netdev - RE: 4.0.4 - bnx2x - DEBUG_LOCKS_WARN_ON(in

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Sun, 31 May 2015 11:34:50 +0000
From:	Yuval Mintz <Yuval.Mintz@...gic.com>
To:	Nikola Ciprich <nikola.ciprich@...uxbox.cz>,
	netdev <netdev@...r.kernel.org>
Subject: RE: 4.0.4 - bnx2x - DEBUG_LOCKS_WARN_ON(in_interrupt())

> Hi,
> 
> just wanted to report, that I'm getting following trace on 4.0.4 kernel:
> 
> [  +0.301517] ------------[ cut here ]------------ [  +0.000227] WARNING: CPU: 0
> PID: 0 at kernel/locking/mutex.c:877 mutex_trylock+0x15d/0x190() [
> +0.000376] DEBUG_LOCKS_WARN_ON(in_interrupt())
...
> 0000000000000001 [  +0.000622] Call Trace:
> [  +0.000215]  <IRQ>  [<ffffffff814ab37b>] dump_stack+0x4f/0x74 [  +0.000351]
> [<ffffffff8104f715>] warn_slowpath_common+0x95/0xe0 [  +0.000224]
> [<ffffffff8104f816>] warn_slowpath_fmt+0x46/0x70 [  +0.000224]
> [<ffffffff814acd3d>] mutex_trylock+0x15d/0x190 [  +0.004247]
> [<ffffffffa055f7d7>] bnx2x_stats_handle+0x47/0x190 [bnx2x] [  +0.000230]
> [<ffffffffa051107d>] bnx2x_timer+0x19d/0x1b0 [bnx2x] [  +0.000228]
...
> 
> It's x86_64 4.0.4 kernel, running on supermicro X10DRW box.

Hi, thanks for the notice Nikola.

his was introduced by dff173de8495
("bnx2x: Fix statistics locking scheme") Which apparently took the
wrong assumption that mutex_trylock() can be called from a timer.

Looks like we'll have to change the implementation into using a
semaphore instead, which can be used from software interrupt context.

[Notice that it's possible that the only thing enforcing this and
Thus causing the warn is CONFIG_DEBUG_MUTEXES; So for now,
You might consider removing such an option as a temporary solution]

Thanks,
Yuval
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html