summaryrefslogtreecommitdiffstats
path: root/net/ipv4
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@g5.osdl.org>2006-03-25 08:39:20 -0800
committerLinus Torvalds <torvalds@g5.osdl.org>2006-03-25 08:39:20 -0800
commitb55813a2e50088ca30df33fa62aeed5d3adb1796 (patch)
treebe50fe404e94869fe854766e190a5753dbc7dd49 /net/ipv4
parent368d17e068f691dba5a4f122c271db5ec9b2ebd6 (diff)
parent9e19bb6d7a0959f5028d46e1ab99c50f0d36eda8 (diff)
Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: [NETFILTER] x_table.c: sem2mutex [IPV4]: Aggregate route entries with different TOS values [TCP]: Mark tcp_*mem[] __read_mostly. [TCP]: Set default max buffers from memory pool size [SCTP]: Fix up sctp_rcv return value [NET]: Take RTNL when unregistering notifier [WIRELESS]: Fix config dependencies. [NET]: Fill in a 32-bit hole in struct sock on 64-bit platforms. [NET]: Ensure device name passed to SO_BINDTODEVICE is NULL terminated. [MODULES]: Don't allow statically declared exports [BRIDGE]: Unaligned accesses in the ethernet bridge
Diffstat (limited to 'net/ipv4')
-rw-r--r--net/ipv4/icmp.c2
-rw-r--r--net/ipv4/route.c45
-rw-r--r--net/ipv4/tcp.c26
3 files changed, 35 insertions, 38 deletions
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index e7bbff4340b..9831fd2c73a 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -753,7 +753,7 @@ static void icmp_redirect(struct sk_buff *skb)
case ICMP_REDIR_HOST:
case ICMP_REDIR_HOSTTOS:
ip_rt_redirect(skb->nh.iph->saddr, ip, skb->h.icmph->un.gateway,
- iph->saddr, iph->tos, skb->dev);
+ iph->saddr, skb->dev);
break;
}
out:
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index fca5fe0cf94..94fcbc5e5a1 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -55,6 +55,8 @@
* Robert Olsson : Added rt_cache statistics
* Arnaldo C. Melo : Convert proc stuff to seq_file
* Eric Dumazet : hashed spinlocks and rt_check_expire() fixes.
+ * Ilia Sotnikov : Ignore TOS on PMTUD and Redirect
+ * Ilia Sotnikov : Removed TOS from hash calculations
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
@@ -247,9 +249,9 @@ static DEFINE_PER_CPU(struct rt_cache_stat, rt_cache_stat);
static int rt_intern_hash(unsigned hash, struct rtable *rth,
struct rtable **res);
-static unsigned int rt_hash_code(u32 daddr, u32 saddr, u8 tos)
+static unsigned int rt_hash_code(u32 daddr, u32 saddr)
{
- return (jhash_3words(daddr, saddr, (u32) tos, rt_hash_rnd)
+ return (jhash_2words(daddr, saddr, rt_hash_rnd)
& rt_hash_mask);
}
@@ -1111,7 +1113,7 @@ static void rt_del(unsigned hash, struct rtable *rt)
}
void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
- u32 saddr, u8 tos, struct net_device *dev)
+ u32 saddr, struct net_device *dev)
{
int i, k;
struct in_device *in_dev = in_dev_get(dev);
@@ -1119,8 +1121,6 @@ void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
u32 skeys[2] = { saddr, 0 };
int ikeys[2] = { dev->ifindex, 0 };
- tos &= IPTOS_RT_MASK;
-
if (!in_dev)
return;
@@ -1141,8 +1141,7 @@ void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
for (i = 0; i < 2; i++) {
for (k = 0; k < 2; k++) {
unsigned hash = rt_hash_code(daddr,
- skeys[i] ^ (ikeys[k] << 5),
- tos);
+ skeys[i] ^ (ikeys[k] << 5));
