tcp: Revert "tcp: tcp_probe: use spin_lock_bh()"
Subsystems (3)
PATCH 1/1 tcp: Revert "tcp: tcp_probe: use spin_lock_bh()"
This reverts commit e70ac171658679ecf6bea4bbd9e9325cd6079d2b.
jtcp_rcv_established() is in fact called with hard irq being disabled.
Initial bug report from Ricardo Nabinger Sanchez [1] still needs to be investigated, but does not look like a TCP bug.
[1] https://www.spinics.net/lists/netdev/msg420960.html
--- net/ipv4/tcp_probe.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv4/tcp_probe.c +2 -2
@@ -117,7 +117,7 @@ static void jtcp_rcv_established(struct sock *sk, struct sk_buff *skb, (fwmark > 0 && skb->mark == fwmark)) && (full || tp->snd_cwnd != tcp_probe.lastcwnd)) { -spin_lock_bh (&tcp_probe.lock); +spin_lock (&tcp_probe.lock);/* If log fills, just silently drop */ if (tcp_probe_avail () >1 ) {struct tcp_log *p = tcp_probe.log + tcp_probe.head;
@@ -157,7 +157,7 @@ static void jtcp_rcv_established(struct sock *sk, struct sk_buff *skb, tcp_probe.head = (tcp_probe.head +1 ) & (bufsize -1 ); } tcp_probe.lastcwnd = tp->snd_cwnd; -spin_unlock_bh (&tcp_probe.lock); +spin_unlock (&tcp_probe.lock);wake_up (&tcp_probe.wait); }
Signature · 1 line
2.11.0.483.g087da7b7c-goog
From: Eric Dumazet <[email protected]>
Date: Tue, 21 Feb 2017 06:21:47 -0800
Quoted 4 lines
This reverts commit
e70ac171658679ecf6bea4bbd9e9325cd6079d2b.
jtcp_rcv_established()is in fact called with hard irq being disabled.
Quoted 8 lines
Initial bug report from Ricardo Nabinger Sanchez [1] still needs to be investigated, but does not look like a TCP bug.
[1] https://www.spinics.net/lists/netdev/msg420960.html
Signed-off-by Eric Dumazet <[email protected]>Reported-by kernel test robot <[email protected]>Cc Ricardo Nabinger Sanchez <[email protected]>
Applied.
But your analysis was quite sound, we can now invoke TCP input path from user context, therefore this jprobe needs to use BH locking. What is the problem.
Quoted 16 lines · David Miller
From: Eric Dumazet <[email protected]>
Date: Tue, 21 Feb 2017 06:21:47 -0800Quoted 4 lines
This reverts commit
e70ac171658679ecf6bea4bbd9e9325cd6079d2b.
jtcp_rcv_established()is in fact called with hard irq being disabled.Quoted 8 lines
Initial bug report from Ricardo Nabinger Sanchez [1] still needs to be investigated, but does not look like a TCP bug.
[1] https://www.spinics.net/lists/netdev/msg420960.html
Signed-off-by Eric Dumazet <[email protected]>Reported-by kernel test robot <[email protected]>Cc Ricardo Nabinger Sanchez <[email protected]>
Quoted 5 lines
Applied.
But your analysis was quite sound, we can now invoke TCP input path from user context, therefore this jprobe needs to use BH locking. What is the problem.
If hard irq are masked, then using spin_lock() should work without a lockdep splat.
The problem here is some false positive, as if one trace_hardirqs_off() was missing.
Peter originally fixed the issue back in this commit :
commit 58dfe883d3bc3b4c08c53a7f39e2ca3ec84f089e
Author: Peter Zijlstra <[email protected]> Date: Thu Oct 11 22:25:25 2007 +0200
So really I have no idea what is happening, because the trace_hardirqs_off() is still in x86 setjmp_pre_handler()
Maybe some FTRACE interaction ?