summaryrefslogtreecommitdiffstats
path: root/net/dcb/dcbnl.c
diff options
context:
space:
mode:
authorThomas Graf <tgraf@suug.ch>2012-06-13 02:54:54 +0000
committerDavid S. Miller <davem@davemloft.net>2012-06-13 15:46:34 -0700
commit33a03aadb52fa05d28aba6d8f0c03c7b3b905897 (patch)
tree4e8fdc1321d516c9a297ece6babde6f164475c80 /net/dcb/dcbnl.c
parent43b03f1f6d6832d744918947d185a7aee89d1e0f (diff)
dcbnl: Prepare framework to shorten handling functions
There is no need to allocate and send the reply message in each handling function separately. Instead, the reply skb can be allocated and sent in dcb_doit() directly. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/dcb/dcbnl.c')
-rw-r--r--net/dcb/dcbnl.c71
1 files changed, 71 insertions, 0 deletions
diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c
index 656c7c75b19..5520e431b07 100644
--- a/net/dcb/dcbnl.c
+++ b/net/dcb/dcbnl.c
@@ -196,6 +196,34 @@ static const struct nla_policy dcbnl_featcfg_nest[DCB_FEATCFG_ATTR_MAX + 1] = {
static LIST_HEAD(dcb_app_list);
static DEFINE_SPINLOCK(dcb_lock);
+static struct sk_buff *dcbnl_newmsg(int type, u8 cmd, u32 port, u32 seq,
+ u32 flags, struct nlmsghdr **nlhp)
+{
+ struct sk_buff *skb;
+ struct dcbmsg *dcb;
+ struct nlmsghdr *nlh;
+
+ skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+ if (!skb)
+ return NULL;
+
+ nlh = nlmsg_put(skb, port, seq, type, sizeof(*dcb), flags);
+ if (!nlh) {
+ /* header should always fit, allocation must be buggy */
+ BUG();
+ }
+
+ dcb = nlmsg_data(nlh);
+ dcb->dcb_family = AF_UNSPEC;
+ dcb->cmd = cmd;
+ dcb->dcb_pad = 0;
+
+ if (nlhp)
+ *nlhp = nlh;
+
+ return skb;
+}
+
/* standard netlink reply call */
static int dcbnl_reply(u8 value, u8 event, u8 cmd, u8 attr, u32 pid,
u32 seq, u16 flags)
@@ -1922,6 +1950,19 @@ static int dcbnl_cee_get(struct net_device *netdev, struct nlattr **tb,
return err;
}
+struct reply_func {
+ /* reply netlink message type */
+ int type;
+
+ /* function to fill message contents */
+ int (*cb)(struct net_device *, struct nlmsghdr *, u32,
+ struct nlattr **, struct sk_buff *);
+};
+
+static const struct reply_func reply_funcs[DCB_CMD_MAX+1] = {
+ /* FIXME: add reply defs */
+};
+
static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
struct net *net = sock_net(skb->sk);
@@ -1930,6 +1971,9 @@ static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
struct nlattr *tb[DCB_ATTR_MAX + 1];
u32 pid = skb ? NETLINK_CB(skb).pid : 0;
int ret = -EINVAL;
+ struct sk_buff *reply_skb;
+ struct nlmsghdr *reply_nlh;
+ const struct reply_func *fn;
if (!net_eq(net, &init_net))
return -EINVAL;
@@ -1939,6 +1983,14 @@ static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
if (ret < 0)
return ret;
+ if (dcb->cmd > DCB_CMD_MAX)
+ return -EINVAL;
+
+ /* check if a reply function has been defined for the command */
+ fn = &reply_funcs[dcb->cmd];
+ if (!fn->cb)
+ return -EOPNOTSUPP;
+
if (!tb[DCB_ATTR_IFNAME])
return -EINVAL;
@@ -1949,6 +2001,25 @@ static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
if (!netdev->dcbnl_ops)
goto errout;
+ reply_skb = dcbnl_newmsg(fn->type, dcb->cmd, pid, nlh->nlmsg_seq,
+ nlh->nlmsg_flags, &reply_nlh);
+ if (!reply_skb) {
+ ret = -ENOBUFS;
+ goto out;
+ }
+
+ ret = fn->cb(netdev, nlh, nlh->nlmsg_seq, tb, reply_skb);
+ if (ret < 0) {
+ nlmsg_free(reply_skb);
+ goto out;
+ }
+
+ nlmsg_end(reply_skb, reply_nlh);
+
+ ret = rtnl_unicast(reply_skb, &init_net, pid);
+ if (ret)
+ goto out;
+
switch (dcb->cmd) {
case DCB_CMD_GSTATE:
ret = dcbnl_getstate(netdev, tb, pid, nlh->nlmsg_seq,