linux-kernel - Re: mm: uninterruptable tasks hanged on mmap

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:	Tue, 2 Feb 2016 22:29:46 +0100
From:	Dmitry Vyukov <dvyukov@...gle.com>
To:	Jiri Kosina <jikos@...nel.org>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
	Oleg Nesterov <oleg@...hat.com>,
	Konstantin Khlebnikov <koct9i@...il.com>,
	"linux-mm@...ck.org" <linux-mm@...ck.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Takashi Iwai <tiwai@...e.de>,
	syzkaller <syzkaller@...glegroups.com>,
	Kostya Serebryany <kcc@...gle.com>,
	Alexander Potapenko <glider@...gle.com>,
	Sasha Levin <sasha.levin@...cle.com>
Subject: Re: mm: uninterruptable tasks hanged on mmap_sem

On Tue, Feb 2, 2016 at 10:24 PM, Jiri Kosina <jikos@...nel.org> wrote:
> On Tue, 2 Feb 2016, Dmitry Vyukov wrote:
>
>> Original log from fuzzer contained the following WARNING in
>> mm/rmap.c:412. But when I tried to reproduce it, I hit these hanged
>> processes instead. I can't reliably detect what program triggered
>> what. So it may be related, or maybe a separate issue.
>>
>> ------------[ cut here ]------------
>> kernel BUG at mm/rmap.c:412!
>
> Are you by any chance in this test sending signals to the fuzzer?
>
> If so, the bug I just fixed in floppy driver can cause all kinds of memory
> corruptions in case you're running multithreaded accessess to /dev/fd0 and
> sending singals to the threads that are trying to access /dev/fd0 at the
> same time.
>
> Could you please double check that the other floppy fix I've sent you a
> couple days ago doesn't fix this as well? (this test makes sense only if
> signals are involved though).

I have "floppy: fix lock_fdc() signal handling" applied (the final,
second version).
The process is multithreaded and it can well receive SIGKILLs.