rthp=&rt_hash_table[hash].chain;
@@ -1152,7 +1151,6 @@ void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
if (rth->fl.fl4_dst != daddr ||
rth->fl.fl4_src != skeys[i] ||
- rth->fl.fl4_tos != tos ||
rth->fl.oif != ikeys[k] ||
rth->fl.iif != 0) {
rthp = &rth->u.rt_next;
@@ -1232,10 +1230,9 @@ reject_redirect:
if (IN_DEV_LOG_MARTIANS(in_dev) && net_ratelimit())
printk(KERN_INFO "Redirect from %u.%u.%u.%u on %s about "
"%u.%u.%u.%u ignored.\n"
- " Advised path = %u.%u.%u.%u -> %u.%u.%u.%u, "
- "tos %02x\n",
+ " Advised path = %u.%u.%u.%u -> %u.%u.%u.%u\n",
NIPQUAD(old_gw), dev->name, NIPQUAD(new_gw),
- NIPQUAD(saddr), NIPQUAD(daddr), tos);
+ NIPQUAD(saddr), NIPQUAD(daddr));
#endif
in_dev_put(in_dev);
}
@@ -1253,8 +1250,7 @@ static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst)
rt->u.dst.expires) {
unsigned hash = rt_hash_code(rt->fl.fl4_dst,
rt->fl.fl4_src ^
- (rt->fl.oif << 5),
- rt->fl.fl4_tos);
+ (rt->fl.oif << 5));
#if RT_CACHE_DEBUG >= 1
printk(KERN_DEBUG "ip_rt_advice: redirect to "
"%u.%u.%u.%u/%02x dropped\n",
@@ -1391,14 +1387,13 @@ unsigned short ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu)
struct rtable *rth;
u32 skeys[2] = { iph->saddr, 0, };
u32 daddr = iph->daddr;
- u8 tos = iph->tos & IPTOS_RT_MASK;
unsigned short est_mtu = 0;
if (ipv4_config.no_pmtu_disc)
return 0;
for (i = 0; i < 2; i++) {
- unsigned hash = rt_hash_code(daddr, skeys[i], tos);
+ unsigned hash = rt_hash_code(daddr, skeys[i]);
rcu_read_lock();
for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
@@ -1407,7 +1402,6 @@ unsigned short ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu)
rth->fl.fl4_src == skeys[i] &&
rth->rt_dst == daddr &&
rth->rt_src == iph->saddr &&
- rth->fl.fl4_tos == tos &&
rth->fl.iif == 0 &&
!(dst_metric_locked(&rth->u.dst, RTAX_MTU))) {
unsigned short mtu = new_mtu;
@@ -1658,7 +1652,7 @@ static int ip_route_input_mc(struct sk_buff *skb, u32 daddr, u32 saddr,
RT_CACHE_STAT_INC(in_slow_mc);
in_dev_put(in_dev);
- hash = rt_hash_code(daddr, saddr ^ (dev->ifindex << 5), tos);
+ hash = rt_hash_code(daddr, saddr ^ (dev->ifindex << 5));
return rt_intern_hash(hash, rth, (struct rtable**) &skb->dst);
e_nobufs:
@@ -1823,7 +1817,7 @@ static inline int ip_mkroute_input_def(struct sk_buff *skb,
return err;
/* put it into the cache */
- hash = rt_hash_code(daddr, saddr ^ (fl->iif << 5), tos);
+ hash = rt_hash_code(daddr, saddr ^ (fl->iif << 5));
return rt_intern_hash(hash, rth, (struct rtable**)&skb->dst);
}
@@ -1864,7 +1858,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb,
return err;
/* put it into the cache */
- hash = rt_hash_code(daddr, saddr ^ (fl->iif << 5), tos);
+ hash = rt_hash_code(daddr, saddr ^ (fl->iif << 5));
err = rt_intern_hash(hash, rth, &rtres);
if (err)
return err;
@@ -2041,7 +2035,7 @@ local_input:
rth->rt_flags &= ~RTCF_LOCAL;
}
rth->rt_type = res.type;
- hash = rt_hash_code(daddr, saddr ^ (fl.iif << 5), tos);
+ hash = rt_hash_code(daddr, saddr ^ (fl.iif << 5));
err = rt_intern_hash(hash, rth, (struct rtable**)&skb->dst);
goto done;
@@ -2088,7 +2082,7 @@ int ip_route_input(struct sk_buff *skb, u32 daddr, u32 saddr,
int iif = dev->ifindex;
