VAVDA: Restore check for frame_mbs_only_flag.

VAVDA: Restore check for frame_mbs_only_flag.

This check was removed due to misinterpretation of the spec, but it
seems it's needed for some streams.

BUG=225787
TEST=manual playback

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=192620

[Pawel Osciak, 2013-04-05 16:06:08]

PTAL

[Ami GONE FROM CHROMIUM, 2013-04-05 16:56:41]

What is the impact of this CL on the bug in question?
(I believe the "after" behavior is that videos using mbs encounter a decode
error; what's the "before" behavior?)

[marcheu, 2013-04-05 16:57:12]

On 2013/04/05 16:56:41, Ami Fischman wrote:
> What is the impact of this CL on the bug in question?
> (I believe the "after" behavior is that videos using mbs encounter a decode
> error; what's the "before" behavior?)

the before behaviour is a locked up gpu :)

[Pawel Osciak, 2013-04-05 16:57:46]

On 2013/04/05 16:56:41, Ami Fischman wrote:
> What is the impact of this CL on the bug in question?
> (I believe the "after" behavior is that videos using mbs encounter a decode
> error; what's the "before" behavior?)

Before this CL: GPU (HW) crash/lockup, after: PLATFORM_FAILURE returned from
VAVDA.

[Ami GONE FROM CHROMIUM, 2013-04-05 17:03:27]

On 2013/04/05 16:57:46, posciak wrote:
> On 2013/04/05 16:56:41, Ami Fischman wrote:
> > What is the impact of this CL on the bug in question?
> > (I believe the "after" behavior is that videos using mbs encounter a decode
> > error; what's the "before" behavior?)
> 
> Before this CL: GPU (HW) crash/lockup, after: PLATFORM_FAILURE returned from
> VAVDA.

OIC; then probably *this* CL should point at BUG=225787 instead.
LGTM

[Pawel Osciak, 2013-04-05 17:09:54]

On 2013/04/05 17:03:27, Ami Fischman wrote:
> On 2013/04/05 16:57:46, posciak wrote:
> > On 2013/04/05 16:56:41, Ami Fischman wrote:
> > > What is the impact of this CL on the bug in question?
> > > (I believe the "after" behavior is that videos using mbs encounter a
decode
> > > error; what's the "before" behavior?)
> > 
> > Before this CL: GPU (HW) crash/lockup, after: PLATFORM_FAILURE returned from
> > VAVDA.
> 
> OIC; then probably *this* CL should point at BUG=225787 instead.

Oh right, that was my intention, copy-pasted wrong number.

> LGTM

[commit-bot: I haz the power, 2013-04-05 17:10:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/posciak@chromium.org/13726007/1

[commit-bot: I haz the power, 2013-04-05 20:06:24]

Change committed as 192620