[Bug 79311] Call trace originating from i915 referencing intel_wait_for_vblank()

[bugzilla-daemon at bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=79311

--- Comment #34 from Alan Bartlett <ajb at elrepo.org> ---
(In reply to Jani Nikula from comment #33)
> (In reply to Alan Bartlett from comment #24)
> > (In reply to Ville Syrjala from comment #22)
> > > Created attachment 149291 [details]
> > > [PATCH] drm/i915: Wait for vblank before enabling the TV encoder
> > > 
> > > Given the bisect result this should get the old behaviour back, but just for
> > > the TV encoder. It doesn't really make any sense to me, but let's try it.
> > 
> > Applying this patch to the linux-3.17-rc3 sources and building results in a
> > kernel that is good.
> 
> What "good" is it if it doesn't fix the problem after all? I'm undecided
> whether I should revert the commit if it doesn't fix the problem.

I only deal in facts. Here are some further facts for you:

(1) If this issue had been "picked up" when it was first reported against
3.16-rc3 then it would not be so difficult to correct.

(2) A warm-boot results in the situation as described in comment 30, with the
traces attached (comments 31 & 32).

(3) A cold boot does not show this issue so, in fact, commit
7a98948f3b536ca9a077e84966ddc0e9f53726df does fix the issue. However . . .

(4) Unfortunately a new problem is now found. This latest problem does also
result in a "vblank wait timed out" message. Its origin is is
"drivers/gpu/drm/i915/intel_display.c:902 g4x_wait_for_vblank+0xd5/0xf0
[i915]()". That is to say from g4x_wait_for_vblank()

Some comments for you:

(1) If you would like to revert a commit, then the correct target for reversion
is commit 56ef52cad5e37fca89638e4bad598a994ecc3d9f and everything that followed
on and was applied on top of it.

(2) As thing stand, all kernels > 3.15.10 are "bad" unless commit
7a98948f3b536ca9a077e84966ddc0e9f53726df is applied.

(3) Applying commit 7a98948f3b536ca9a077e84966ddc0e9f53726df causes a new but
similar problem in that another "vblank wait timed out" trace is now output.

A question for you:

(1) Do you want a fresh bug report to be opened for the new "vblank wait timed
out" originating from "drivers/gpu/drm/i915/intel_display.c:902
g4x_wait_for_vblank+0xd5/0xf0 [i915]()" ?

-- 
You are receiving this mail because:
You are on the CC list for the bug.