ (pos) = (typeof(pos))((char *)(pos) + (pos)->next_offset)) +/* can only be xt_entry_match, so no use of typeof here */ +#define xt_ematch_foreach(pos, entry) \ + for ((pos) = (struct xt_entry_match *)entry->elems; \ + (pos) < (struct xt_entry_match *)((char *)(entry) + \ + (entry)->target_offset); \ + (pos) = (struct xt_entry_match *)((char *)(pos) + \ + (pos)->u.match_size)) + #ifdef __KERNEL__ #include diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 5b20ae724b4..704a7b6e816 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -223,11 +223,11 @@ ipt_get_target(struct ipt_entry *e) return (void *)e + e->target_offset; } +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IPT_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct ipt_entry, e, fn, ## args) -#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IPT_ENTRY_ITERATE(entries, size, fn, args...) \ XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args) @@ -315,10 +315,6 @@ compat_ipt_get_target(struct compat_ipt_entry *e) #define COMPAT_IPT_ALIGN(s) COMPAT_XT_ALIGN(s) -/* fn returns 0 to continue iteration */ -#define COMPAT_IPT_MATCH_ITERATE(e, fn, args...) \ - XT_MATCH_ITERATE(struct compat_ipt_entry, e, fn, ## args) - #endif /* CONFIG_COMPAT */ #endif /*__KERNEL__*/ #endif /* _IPTABLES_H */ diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index 8bb3f5ba5ff..e5ba03d783c 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -280,11 +280,11 @@ ip6t_get_target(struct ip6t_entry *e) return (void *)e + e->target_offset; } +#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IP6T_MATCH_ITERATE(e, fn, args...) \ XT_MATCH_ITERATE(struct ip6t_entry, e, fn, ## args) -#ifndef __KERNEL__ /* fn returns 0 to continue iteration */ #define IP6T_ENTRY_ITERATE(entries, size, fn, args...) \ XT_ENTRY_ITERATE(struct ip6t_entry, entries, size, fn, ## args) @@ -343,10 +343,6 @@ compat_ip6t_get_target(struct compat_ip6t_entry *e) #define COMPAT_IP6T_ALIGN(s) COMPAT_XT_ALIGN(s) -/* fn returns 0 to continue iteration */ -#define COMPAT_IP6T_MATCH_ITERATE(e, fn, args...) \ - XT_MATCH_ITERATE(struct compat_ip6t_entry, e, fn, ## args) - #endif /* CONFIG_COMPAT */ #endif /*__KERNEL__*/ #endif /* _IP6_TABLES_H */ diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 9c8aa394c51..3a7fc732b91 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -366,16 +366,21 @@ ipt_do_table(struct sk_buff *skb, do { const struct ipt_entry_target *t; + const struct xt_entry_match *ematch; IP_NF_ASSERT(e); IP_NF_ASSERT(back); if (!ip_packet_match(ip, indev, outdev, - &e->ip, mtpar.fragoff) || - IPT_MATCH_ITERATE(e, do_match, skb, &mtpar) != 0) { + &e->ip, mtpar.fragoff)) { + no_match: e = ipt_next_entry(e); continue; } + xt_ematch_foreach(ematch, e) + if (do_match(ematch, skb, &mtpar) != 0) + goto no_match; + ADD_COUNTER(e->counters, ntohs(ip->tot_len), 1); t = ipt_get_target(e); @@ -686,6 +691,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, int ret; unsigned int j; struct xt_mtchk_param mtpar; + struct xt_entry_match *ematch; ret = check_entry(e, name); if (ret) @@ -697,7 +703,11 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, mtpar.entryinfo = &e->ip; mtpar.hook_mask = e->comefrom; mtpar.family = NFPROTO_IPV4; - ret = IPT_MATCH_ITERATE(e, find_check_match, &mtpar, &j); + xt_ematch_foreach(ematch, e) { + ret = find_check_match(ematch, &mtpar, &j); + if (ret != 0) + break; + } if (ret != 0) goto cleanup_matches; @@ -720,7 +730,9 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, err: module_put(t->u.kernel.target->me); cleanup_matches: - IPT_MATCH_ITERATE(e, cleanup_match, net, &j); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, &j) != 0) + break; return ret; } @@ -791,9 +803,12 @@ cleanup_entry(struct ipt_entry *e, struct net *net) { struct xt_tgdtor_param par; struct ipt_entry_target *t; + struct xt_entry_match *ematch; /* Cleanup all matches */ - IPT_MATCH_ITERATE(e, cleanup_match, net, NULL); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, NULL) != 0) + break; t = ipt_get_target(e); par.net = net; @@ -1060,13 +1075,16 @@ static int compat_calc_entry(const struct ipt_entry *e, const struct xt_table_info *info, const void *base, struct xt_table_info *newinfo) { + const struct xt_entry_match *ematch; const struct ipt_entry_target *t; unsigned int entry_offset; int off, i, ret; off = sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry); entry_offset = (void *)e - base; - IPT_MATCH_ITERATE(e, compat_calc_match, &off); + xt_ematch_foreach(ematch, e) + if (compat_calc_match(ematch, &off) != 0) + break; t = ipt_get_target_c(e); off += xt_compat_target_offset(t->u.kernel.target); newinfo->size -= off; @@ -1441,7 +1459,8 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr, struct compat_ipt_entry __user *ce; u_int16_t target_offset, next_offset; compat_uint_t origsize; - int ret; + const struct xt_entry_match *ematch; + int ret = 0; origsize = *size; ce = (struct compat_ipt_entry __user *)*dstptr; @@ -1453,7 +1472,11 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr, *dstptr += sizeof(struct compat_ipt_entry); *size -= sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry); - ret = IPT_MATCH_ITERATE(e, xt_compat_match_to_user, dstptr, size); + xt_ematch_foreach(ematch, e) { + ret = xt_compat_match_to_user(ematch, dstptr, size); + if (ret != 0) + break; + } target_offset = e->target_offset - (origsize - *size); if (ret) return ret; @@ -1505,9 +1528,12 @@ compat_release_match(struct ipt_entry_match *m, unsigned int *i) static void compat_release_entry(struct compat_ipt_entry *e) { struct ipt_entry_target *t; + struct xt_entry_match *ematch; /* Cleanup all matches */ - COMPAT_IPT_MATCH_ITERATE(e, compat_release_match, NULL); + xt_ematch_foreach(ematch, e) + if (compat_release_match(ematch, NULL) != 0) + break; t = compat_ipt_get_target(e); module_put(t->u.kernel.target->me); } @@ -1522,6 +1548,7 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e, const unsigned int *underflows, const char *name) { + struct xt_entry_match *ematch; struct ipt_entry_target *t; struct xt_target *target; unsigned int entry_offset; @@ -1550,8 +1577,12 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e, off = sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry); entry_offset = (void *)e - (void *)base; j = 0; - ret = COMPAT_IPT_MATCH_ITERATE(e, compat_find_calc_match, name, - &e->ip, e->comefrom, &off, &j); + xt_ematch_foreach(ematch, e) { + ret = compat_find_calc_match(ematch, name, + &e->ip, e->comefrom, &off, &j); + if (ret != 0) + break; + } if (ret != 0) goto release_matches; @@ -1590,7 +1621,9 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e, out: module_put(t->u.kernel.target->me); release_matches: - IPT_MATCH_ITERATE(e, compat_release_match, &j); + xt_ematch_foreach(ematch, e) + if (compat_release_match(ematch, &j) != 0) + break; return ret; } @@ -1604,6 +1637,7 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, struct ipt_entry *de; unsigned int origsize; int ret, h; + struct xt_entry_match *ematch; ret = 0; origsize = *size; @@ -1614,8 +1648,11 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, *dstptr += sizeof(struct ipt_entry); *size += sizeof(struct ipt_entry) - sizeof(struct compat_ipt_entry); - ret = COMPAT_IPT_MATCH_ITERATE(e, xt_compat_match_from_user, - dstptr, size); + xt_ematch_foreach(ematch, e) { + ret = xt_compat_match_from_user(ematch, dstptr, size); + if (ret != 0) + break; + } if (ret) return ret; de->target_offset = e->target_offset - (origsize - *size); @@ -1636,9 +1673,10 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr, static int compat_check_entry(struct ipt_entry *e, struct net *net, const char *name) { + struct xt_entry_match *ematch; struct xt_mtchk_param mtpar; unsigned int j; - int ret; + int ret = 0; j = 0; mtpar.net = net; @@ -1646,7 +1684,11 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name) mtpar.entryinfo = &e->ip; mtpar.hook_mask = e->comefrom; mtpar.family = NFPROTO_IPV4; - ret = IPT_MATCH_ITERATE(e, check_match, &mtpar, &j); + xt_ematch_foreach(ematch, e) { + ret = check_match(ematch, &mtpar, &j); + if (ret != 0) + break; + } if (ret) goto cleanup_matches; @@ -1656,7 +1698,9 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name) return 0; cleanup_matches: - IPT_MATCH_ITERATE(e, cleanup_match, net, &j); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, &j) != 0) + break; return ret; } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index b7e27c19c7a..1537e6bad5d 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -393,16 +393,21 @@ ip6t_do_table(struct sk_buff *skb, do { const struct ip6t_entry_target *t; + const struct xt_entry_match *ematch; IP_NF_ASSERT(e); IP_NF_ASSERT(back); if (!ip6_packet_match(skb, indev, outdev, &e->ipv6, - &mtpar.thoff, &mtpar.fragoff, &hotdrop) || - IP6T_MATCH_ITERATE(e, do_match, skb, &mtpar) != 0) { + &mtpar.thoff, &mtpar.fragoff, &hotdrop)) { + no_match: e = ip6t_next_entry(e); continue; } + xt_ematch_foreach(ematch, e) + if (do_match(ematch, skb, &mtpar) != 0) + goto no_match; + ADD_COUNTER(e->counters, ntohs(ipv6_hdr(skb)->payload_len) + sizeof(struct ipv6hdr), 1); @@ -717,6 +722,7 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, int ret; unsigned int j; struct xt_mtchk_param mtpar; + struct xt_entry_match *ematch; ret = check_entry(e, name); if (ret) @@ -728,7 +734,11 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, mtpar.entryinfo = &e->ipv6; mtpar.hook_mask = e->comefrom; mtpar.family = NFPROTO_IPV6; - ret = IP6T_MATCH_ITERATE(e, find_check_match, &mtpar, &j); + xt_ematch_foreach(ematch, e) { + ret = find_check_match(ematch, &mtpar, &j); + if (ret != 0) + break; + } if (ret != 0) goto cleanup_matches; @@ -751,7 +761,9 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, err: module_put(t->u.kernel.target->me); cleanup_matches: - IP6T_MATCH_ITERATE(e, cleanup_match, net, &j); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, &j) != 0) + break; return ret; } @@ -821,9 +833,12 @@ static void cleanup_entry(struct ip6t_entry *e, struct net *net) { struct xt_tgdtor_param par; struct ip6t_entry_target *t; + struct xt_entry_match *ematch; /* Cleanup all matches */ - IP6T_MATCH_ITERATE(e, cleanup_match, net, NULL); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, NULL) != 0) + break; t = ip6t_get_target(e); par.net = net; @@ -1090,13 +1105,16 @@ static int compat_calc_entry(const struct ip6t_entry *e, const struct xt_table_info *info, const void *base, struct xt_table_info *newinfo) { + const struct xt_entry_match *ematch; const struct ip6t_entry_target *t; unsigned int entry_offset; int off, i, ret; off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry); entry_offset = (void *)e - base; - IP6T_MATCH_ITERATE(e, compat_calc_match, &off); + xt_ematch_foreach(ematch, e) + if (compat_calc_match(ematch, &off) != 0) + break; t = ip6t_get_target_c(e); off += xt_compat_target_offset(t->u.kernel.target); newinfo->size -= off; @@ -1474,7 +1492,8 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr, struct compat_ip6t_entry __user *ce; u_int16_t target_offset, next_offset; compat_uint_t origsize; - int ret; + const struct xt_entry_match *ematch; + int ret = 0; origsize = *size; ce = (struct compat_ip6t_entry __user *)*dstptr; @@ -1486,7 +1505,11 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr, *dstptr += sizeof(struct compat_ip6t_entry); *size -= sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry); - ret = IP6T_MATCH_ITERATE(e, xt_compat_match_to_user, dstptr, size); + xt_ematch_foreach(ematch, e) { + ret = xt_compat_match_to_user(ematch, dstptr, size); + if (ret != 0) + break; + } target_offset = e->target_offset - (origsize - *size); if (ret) return ret; @@ -1538,9 +1561,12 @@ compat_release_match(struct ip6t_entry_match *m, unsigned int *i) static void compat_release_entry(struct compat_ip6t_entry *e) { struct ip6t_entry_target *t; + struct xt_entry_match *ematch; /* Cleanup all matches */ - COMPAT_IP6T_MATCH_ITERATE(e, compat_release_match, NULL); + xt_ematch_foreach(ematch, e) + if (compat_release_match(ematch, NULL) != 0) + break; t = compat_ip6t_get_target(e); module_put(t->u.kernel.target->me); } @@ -1555,6 +1581,7 @@ check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e, const unsigned int *underflows, const char *name) { + struct xt_entry_match *ematch; struct ip6t_entry_target *t; struct xt_target *target; unsigned int entry_offset; @@ -1583,8 +1610,12 @@ check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e, off = sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry); entry_offset = (void *)e - (void *)base; j = 0; - ret = COMPAT_IP6T_MATCH_ITERATE(e, compat_find_calc_match, name, - &e->ipv6, e->comefrom, &off, &j); + xt_ematch_foreach(ematch, e) { + ret = compat_find_calc_match(ematch, name, + &e->ipv6, e->comefrom, &off, &j); + if (ret != 0) + break; + } if (ret != 0) goto release_matches; @@ -1623,7 +1654,9 @@ check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e, out: module_put(t->u.kernel.target->me); release_matches: - IP6T_MATCH_ITERATE(e, compat_release_match, &j); + xt_ematch_foreach(ematch, e) + if (compat_release_match(ematch, &j) != 0) + break; return ret; } @@ -1637,6 +1670,7 @@ compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr, struct ip6t_entry *de; unsigned int origsize; int ret, h; + struct xt_entry_match *ematch; ret = 0; origsize = *size; @@ -1647,8 +1681,11 @@ compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr, *dstptr += sizeof(struct ip6t_entry); *size += sizeof(struct ip6t_entry) - sizeof(struct compat_ip6t_entry); - ret = COMPAT_IP6T_MATCH_ITERATE(e, xt_compat_match_from_user, - dstptr, size); + xt_ematch_foreach(ematch, e) { + ret = xt_compat_match_from_user(ematch, dstptr, size); + if (ret != 0) + break; + } if (ret) return ret; de->target_offset = e->target_offset - (origsize - *size); @@ -1670,8 +1707,9 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, const char *name) { unsigned int j; - int ret; + int ret = 0; struct xt_mtchk_param mtpar; + struct xt_entry_match *ematch; j = 0; mtpar.net = net; @@ -1679,7 +1717,11 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, mtpar.entryinfo = &e->ipv6; mtpar.hook_mask = e->comefrom; mtpar.family = NFPROTO_IPV6; - ret = IP6T_MATCH_ITERATE(e, check_match, &mtpar, &j); + xt_ematch_foreach(ematch, e) { + ret = check_match(ematch, &mtpar, &j); + if (ret != 0) + break; + } if (ret) goto cleanup_matches; @@ -1689,7 +1731,9 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, return 0; cleanup_matches: - IP6T_MATCH_ITERATE(e, cleanup_match, net, &j); + xt_ematch_foreach(ematch, e) + if (cleanup_match(ematch, net, &j) != 0) + break; return ret; } diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index 6f21b4377db..0e357ac9a2a 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -239,6 +239,7 @@ static bool tcpmss_tg4_check(const struct xt_tgchk_param *par) { const struct xt_tcpmss_info *info = par->targinfo; const struct ipt_entry *e = par->entryinfo; + const struct xt_entry_match *ematch; if (info->mss == XT_TCPMSS_CLAMP_PMTU && (par->hook_mask & ~((1 << NF_INET_FORWARD) | @@ -248,8 +249,9 @@ static bool tcpmss_tg4_check(const struct xt_tgchk_param *par) "FORWARD, OUTPUT and POSTROUTING hooks\n"); return false; } - if (IPT_MATCH_ITERATE(e, find_syn_match)) - return true; + xt_ematch_foreach(ematch, e) + if (find_syn_match(ematch)) + return true; printk("xt_TCPMSS: Only works on TCP SYN packets\n"); return false; } @@ -259,6 +261,7 @@ static bool tcpmss_tg6_check(const struct xt_tgchk_param *par) { const struct xt_tcpmss_info *info = par->targinfo; const struct ip6t_entry *e = par->entryinfo; + const struct xt_entry_match *ematch; if (info->mss == XT_TCPMSS_CLAMP_PMTU && (par->hook_mask & ~((1 << NF_INET_FORWARD) | @@ -268,8 +271,9 @@ static bool tcpmss_tg6_check(const struct xt_tgchk_param *par) "FORWARD, OUTPUT and POSTROUTING hooks\n"); return false; } - if (IP6T_MATCH_ITERATE(e, find_syn_match)) - return true; + xt_ematch_foreach(ematch, e) + if (find_syn_match(ematch)) + return true; printk("xt_TCPMSS: Only works on TCP SYN packets\n"); return false; } -- cgit v1.2.3-70-g09d2 From 52c793f24054f5dc30d228e37e0e19cc8313f086 Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Mon, 22 Feb 2010 22:21:17 +0000 Subject: can: netlink support for bus-error reporting and counters This patch makes the bus-error reporting configurable and allows to retrieve the CAN TX and RX bus error counters via netlink interface. I have added support for the SJA1000. The TX and RX bus error counters are also copied to the data fields 6..7 of error messages when state changes are reported. Signed-off-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/dev.c | 6 ++++++ drivers/net/can/sja1000/sja1000.c | 25 ++++++++++++++++++++++--- include/linux/can/dev.h | 2 ++ include/linux/can/netlink.h | 18 ++++++++++++++---- 4 files changed, 44 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index f08f1202ff0..904aa369f80 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -574,6 +574,7 @@ static const struct nla_policy can_policy[IFLA_CAN_MAX + 1] = { [IFLA_CAN_BITTIMING_CONST] = { .len = sizeof(struct can_bittiming_const) }, [IFLA_CAN_CLOCK] = { .len = sizeof(struct can_clock) }, + [IFLA_CAN_BERR_COUNTER] = { .len = sizeof(struct can_berr_counter) }, }; static int can_changelink(struct net_device *dev, @@ -649,6 +650,8 @@ static size_t can_get_size(const struct net_device *dev) size += nla_total_size(sizeof(u32)); /* IFLA_CAN_RESTART_MS */ size += sizeof(struct can_bittiming); /* IFLA_CAN_BITTIMING */ size += sizeof(struct can_clock); /* IFLA_CAN_CLOCK */ + if (priv->do_get_berr_counter) /* IFLA_CAN_BERR_COUNTER */ + size += sizeof(struct can_berr_counter); if (priv->bittiming_const) /* IFLA_CAN_BITTIMING_CONST */ size += sizeof(struct can_bittiming_const); @@ -659,6 +662,7 @@ static int can_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct can_priv *priv = netdev_priv(dev); struct can_ctrlmode cm = {.flags = priv->ctrlmode}; + struct can_berr_counter bec; enum can_state state = priv->state; if (priv->do_get_state) @@ -669,6 +673,8 @@ static int can_fill_info(struct sk_buff *skb, const struct net_device *dev) NLA_PUT(skb, IFLA_CAN_BITTIMING, sizeof(priv->bittiming), &priv->bittiming); NLA_PUT(skb, IFLA_CAN_CLOCK, sizeof(cm), &priv->clock); + if (priv->do_get_berr_counter && !priv->do_get_berr_counter(dev, &bec)) + NLA_PUT(skb, IFLA_CAN_BERR_COUNTER, sizeof(bec), &bec); if (priv->bittiming_const) NLA_PUT(skb, IFLA_CAN_BITTIMING_CONST, sizeof(*priv->bittiming_const), priv->bittiming_const); diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index ace103a4483..145b1a731a5 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -130,8 +130,12 @@ static void set_normal_mode(struct net_device *dev) /* check reset bit */ if ((status & MOD_RM) == 0) { priv->can.state = CAN_STATE_ERROR_ACTIVE; - /* enable all interrupts */ - priv->write_reg(priv, REG_IER, IRQ_ALL); + /* enable interrupts */ + if (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING) + priv->write_reg(priv, REG_IER, IRQ_ALL); + else + priv->write_reg(priv, REG_IER, + IRQ_ALL & ~IRQ_BEI); return; } @@ -203,6 +207,17 @@ static int sja1000_set_bittiming(struct net_device *dev) return 0; } +static int sja1000_get_berr_counter(const struct net_device *dev, + struct can_berr_counter *bec) +{ + struct sja1000_priv *priv = netdev_priv(dev); + + bec->txerr = priv->read_reg(priv, REG_TXERR); + bec->rxerr = priv->read_reg(priv, REG_RXERR); + + return 0; +} + /* * initialize SJA1000 chip: * - reset chip @@ -437,6 +452,8 @@ static int sja1000_err(struct net_device *dev, uint8_t isrc, uint8_t status) CAN_ERR_CRTL_TX_PASSIVE : CAN_ERR_CRTL_RX_PASSIVE; } + cf->data[6] = txerr; + cf->data[7] = rxerr; } priv->can.state = state; @@ -567,7 +584,9 @@ struct net_device *alloc_sja1000dev(int sizeof_priv) priv->can.bittiming_const = &sja1000_bittiming_const; priv->can.do_set_bittiming = sja1000_set_bittiming; priv->can.do_set_mode = sja1000_set_mode; - priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; + priv->can.do_get_berr_counter = sja1000_get_berr_counter; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES | + CAN_CTRLMODE_BERR_REPORTING; if (sizeof_priv) priv->priv = (void *)priv + sizeof(struct sja1000_priv); diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index c8c660a79f9..6e5a7f00223 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -47,6 +47,8 @@ struct can_priv { int (*do_set_mode)(struct net_device *dev, enum can_mode mode); int (*do_get_state)(const struct net_device *dev, enum can_state *state); + int (*do_get_berr_counter)(const struct net_device *dev, + struct can_berr_counter *bec); unsigned int echo_skb_max; struct sk_buff **echo_skb; diff --git a/include/linux/can/netlink.h b/include/linux/can/netlink.h index c818335fbb1..3250de935e1 100644 --- a/include/linux/can/netlink.h +++ b/include/linux/can/netlink.h @@ -69,6 +69,14 @@ enum can_state { CAN_STATE_MAX }; +/* + * CAN bus error counters + */ +struct can_berr_counter { + __u16 txerr; + __u16 rxerr; +}; + /* * CAN controller mode */ @@ -77,10 +85,11 @@ struct can_ctrlmode { __u32 flags; }; -#define CAN_CTRLMODE_LOOPBACK 0x1 /* Loopback mode */ -#define CAN_CTRLMODE_LISTENONLY 0x2 /* Listen-only mode */ -#define CAN_CTRLMODE_3_SAMPLES 0x4 /* Triple sampling mode */ -#define CAN_CTRLMODE_ONE_SHOT 0x8 /* One-Shot mode */ +#define CAN_CTRLMODE_LOOPBACK 0x01 /* Loopback mode */ +#define CAN_CTRLMODE_LISTENONLY 0x02 /* Listen-only mode */ +#define CAN_CTRLMODE_3_SAMPLES 0x04 /* Triple sampling mode */ +#define CAN_CTRLMODE_ONE_SHOT 0x08 /* One-Shot mode */ +#define CAN_CTRLMODE_BERR_REPORTING 0x10 /* Bus-error reporting */ /* * CAN device statistics @@ -106,6 +115,7 @@ enum { IFLA_CAN_CTRLMODE, IFLA_CAN_RESTART_MS, IFLA_CAN_RESTART, + IFLA_CAN_BERR_COUNTER, __IFLA_CAN_MAX }; -- cgit v1.2.3-70-g09d2 From 45bb00609022ecf1d97e083666c68c74d237b799 Mon Sep 17 00:00:00 2001 From: Ulrich Weber Date: Thu, 25 Feb 2010 23:28:58 +0000 Subject: ipv6: Remove IPV6_ADDR_RESERVED RFC 4291 section 2.4 states that all uncategorized addresses should be considered as Global Unicast. This will remove IPV6_ADDR_RESERVED completely and return IPV6_ADDR_UNICAST in ipv6_addr_type() instead. Signed-off-by: Ulrich Weber Signed-off-by: David S. Miller --- include/net/ipv6.h | 1 - net/ipv6/addrconf.c | 3 +-- net/ipv6/addrconf_core.c | 2 +- net/ipv6/route.c | 2 +- 4 files changed, 3 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index d067db1f88c..e72fb10ce57 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -73,7 +73,6 @@ #define IPV6_ADDR_SCOPE_MASK 0x00f0U #define IPV6_ADDR_MAPPED 0x1000U -#define IPV6_ADDR_RESERVED 0x2000U /* reserved address space */ /* * Addr