Re: [PATCH] net: fix race in the receive/select

view
thread

!MAILaRCHIVE_VOTE_RePLACE

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

[view in full thread]

From: Jiri Olsa <jolsa@...>

To: Andi Kleen <andi@...>

Cc: <netdev@...>, <linux-kernel@...>, <fbl@...>, <nhorman@...>, <davem@...>, <oleg@...>, <eric.dumazet@...>

Subject: Re: [PATCH] net: fix race in the receive/select

Date: Monday, June 29, 2009 - 5:24 am

On Mon, Jun 29, 2009 at 11:12:33AM +0200, Andi Kleen wrote:
quoted text> Jiri Olsa <jolsa@redhat.com> writes:
> 
> > Adding memory barrier to the __pollwait function paired with 
> > receive callbacks.  The smp_mb__after_lock define is added, 
> > since {read|write|spin}_lock() on x86 are full memory barriers.
> 
> I was wondering did you see that race actually happening in practice?
> If yes on which system?
> 
> At least on x86 I can't see how it happens. mb() is only a compile
> time barrier and the compiler doesn't optimize over indirect callbacks
> like __pollwait() anyways.
> 
> It might be still needed on some weaker ordered architectures, but did you
> actually see it there?
> 
> -Andi

yes, we have a customer that has been able to reproduce this problem on 
x86_64 CPU model Xeon E5345*2, but they didn't reproduce on XEON MV, for example.

they were able to capture a backtrace when the race happened:
https://bugzilla.redhat.com/show_bug.cgi?id=494404#c1

jirka

quoted text> 
> -- 
> ak@linux.intel.com -- Speaking for myself only.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:

[PATCH] net: fix race in the receive/select, Jiri Olsa, (Thu Jun 25, 8:25 am)

Re: [PATCH] net: fix race in the receive/select, Andi Kleen, (Mon Jun 29, 5:12 am)

Re: [PATCH] net: fix race in the receive/select, Herbert Xu, (Tue Jun 30, 11:39 pm)

Re: [PATCH] net: fix race in the receive/select, Andi Kleen, (Wed Jul 1, 2:27 am)

Re: [PATCH] net: fix race in the receive/select, Herbert Xu, (Wed Jul 1, 3:03 am)

Re: [PATCH] net: fix race in the receive/select, Andi Kleen, (Wed Jul 1, 3:22 am)

Re: [PATCH] net: fix race in the receive/select, Herbert Xu, (Wed Jul 1, 4:31 am)

Re: [PATCH] net: fix race in the receive/select, Jiri Olsa, (Wed Jul 1, 4:44 am)

Re: [PATCH] net: fix race in the receive/select, Herbert Xu, (Wed Jul 1, 6:58 am)

Re: [PATCH] net: fix race in the receive/select, Herbert Xu, (Wed Jul 1, 9:07 am)

Re: [PATCH] net: fix race in the receive/select, Zan Lynx, (Mon Jun 29, 12:59 pm)

Re: [PATCH] net: fix race in the receive/select, Andi Kleen, (Mon Jun 29, 1:29 pm)

Re: [PATCH] net: fix race in the receive/select, Jiri Olsa, (Mon Jun 29, 5:24 am)

Re: [PATCH] net: fix race in the receive/select, Tejun Heo, (Thu Jun 25, 9:50 pm)

Re: [PATCH] net: fix race in the receive/select, Eric Dumazet, (Thu Jun 25, 7:18 pm)

Re: [PATCH] net: fix race in the receive/select, Oleg Nesterov, (Thu Jun 25, 8:24 am)

Re: [PATCH] net: fix race in the receive/select, Davide Libenzi, (Thu Jun 25, 9:31 pm)

Re: [PATCH] net: fix race in the receive/select, Eric Dumazet, (Thu Jun 25, 9:59 pm)

Re: [PATCH] net: fix race in the receive/select, Davide Libenzi, (Thu Jun 25, 10:04 pm)

Re: [PATCH] net: fix race in the receive/select, Eric Dumazet, (Thu Jun 25, 10:19 pm)

Re: [PATCH] net: fix race in the receive/select, Oleg Nesterov, (Fri Jun 26, 9:46 am)

Re: [PATCH] net: fix race in the receive/select, Davide Libenzi, (Thu Jun 25, 11:14 pm)

Re: [PATCH] net: fix race in the receive/select, Oleg Nesterov, (Fri Jun 26, 9:57 am)

Re: [PATCH] net: fix race in the receive/select, Davide Libenzi, (Fri Jun 26, 1:32 pm)

Re: [PATCH] net: fix race in the receive/select, Oleg Nesterov, (Fri Jun 26, 10:50 am)

Re: [PATCH] net: fix race in the receive/select, Jarek Poplawski, (Sun Jun 28, 7:10 am)

Re: [PATCH] net: fix race in the receive/select, Jarek Poplawski, (Sun Jun 28, 7:22 am)

Re: [PATCH] net: fix race in the receive/select, Oleg Nesterov, (Sun Jun 28, 2:04 pm)

Re: [PATCH] net: fix race in the receive/select, Jarek Poplawski, (Sun Jun 28, 5:48 pm)

Re: [PATCH] net: fix race in the receive/select, Jiri Olsa, (Mon Jun 29, 5:27 am)

Re: [PATCH] net: fix race in the receive/select, Davide Libenzi, (Fri Jun 26, 2:12 pm)

Re: [PATCH] net: fix race in the receive/select, David Miller, (Fri Jun 26, 2:17 pm)

Re: [PATCH] net: fix race in the receive/select, Eric Dumazet, (Fri Jun 26, 3:35 pm)

Re: [PATCH] net: fix race in the receive/select, Jiri Olsa, (Mon Jun 29, 5:34 am)

Re: [PATCH] net: fix race in the receive/select, Jarek Poplawski, (Fri Jun 26, 1:42 am)

Re: [PATCH] net: fix race in the receive/select, Jarek Poplawski, (Fri Jun 26, 4:10 am)

Re: [PATCH] net: fix race in the receive/select, David Miller, (Thu Jun 25, 10:11 pm)

Navigation

Popular discussions


linux-kernel:
Amit K. Arora	[RFC] Heads up on sys_fallocate()
H. Peter Anvin	Re: [RFC 00/15] x86_64: Optimize percpu accesses
Nicolas Pitre	Re: [RFC patch 08/18] cnt32_to_63 should use smp_rmb()
Bart Van Assche	Integration of SCST in the mainstream Linux kernel
git:

linux-netdev:
Jarek Poplawski	[PATCH] pkt_sched: Destroy gen estimators under rtnl_lock().
David Miller	[GIT]: Networking
Gerrit Renker	[PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side)
Natalie Protasevich	[BUG] New Kernel Bugs
openbsd-misc:

Colocation donated by:

Online users

Syndicate