From: Ingo Molnar <mingo@elte.hu>
The attached patch fixes long scheduling latencies caused by access to the
/proc/net/tcp file. The seqfile functions keep softirqs disabled for a
very long time (i've seen reports of 20+ msecs, if there are enough sockets
in the system). With the attached patch it's below 100 usecs.
The cond_resched_softirq() relies on the implicit knowledge that this code
executes in process context and runs with softirqs disabled.
Potentially enabling softirqs means that the socket list might change
between buckets - but this is not an issue since seqfiles have a 4K
iteration granularity anyway and /proc/net/tcp is often (much) larger than
that.
This patch has been in the -VP patchset for weeks.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
25-akpm/net/ipv4/tcp_ipv4.c | 9 ++++++++-
1 files changed, 8 insertions(+), 1 deletion(-)
diff -puN net/ipv4/tcp_ipv4.c~sched-net-fix-scheduling-latencies-in-netstat net/ipv4/tcp_ipv4.c
--- 25/net/ipv4/tcp_ipv4.c~sched-net-fix-scheduling-latencies-in-netstat Tue Sep 14 17:39:47 2004
+++ 25-akpm/net/ipv4/tcp_ipv4.c Tue Sep 14 17:39:47 2004
@@ -2223,7 +2223,10 @@ static void *established_get_first(struc
struct sock *sk;
struct hlist_node *node;
struct tcp_tw_bucket *tw;
-
+
+ /* We can reschedule _before_ having picked the target: */
+ cond_resched_softirq();
+
read_lock(&tcp_ehash[st->bucket].lock);
sk_for_each(sk, node, &tcp_ehash[st->bucket].chain) {
if (sk->sk_family != st->family) {
@@ -2270,6 +2273,10 @@ get_tw:
}
read_unlock(&tcp_ehash[st->bucket].lock);
st->state = TCP_SEQ_STATE_ESTABLISHED;
+
+ /* We can reschedule between buckets: */
+ cond_resched_softirq();
+
if (++st->bucket < tcp_ehash_size) {
read_lock(&tcp_ehash[st->bucket].lock);
sk = sk_head(&tcp_ehash[st->bucket].chain);
_