scopes diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 1b327f15e7e..88fd8c5877e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -992,8 +992,7 @@ struct ipv6_saddr_dst { static inline int ipv6_saddr_preferred(int type) { - if (type & (IPV6_ADDR_MAPPED|IPV6_ADDR_COMPATv4| - IPV6_ADDR_LOOPBACK|IPV6_ADDR_RESERVED)) + if (type & (IPV6_ADDR_MAPPED|IPV6_ADDR_COMPATv4|IPV6_ADDR_LOOPBACK)) return 1; return 0; } diff --git a/net/ipv6/addrconf_core.c b/net/ipv6/addrconf_core.c index 3f82e9542ed..6b03826552e 100644 --- a/net/ipv6/addrconf_core.c +++ b/net/ipv6/addrconf_core.c @@ -72,7 +72,7 @@ int __ipv6_addr_type(const struct in6_addr *addr) IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); /* addr-select 3.3 */ } - return (IPV6_ADDR_RESERVED | + return (IPV6_ADDR_UNICAST | IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); /* addr-select 3.4 */ } EXPORT_SYMBOL(__ipv6_addr_type); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 88c0a5c49ae..b08879e97f2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1873,7 +1873,7 @@ static int ip6_pkt_drop(struct sk_buff *skb, u8 code, int ipstats_mib_noroutes) switch (ipstats_mib_noroutes) { case IPSTATS_MIB_INNOROUTES: type = ipv6_addr_type(&ipv6_hdr(skb)->daddr); - if (type == IPV6_ADDR_ANY || type == IPV6_ADDR_RESERVED) { + if (type == IPV6_ADDR_ANY) { IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst), IPSTATS_MIB_INADDRERRORS); break; -- cgit v1.2.3-70-g09d2 From c79c5ffdce14abb4de3878c5aa8c3c6e5093c69b Mon Sep 17 00:00:00 2001 From: Peter Waskiewicz Date: Fri, 26 Feb 2010 01:54:20 +0000 Subject: ethtool: Add n-tuple string length to drvinfo and return it The drvinfo struct should include the number of strings that get_rx_ntuple will return. It will be variable if an underlying driver implements its own get_rx_ntuple routine, so userspace needs to know how much data is coming. Signed-off-by: Peter P Waskiewicz Jr Signed-off-by: Jeff Kirsher Signed-off-by: David S. Miller --- include/linux/ethtool.h | 1 + net/core/ethtool.c | 3 +++ 2 files changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index cca1c3de140..f7992a256b7 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -63,6 +63,7 @@ struct ethtool_drvinfo { char reserved2[12]; __u32 n_priv_flags; /* number of flags valid in ETHTOOL_GPFLAGS */ __u32 n_stats; /* number of u64's from ETHTOOL_GSTATS */ + __u32 n_ntuples; /* number of n-tuple filters from GSTRINGS */ __u32 testinfo_len; __u32 eedump_len; /* Size of data from ETHTOOL_GEEPROM (bytes) */ __u32 regdump_len; /* Size of data from ETHTOOL_GREGS (bytes) */ diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 31b1eddc1b8..1c94f48e27b 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -224,6 +224,9 @@ static noinline int ethtool_get_drvinfo(struct net_device *dev, void __user *use rc = ops->get_sset_count(dev, ETH_SS_PRIV_FLAGS); if (rc >= 0) info.n_priv_flags = rc; + rc = ops->get_sset_count(dev, ETH_SS_NTUPLE_FILTERS); + if (rc >= 0) + info.n_ntuples = rc; } if (ops->get_regs_len) info.regdump_len = ops->get_regs_len(dev); -- cgit v1.2.3-70-g09d2 From 738b0343e73604750feb107e063c28b3ca36cb84 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 26 Feb 2010 05:12:02 -0800 Subject: Revert "ethtool: Add n-tuple string length to drvinfo and return it" This reverts commit c79c5ffdce14abb4de3878c5aa8c3c6e5093c69b. As Jeff points out we can't break the user visible interface like this, we need to add this into the reserved[] thing. Signed-off-by: David S. Miller --- include/linux/ethtool.h | 1 - net/core/ethtool.c | 3 --- 2 files changed, 4 deletions(-) (limited to 'include') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index f7992a256b7..cca1c3de140 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -63,7 +63,6 @@ struct ethtool_drvinfo { char reserved2[12]; __u32 n_priv_flags; /* number of flags valid in ETHTOOL_GPFLAGS */ __u32 n_stats; /* number of u64's from ETHTOOL_GSTATS */ - __u32 n_ntuples; /* number of n-tuple filters from GSTRINGS */ __u32 testinfo_len; __u32 eedump_len; /* Size of data from ETHTOOL_GEEPROM (bytes) */ __u32 regdump_len; /* Size of data from ETHTOOL_GREGS (bytes) */ diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 1c94f48e27b..31b1eddc1b8 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -224,9 +224,6 @@ static noinline int ethtool_get_drvinfo(struct net_device *dev, void __user *use rc = ops->get_sset_count(dev, ETH_SS_PRIV_FLAGS); if (rc >= 0) info.n_priv_flags = rc; - rc = ops->get_sset_count(dev, ETH_SS_NTUPLE_FILTERS); - if (rc >= 0) - info.n_ntuples = rc; } if (ops->get_regs_len) info.regdump_len = ops->get_regs_len(dev); -- cgit v1.2.3-70-g09d2 From 773c3e75d1fc7ea5058bfeab5d82bac5b85f8cd8 Mon Sep 17 00:00:00 2001 From: Sriramakrishnan Date: Fri, 26 Feb 2010 05:22:03 -0800 Subject: can: ti hecc module : add platform specific initialization callback. CAN module on AM3517 requires programming of IO expander as part of init sequence - to enable CAN PHY. Added platform specific callback to handle phy control(switch on /off). Signed-off-by: Sriramakrishnan Signed-off-by: David S. Miller --- drivers/net/can/ti_hecc.c | 17 ++++++++++++++++- include/linux/can/platform/ti_hecc.h | 8 ++++++-- 2 files changed, 22 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index b861cd56107..0c3d2ba0d17 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -28,9 +28,11 @@ * .mbx_offset = 0x2000, * .int_line = 0, * .revision = 1, + * .transceiver_switch = hecc_phy_control, * }; * - * Please see include/can/platform/ti_hecc.h for description of above fields + * Please see include/linux/can/platform/ti_hecc.h for description of + * above fields. * */ @@ -220,6 +222,7 @@ struct ti_hecc_priv { u32 tx_head; u32 tx_tail; u32 rx_next; + void (*transceiver_switch)(int); }; static inline int get_tx_head_mb(struct ti_hecc_priv *priv) @@ -317,6 +320,13 @@ static int ti_hecc_set_btc(struct ti_hecc_priv *priv) return 0; } +static void ti_hecc_transceiver_switch(const struct ti_hecc_priv *priv, + int on) +{ + if (priv->transceiver_switch) + priv->transceiver_switch(on); +} + static void ti_hecc_reset(struct net_device *ndev) { u32 cnt; @@ -818,10 +828,13 @@ static int ti_hecc_open(struct net_device *ndev) return err; } + ti_hecc_transceiver_switch(priv, 1); + /* Open common can device */ err = open_candev(ndev); if (err) { dev_err(ndev->dev.parent, "open_candev() failed %d\n", err); + ti_hecc_transceiver_switch(priv, 0); free_irq(ndev->irq, ndev); return err; } @@ -842,6 +855,7 @@ static int ti_hecc_close(struct net_device *ndev) ti_hecc_stop(ndev); free_irq(ndev->irq, ndev); close_candev(ndev); + ti_hecc_transceiver_switch(priv, 0); return 0; } @@ -903,6 +917,7 @@ static int ti_hecc_probe(struct platform_device *pdev) priv->hecc_ram_offset = pdata->hecc_ram_offset; priv->mbx_offset = pdata->mbx_offset; priv->int_line = pdata->int_line; + priv->transceiver_switch = pdata->transceiver_switch; priv->can.bittiming_const = &ti_hecc_bittiming_const; priv->can.do_set_mode = ti_hecc_do_set_mode; diff --git a/include/linux/can/platform/ti_hecc.h b/include/linux/can/platform/ti_hecc.h index 4688c7bb1bd..af17cb3f7a8 100644 --- a/include/linux/can/platform/ti_hecc.h +++ b/include/linux/can/platform/ti_hecc.h @@ -1,3 +1,6 @@ +#ifndef __CAN_PLATFORM_TI_HECC_H__ +#define __CAN_PLATFORM_TI_HECC_H__ + /* * TI HECC (High End CAN Controller) driver platform header * @@ -23,6 +26,7 @@ * @mbx_offset: Mailbox RAM offset * @int_line: Interrupt line to use - 0 or 1 * @version: version for future use + * @transceiver_switch: platform specific callback fn for transceiver control * * Platform data structure to get all platform specific settings. * this structure also accounts the fact that the IP may have different @@ -35,6 +39,6 @@ struct ti_hecc_platform_data { u32 mbx_offset; u32 int_line; u32 version; + void (*transceiver_switch) (int); }; - - +#endif -- cgit v1.2.3-70-g09d2 From caf66e581172dc5032bb84841a91bc7b77ad9876 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Thu, 25 Feb 2010 12:02:45 -0500 Subject: netdevice.h: check for CONFIG_WLAN instead of CONFIG_WLAN_80211 In "wireless: remove WLAN_80211 and WLAN_PRE80211 from Kconfig" I inadvertantly missed a line in include/linux/netdevice.h. I thereby effectively reverted "net: Set LL_MAX_HEADER properly for wireless." by accident. :-( Now we should check there for CONFIG_WLAN instead. Signed-off-by: John W. Linville Reported-by: Christoph Egger Cc: stable@kernel.org --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a3fccc85b1a..99914e6fa3f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -136,7 +136,7 @@ static inline bool dev_xmit_complete(int rc) * used. */ -#if defined(CONFIG_WLAN_80211) || defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE) +#if defined(CONFIG_WLAN) || defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE) # if defined(CONFIG_MAC80211_MESH) # define LL_MAX_HEADER 128 # else -- cgit v1.2.3-70-g09d2 From a2835763e130c343ace5320c20d33c281e7097b7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Feb 2010 06:34:51 +0000 Subject: rtnetlink: handle rtnl_link netlink notifications manually In order to support specifying device flags during device creation, we must be able to roll back device registration in case setting the flags fails without sending any notifications related to the device to userspace. This patch changes rollback_registered_many() and register_netdevice() to manually send netlink notifications for devices not handled by rtnl_link and allows to defer notifications for devices handled by rtnl_link until setup is complete. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netdevice.h | 7 ++++++- net/core/dev.c | 8 +++++++- net/core/rtnetlink.c | 4 +--- 3 files changed, 14 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7a2aea56f19..1bfda90c262 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -924,7 +924,12 @@ struct net_device { NETREG_UNREGISTERED, /* completed unregister todo */ NETREG_RELEASED, /* called free_netdev */ NETREG_DUMMY, /* dummy device for NAPI poll */ - } reg_state; + } reg_state:16; + + enum { + RTNL_LINK_INITIALIZED, + RTNL_LINK_INITIALIZING, + } rtnl_link_state:16; /* Called from unregister, can be used to call free_netdev */ void (*destructor)(struct net_device *dev); diff --git a/net/core/dev.c b/net/core/dev.c index eb7f1a4fefc..75332b08952 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4865,6 +4865,10 @@ static void rollback_registered_many(struct list_head *head) */ call_netdevice_notifiers(NETDEV_UNREGISTER, dev); + if (!dev->rtnl_link_ops || + dev->rtnl_link_state == RTNL_LINK_INITIALIZED) + rtmsg_ifinfo(RTM_DELLINK, dev, ~0U); + /* * Flush the unicast and multicast chains */ @@ -5091,7 +5095,9 @@ int register_netdevice(struct net_device *dev) * Prevent userspace races by waiting until the network * device is fully setup before sending notifications. */ - rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U); + if (!dev->rtnl_link_ops || + dev->rtnl_link_state == RTNL_LINK_INITIALIZED) + rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U); out: return ret; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b7c7dfd8650..020e43bfef5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1425,9 +1425,6 @@ static int rtnetlink_event(struct notifier_block *this, unsigned long event, voi struct net_device *dev = ptr; switch (event) { - case NETDEV_UNREGISTER: - rtmsg_ifinfo(RTM_DELLINK, dev, ~0U); - break; case NETDEV_UP: case NETDEV_DOWN: rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING); @@ -1437,6 +1434,7 @@ static int rtnetlink_event(struct notifier_block *this, unsigned long event, voi case NETDEV_REGISTER: case NETDEV_CHANGE: case NETDEV_GOING_DOWN: + case NETDEV_UNREGISTER: case NETDEV_UNREGISTER_BATCH: break; default: -- cgit v1.2.3-70-g09d2 From bd38081160bb3d036db98472e537b6a7dd4da51a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Feb 2010 06:34:53 +0000 Subject: dev: support deferring device flag change notifications Split dev_change_flags() into two functions: __dev_change_flags() to perform the actual changes and __dev_notify_flags() to invoke netdevice notifiers. This will be used by rtnl_link to defer netlink notifications until the device has been fully configured. This changes ordering of some operations, in particular: - netlink notifications are sent after all changes have been performed. As a side effect this surpresses one unnecessary netlink message when the IFF_UP and other flags are changed simultaneously. - The NETDEV_UP/NETDEV_DOWN and NETDEV_CHANGE notifiers are invoked after all changes have been performed. Their relative is unchanged. - net_dmaengine_put() is invoked before the NETDEV_DOWN notifier instead of afterwards. This should not make any difference since both RX and TX are already shut down at this point. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 + net/core/dev.c | 163 +++++++++++++++++++++++++++++----------------- net/core/rtnetlink.c | 2 - 3 files changed, 106 insertions(+), 61 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1bfda90c262..c79a88be7c3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1587,7 +1587,9 @@ extern int dev_valid_name(const char *name); extern int dev_ioctl(struct net *net, unsigned int cmd, void __user *); extern int dev_ethtool(struct net *net, struct ifreq *); extern unsigned dev_get_flags(const struct net_device *); +extern int __dev_change_flags(struct net_device *, unsigned int flags); extern int dev_change_flags(struct net_device *, unsigned); +extern void __dev_notify_flags(struct net_device *, unsigned int old_flags); extern int dev_change_name(struct net_device *, const char *); extern int dev_set_alias(struct net_device *, const char *, size_t); extern int dev_change_net_namespace(struct net_device *, diff --git a/net/core/dev.c b/net/core/dev.c index 75332b08952..e5972f7f7e1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1113,32 +1113,13 @@ void dev_load(struct net *net, const char *name) } EXPORT_SYMBOL(dev_load); -/** - * dev_open - prepare an interface for use. - * @dev: device to open - * - * Takes a device from down to up state. The device's private open - * function is invoked and then the multicast lists are loaded. Finally - * the device is moved into the up state and a %NETDEV_UP message is - * sent to the netdev notifier chain. - * - * Calling this function on an active interface is a nop. On a failure - * a negative errno code is returned. - */ -int dev_open(struct net_device *dev) +static int __dev_open(struct net_device *dev) { const struct net_device_ops *ops = dev->netdev_ops; int ret; ASSERT_RTNL(); - /* - * Is it already up? - */ - - if (dev->flags & IFF_UP) - return 0; - /* * Is it even present? */ @@ -1187,36 +1168,57 @@ int dev_open(struct net_device *dev) * Wakeup transmit queue engine */ dev_activate(dev); - - /* - * ... and announce new interface. - */ - call_netdevice_notifiers(NETDEV_UP, dev); } return ret; } -EXPORT_SYMBOL(dev_open); /** - * dev_close - shutdown an interface. - * @dev: device to shutdown + * dev_open - prepare an interface for use. + * @dev: device to open * - * This function moves an active device into down state. A - * %NETDEV_GOING_DOWN is sent to the netdev notifier chain. The device - * is then deactivated and finally a %NETDEV_DOWN is sent to the notifier - * chain. + * Takes a device from down to up state. The device's private open + * function is invoked and then the multicast lists are loaded. Finally + * the device is moved into the up state and a %NETDEV_UP message is + * sent to the netdev notifier chain. + * + * Calling this function on an active interface is a nop. On a failure + * a negative errno code is returned. */ -int dev_close(struct net_device *dev) +int dev_open(struct net_device *dev) +{ + int ret; + + /* + * Is it already up? + */ + if (dev->flags & IFF_UP) + return 0; + + /* + * Open device + */ + ret = __dev_open(dev); + if (ret < 0) + return ret; + + /* + * ... and announce new interface. + */ + rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING); + call_netdevice_notifiers(NETDEV_UP, dev); + + return ret; +} +EXPORT_SYMBOL(dev_open); + +static int __dev_close(struct net_device *dev) { const struct net_device_ops *ops = dev->netdev_ops; - ASSERT_RTNL(); + ASSERT_RTNL(); might_sleep(); - if (!(dev->flags & IFF_UP)) - return 0; - /* * Tell people we are going down, so that they can * prepare to death, when device is still operating. @@ -1252,14 +1254,34 @@ int dev_close(struct net_device *dev) dev->flags &= ~IFF_UP; /* - * Tell people we are down + * Shutdown NET_DMA */ - call_netdevice_notifiers(NETDEV_DOWN, dev); + net_dmaengine_put(); + + return 0; +} + +/** + * dev_close - shutdown an interface. + * @dev: device to shutdown + * + * This function moves an active device into down state. A + * %NETDEV_GOING_DOWN is sent to the netdev notifier chain. The device + * is then deactivated and finally a %NETDEV_DOWN is sent to the notifier + * chain. + */ +int dev_close(struct net_device *dev) +{ + if (!(dev->flags & IFF_UP)) + return 0; + + __dev_close(dev); /* - * Shutdown NET_DMA + * Tell people we are down */ - net_dmaengine_put(); + rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING); + call_netdevice_notifiers(NETDEV_DOWN, dev); return 0; } @@ -4299,18 +4321,10 @@ unsigned dev_get_flags(const struct net_device *dev) } EXPORT_SYMBOL(dev_get_flags); -/** - * dev_change_flags - change device settings - * @dev: device - * @flags: device state flags - * - * Change settings on device based state flags. The flags are - * in the userspace exported format. - */ -int dev_change_flags(struct net_device *dev, unsigned flags) +int __dev_change_flags(struct net_device *dev, unsigned int flags) { - int ret, changes; int old_flags = dev->flags; + int ret; ASSERT_RTNL(); @@ -4341,17 +4355,12 @@ int dev_change_flags(struct net_device *dev, unsigned flags) ret = 0; if ((old_flags ^ flags) & IFF_UP) { /* Bit is different ? */ - ret = ((old_flags & IFF_UP) ? dev_close : dev_open)(dev); + ret = ((old_flags & IFF_UP) ? __dev_close : __dev_open)(dev); if (!ret) dev_set_rx_mode(dev); } - if (dev->flags & IFF_UP && - ((old_flags ^ dev->flags) & ~(IFF_UP | IFF_PROMISC | IFF_ALLMULTI | - IFF_VOLATILE))) - call_netdevice_notifiers(NETDEV_CHANGE, dev); - if ((flags ^ dev->gflags) & IFF_PROMISC) { int inc = (flags & IFF_PROMISC) ? 1 : -1; @@ -4370,11 +4379,47 @@ int dev_change_flags(struct net_device *dev, unsigned flags) dev_set_allmulti(dev, inc); } - /* Exclude state transition flags, already notified */ - changes = (old_flags ^ dev->flags) & ~(IFF_UP | IFF_RUNNING); + return ret; +} + +void __dev_notify_flags(struct net_device *dev, unsigned int old_flags) +{ + unsigned int changes = dev->flags ^ old_flags; + + if (changes & IFF_UP) { + if (dev->flags & IFF_UP) + call_netdevice_notifiers(NETDEV_UP, dev); + else + call_netdevice_notifiers(NETDEV_DOWN, dev); + } + + if (dev->flags & IFF_UP && + (changes & ~(IFF_UP | IFF_PROMISC | IFF_ALLMULTI | IFF_VOLATILE))) + call_netdevice_notifiers(NETDEV_CHANGE, dev); +} + +/** + * dev_change_flags - change device settings + * @dev: device + * @flags: device state flags + * + * Change settings on device based state flags. The flags are + * in the userspace exported format. + */ +int dev_change_flags(struct net_device *dev, unsigned flags) +{ + int ret, changes; + int old_flags = dev->flags; + + ret = __dev_change_flags(dev, flags); + if (ret < 0) + return ret; + + changes = old_flags ^ dev->flags; if (changes) rtmsg_ifinfo(RTM_NEWLINK, dev, changes); + __dev_notify_flags(dev, old_flags); return ret; } EXPORT_SYMBOL(dev_change_flags); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 020e43bfef5..c21ec4236dd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1427,8 +1427,6 @@ static int rtnetlink_event(struct notifier_block *this, unsigned long event, voi switch (event) { case NETDEV_UP: case NETDEV_DOWN: - rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING); - break; case NETDEV_PRE_UP: case NETDEV_POST_INIT: case NETDEV_REGISTER: -- cgit v1.2.3-70-g09d2 From 3729d5021257b283f7fce33d957893162ccb2c9d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Feb 2010 06:34:54 +0000 Subject: rtnetlink: support specifying device flags on device creation commit e8469ed959c373c2ff9e6f488aa5a14971aebe1f Author: Patrick McHardy Date: Tue Feb 23 20:41:30 2010 +0100 Support specifying the initial device flags when creating a device though rtnl_link. Devices allocated by rtnl_create_link() are marked as INITIALIZING in order to surpress netlink registration notifications. To complete setup, rtnl_configure_link() must be called, which performs the device flag changes and invokes the deferred notifiers if everything went well. Two examples: # add macvlan to eth0 # $ ip link add link eth0 up allmulticast on type macvlan [LINK]11: macvlan0@eth0: mtu 1500 qdisc noqueue state UNKNOWN link/ether 26:f8:84:02:f9:2a brd ff:ff:ff:ff:ff:ff [ROUTE]ff00::/8 dev macvlan0 table local metric 256 mtu 1500 advmss 1440 hoplimit 0 [ROUTE]fe80::/64 dev macvlan0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 0 [LINK]11: macvlan0@eth0: mtu 1500 link/ether 26:f8:84:02:f9:2a [ADDR]11: macvlan0 inet6 fe80::24f8:84ff:fe02:f92a/64 scope link valid_lft forever preferred_lft forever [ROUTE]local fe80::24f8:84ff:fe02:f92a via :: dev lo table local proto none metric 0 mtu 16436 advmss 16376 hoplimit 0 [ROUTE]default via fe80::215:e9ff:fef0:10f8 dev macvlan0 proto kernel metric 1024 mtu 1500 advmss 1440 hoplimit 0 [NEIGH]fe80::215:e9ff:fef0:10f8 dev macvlan0 lladdr 00:15:e9:f0:10:f8 router STALE [ROUTE]2001:6f8:974::/64 dev macvlan0 proto kernel metric 256 expires 0sec mtu 1500 advmss 1440 hoplimit 0 [PREFIX]prefix 2001:6f8:974::/64 dev macvlan0 onlink autoconf valid 14400 preferred 131084 [ADDR]11: macvlan0 inet6 2001:6f8:974:0:24f8:84ff:fe02:f92a/64 scope global dynamic valid_lft 86399sec preferred_lft 14399sec # add VLAN to eth1, eth1 is down # $ ip link add link eth1 up type vlan id 1000 RTNETLINK answers: Network is down Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- drivers/net/veth.c | 15 +++++++++----- include/net/rtnetlink.h | 2 ++ net/core/rtnetlink.c | 52 ++++++++++++++++++++++++++++++++++++++++--------- 3 files changed, 55 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 35609e64f6f..b583d4968ad 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -333,19 +333,17 @@ static int veth_newlink(struct net *src_net, struct net_device *dev, struct veth_priv *priv; char ifname[IFNAMSIZ]; struct nlattr *peer_tb[IFLA_MAX + 1], **tbp; + struct ifinfomsg *ifmp; struct net *net; /* * create and register peer first - * - * struct ifinfomsg is at the head of VETH_INFO_PEER, but we - * skip it since no info from it is useful yet */ - if (data != NULL && data[VETH_INFO_PEER] != NULL) { struct nlattr *nla_peer; nla_peer = data[VETH_INFO_PEER]; + ifmp = nla_data(nla_peer); err = nla_parse(peer_tb, IFLA_MAX, nla_data(nla_peer) + sizeof(struct ifinfomsg), nla_len(nla_peer) - sizeof(struct ifinfomsg), @@ -358,8 +356,10 @@ static int veth_newlink(struct net *src_net, struct net_device *dev, return err; tbp = peer_tb; - } else + } else { + ifmp = NULL; tbp = tb; + } if (tbp[IFLA_IFNAME]) nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ); @@ -387,6 +387,10 @@ static int veth_newlink(struct net *src_net, struct net_device *dev, netif_carrier_off(peer); + err = rtnl_configure_link(peer, ifmp); + if (err < 0) + goto err_configure_peer; + /* * register dev last * @@ -428,6 +432,7 @@ static int veth_newlink(struct net *src_net, struct net_device *dev, err_register_dev: /* nothing to do */ err_alloc_name: +err_configure_peer: unregister_netdevice(peer); return err; diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 48d3efcb088..af60fd05084 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -87,6 +87,8 @@ extern void rtnl_link_unregister(struct rtnl_link_ops *ops); extern struct net *rtnl_link_get_net(struct net *src_net, struct nlattr *tb[]); extern struct net_device *rtnl_create_link(struct net *src_net, struct net *net, char *ifname, const struct rtnl_link_ops *ops, struct nlattr *tb[]); +extern int rtnl_configure_link(struct net_device *dev, + const struct ifinfomsg *ifm); extern const struct nla_policy ifla_policy[IFLA_MAX+1]; #define MODULE_ALIAS_RTNL_LINK(kind) MODULE_ALIAS("rtnl-link-" kind) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index c21ec4236dd..d1472a42332 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -549,6 +549,19 @@ static void set_operstate(struct net_device *dev, unsigned char transition) } } +static unsigned int rtnl_dev_combine_flags(const struct net_device *dev, + const struct ifinfomsg *ifm) +{ + unsigned int flags = ifm->ifi_flags; + + /* bugwards compatibility: ifi_change == 0 is treated as ~0 */ + if (ifm->ifi_change) + flags = (flags & ifm->ifi_change) | + (dev->flags & ~ifm->ifi_change); + + return flags; +} + static void copy_rtnl_link_stats(struct rtnl_link_stats *a, const struct net_device_stats *b) { @@ -904,13 +917,7 @@ static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, } if (ifm->ifi_flags || ifm->ifi_change) { - unsigned int flags = ifm->ifi_flags; - - /* bugwards compatibility: ifi_change == 0 is treated as ~0 */ - if (ifm->ifi_change) - flags = (flags & ifm->ifi_change) | - (dev->flags & ~ifm->ifi_change); - err = dev_change_flags(dev, flags); + err = dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm)); if (err < 0) goto errout; } @@ -1053,6 +1060,26 @@ static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) return 0; } +int rtnl_configure_link(struct net_device *dev, const struct ifinfomsg *ifm) +{ + unsigned int old_flags; + int err; + + old_flags = dev->flags; + if (ifm && (ifm->ifi_flags || ifm->ifi_change)) { + err = __dev_change_flags(dev, rtnl_dev_combine_flags(dev, ifm)); + if (err < 0) + return err; + } + + dev->rtnl_link_state = RTNL_LINK_INITIALIZED; + rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U); + + __dev_notify_flags(dev, old_flags); + return 0; +} +EXPORT_SYMBOL(rtnl_configure_link); + struct net_device *rtnl_create_link(struct net *src_net, struct net *net, char *ifname, const struct rtnl_link_ops *ops, struct nlattr *tb[]) { @@ -1074,6 +1101,7 @@ struct net_device *rtnl_create_link(struct net *src_net, struct net *net, dev_net_set(dev, net); dev->rtnl_link_ops = ops; + dev->rtnl_link_state = RTNL_LINK_INITIALIZING; dev->real_num_tx_queues = real_num_queues; if (strchr(dev->name, '%')) { @@ -1203,7 +1231,7 @@ replay: if (!(nlh->nlmsg_flags & NLM_F_CREATE)) return -ENODEV; - if (ifm->ifi_index || ifm->ifi_flags || ifm->ifi_change) + if (ifm->ifi_index) return -EOPNOTSUPP; if (tb[IFLA_MAP] || tb[IFLA_MASTER] || tb[IFLA_PROTINFO]) return -EOPNOTSUPP; @@ -1234,9 +1262,15 @@ replay: err = ops->newlink(net, dev, tb, data); else err = register_netdevice(dev); - if (err < 0 && !IS_ERR(dev)) + if (err < 0 && !IS_ERR(dev)) { free_netdev(dev); + goto out; + } + err = rtnl_configure_link(dev, ifm); + if (err < 0) + unregister_netdevice(dev); +out: put_net(dest_net); return err; } -- cgit v1.2.3-70-g09d2 From da3f5cf1f8ebb0fab5c5fd09adb189166594ad6c Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 23 Feb 2010 11:45:51 +0000 Subject: skbuff: align sk_buff::cb to 64 bit and close some potential holes The alignment requirement for 64-bit load/store instructions on ARM is implementation defined. Some CPUs (such as Marvell Feroceon) do not generate an exception, if such an instruction is executed with an address that is not 64 bit aligned. In such a case, the Feroceon corrupts adjacent memory, which showed up in my tests as a crash in the rx path of ath9k that only occured with CONFIG_XFRM set. This crash happened, because the first field of the mac80211 rx status info in the cb is an u64, and changing it corrupted the skb->sp field. This patch also closes some potential pre-existing holes in the sk_buff struct surrounding the cb[] area. Signed-off-by: Felix Fietkau Cc: stable@kernel.org Signed-off-by: David S. Miller --- include/linux/skbuff.h | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ba0f8e3a9cd..d266eeef522 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -315,22 +315,23 @@ struct sk_buff { struct sk_buff *next; struct sk_buff *prev; - struct sock *sk; ktime_t tstamp; + + struct sock *sk; struct net_device *dev; - unsigned long _skb_dst; -#ifdef CONFIG_XFRM - struct sec_path *sp; -#endif /* * This is the control buffer. It is free to use for every * layer. Please put your private variables there. If you * want to keep them across layers you have to do a skb_clone() * first. This is owned by whoever has the skb queued ATM. */ - char cb[48]; + char cb[48] __aligned(8); + unsigned long _skb_dst; +#ifdef CONFIG_XFRM + struct sec_path *sp; +#endif unsigned int len, data_len; __u16 mac_len, -- cgit v1.2.3-70-g09d2 From c13854cef4751000b968d4e8ac95796562d5b96f Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 8 Feb 2010 15:27:07 +0100 Subject: Bluetooth: Convert controller hdev->type to hdev->bus The hdev->type is misnamed and should be actually hdev->bus instead. So convert it now. Signed-off-by: Marcel Holtmann --- drivers/bluetooth/bfusb.c | 2 +- drivers/bluetooth/bluecard_cs.c | 2 +- drivers/bluetooth/bpa10x.c | 2 +- drivers/bluetooth/bt3c_cs.c | 2 +- drivers/bluetooth/btmrvl_main.c | 2 +- drivers/bluetooth/btsdio.c | 2 +- drivers/bluetooth/btuart_cs.c | 2 +- drivers/bluetooth/btusb.c | 2 +- drivers/bluetooth/dtl1_cs.c | 2 +- drivers/bluetooth/hci_ldisc.c | 2 +- drivers/bluetooth/hci_vhci.c | 2 +- include/net/bluetooth/hci.h | 2 +- include/net/bluetooth/hci_core.h | 2 +- net/bluetooth/hci_core.c | 8 ++++---- net/bluetooth/hci_sysfs.c | 16 ++++++++-------- 15 files changed, 25 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/drivers/bluetooth/bfusb.c b/drivers/bluetooth/bfusb.c index 2a00707aba3..005919ab043 100644 --- a/drivers/bluetooth/bfusb.c +++ b/drivers/bluetooth/bfusb.c @@ -703,7 +703,7 @@ static int bfusb_probe(struct usb_interface *intf, const struct usb_device_id *i data->hdev = hdev; - hdev->type = HCI_USB; + hdev->bus = HCI_USB; hdev->driver_data = data; SET_HCIDEV_DEV(hdev, &intf->dev); diff --git a/drivers/bluetooth/bluecard_cs.c b/drivers/bluetooth/bluecard_cs.c index c2cf8114471..d9bf87ca9e8 100644 --- a/drivers/bluetooth/bluecard_cs.c +++ b/drivers/bluetooth/bluecard_cs.c @@ -736,7 +736,7 @@ static int bluecard_open(bluecard_info_t *info) info->hdev = hdev; - hdev->type = HCI_PCCARD; + hdev->bus = HCI_PCCARD; hdev->driver_data = info; SET_HCIDEV_DEV(hdev, &info->p_dev->dev); diff --git a/drivers/bluetooth/bpa10x.c b/drivers/bluetooth/bpa10x.c index c115285867c..d945cd12433 100644 --- a/drivers/bluetooth/bpa10x.c +++ b/drivers/bluetooth/bpa10x.c @@ -469,7 +469,7 @@ static int bpa10x_probe(struct usb_interface *intf, const struct usb_device_id * return -ENOMEM; } - hdev->type = HCI_USB; + hdev->bus = HCI_USB; hdev->driver_data = data; data->hdev = hdev; diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c index 9f5926aaf57..027cb8bf650 100644 --- a/drivers/bluetooth/bt3c_cs.c +++ b/drivers/bluetooth/bt3c_cs.c @@ -582,7 +582,7 @@ static int bt3c_open(bt3c_info_t *info) info->hdev = hdev; - hdev->type = HCI_PCCARD; + hdev->bus = HCI_PCCARD; hdev->driver_data = info; SET_HCIDEV_DEV(hdev, &info->p_dev->dev); diff --git a/drivers/bluetooth/btmrvl_main.c b/drivers/bluetooth/btmrvl_main.c index f97771ce432..53a43adf2e2 100644 --- a/drivers/bluetooth/btmrvl_main.c +++ b/drivers/bluetooth/btmrvl_main.c @@ -563,7 +563,7 @@ struct btmrvl_private *btmrvl_add_card(void *card) priv->btmrvl_dev.tx_dnld_rdy = true; - hdev->type = HCI_SDIO; + hdev->bus = HCI_SDIO; hdev->open = btmrvl_open; hdev->close = btmrvl_close; hdev->flush = btmrvl_flush; diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c index 7e298275c8f..76e5127884f 100644 --- a/drivers/bluetooth/btsdio.c +++ b/drivers/bluetooth/btsdio.c @@ -326,7 +326,7 @@ static int btsdio_probe(struct sdio_func *func, return -ENOMEM; } - hdev->type = HCI_SDIO; + hdev->bus = HCI_SDIO; hdev->driver_data = data; data->hdev = hdev; diff --git a/drivers/bluetooth/btuart_cs.c b/drivers/bluetooth/btuart_cs.c index 91c52309980..60c0953d7d0 100644 --- a/drivers/bluetooth/btuart_cs.c +++ b/drivers/bluetooth/btuart_cs.c @@ -500,7 +500,7 @@ static int btuart_open(btuart_info_t *info) info->hdev = hdev; - hdev->type = HCI_PCCARD; + hdev->bus = HCI_PCCARD; hdev->driver_data = info; SET_HCIDEV_DEV(hdev, &info->p_dev->dev); diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index a699f09ddf7..5d9cc53bd64 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -939,7 +939,7 @@ static int btusb_probe(struct usb_interface *intf, return -ENOMEM; } - hdev->type = HCI_USB; + hdev->bus = HCI_USB; hdev->driver_data = data; data->hdev = hdev; diff --git a/drivers/bluetooth/dtl1_cs.c b/drivers/bluetooth/dtl1_cs.c index 697591941e1..17788317c51 100644 --- a/drivers/bluetooth/dtl1_cs.c +++ b/drivers/bluetooth/dtl1_cs.c @@ -485,7 +485,7 @@ static int dtl1_open(dtl1_info_t *info) info->hdev = hdev; - hdev->type = HCI_PCCARD; + hdev->bus = HCI_PCCARD; hdev->driver_data = info; SET_HCIDEV_DEV(hdev, &info->p_dev->dev); diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c index aa0919386b8..76a1abb8f21 100644 --- a/drivers/bluetooth/hci_ldisc.c +++ b/drivers/bluetooth/hci_ldisc.c @@ -383,7 +383,7 @@ static int hci_uart_register_dev(struct hci_uart *hu) hu->hdev = hdev; - hdev->type = HCI_UART; + hdev->bus = HCI_UART; hdev->driver_data = hu; hdev->open = hci_uart_open; diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index 7595274103f..bb0aefdb426 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -236,7 +236,7 @@ static int vhci_open(struct inode *inode, struct file *file) data->hdev = hdev; - hdev->type = HCI_VIRTUAL; + hdev->bus = HCI_VIRTUAL; hdev->driver_data = data; hdev->open = vhci_open_dev; diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index ed3aea1605e..3350a665180 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -43,7 +43,7 @@ #define HCI_NOTIFY_CONN_DEL 2 #define HCI_NOTIFY_VOICE_SETTING 3 -/* HCI device types */ +/* HCI bus types */ #define HCI_VIRTUAL 0 #define HCI_USB 1 #define HCI_PCCARD 2 diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 7b86094a894..7e65885290d 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -70,7 +70,7 @@ struct hci_dev { char name[8]; unsigned long flags; __u16 id; - __u8 type; + __u8 bus; bdaddr_t bdaddr; __u8 dev_name[248]; __u8 dev_class[3]; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 94ba3498202..4b62ed01ddc 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -797,7 +797,7 @@ int hci_get_dev_info(void __user *arg) strcpy(di.name, hdev->name); di.bdaddr = hdev->bdaddr; - di.type = hdev->type; + di.type = hdev->bus; di.flags = hdev->flags; di.pkt_type = hdev->pkt_type; di.acl_mtu = hdev->acl_mtu; @@ -869,8 +869,8 @@ int hci_register_dev(struct hci_dev *hdev) struct list_head *head = &hci_dev_list, *p; int i, id = 0; - BT_DBG("%p name %s type %d owner %p", hdev, hdev->name, - hdev->type, hdev->owner); + BT_DBG("%p name %s bus %d owner %p", hdev, hdev->name, + hdev->bus, hdev->owner); if (!hdev->open || !hdev->close || !hdev->destruct) return -EINVAL; @@ -946,7 +946,7 @@ int hci_unregister_dev(struct hci_dev *hdev) { int i; - BT_DBG("%p name %s type %d", hdev, hdev->name, hdev->type); + BT_DBG("%p name %s bus %d", hdev, hdev->name, hdev->bus); write_lock_bh(&hci_dev_list_lock); list_del(&hdev->list); diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 2bc6f6a8de6..9b5f376813b 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -166,9 +166,9 @@ void hci_conn_del_sysfs(struct hci_conn *conn) queue_work(bt_workq, &conn->work_del); } -static inline char *host_typetostr(int type) +static inline char *host_bustostr(int bus) { - switch (type) { + switch (bus) { case HCI_VIRTUAL: return "VIRTUAL"; case HCI_USB: @@ -188,10 +188,10 @@ static inline char *host_typetostr(int type) } } -static ssize_t show_type(struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t show_bus(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); - return sprintf(buf, "%s\n", host_typetostr(hdev->type)); + return sprintf(buf, "%s\n", host_bustostr(hdev->bus)); } static ssize_t show_name(struct device *dev, struct device_attribute *attr, char *buf) @@ -355,7 +355,7 @@ static ssize_t store_sniff_min_interval(struct device *dev, struct device_attrib return count; } -static DEVICE_ATTR(type, S_IRUGO, show_type, NULL); +static DEVICE_ATTR(bus, S_IRUGO, show_bus, NULL); static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); static DEVICE_ATTR(class, S_IRUGO, show_class, NULL); static DEVICE_ATTR(address, S_IRUGO, show_address, NULL); @@ -373,7 +373,7 @@ static DEVICE_ATTR(sniff_min_interval, S_IRUGO | S_IWUSR, show_sniff_min_interval, store_sniff_min_interval); static struct attribute *bt_host_attrs[] = { - &dev_attr_type.attr, + &dev_attr_bus.attr, &dev_attr_name.attr, &dev_attr_class.attr, &dev_attr_address.attr, @@ -414,7 +414,7 @@ int hci_register_sysfs(struct hci_dev *hdev) struct device *dev = &hdev->dev; int err; - BT_DBG("%p name %s type %d", hdev, hdev->name, hdev->type); + BT_DBG("%p name %s bus %d", hdev, hdev->name, hdev->bus); dev->type = &bt_host; dev->class = bt_class; @@ -433,7 +433,7 @@ int hci_register_sysfs(struct hci_dev *hdev) void hci_unregister_sysfs(struct hci_dev *hdev) { - BT_DBG("%p name %s type %d", hdev, hdev->name, hdev->type); + BT_DBG("%p name %s bus %d", hdev, hdev->name, hdev->bus); device_del(&hdev->dev); } -- cgit v1.2.3-70-g09d2 From ca325f698996c1a0770a67f41e7dc97a007d8bc2 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 8 Feb 2010 16:22:31 +0100 Subject: Bluetooth: Convert inquiry cache to use debugfs instead of sysfs The output of the inquiry cache is only useful for debugging purposes and so move it into debugfs. Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 2 + net/bluetooth/hci_sysfs.c | 92 +++++++++++++++++++++++++++------------- 2 files changed, 64 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 7e65885290d..4c94c1e233a 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -134,6 +134,8 @@ struct hci_dev { atomic_t promisc; + struct dentry *debugfs; + struct device *parent; struct device dev; diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 9b5f376813b..f9d93f9cbcd 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -2,6 +2,7 @@ #include #include +#include #include #include @@ -9,6 +10,9 @@ struct class *bt_class = NULL; EXPORT_SYMBOL_GPL(bt_class); +struct dentry *bt_debugfs = NULL; +EXPORT_SYMBOL_GPL(bt_debugfs); + static struct workqueue_struct *bt_workq; static inline char *link_typetostr(int type) @@ -251,32 +255,6 @@ static ssize_t show_hci_revision(struct device *dev, struct device_attribute *at return sprintf(buf, "%d\n", hdev->hci_rev); } -static ssize_t show_inquiry_cache(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct hci_dev *hdev = dev_get_drvdata(dev); - struct inquiry_cache *cache = &hdev->inq_cache; - struct inquiry_entry *e; - int n = 0; - - hci_dev_lock_bh(hdev); - - for (e = cache->list; e; e = e->next) { - struct inquiry_data *data = &e->data; - bdaddr_t bdaddr; - baswap(&bdaddr, &data->bdaddr); - n += sprintf(buf + n, "%s %d %d %d 0x%.2x%.2x%.2x 0x%.4x %d %d %u\n", - batostr(&bdaddr), - data->pscan_rep_mode, data->pscan_period_mode, - data->pscan_mode, data->dev_class[2], - data->dev_class[1], data->dev_class[0], - __le16_to_cpu(data->clock_offset), - data->rssi, data->ssp_mode, e->timestamp); - } - - hci_dev_unlock_bh(hdev); - return n; -} - static ssize_t show_idle_timeout(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); @@ -363,7 +341,6 @@ static DEVICE_ATTR(features, S_IRUGO, show_features, NULL); static DEVICE_ATTR(manufacturer, S_IRUGO, show_manufacturer, NULL); static DEVICE_ATTR(hci_version, S_IRUGO, show_hci_version, NULL); static DEVICE_ATTR(hci_revision, S_IRUGO, show_hci_revision, NULL); -static DEVICE_ATTR(inquiry_cache, S_IRUGO, show_inquiry_cache, NULL); static DEVICE_ATTR(idle_timeout, S_IRUGO | S_IWUSR, show_idle_timeout, store_idle_timeout); @@ -381,7 +358,6 @@ static struct attribute *bt_host_attrs[] = { &dev_attr_manufacturer.attr, &dev_attr_hci_version.attr, &dev_attr_hci_revision.attr, - &dev_attr_inquiry_cache.attr, &dev_attr_idle_timeout.attr, &dev_attr_sniff_max_interval.attr, &dev_attr_sniff_min_interval.attr, @@ -409,6 +385,46 @@ static struct device_type bt_host = { .release = bt_host_release, }; +static int inquiry_cache_open(struct inode *inode, struct file *file) +{ + file->private_data = inode->i_private; + return 0; +} + +static ssize_t inquiry_cache_read(struct file *file, char __user *userbuf, + size_t count, loff_t *ppos) +{ + struct hci_dev *hdev = file->private_data; + struct inquiry_cache *cache = &hdev->inq_cache; + struct inquiry_entry *e; + char buf[4096]; + int n = 0; + + hci_dev_lock_bh(hdev); + + for (e = cache->list; e; e = e->next) { + struct inquiry_data *data = &e->data; + bdaddr_t bdaddr; + baswap(&bdaddr, &data->bdaddr); + n += sprintf(buf + n, "%s %d %d %d 0x%.2x%.2x%.2x 0x%.4x %d %d %u\n", + batostr(&bdaddr), + data->pscan_rep_mode, data->pscan_period_mode, + data->pscan_mode, data->dev_class[2], + data->dev_class[1], data->dev_class[0], + __le16_to_cpu(data->clock_offset), + data->rssi, data->ssp_mode, e->timestamp); + } + + hci_dev_unlock_bh(hdev); + + return simple_read_from_buffer(userbuf, count, ppos, buf, n); +} + +static const struct file_operations inquiry_cache_fops = { + .open = inquiry_cache_open, + .read = inquiry_cache_read, +}; + int hci_register_sysfs(struct hci_dev *hdev) { struct device *dev = &hdev->dev; @@ -428,6 +444,16 @@ int hci_register_sysfs(struct hci_dev *hdev) if (err < 0) return err; + if (!bt_debugfs) + return 0; + + hdev->debugfs = debugfs_create_dir(hdev->name, bt_debugfs); + if (!hdev->debugfs) + return 0; + + debugfs_create_file("inquiry_cache", 0444, hdev->debugfs, + hdev, &inquiry_cache_fops); + return 0; } @@ -435,6 +461,8 @@ void hci_unregister_sysfs(struct hci_dev *hdev) { BT_DBG("%p name %s bus %d", hdev, hdev->name, hdev->bus); + debugfs_remove_recursive(hdev->debugfs); + device_del(&hdev->dev); } @@ -444,6 +472,8 @@ int __init bt_sysfs_init(void) if (!bt_workq) return -ENOMEM; + bt_debugfs = debugfs_create_dir("bluetooth", NULL); + bt_class = class_create(THIS_MODULE, "bluetooth"); if (IS_ERR(bt_class)) { destroy_workqueue(bt_workq); @@ -455,7 +485,9 @@ int __init bt_sysfs_init(void) void bt_sysfs_cleanup(void) { - destroy_workqueue(bt_workq); - class_destroy(bt_class); + + debugfs_remove_recursive(bt_debugfs); + + destroy_workqueue(bt_workq); } -- cgit v1.2.3-70-g09d2 From 943da25d95c7e8fd8c39dbf09e030f5da46f5d85 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 13 Feb 2010 02:28:41 +0100 Subject: Bluetooth: Add controller types for BR/EDR and 802.11 AMP With the Bluetooth 3.0 specification and the introduction of alternate MAC/PHY (AMP) support, it is required to differentiate between primary BR/EDR controllers and 802.11 AMP controllers. So introduce a special type inside HCI device for differentiation. For now all AMP controllers will be treated as raw devices until an AMP manager has been implemented. Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 4 ++++ include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_core.c | 6 +++++- net/bluetooth/hci_sysfs.c | 20 ++++++++++++++++++++ 4 files changed, 30 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 3350a665180..fc0c502d9fd 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -52,6 +52,10 @@ #define HCI_PCI 5 #define HCI_SDIO 6 +/* HCI controller types */ +#define HCI_BREDR 0x00 +#define HCI_80211 0x01 + /* HCI device quirks */ enum { HCI_QUIRK_NO_RESET, diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 4c94c1e233a..ce3c99e5fa2 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -71,6 +71,7 @@ struct hci_dev { unsigned long flags; __u16 id; __u8 bus; + __u8 dev_type; bdaddr_t bdaddr; __u8 dev_name[248]; __u8 dev_class[3]; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 4b62ed01ddc..4ad23192c7a 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -491,6 +491,10 @@ int hci_dev_open(__u16 dev) if (test_bit(HCI_QUIRK_RAW_DEVICE, &hdev->quirks)) set_bit(HCI_RAW, &hdev->flags); + /* Treat all non BR/EDR controllers as raw devices for now */ + if (hdev->dev_type != HCI_BREDR) + set_bit(HCI_RAW, &hdev->flags); + if (hdev->open(hdev)) { ret = -EIO; goto done; @@ -797,7 +801,7 @@ int hci_get_dev_info(void __user *arg) strcpy(di.name, hdev->name); di.bdaddr = hdev->bdaddr; - di.type = hdev->bus; + di.type = (hdev->bus & 0x0f) | (hdev->dev_type << 4); di.flags = hdev->flags; di.pkt_type = hdev->pkt_type; di.acl_mtu = hdev->acl_mtu; diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index f9d93f9cbcd..1a79a6c7e30 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -192,12 +192,30 @@ static inline char *host_bustostr(int bus) } } +static inline char *host_typetostr(int type) +{ + switch (type) { + case HCI_BREDR: + return "BR/EDR"; + case HCI_80211: + return "802.11"; + default: + return "UNKNOWN"; + } +} + static ssize_t show_bus(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); return sprintf(buf, "%s\n", host_bustostr(hdev->bus)); } +static ssize_t show_type(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct hci_dev *hdev = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", host_typetostr(hdev->dev_type)); +} + static