[Bug 93782] [i9xx TV][BISECT] vblank wait timeout on crtc

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93782

--- Comment #41 from Felix Schwarz <felix.schwarz at oss.schwarz.eu> ---
(In reply to andram from comment #40)
> FWIW,  for me I can get around this problem by disabling the SVIDEO-1
> connector. The laptop in question does not even have a physical svideo
> connector, so no functionality is lost for me.  
> 
> I do this either by my patch above, or alternatively by adding 
> 
> video=SVIDEO-1:d
> 
> to the kernel command line.

That works for me too. :-)


In the mean time I finally automated my kernel bisection (and trimmed down the
number of modules to build) so it did not require so much personal time.

As I suspected before the problem is in the atomic conversion:

commit 74c090b1bdc57b1c9f1361908cca5a3d8a80fb08
Author: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
Date:   Mon Jul 13 16:30:30 2015 +0200

    drm/i915: Use full atomic modeset.

    Huzzah! \o/

    Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
    Reviewed-by: Daniel Stone <daniels at collabora.com>
    Signed-off-by: Daniel Vetter <daniel.vetter at ffwll.ch>


Please note that there are few caveats if you try to reproduce this with a
GM965 chipset:
- I was unable to compile these kernel versions with GCC 6 (had to resort to
GCC 5 instead)
- There was an unrelated bug introduced in git rev eddfcbcd (and fixed about 1
½ months later in git rev 80aa9312) which prevents the kernel to boot with a
GM965 chipset. In order to bisect I had to apply the fix to all commits in
betwen.
- For some time before (and exactly up to) the bisection commit 74c090b1 I
always see this error in dmesg:
>> [drm:drm_calc_timestamping_constants] *ERROR* crtc 24: Can't calculate constants, dotclock = 0!

I chose to treat the dotclock error as "good" but that might be a hint where to
look.


Anyway I'll provide some hardware to Lyude Paul from Red Hat and he was
optimistic to find the problem pretty quickly. However if anyone has an idea
based on the bisected commit I'd be glad to try :-)

-- 
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/20170107/78e5ee8a/attachment.html>