[Bug 101246] [BAT][BRW] NULL pointer dereference in snd_hda_codec_generic

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Fri Jun 9 13:08:34 UTC 2017


https://bugs.freedesktop.org/show_bug.cgi?id=101246

--- Comment #9 from Martin Peres <martin.peres at free.fr> ---
(In reply to Libin Yang from comment #8)
> (In reply to Martin Peres from comment #7)
> > (In reply to Libin Yang from comment #6)
> > > My suggestion test is to try to find why it is using generic driver.
> > 
> > Hmm, I get your point, but whatever the reason why, it should still not make
> > any null pointer de-reference.
> 
> Yes, there seems 2 issues: 1) why using generic; 2) the null pointer.
> 
> For the second issue, it seems you've already find a solution. I recommend
> you submit a patch.

I will wait to actually catch the bug red-handed. I pushed the patch in our CI,
we should have a clear culprit in the coming days.

> 
> > 
> > However, did you see that this is for the i965G platform, which is 10 years
> > old? Is the non-generic driver supporting this platform?
> 
> We did not do any test on i965G platform. It's not our scope.

That's why automated CI is good ;)

> 
> > 
> > > However, if you can fix it in generic driver, please make a patch and send
> > > it to community.
> > 
> > Well, I already wrote a patch that papers around the issue and will still
> > report it in the logs[1], but this is not a proper patch.
> > 
> > I do not have any background in the sound subsystem, so I do not know what
> > are the correct assumptions to have, especially around the binding and
> > unbinding of the driver. I am merely doing some QA/CI job here, and
> > reporting the issue to you, who own this code.
> > 
> > I am however willing to help, but not by configuring the system differently
> > in order to get rid of the problem. This would be a workaround, not a fix.
> > 
> > [1] https://patchwork.freedesktop.org/patch/160525/
> 
> Please feel free to make the patch and submit it. Takashi is very
> professional and nice. I did think we need check the null pointer. But maybe
> it is not as your patch. Maybe it is in another place.

What mailing list should I start the conversation in? I really think that the
unbind in hda-generic.c is very sketchy, but I may be missing out on something.

Thanks for your feedback!

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20170609/00376220/attachment.html>


More information about the intel-gfx-bugs mailing list