[Bug 94515] xorg-server 1.18.2 freeze desktop on i915

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Thu Mar 24 16:36:22 UTC 2016


https://bugs.freedesktop.org/show_bug.cgi?id=94515

--- Comment #39 from Joakim Tjernlund <joakim.tjernlund at infinera.com> ---
(In reply to Chris Wilson from comment #35)
> It looks like the Client sent PresentPixmap with a divisor=1<<32, or even
> target_msc of 1<<32.
> 
> I think
> 
> commit f656f6afa288c63968f45d6d32c96a4cf6213117
> Author: Chris Wilson <chris at chris-wilson.co.uk>
> Date:   Tue Mar 22 22:11:18 2016 +0000
> 
>     sna/present: Report BadValue if target_msc exceeds last known msc by
> 1<<32
>     
>     As we cannot queue a vblank further than 1<<32 into the future, nor can
>     we set a Timer to expire that far into the future, creating such a
>     vblank event is futile. Report BadValue and hope for the best.
>     
>     References: https://bugs.freedesktop.org/show_bug.cgi?id=94515
>     Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> 
> should hide the immediate issue with the freeze / busy spin.
> 
> Can someone attach gdb and put a breakpoint on present_pixmap() line 953,
> just after present_queue_vblank() returns an error and print the locals? I
> am most interested in getting the arguments to the function so we can
> compare what the client requested with the computed vblank.

+if (warn_unless(msc - swap->msc < 1ull<<32))
Should that be <= ?

or possibly just cap the time to 1<<32 ?

-- 
You are receiving this mail because:
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/20160324/7cce3b3c/attachment.html>


More information about the intel-gfx-bugs mailing list