Always expose libdrm dependency when chromeos==1

Always expose libdrm dependency when chromeos==1

This fixes a build break for linux chromeos:

gyp: Dependency '/b/build/slave/linux_chromeos/build/src/build/linux/system.gyp:libdrm#target' not found while trying to load target /b/build/slave/linux_chromeos/build/src/content/content_shell_and_tests.gyp:vaapi_jpeg_decoder_unittest#target

BUG=531814
TEST=gclient runhooks with "use_ozone=1 chromeos=1"

Committed: https://crrev.com/35ecf40fb75893838295418d07fa5d37bcca10d7
Cr-Commit-Position: refs/heads/master@{#349057}

[spang, 2015-09-16 01:11:06]

spang@chromium.org changed reviewers:
+ dpranke@chromium.org

[commit-bot: I haz the power, 2015-09-16 01:14:17]

Patchset 2 (id:??) landed as
https://crrev.com/35ecf40fb75893838295418d07fa5d37bcca10d7
Cr-Commit-Position: refs/heads/master@{#349057}

[spang, 2015-09-16 01:14:34]

Committed patchset #2 (id:20001) manually as
35ecf40fb75893838295418d07fa5d37bcca10d7 (tree was closed).

[Dirk Pranke, 2015-09-16 01:16:09]

doesn't use_ozone=1 (and ozone_auto_platforms==1, which is the default) imply
ozone_platform_gbm=1 ? i.e., does setting chromeos=1 actually have any effect if
you're also setting use_ozone=1 ?

(I'm questioning your TEST= line, not the change itself)

[spang, 2015-09-16 01:19:11]

On 2015/09/16 01:16:09, Dirk Pranke wrote:
> doesn't use_ozone=1 (and ozone_auto_platforms==1, which is the default) imply
> ozone_platform_gbm=1 ? i.e., does setting chromeos=1 actually have any effect
if
> you're also setting use_ozone=1 ?
> 
> (I'm questioning your TEST= line, not the change itself)

I had to revert https://codereview.chromium.org/1336023002 to mitigate a
different issue.

It did imply ozone_platform_gbm=1, and it will imply ozone_platform_gbm=1, but
until the other issue is solved it doesn't :(

[Dirk Pranke, 2015-09-16 01:20:41]

On 2015/09/16 01:19:11, spang wrote:
> On 2015/09/16 01:16:09, Dirk Pranke wrote:
> > doesn't use_ozone=1 (and ozone_auto_platforms==1, which is the default)
imply
> > ozone_platform_gbm=1 ? i.e., does setting chromeos=1 actually have any
effect
> if
> > you're also setting use_ozone=1 ?
> > 
> > (I'm questioning your TEST= line, not the change itself)
> 
> I had to revert https://codereview.chromium.org/1336023002 to mitigate a
> different issue.
> 
> It did imply ozone_platform_gbm=1, and it will imply ozone_platform_gbm=1, but
> until the other issue is solved it doesn't :(

ah, maybe the version of the code I was looking at had your change.

[Dirk Pranke, 2015-09-16 01:20:48]

lgtm

[commit-bot: I haz the power, 2015-09-23 12:52:07]

Patchset 2 (id:??) landed as
https://crrev.com/35ecf40fb75893838295418d07fa5d37bcca10d7
Cr-Commit-Position: refs/heads/master@{#349057}