linux-kernel - RE: Re: [PATCH 1/1] Preventive fix in sound module

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: <20180718105820epcms5p173879897edd4cc93b381f8125c11b91b@epcms5p1>
Date:   Wed, 18 Jul 2018 16:28:20 +0530
From:   Srikanth Korangala Hari <srikanth.h@...sung.com>
To:     Takashi Iwai <tiwai@...e.de>
CC:     "alsa-devel@...a-project.org" <alsa-devel@...a-project.org>,
        "keescook@...omium.org" <keescook@...omium.org>,
        "ben.hutchings@...ethink.co.uk" <ben.hutchings@...ethink.co.uk>,
        "perex@...ex.cz" <perex@...ex.cz>,
        "elfring@...rs.sourceforge.net" <elfring@...rs.sourceforge.net>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "viro@...iv.linux.org.uk" <viro@...iv.linux.org.uk>,
        CPGS <cpgs@...sung.com>
Subject: RE: Re: [PATCH 1/1] Preventive fix in sound module


>> 
>> Signed-off-by: Srikanth K H <srikanth.h@...sung.com>
 
>What does this fix, and above all, why is this needed?

Hi,

When the sound driver creates the timer without sound card object, then while reading the sound info entry the timer object’s card information is dereferenced without checking for NULL pointer which will result for kernel panic. I tried to simulate this scenario and got below call stack,
[   36.668] E/DEVKMSG (P    0, T    0): Unable to handle kernel NULL pointer dereference at virtual address 00000000
[   36.668] E/DEVKMSG (P    0, T    0): pgd = e52f0000
[   36.668] E/DEVKMSG (P    0, T    0): [00000000] *pgd=00000000
[   36.668] E/DEVKMSG (P    0, T    0): Internal error: Oops: 5 [#1] PREEMPT SMP ARM
[   36.668] E/DEVKMSG (P    0, T    0): Modules linked in:
[   36.668] E/DEVKMSG (P    0, T    0): CPU: 1 PID: 1258 Comm: cat Tainted: G        W    3.10.65-00121-g83e9b9b-dirty #54-Tizen
[   36.668] E/DEVKMSG (P    0, T    0): task: e653aec0 ti: e52ec000 task.ti: e52ec000
[   36.668] E/DEVKMSG (P    0, T    0): PC is at snd_timer_proc_read+0x104/0x278
[   36.668] E/DEVKMSG (P    0, T    0): LR is at snd_timer_proc_read+0xec/0x278
[   36.668] E/DEVKMSG (P    0, T    0): pc : [<c0527cfc>]    lr : [<c0527ce4>]    psr: 60040013\x0asp : e52eded0  ip : 00000000  fp : 10624dd3
[   36.668] E/DEVKMSG (P    0, T    0): r10: c08ded6c  r9 : e49e3bd8  r8 : c074f518
[   36.668] E/DEVKMSG (P    0, T    0): r7 : c0afbae4  r6 : eb95a000  r5 : e49e3240  r4 : eb257e00
[   36.668] E/DEVKMSG (P    0, T    0): r3 : 00000000  r2 : 00000000  r1 : c0987cd7  r0 : e49e3240
[   36.668] E/DEVKMSG (P    0, T    0): Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
[   36.668] E/DEVKMSG (P    0, T    0): Control: 10c53c7d  Table: a52f006a  DAC: 00000015

Hence this is a preventive patch to avoid kernel panic in case if the card object passed to timer function is NULL. This would not happen in normal case, but in case of buggy scenario this would results in kernel panic rather than graceful exit.

thanks,
srikanth