ssize_t show_name(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); @@ -334,6 +352,7 @@ static ssize_t store_sniff_min_interval(struct device *dev, struct device_attrib } static DEVICE_ATTR(bus, S_IRUGO, show_bus, NULL); +static DEVICE_ATTR(type, S_IRUGO, show_type, NULL); static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); static DEVICE_ATTR(class, S_IRUGO, show_class, NULL); static DEVICE_ATTR(address, S_IRUGO, show_address, NULL); @@ -351,6 +370,7 @@ static DEVICE_ATTR(sniff_min_interval, S_IRUGO | S_IWUSR, static struct attribute *bt_host_attrs[] = { &dev_attr_bus.attr, + &dev_attr_type.attr, &dev_attr_name.attr, &dev_attr_class.attr, &dev_attr_address.attr, -- cgit v1.2.3-70-g09d2 From a2ce766238f72ff7337606c0bc96803c30c9e05c Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Fri, 26 Feb 2010 14:04:39 +0000 Subject: pci: Add PCI LRDT tag size and section size This patch adds a preprocessor constant to describe the PCI VPD large resource data type tag size and an inline function to extract the large resource section size from the large resource data type tag. Signed-off-by: Matt Carlson Signed-off-by: Michael Chan Acked-by: Jesse Barnes Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 8 +++++--- drivers/net/tg3.c | 13 ++++++------- include/linux/pci.h | 14 ++++++++++++++ 3 files changed, 25 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index d3f739a295d..bb403887b54 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -7773,15 +7773,17 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) unsigned int block_end; if (val == 0x82 || val == 0x91) { - i = (i + 3 + (data[i + 1] + (data[i + 2] << 8))); + i += PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&data[i]); continue; } if (val != 0x90) goto vpd_done; - block_end = (i + 3 + (data[i + 1] + (data[i + 2] << 8))); - i += 3; + block_end = (i + PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&data[i])); + i += PCI_VPD_LRDT_TAG_SIZE; if (block_end > BNX2_VPD_LEN) goto vpd_done; diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 0d06e4007f4..5fccbe45994 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -12590,19 +12590,18 @@ static void __devinit tg3_read_partno(struct tg3 *tp) unsigned int block_end; if (val == 0x82 || val == 0x91) { - i = (i + 3 + - (vpd_data[i + 1] + - (vpd_data[i + 2] << 8))); + i += PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&vpd_data[i]); continue; } if (val != 0x90) goto out_not_found; - block_end = (i + 3 + - (vpd_data[i + 1] + - (vpd_data[i + 2] << 8))); - i += 3; + block_end = i + PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&vpd_data[i]); + + i += PCI_VPD_LRDT_TAG_SIZE; if (block_end > TG3_NVM_VPD_LEN) goto out_not_found; diff --git a/include/linux/pci.h b/include/linux/pci.h index e2575f86133..6f62a499023 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1349,5 +1349,19 @@ static inline bool pci_is_pcie(struct pci_dev *dev) void pci_request_acs(void); + +#define PCI_VPD_LRDT_TAG_SIZE 3 + +/** + * pci_vpd_lrdt_size - Extracts the Large Resource Data Type length + * @lrdt: Pointer to the beginning of the Large Resource Data Type tag + * + * Returns the extracted Large Resource Data Type length. + */ +static inline u16 pci_vpd_lrdt_size(const u8 *lrdt) +{ + return (u16)lrdt[1] + ((u16)lrdt[2] << 8); +} + #endif /* __KERNEL__ */ #endif /* LINUX_PCI_H */ -- cgit v1.2.3-70-g09d2 From 7ad506fa1adc2da3d394c562f09b8e1b3026c402 Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Fri, 26 Feb 2010 14:04:40 +0000 Subject: pci: Add large and small resource data type code This patch introduces more VPD preprocessor definitions to identify some small and large resource data type item names. The patch then continues to correct how the tg3 and bnx2 drivers search for the "read-only data" large resource data type. Signed-off-by: Matt Carlson Signed-off-by: Michael Chan Acked-by: Jesse Barnes Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 23 +++++++++++++++++------ drivers/net/tg3.c | 23 +++++++++++++++++------ include/linux/pci.h | 34 +++++++++++++++++++++++++++++++++- 3 files changed, 67 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index bb403887b54..084ef102b8c 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -7772,15 +7772,26 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) unsigned char val = data[i]; unsigned int block_end; - if (val == 0x82 || val == 0x91) { - i += PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&data[i]); + if (val & PCI_VPD_LRDT) { + if (i + PCI_VPD_LRDT_TAG_SIZE > BNX2_VPD_LEN) + break; + + if (val != PCI_VPD_LRDT_RO_DATA) { + i += PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&data[i]); + + continue; + } + } else { + if ((val & PCI_VPD_SRDT_TIN_MASK) == PCI_VPD_STIN_END) + break; + + i += PCI_VPD_SRDT_TAG_SIZE + + pci_vpd_srdt_size(&data[i]); + continue; } - if (val != 0x90) - goto vpd_done; - block_end = (i + PCI_VPD_LRDT_TAG_SIZE + pci_vpd_lrdt_size(&data[i])); i += PCI_VPD_LRDT_TAG_SIZE; diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 5fccbe45994..ed57a62b3ac 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -12589,15 +12589,26 @@ static void __devinit tg3_read_partno(struct tg3 *tp) unsigned char val = vpd_data[i]; unsigned int block_end; - if (val == 0x82 || val == 0x91) { - i += PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&vpd_data[i]); + if (val & PCI_VPD_LRDT) { + if (i + PCI_VPD_LRDT_TAG_SIZE > TG3_NVM_VPD_LEN) + break; + + if (val != PCI_VPD_LRDT_RO_DATA) { + i += PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&vpd_data[i]); + + continue; + } + } else { + if ((val & PCI_VPD_SRDT_TIN_MASK) == PCI_VPD_STIN_END) + break; + + i += PCI_VPD_SRDT_TAG_SIZE + + pci_vpd_srdt_size(&vpd_data[i]); + continue; } - if (val != 0x90) - goto out_not_found; - block_end = i + PCI_VPD_LRDT_TAG_SIZE + pci_vpd_lrdt_size(&vpd_data[i]); diff --git a/include/linux/pci.h b/include/linux/pci.h index 6f62a499023..198d062640b 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1350,7 +1350,28 @@ static inline bool pci_is_pcie(struct pci_dev *dev) void pci_request_acs(void); -#define PCI_VPD_LRDT_TAG_SIZE 3 +#define PCI_VPD_LRDT 0x80 /* Large Resource Data Type */ +#define PCI_VPD_LRDT_ID(x) (x | PCI_VPD_LRDT) + +/* Large Resource Data Type Tag Item Names */ +#define PCI_VPD_LTIN_ID_STRING 0x02 /* Identifier String */ +#define PCI_VPD_LTIN_RO_DATA 0x10 /* Read-Only Data */ +#define PCI_VPD_LTIN_RW_DATA 0x11 /* Read-Write Data */ + +#define PCI_VPD_LRDT_ID_STRING PCI_VPD_LRDT_ID(PCI_VPD_LTIN_ID_STRING) +#define PCI_VPD_LRDT_RO_DATA PCI_VPD_LRDT_ID(PCI_VPD_LTIN_RO_DATA) +#define PCI_VPD_LRDT_RW_DATA PCI_VPD_LRDT_ID(PCI_VPD_LTIN_RW_DATA) + +/* Small Resource Data Type Tag Item Names */ +#define PCI_VPD_STIN_END 0x78 /* End */ + +#define PCI_VPD_SRDT_END PCI_VPD_STIN_END + +#define PCI_VPD_SRDT_TIN_MASK 0x78 +#define PCI_VPD_SRDT_LEN_MASK 0x07 + +#define PCI_VPD_LRDT_TAG_SIZE 3 +#define PCI_VPD_SRDT_TAG_SIZE 1 /** * pci_vpd_lrdt_size - Extracts the Large Resource Data Type length @@ -1363,5 +1384,16 @@ static inline u16 pci_vpd_lrdt_size(const u8 *lrdt) return (u16)lrdt[1] + ((u16)lrdt[2] << 8); } +/** + * pci_vpd_srdt_size - Extracts the Small Resource Data Type length + * @lrdt: Pointer to the beginning of the Small Resource Data Type tag + * + * Returns the extracted Small Resource Data Type length. + */ +static inline u8 pci_vpd_srdt_size(const u8 *srdt) +{ + return (*srdt) & PCI_VPD_SRDT_LEN_MASK; +} + #endif /* __KERNEL__ */ #endif /* LINUX_PCI_H */ -- cgit v1.2.3-70-g09d2 From b55ac1b22690d2e5b02a61cf6d69c2d66969c79d Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Fri, 26 Feb 2010 14:04:41 +0000 Subject: pci: Add helper to find a VPD resource data type This patch adds the pci_vpd_find_tag() helper function to find VPD resource data types in a buffer. Signed-off-by: Matt Carlson Signed-off-by: Michael Chan Acked-by: Jesse Barnes Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 24 ++++-------------------- drivers/net/tg3.c | 26 +++++--------------------- drivers/pci/Makefile | 2 +- drivers/pci/vpd.c | 43 +++++++++++++++++++++++++++++++++++++++++++ include/linux/pci.h | 12 ++++++++++++ 5 files changed, 65 insertions(+), 42 deletions(-) create mode 100644 drivers/pci/vpd.c (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 084ef102b8c..fd43feb5a35 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -7769,28 +7769,12 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) } for (i = 0; i <= BNX2_VPD_LEN - 3; ) { - unsigned char val = data[i]; unsigned int block_end; - if (val & PCI_VPD_LRDT) { - if (i + PCI_VPD_LRDT_TAG_SIZE > BNX2_VPD_LEN) - break; - - if (val != PCI_VPD_LRDT_RO_DATA) { - i += PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&data[i]); - - continue; - } - } else { - if ((val & PCI_VPD_SRDT_TIN_MASK) == PCI_VPD_STIN_END) - break; - - i += PCI_VPD_SRDT_TAG_SIZE + - pci_vpd_srdt_size(&data[i]); - - continue; - } + i = pci_vpd_find_tag(data, i, BNX2_VPD_LEN, + PCI_VPD_LRDT_RO_DATA); + if (i < 0) + break; block_end = (i + PCI_VPD_LRDT_TAG_SIZE + pci_vpd_lrdt_size(&data[i])); diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index ed57a62b3ac..76ad141ab44 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -12547,7 +12547,7 @@ skip_phy_reset: static void __devinit tg3_read_partno(struct tg3 *tp) { unsigned char vpd_data[TG3_NVM_VPD_LEN]; /* in little-endian format */ - unsigned int i; + int i; u32 magic; if ((tp->tg3_flags3 & TG3_FLG3_NO_NVRAM) || @@ -12586,28 +12586,12 @@ static void __devinit tg3_read_partno(struct tg3 *tp) /* Now parse and find the part number. */ for (i = 0; i < TG3_NVM_VPD_LEN - 2; ) { - unsigned char val = vpd_data[i]; unsigned int block_end; - if (val & PCI_VPD_LRDT) { - if (i + PCI_VPD_LRDT_TAG_SIZE > TG3_NVM_VPD_LEN) - break; - - if (val != PCI_VPD_LRDT_RO_DATA) { - i += PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&vpd_data[i]); - - continue; - } - } else { - if ((val & PCI_VPD_SRDT_TIN_MASK) == PCI_VPD_STIN_END) - break; - - i += PCI_VPD_SRDT_TAG_SIZE + - pci_vpd_srdt_size(&vpd_data[i]); - - continue; - } + i = pci_vpd_find_tag(vpd_data, i, TG3_NVM_VPD_LEN, + PCI_VPD_LRDT_RO_DATA); + if (i < 0) + break; block_end = i + PCI_VPD_LRDT_TAG_SIZE + pci_vpd_lrdt_size(&vpd_data[i]); diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile index 4df48d58eaa..b2f6d777a08 100644 --- a/drivers/pci/Makefile +++ b/drivers/pci/Makefile @@ -4,7 +4,7 @@ obj-y += access.o bus.o probe.o remove.o pci.o quirks.o \ pci-driver.o search.o pci-sysfs.o rom.o setup-res.o \ - irq.o + irq.o vpd.o obj-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_SYSFS) += slot.o diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c new file mode 100644 index 00000000000..6bc554576f8 --- /dev/null +++ b/drivers/pci/vpd.c @@ -0,0 +1,43 @@ +/* + * File: vpd.c + * Purpose: Provide PCI VPD support + * + * Copyright (C) 2010 Broadcom Corporation. + */ + +#include + +int pci_vpd_find_tag(const u8 *buf, unsigned int off, unsigned int len, u8 rdt) +{ + int i; + + for (i = off; i < len; ) { + u8 val = buf[i]; + + if (val & PCI_VPD_LRDT) { + /* Don't return success of the tag isn't complete */ + if (i + PCI_VPD_LRDT_TAG_SIZE > len) + break; + + if (val == rdt) + return i; + + i += PCI_VPD_LRDT_TAG_SIZE + + pci_vpd_lrdt_size(&buf[i]); + } else { + u8 tag = val & ~PCI_VPD_SRDT_LEN_MASK; + + if (tag == rdt) + return i; + + if (tag == PCI_VPD_SRDT_END) + break; + + i += PCI_VPD_SRDT_TAG_SIZE + + pci_vpd_srdt_size(&buf[i]); + } + } + + return -ENOENT; +} +EXPORT_SYMBOL_GPL(pci_vpd_find_tag); diff --git a/include/linux/pci.h b/include/linux/pci.h index 198d062640b..e30ceea7345 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1395,5 +1395,17 @@ static inline u8 pci_vpd_srdt_size(const u8 *srdt) return (*srdt) & PCI_VPD_SRDT_LEN_MASK; } +/** + * pci_vpd_find_tag - Locates the Resource Data Type tag provided + * @buf: Pointer to buffered vpd data + * @off: The offset into the buffer at which to begin the search + * @len: The length of the vpd buffer + * @rdt: The Resource Data Type to search for + * + * Returns the index where the Resource Data Type was found or + * -ENOENT otherwise. + */ +int pci_vpd_find_tag(const u8 *buf, unsigned int off, unsigned int len, u8 rdt); + #endif /* __KERNEL__ */ #endif /* LINUX_PCI_H */ -- cgit v1.2.3-70-g09d2 From e1d5bdabb94da89bdb3c3f2ee105cf61fca88ec8 Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Fri, 26 Feb 2010 14:04:42 +0000 Subject: pci: Add VPD information field helper functions This patch adds a preprocessor constant to describe the PCI VPD information field header size and an inline function to extract the size of the information field itself. Signed-off-by: Matt Carlson Signed-off-by: Michael Chan Acked-by: Jesse Barnes Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 11 ++++++----- drivers/net/tg3.c | 7 ++++--- include/linux/pci.h | 13 +++++++++++++ 3 files changed, 23 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index fd43feb5a35..b808707f83f 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -7784,14 +7784,15 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) goto vpd_done; while (i < (block_end - 2)) { - int len = data[i + 2]; + int len = pci_vpd_info_field_size(&data[i]); - if (i + 3 + len > block_end) + if (i + PCI_VPD_INFO_FLD_HDR_SIZE + len > block_end) goto vpd_done; if (data[i] == 'M' && data[i + 1] == 'N') { if (len != 4 || - memcmp(&data[i + 3], "1028", 4)) + memcmp(&data[i + PCI_VPD_INFO_FLD_HDR_SIZE], + "1028", 4)) goto vpd_done; mn_match = true; @@ -7800,9 +7801,9 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) goto vpd_done; v0_len = len; - v0_str = &data[i + 3]; + v0_str = &data[i + PCI_VPD_INFO_FLD_HDR_SIZE]; } - i += 3 + len; + i += PCI_VPD_INFO_FLD_HDR_SIZE + len; if (mn_match && v0_str) { memcpy(bp->fw_version, v0_str, v0_len); diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 76ad141ab44..f59f36910e9 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -12604,9 +12604,9 @@ static void __devinit tg3_read_partno(struct tg3 *tp) while (i < (block_end - 2)) { if (vpd_data[i + 0] == 'P' && vpd_data[i + 1] == 'N') { - int partno_len = vpd_data[i + 2]; + int partno_len = pci_vpd_info_field_size(&vpd_data[i]); - i += 3; + i += PCI_VPD_INFO_FLD_HDR_SIZE; if (partno_len > TG3_BPN_SIZE || (partno_len + i) > TG3_NVM_VPD_LEN) goto out_not_found; @@ -12617,7 +12617,8 @@ static void __devinit tg3_read_partno(struct tg3 *tp) /* Success. */ return; } - i += 3 + vpd_data[i + 2]; + i += PCI_VPD_INFO_FLD_HDR_SIZE + + pci_vpd_info_field_size(&vpd_data[i]); } /* Part number not found. */ diff --git a/include/linux/pci.h b/include/linux/pci.h index e30ceea7345..cfff32fc6e3 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1373,6 +1373,8 @@ void pci_request_acs(void); #define PCI_VPD_LRDT_TAG_SIZE 3 #define PCI_VPD_SRDT_TAG_SIZE 1 +#define PCI_VPD_INFO_FLD_HDR_SIZE 3 + /** * pci_vpd_lrdt_size - Extracts the Large Resource Data Type length * @lrdt: Pointer to the beginning of the Large Resource Data Type tag @@ -1395,6 +1397,17 @@ static inline u8 pci_vpd_srdt_size(const u8 *srdt) return (*srdt) & PCI_VPD_SRDT_LEN_MASK; } +/** + * pci_vpd_info_field_size - Extracts the information field length + * @lrdt: Pointer to the beginning of an information field header + * + * Returns the extracted information field length. + */ +static inline u8 pci_vpd_info_field_size(const u8 *info_field) +{ + return info_field[2]; +} + /** * pci_vpd_find_tag - Locates the Resource Data Type tag provided * @buf: Pointer to buffered vpd data -- cgit v1.2.3-70-g09d2 From 4067a8541d397e9d6b443dd2ce0ecb78bfd991db Mon Sep 17 00:00:00 2001 From: Matt Carlson Date: Fri, 26 Feb 2010 14:04:43 +0000 Subject: pci: Add helper to search for VPD keywords This patch adds the pci_vpd_find_info_keyword() helper function to find information field keywords within read-only and read-write large resource data type sections. Signed-off-by: Matt Carlson Signed-off-by: Michael Chan Acked-by: Jesse Barnes Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 54 +++++++++++++++++++++++++++++------------------------ drivers/net/tg3.c | 32 +++++++++++++------------------ drivers/pci/vpd.c | 18 ++++++++++++++++++ include/linux/pci.h | 17 +++++++++++++++++ 4 files changed, 78 insertions(+), 43 deletions(-) (limited to 'include') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index b808707f83f..bc23bfb687c 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -7769,48 +7769,54 @@ bnx2_read_vpd_fw_ver(struct bnx2 *bp) } for (i = 0; i <= BNX2_VPD_LEN - 3; ) { - unsigned int block_end; + int j; + unsigned int block_end, rosize; i = pci_vpd_find_tag(data, i, BNX2_VPD_LEN, PCI_VPD_LRDT_RO_DATA); if (i < 0) break; - block_end = (i + PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&data[i])); + rosize = pci_vpd_lrdt_size(&data[i]); + block_end = i + PCI_VPD_LRDT_TAG_SIZE + rosize; i += PCI_VPD_LRDT_TAG_SIZE; if (block_end > BNX2_VPD_LEN) goto vpd_done; - while (i < (block_end - 2)) { - int len = pci_vpd_info_field_size(&data[i]); + j = pci_vpd_find_info_keyword(data, i, rosize, + PCI_VPD_RO_KEYWORD_MFR_ID); + if (j > 0) { + int len = pci_vpd_info_field_size(&data[j]); - if (i + PCI_VPD_INFO_FLD_HDR_SIZE + len > block_end) + if (j + PCI_VPD_INFO_FLD_HDR_SIZE + len > block_end || + len != 4 || + memcmp(&data[j + PCI_VPD_INFO_FLD_HDR_SIZE], + "1028", 4)) goto vpd_done; - if (data[i] == 'M' && data[i + 1] == 'N') { - if (len != 4 || - memcmp(&data[i + PCI_VPD_INFO_FLD_HDR_SIZE], - "1028", 4)) - goto vpd_done; - mn_match = true; + mn_match = true; + } - } else if (data[i] == 'V' && data[i + 1] == '0') { - if (len > BNX2_MAX_VER_SLEN) - goto vpd_done; + j = pci_vpd_find_info_keyword(data, i, rosize, + PCI_VPD_RO_KEYWORD_VENDOR0); + if (j > 0) { + int len = pci_vpd_info_field_size(&data[j]); - v0_len = len; - v0_str = &data[i + PCI_VPD_INFO_FLD_HDR_SIZE]; - } - i += PCI_VPD_INFO_FLD_HDR_SIZE + len; - - if (mn_match && v0_str) { - memcpy(bp->fw_version, v0_str, v0_len); - bp->fw_version[v0_len] = ' '; + j += PCI_VPD_INFO_FLD_HDR_SIZE; + if (j + len > block_end || len > BNX2_MAX_VER_SLEN) goto vpd_done; - } + + v0_len = len; + v0_str = &data[j]; } + + if (mn_match && v0_str) { + memcpy(bp->fw_version, v0_str, v0_len); + bp->fw_version[v0_len] = ' '; + goto vpd_done; + } + goto vpd_done; } diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index f59f36910e9..204c565caa5 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -12586,39 +12586,33 @@ static void __devinit tg3_read_partno(struct tg3 *tp) /* Now parse and find the part number. */ for (i = 0; i < TG3_NVM_VPD_LEN - 2; ) { - unsigned int block_end; + unsigned int block_end, rosize; i = pci_vpd_find_tag(vpd_data, i, TG3_NVM_VPD_LEN, PCI_VPD_LRDT_RO_DATA); if (i < 0) break; - block_end = i + PCI_VPD_LRDT_TAG_SIZE + - pci_vpd_lrdt_size(&vpd_data[i]); - + rosize = pci_vpd_lrdt_size(&vpd_data[i]); + block_end = i + PCI_VPD_LRDT_TAG_SIZE + rosize; i += PCI_VPD_LRDT_TAG_SIZE; if (block_end > TG3_NVM_VPD_LEN) goto out_not_found; - while (i < (block_end - 2)) { - if (vpd_data[i + 0] == 'P' && - vpd_data[i + 1] == 'N') { - int partno_len = pci_vpd_info_field_size(&vpd_data[i]); + i = pci_vpd_find_info_keyword(vpd_data, i, rosize, + PCI_VPD_RO_KEYWORD_PARTNO); + if (i > 0) { + u8 len = pci_vpd_info_field_size(&vpd_data[i]); - i += PCI_VPD_INFO_FLD_HDR_SIZE; - if (partno_len > TG3_BPN_SIZE || - (partno_len + i) > TG3_NVM_VPD_LEN) - goto out_not_found; + i += PCI_VPD_INFO_FLD_HDR_SIZE; + if (len > TG3_BPN_SIZE || + (len + i) > TG3_NVM_VPD_LEN) + break; - memcpy(tp->board_part_number, - &vpd_data[i], partno_len); + memcpy(tp->board_part_number, &vpd_data[i], len); - /* Success. */ - return; - } - i += PCI_VPD_INFO_FLD_HDR_SIZE + - pci_vpd_info_field_size(&vpd_data[i]); + return; } /* Part number not found. */ diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c index 6bc554576f8..a5a5ca17cfe 100644 --- a/drivers/pci/vpd.c +++ b/drivers/pci/vpd.c @@ -41,3 +41,21 @@ int pci_vpd_find_tag(const u8 *buf, unsigned int off, unsigned int len, u8 rdt) return -ENOENT; } EXPORT_SYMBOL_GPL(pci_vpd_find_tag); + +int pci_vpd_find_info_keyword(const u8 *buf, unsigned int off, + unsigned int len, const char *kw) +{ + int i; + + for (i = off; i + PCI_VPD_INFO_FLD_HDR_SIZE <= off + len;) { + if (buf[i + 0] == kw[0] && + buf[i + 1] == kw[1]) + return i; + + i += PCI_VPD_INFO_FLD_HDR_SIZE + + pci_vpd_info_field_size(&buf[i]); + } + + return -ENOENT; +} +EXPORT_SYMBOL_GPL(pci_vpd_find_info_keyword); diff --git a/include/linux/pci.h b/include/linux/pci.h index cfff32fc6e3..1f4a52131c9 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1375,6 +1375,10 @@ void pci_request_acs(void); #define PCI_VPD_INFO_FLD_HDR_SIZE 3 +#define PCI_VPD_RO_KEYWORD_PARTNO "PN" +#define PCI_VPD_RO_KEYWORD_MFR_ID "MN" +#define PCI_VPD_RO_KEYWORD_VENDOR0 "V0" + /** * pci_vpd_lrdt_size - Extracts the Large Resource Data Type length * @lrdt: Pointer to the beginning of the Large Resource Data Type tag @@ -1420,5 +1424,18 @@ static inline u8 pci_vpd_info_field_size(const u8 *info_field) */ int pci_vpd_find_tag(const u8 *buf, unsigned int off, unsigned int len, u8 rdt); +/** + * pci_vpd_find_info_keyword - Locates an information field keyword in the VPD + * @buf: Pointer to buffered vpd data + * @off: The offset into the buffer at which to begin the search + * @len: The length of the buffer area, relative to off, in which to search + * @kw: The keyword to search for + * + * Returns the index where the information field keyword was found or + * -ENOENT otherwise. + */ +int pci_vpd_find_info_keyword(const u8 *buf, unsigned int off, + unsigned int len, const char *kw); + #endif /* __KERNEL__ */ #endif /* LINUX_PCI_H */ -- cgit v1.2.3-70-g09d2 From 4ab408dea0f0dba4dec0555f4f35b7ae703f5e91 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 1 Mar 2010 03:09:26 +0000 Subject: net: fix protocol sk_buff field Commit e992cd9b72a18 (kmemcheck: make bitfield annotations truly no-ops when disabled) allows us to revert a workaround we did in the past to not add holes in sk_buff structure. This patch partially reverts commit 14d18a81b5171 (net: fix kmemcheck annotations) so that sparse doesnt complain: include/linux/skbuff.h:357:41: error: invalid bitfield specifier for type restricted __be16. Reported-by: Stefan Richter Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index d266eeef522..03f816a9b65 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -355,8 +355,8 @@ struct sk_buff { ipvs_property:1, peeked:1, nf_trace:1; - __be16 protocol:16; kmemcheck_bitfield_end(flags1); + __be16 protocol; void (*destructor)(struct sk_buff *skb); #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) -- cgit v1.2.3-70-g09d2