[Bug 97666] Kernel "NULL pointer dereference" with MST monitor on HSW

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97666

--- Comment #11 from Carlos Santa <carlos.santa at intel.com> ---
unfortunately the steps to repro from comment #1 are not working for me, so
it's making progress on this bug rather slow. I was able to see the null
pointer only one time, however the stack trace although similar it wasn't
exactly the same (see attachment). Specifically, step #4, where I am supposed
to get the screen back is not happening for me, it stays black the entire time
with no crash. Step #5 above, where another mode setting occurs I am able to
get the screen back but no crash either. (I am on skylake + nightly). I have
been trying the above steps consistently 10+ times.

Looking at the stack trace and the problem appears to be starting when
drm_probe_ddc() gets called from drm_get_edid() inside drm_edid.c. The adapter
as a parameter is being passed but never check for null, specifically the call
to drm_do_probe_ddc_edid() derefences some of the fields of adapter struct when
the call to i2c_transfer() gets called, most likely causing the crash. I am
attaching a debug patch as reference.

-- 
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/20161009/c994d64a/attachment.html>