tos &= IPTOS_RT_MASK;
- hash = rt_hash_code(daddr, saddr ^ (iif << 5), tos);
+ hash = rt_hash_code(daddr, saddr ^ (iif << 5));
rcu_read_lock();
for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
@@ -2286,10 +2280,8 @@ static inline int ip_mkroute_output_def(struct rtable **rp,
int err = __mkroute_output(&rth, res, fl, oldflp, dev_out, flags);
unsigned hash;
if (err == 0) {
- u32 tos = RT_FL_TOS(oldflp);
-
hash = rt_hash_code(oldflp->fl4_dst,
- oldflp->fl4_src ^ (oldflp->oif << 5), tos);
+ oldflp->fl4_src ^ (oldflp->oif << 5));
err = rt_intern_hash(hash, rth, rp);
}
@@ -2304,7 +2296,6 @@ static inline int ip_mkroute_output(struct rtable** rp,
unsigned flags)
{
#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- u32 tos = RT_FL_TOS(oldflp);
unsigned char hop;
unsigned hash;
int err = -EINVAL;
@@ -2334,7 +2325,7 @@ static inline int ip_mkroute_output(struct rtable** rp,
hash = rt_hash_code(oldflp->fl4_dst,
oldflp->fl4_src ^
- (oldflp->oif << 5), tos);
+ (oldflp->oif << 5));
err = rt_intern_hash(hash, rth, rp);
/* forward hop information to multipath impl. */
@@ -2563,7 +2554,7 @@ int __ip_route_output_key(struct rtable **rp, const struct flowi *flp)
unsigned hash;
struct rtable *rth;
- hash = rt_hash_code(flp->fl4_dst, flp->fl4_src ^ (flp->oif << 5), flp->fl4_tos);
+ hash = rt_hash_code(flp->fl4_dst, flp->fl4_src ^ (flp->oif << 5));
rcu_read_lock_bh();
for (rth = rcu_dereference(rt_hash_table[hash].chain); rth;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 19ea5c0b094..87f68e787d0 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -257,6 +257,7 @@
#include <linux/fs.h>
#include <linux/random.h>
#include <linux/bootmem.h>
+#include <linux/cache.h>
#include <net/icmp.h>
#include <net/tcp.h>
@@ -275,9 +276,9 @@ atomic_t tcp_orphan_count = ATOMIC_INIT(0);
EXPORT_SYMBOL_GPL(tcp_orphan_count);
-int sysctl_tcp_mem[3];
-int sysctl_tcp_wmem[3] = { 4 * 1024, 16 * 1024, 128 * 1024 };
-int sysctl_tcp_rmem[3] = { 4 * 1024, 87380, 87380 * 2 };
+int sysctl_tcp_mem[3] __read_mostly;
+int sysctl_tcp_wmem[3] __read_mostly;
+int sysctl_tcp_rmem[3] __read_mostly;
EXPORT_SYMBOL(sysctl_tcp_mem);
EXPORT_SYMBOL(sysctl_tcp_rmem);
@@ -2081,7 +2082,8 @@ __setup("thash_entries=", set_thash_entries);
void __init tcp_init(void)
{
struct sk_buff *skb = NULL;
- int order, i;
+ unsigned long limit;
+ int order, i, max_share;
if (sizeof(struct tcp_skb_cb) > sizeof(skb->cb))
__skb_cb_too_small_for_tcp(sizeof(struct tcp_skb_cb),
@@ -2155,12 +2157,16 @@ void __init tcp_init(void)
sysctl_tcp_mem[1] = 1024 << order;
sysctl_tcp_mem[2] = 1536 << order;
- if (order < 3) {
- sysctl_tcp_wmem[2] = 64 * 1024;
- sysctl_tcp_rmem[0] = PAGE_SIZE;
- sysctl_tcp_rmem[1] = 43689;
- sysctl_tcp_rmem[2] = 2 * 43689;
- }
+ limit = ((unsigned long)sysctl_tcp_mem[1]) << (PAGE_SHIFT - 7);
+ max_share = min(4UL*1024*1024, limit);
+
+ sysctl_tcp_wmem[0] = SK_STREAM_MEM_QUANTUM;
+ sysctl_tcp_wmem[1] = 16*1024;
+ sysctl_tcp_wmem[2] = max(64*1024, max_share);
+
+ sysctl_tcp_rmem[0] = SK_STREAM_MEM_QUANTUM;
+ sysctl_tcp_rmem[1] = 87380;
+ sysctl_tcp_rmem[2] = max(87380, max_share);
printk(KERN_INFO "TCP: Hash tables configured "
"(established %d bind %d)\n",