linux-kernel - Re: [PATCH] bio: decrease bi_iter.bi

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]

Message-ID: <CACVXFVPQR6xdzJxakPhA_rMiNyquooEGt_XS50-nZmyY0P+0CA@mail.gmail.com>
Date:	Thu, 29 May 2014 01:44:41 +0800
From:	Ming Lei <tom.leiming@...il.com>
To:	Maurizio Lombardi <mlombard@...hat.com>
Cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Jens Axboe <axboe@...nel.dk>, Jet Chen <jet.chen@...el.com>,
	Dongsu Park <dongsu.park@...fitbricks.com>
Subject: Re: [PATCH] bio: decrease bi_iter.bi_size by len in the fail path

On Thu, May 29, 2014 at 1:21 AM, Maurizio Lombardi <mlombard@...hat.com> wrote:
> Hi Ming,
>
> On Thu, May 29, 2014 at 12:59:19AM +0800, Ming Lei wrote:
>>
>> Actually, the correct thing may be like what did in the
>> attached patch, as Maurizio discussed with me[1].
>>
>> Very interestingly, I have reproduced the problem one time
>> with ext4/271 ext4/301 ext4/305, but won't with the attached
>> patch after running it for 3 rounds.
>>
>> [tom@...alhost xfstests]$ sudo ./check ext4/271 ext4/301 ext4/305
>> FSTYP         -- ext4
>> PLATFORM      -- Linux/x86_64 localhost 3.15.0-rc7-next-20140527+
>> MKFS_OPTIONS  -- /dev/vdc
>> MOUNT_OPTIONS -- -o acl,user_xattr /dev/vdc /mnt/scratch
>>
>> ext4/271 1s ... 1s
>> ext4/301 31s ... 32s
>> ext4/305 181s ... 180s
>> Ran: ext4/271 ext4/301 ext4/305
>> Passed all 3 tests
>>
>> Jet, could you test the attached patch?
>>
>> [1], https://lkml.org/lkml/2014/5/27/327
>
> There is a little mistake in your patch, you removed bio->bi_iter.bi_size += len;
> after the "done" label,
> but be careful that at line 747 there is a "goto done"... bi_size should be incremented
> before jumping there.

Good catch, thanks Maurizio.

Jet, please test the attached patch in this mail and ignore previous
one.

The story behind the patch should be like below:

- one page is added in __bio_add_page() 'successfully',
and bio->bi_phys_segments is equal to queue_max_segments(q),
but it should have been rejected since the last vector isn't covered

- next time, __bio_add_page() is called to add one page, but this
time blk_recount_segments() can figure out the actual physical
segments and find it is more than max segments, so failure is
triggered, but the bio->bi_phys_segments is updated with
max segments plus one

- the oops is triggered and reported by Jet, :-)


Thanks,
-- 
Ming Lei

View attachment "fix_compute_segments.patch" of type "text/x-patch" (997 bytes)