Clean up V4L2 codec-related flags and defines

Clean up V4L2 codec-related flags and defines

All arm platforms support V4L2 SVDA by default and use_v4l2_codec also governs V4L2 VDA/VEA now.
Generates libv4l2 stub in GN file.

BUG=449883
TEST=build chrome and vda unittest on peach_pit, veyron_pinky, squawks,
and panther_freon

Committed: https://crrev.com/d5eeecc577f274bb8ed9d1954f263d83b5f00050
Cr-Commit-Position: refs/heads/master@{#314536}

[henryhsu, 2015-01-20 07:25:11]

henryhsu@chromium.org changed reviewers:
+ piman@chromium.org, posciak@chromium.org, wuchengli@chromium.org

[henryhsu, 2015-01-20 07:25:11]

posciak@chromium.org: Please review changes in 

piman@chromium.org: Please review changes in 

PTAL

[piman, 2015-01-20 19:22:49]

lgtm

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi
File content/content_common.gypi (right):

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi#...
content/content_common.gypi:810: 'USE_V4L2_CODEC'
Why is this needed? USE_V4L2_CODEC is not used in headers...

[henryhsu, 2015-01-21 01:36:06]

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi
File content/content_common.gypi (right):

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi#...
content/content_common.gypi:810: 'USE_V4L2_CODEC'
On 2015/01/20 19:22:49, piman (Very slow to review) wrote:
> Why is this needed? USE_V4L2_CODEC is not used in headers...

video_decode_accelerator_unittest uses this definition.
Therefore, content_test.gypi will inherit this definition then we don't have to
declare use_v4l2_codec variable again.

[Pawel Osciak, 2015-01-21 09:07:02]

https://chromiumcodereview.appspot.com/828063009/diff/1/content/content_commo...
File content/content_common.gypi (right):

https://chromiumcodereview.appspot.com/828063009/diff/1/content/content_commo...
content/content_common.gypi:808: 'direct_dependent_settings': {
Could you please split the CL into two, one including this chunk only? Otherwise
the test will not pick up V4L2 VDA on x86, which we need for M41.
Thanks.

[wuchengli, 2015-01-21 12:50:42]

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
File content/common/common.gni (right):

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This is
used for x86 CrOS
Update the comment. This will be used by both x86 and arm. Right?

[henryhsu, 2015-01-21 13:00:40]

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
File content/common/common.gni (right):

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This is
used for x86 CrOS
On 2015/01/21 12:50:42, wuchengli wrote:
> Update the comment. This will be used by both x86 and arm. Right?

No. The condition is (arm || (use_ozone && use_v4l2_codec)). So use_v4l2_codec
is only for x86 CrOS.

[Pawel Osciak, 2015-01-22 08:30:02]

https://codereview.chromium.org/828063009/diff/1/content/common/BUILD.gn
File content/common/BUILD.gn (right):

https://codereview.chromium.org/828063009/diff/1/content/common/BUILD.gn#newc...
content/common/BUILD.gn:339: "gpu/media/accelerated_video_decoder.h",
Please keep lexicographical order.

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
File content/common/common.gni (right):

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This is
used for x86 CrOS
On 2015/01/21 13:00:40, henryhsu wrote:
> On 2015/01/21 12:50:42, wuchengli wrote:
> > Update the comment. This will be used by both x86 and arm. Right?
> 
> No. The condition is (arm || (use_ozone && use_v4l2_codec)). So use_v4l2_codec
> is only for x86 CrOS.

Yes, but the name would be misleading then, and that's not the intention. I'd
rather we had use_v4l2_codec on all platforms, since it's used to save us from
non-CrOS builds, and should be enabled on all CrOS.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
File content/common/gpu/media/gpu_video_decode_accelerator.cc (right):

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
content/common/gpu/media/gpu_video_decode_accelerator.cc:36: #include
"content/common/gpu/media/v4l2_slice_video_decode_accelerator.h"
Lexicographical order please.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
content/common/gpu/media/gpu_video_decode_accelerator.cc:37: //
defined(ARCH_CPU_ARMEL) || (defined(USE_OZONE) && defined(USE_V4L2_CODEC))
I think this is more confusing than helpful. I'd remove this comment.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/vid...
File content/common/gpu/media/video_decode_accelerator_unittest.cc (right):

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/vid...
content/common/gpu/media/video_decode_accelerator_unittest.cc:63: #include
"content/common/gpu/media/v4l2_slice_video_decode_accelerator.h"
Order please.

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi
File content/content_common.gypi (right):

https://codereview.chromium.org/828063009/diff/1/content/content_common.gypi#...
content/content_common.gypi:832: 'common/gpu/media/accelerated_video_decoder.h',
Order please.

[henryhsu, 2015-01-22 09:09:09]

all done

https://codereview.chromium.org/828063009/diff/1/content/common/BUILD.gn
File content/common/BUILD.gn (right):

https://codereview.chromium.org/828063009/diff/1/content/common/BUILD.gn#newc...
content/common/BUILD.gn:339: "gpu/media/accelerated_video_decoder.h",
On 2015/01/22 08:30:01, Pawel Osciak wrote:
> Please keep lexicographical order.

Done.

