linux-kernel - Re: [BUG 3.3-rc2] spinlock trylock failure on UP on CPU#0

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

Date:	Sun, 5 Feb 2012 14:55:47 -0800
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Knut Petersen <Knut_Petersen@...nline.de>,
	Jens Axboe <axboe@...nel.dk>, Tejun Heo <tj@...nel.org>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: [BUG 3.3-rc2] spinlock trylock failure on UP on CPU#0

Ok, interesting. You're triggering the issue because you have a UP
build with the spinlock debugging code enabled.

On Sun, Feb 5, 2012 at 12:57 PM, Knut Petersen
<Knut_Petersen@...nline.de> wrote:
> Booting  23783f817bceedd6d4e549385e3f400ea64059e5 I get the following trace
> an AOpen i915GMm-hfs with Pentium-M Dothan:

This is probably the same "technically legal" code that LOCKDEP also
is unhappy about - the nested fast path release. We do a trylock on
the queue lock while we already are holding the queue lock
recursively.

Jens, Tejun - the crazy-ass games that this code is playing is clearly
messing not just with lockdep, it's messing with some very reasonable
spinlock sanity checks on UP (which just boil down to "a spinlock
should never have contention on UP")

Any chance of sanitizing this all?

                       Linus

> [   12.526390] BUG: spinlock trylock failure on UP on CPU#0, scsi_id/543
> [   12.528009]  lock: f4ed8830, .magic: dead4ead, .owner: scsi_id/543, .owner_cpu: 0
> [   12.528009] Pid: 543, comm: scsi_id Not tainted 3.3.0-rc2-main+ #21
> [   12.528009] Call Trace:
> [   12.528009]  [<c043e56d>] spin_dump+0x6c/0x76
> [   12.528009]  [<c043e594>] spin_bug+0x1d/0x21
> [   12.528009]  [<c027739a>] do_raw_spin_trylock+0x33/0x38
> [   12.528009]  [<c04429c2>] _raw_spin_trylock+0x17/0x63
> [   12.528009]  [<c025fafa>] put_io_context+0xa1/0x10a
> [   12.528009]  [<c0269d85>] __cfq_slice_expired+0x2f6/0x319
> [   12.528009]  [<c026a8b2>] cfq_dispatch_requests+0x39/0x714
> [   12.528009]  [<c025b328>] elv_drain_elevator+0x3f/0x7d
> [   12.528009]  [<c025b3f0>] __elv_add_request+0x8a/0x169
> [   12.528009]  [<c02604a4>] blk_execute_rq_nowait+0x88/0xab
> [   12.528009]  [<c0260573>] blk_execute_rq+0xac/0xcd
> [   12.528009]  [<c0263cf5>] sg_io+0x23c/0x325
> [   12.528009]  [<c02642a3>] scsi_cmd_ioctl+0x1b1/0x332
> [   12.528009]  [<c0264452>] scsi_cmd_blk_ioctl+0x2e/0x38
> [   12.528009]  [<c035c15e>] sd_ioctl+0xd4/0xf4
> [   12.528009]  [<c02618ce>] __blkdev_driver_ioctl+0x26/0x2e
> [   12.528009]  [<c026226a>] blkdev_ioctl+0x70e/0x728
> [   12.528009]  [<c01b3f87>] block_ioctl+0x31/0x39
> [   12.528009]  [<c019ebb9>] do_vfs_ioctl+0x47b/0x4af
> [   12.528009]  [<c019ec2e>] sys_ioctl+0x41/0x63
> [   12.528009]  [<c0442c05>] syscall_call+0x7/0xb
>
> cu,
>  -kp
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/