linux-kernel - Re: Preempt & smp_processor_id in __make

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:	Tue, 26 Jul 2011 09:15:56 -0700
From:	"Williams, Dan J" <dan.j.williams@...el.com>
To:	Sergey Senozhatsky <sergey.senozhatsky@...il.com>
Cc:	Steven Whitehouse <swhiteho@...hat.com>,
	Jens Axboe <axboe@...nel.dk>, linux-kernel@...r.kernel.org,
	Christoph Hellwig <hch@...radead.org>,
	Roland Dreier <roland@...estorage.com>
Subject: Re: Preempt & smp_processor_id in __make_request

On Tue, Jul 26, 2011 at 3:56 AM, Sergey Senozhatsky
<sergey.senozhatsky@...il.com> wrote:
> On (07/26/11 10:32), Steven Whitehouse wrote:
>> Jul 26 09:54:04 chywoon kernel: BUG: using smp_processor_id() in preemptible [00
>> 000000] code: jbd2/dm-0-8/1546
>> Jul 26 09:54:04 chywoon kernel: caller is __make_request+0x209/0x350
>> Jul 26 09:54:04 chywoon kernel: Pid: 1546, comm: jbd2/dm-0-8 Tainted: G        W
>>    3.0.0+ #252
>> Jul 26 09:54:04 chywoon kernel: Call Trace:
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff813db897>] debug_smp_processor_id+0xe7
>> /0x100
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff813b6f29>] __make_request+0x209/0x350
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff8154635e>] ? dm_request+0x2e/0x280
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff813b3ffb>] generic_make_request+0x27b/
>> 0x550
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff8123ef7e>] ? jbd2_journal_file_buffer+
>> 0x8e/0x130
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff813b432f>] submit_bio+0x5f/0xd0
>> Jul 26 09:54:04 chywoon kernel: [<ffffffff811b14a6>] submit_bh+0xe6/0x120
>> etc.
>>
>> The (trivial) fix appears to be the following:
>>
>> diff --git a/block/blk-core.c b/block/blk-core.c
>> index f8cb099..f925581 100644
>> --- a/block/blk-core.c
>> +++ b/block/blk-core.c
>> @@ -1283,7 +1283,7 @@ get_rq:
>>
>>       if (test_bit(QUEUE_FLAG_SAME_COMP, &q->queue_flags) ||
>>           bio_flagged(bio, BIO_CPU_AFFINE))
>> -             req->cpu = smp_processor_id();
>> +             req->cpu = raw_smp_processor_id();
>>
>>       plug = current->plug;
>>       if (plug) {
>>
>> However this fixes the symptoms, rather than the cause, so I'm not at
>> all sure that this is the correct solution,

I did not realize smp_processor_id() enforced that restriction.  It
seems like we want raw_smp_processor_id() since it is an affinity
hint, and not an affinity guarantee.

--
Dan
--
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/