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
File content/common/common.gni (right):

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This is
used for x86 CrOS
On 2015/01/22 08:30:02, Pawel Osciak wrote:
> On 2015/01/21 13:00:40, henryhsu wrote:
> > On 2015/01/21 12:50:42, wuchengli wrote:
> > > Update the comment. This will be used by both x86 and arm. Right?
> > 
> > No. The condition is (arm || (use_ozone && use_v4l2_codec)). So
use_v4l2_codec
> > is only for x86 CrOS.
> 
> Yes, but the name would be misleading then, and that's not the intention. I'd
> rather we had use_v4l2_codec on all platforms, since it's used to save us from
> non-CrOS builds, and should be enabled on all CrOS.

Done.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
File content/common/gpu/media/gpu_video_decode_accelerator.cc (right):

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
content/common/gpu/media/gpu_video_decode_accelerator.cc:36: #include
"content/common/gpu/media/v4l2_slice_video_decode_accelerator.h"
On 2015/01/22 08:30:02, Pawel Osciak wrote:
> Lexicographical order please.

Done.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/gpu...
content/common/gpu/media/gpu_video_decode_accelerator.cc:37: //
defined(ARCH_CPU_ARMEL) || (defined(USE_OZONE) && defined(USE_V4L2_CODEC))
On 2015/01/22 08:30:02, Pawel Osciak wrote:
> I think this is more confusing than helpful. I'd remove this comment.

Done.

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/vid...
File content/common/gpu/media/video_decode_accelerator_unittest.cc (right):

https://codereview.chromium.org/828063009/diff/1/content/common/gpu/media/vid...
content/common/gpu/media/video_decode_accelerator_unittest.cc:63: #include
"content/common/gpu/media/v4l2_slice_video_decode_accelerator.h"
On 2015/01/22 08:30:02, Pawel Osciak wrote:
> Order please.

Done.

[Pawel Osciak, 2015-01-23 05:50:47]

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
File content/common/common.gni (right):

https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This is
used for x86 CrOS
On 2015/01/22 09:09:09, henryhsu wrote:
> On 2015/01/22 08:30:02, Pawel Osciak wrote:
> > On 2015/01/21 13:00:40, henryhsu wrote:
> > > On 2015/01/21 12:50:42, wuchengli wrote:
> > > > Update the comment. This will be used by both x86 and arm. Right?
> > > 
> > > No. The condition is (arm || (use_ozone && use_v4l2_codec)). So
> use_v4l2_codec
> > > is only for x86 CrOS.
> > 
> > Yes, but the name would be misleading then, and that's not the intention.
I'd
> > rather we had use_v4l2_codec on all platforms, since it's used to save us
from
> > non-CrOS builds, and should be enabled on all CrOS.
> 
> Done.

If done, then I think we should not have to check for ARM for V4L2(S)VDA, but
for use_v4l2_codec instead?

[henryhsu, 2015-01-23 05:58:59]

On 2015/01/23 05:50:47, Pawel Osciak wrote:
> https://codereview.chromium.org/828063009/diff/1/content/common/common.gni
> File content/common/common.gni (right):
> 
>
https://codereview.chromium.org/828063009/diff/1/content/common/common.gni#ne...
> content/common/common.gni:19: # Indicates if Video4Linux2 codec is used. This
is
> used for x86 CrOS
> On 2015/01/22 09:09:09, henryhsu wrote:
> > On 2015/01/22 08:30:02, Pawel Osciak wrote:
> > > On 2015/01/21 13:00:40, henryhsu wrote:
> > > > On 2015/01/21 12:50:42, wuchengli wrote:
> > > > > Update the comment. This will be used by both x86 and arm. Right?
> > > > 
> > > > No. The condition is (arm || (use_ozone && use_v4l2_codec)). So
> > use_v4l2_codec
> > > > is only for x86 CrOS.
> > > 
> > > Yes, but the name would be misleading then, and that's not the intention.
> I'd
> > > rather we had use_v4l2_codec on all platforms, since it's used to save us
> from
> > > non-CrOS builds, and should be enabled on all CrOS.
> > 
> > Done.
> 
> If done, then I think we should not have to check for ARM for V4L2(S)VDA, but
> for use_v4l2_codec instead?

If so, we should add this flag for all ARM platforms. 
Hmm.. I'll upload another CL to add this flag. But I think we may need to mark
all arm platform PFQs as non-important

[henryhsu, 2015-02-04 03:45:17]

Chromium version is greater than
chromeos-base/chromeos-chrome-42.0.2291.0_rc-r1. So all arm platforms have
use_v4l2_codec=1 now. PTAL.

[Pawel Osciak, 2015-02-04 10:07:54]

lgtm

[henryhsu, 2015-02-04 10:08:14]

The CQ bit was checked by henryhsu@chromium.org

[commit-bot: I haz the power, 2015-02-04 10:08:37]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/828063009/40001

[commit-bot: I haz the power, 2015-02-04 10:56:01]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-02-04 10:57:44]

Patchset 3 (id:??) landed as
https://crrev.com/d5eeecc577f274bb8ed9d1954f263d83b5f00050
Cr-Commit-Position: refs/heads/master@{#314536}

[Pawel Osciak, 2015-02-05 05:06:12]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/902773003/ by posciak@chromium.org.

The reason for reverting is: x86 Chrome compile failure due to
https://code.google.com/p/chromium/issues/detail?id=455545..