libvpx: enable high bit depth for vp9

libvpx: enable high bit depth for vp9

Allow the library to process 8 bit (plus high bitdepth internal), 10
bit, and 12 bit files.

Only for x86 targets. Performance for 8 bit (regular internal) is within
1-2% for these targets.

BUG=445071
R=jzern

Committed: https://crrev.com/8d5b2859f6245d3e84a53e985de4b8a9f64afc9d
Cr-Commit-Position: refs/heads/master@{#420523}

[Johann, 2016-09-22 22:17:57]

The CQ bit was checked by johannkoenig@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-09-22 22:18:20]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Johann, 2016-09-22 22:18:49]

http://imgur.com/7drHiqr

[Johann, 2016-09-22 23:07:34]

The CQ bit was unchecked by johannkoenig@google.com

[jzern, 2016-09-22 23:12:13]

On 2016/09/22 22:18:49, Johann wrote:
> http://imgur.com/7drHiqr

lgtm.
this will increase binary size so may end up getting reverted in its current
state.

[Johann, 2016-09-22 23:14:26]

win_clang *seems* unrelated but it's tricky:
FAILED: angle_unittests.exe angle_unittests.exe.pdb 
E:/b/depot_tools/python276_bin/python.exe
../../build/toolchain/win/tool_wrapper.py link-wrapper environment.x64 False
link.exe /nologo /OUT:./angle_unittests.exe /PDB:./angle_unittests.exe.pdb
@./angle_unittests.exe.rsp
translator_lib.lib(RewriteUnaryMinusOperatorInt.obj) : fatal error LNK1236:
corrupt or invalid COFF sections

I don't think libvpx is used by angle or it's unit tests. Suspecting flake? CQ
will sort it out.

[Johann, 2016-09-22 23:14:44]

The CQ bit was checked by johannkoenig@google.com

[commit-bot: I haz the power, 2016-09-22 23:15:24]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-23 01:07:10]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-09-23 01:10:42]

Description was changed from

==========
libvpx: enable high bit depth for vp9

Allow the library to process 8 bit (plus high bitdepth internal), 10
bit, and 12 bit files.

Only for x86 targets. Performance for 8 bit (regular internal) is within
1-2% for these targets.

BUG=445071
R=jzern
==========

to

==========
libvpx: enable high bit depth for vp9

Allow the library to process 8 bit (plus high bitdepth internal), 10
bit, and 12 bit files.

Only for x86 targets. Performance for 8 bit (regular internal) is within
1-2% for these targets.

BUG=445071
R=jzern

Committed: https://crrev.com/8d5b2859f6245d3e84a53e985de4b8a9f64afc9d
Cr-Commit-Position: refs/heads/master@{#420523}
==========

[commit-bot: I haz the power, 2016-09-23 01:10:43]

Patchset 1 (id:??) landed as
https://crrev.com/8d5b2859f6245d3e84a53e985de4b8a9f64afc9d
Cr-Commit-Position: refs/heads/master@{#420523}

[Nico, 2017-01-13 19:14:02]

thakis@chromium.org changed reviewers:
+ thakis@chromium.org

[Nico, 2017-01-13 19:14:03]

This apparently increased binary size of chrome by half a meg
(https://bugs.chromium.org/p/chromium/issues/detail?id=457078#c78). Was that
expected?

[Johann, 2017-01-13 19:15:41]

On 2017/01/13 19:14:03, Nico wrote:
> This apparently increased binary size of chrome by half a meg
> (https://bugs.chromium.org/p/chromium/issues/detail?id=457078#c78). Was that
> expected?

Yes:
https://bugs.chromium.org/p/chromium/issues/detail?id=650028

[Nico, 2017-01-13 19:20:55]

Thanks for the fast reply!

On Fri, Jan 13, 2017 at 2:15 PM, <johannkoenig@google.com> wrote:

> On 2017/01/13 19:14:03, Nico wrote:
> > This apparently increased binary size of chrome by half a meg
> > (https://bugs.chromium.org/p/chromium/issues/detail?id=457078#c78). Was
> that
> > expected?
>
> Yes:
> https://bugs.chromium.org/p/chromium/issues/detail?id=650028
>
> https://codereview.chromium.org/2361263002/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Johann, 2017-01-13 19:22:07]

On 2017/01/13 19:20:55, Nico wrote:
> Thanks for the fast reply!

No problem. Sorry it's not better news ...