openwrtv4/target/linux/generic/backport-4.9/024-9-tcp-add-a-missing-barrier-in-tcp_tasklet_func.patch
Pavel Kubelun b2ea46fe23 generic: net: tcp: backport tcp tx performance patches
An overall throughput gain of 22 % for heavy TCP use over a single TX queue.

Original patchset comment
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v4.13&id=3f4888adae7c1619b990d98a9b967536f71822b8

Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
2017-10-08 15:19:07 +03:00

40 lines
1.5 KiB
Diff

From 0a9648f1293966c838dc570da73c15a76f4c89d6 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 21 Dec 2016 05:42:43 -0800
Subject: [PATCH 09/10] tcp: add a missing barrier in tcp_tasklet_func()
Madalin reported crashes happening in tcp_tasklet_func() on powerpc64
Before TSQ_QUEUED bit is cleared, we must ensure the changes done
by list_del(&tp->tsq_node); are committed to memory, otherwise
corruption might happen, as an other cpu could catch TSQ_QUEUED
clearance too soon.
We can notice that old kernels were immune to this bug, because
TSQ_QUEUED was cleared after a bh_lock_sock(sk)/bh_unlock_sock(sk)
section, but they could have missed a kick to write additional bytes,
when NIC interrupts for a given flow are spread to multiple cpus.
Affected TCP flows would need an incoming ACK or RTO timer to add more
packets to the pipe. So overall situation should be better now.
Fixes: b223feb9de2a ("tcp: tsq: add shortcut in tcp_tasklet_func()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Madalin Bucur <madalin.bucur@nxp.com>
Tested-by: Madalin Bucur <madalin.bucur@nxp.com>
Tested-by: Xing Lei <xing.lei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
net/ipv4/tcp_output.c | 1 +
1 file changed, 1 insertion(+)
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -769,6 +769,7 @@ static void tcp_tasklet_func(unsigned lo
list_del(&tp->tsq_node);
sk = (struct sock *)tp;
+ smp_mb__before_atomic();
clear_bit(TSQ_QUEUED, &sk->sk_tsq_flags);
if (!sk->sk_lock.owned &&