Move GPU device/driver info related code from content to gpu.

Move GPU device/driver info related code from content to gpu.

This has been suggested by gman, and agreed by kbr and jam, for the following reasons:

1) These are gpu related code, and are independent of content / browser, so putting them under gpu/ is the right thing to do conceptually.

2) This enables us to set up tests in various places with the correct blacklisting/driver_bug_workarounds information.  Otherwise, for the moment, gpu/ has no visibility into content/ side, so we have to duplicate the driver_bug_workarounds code and hardwire them for testing purpose.  This is going to cause a lot of bugs in the future, as we have the two pieces of code for the same thing (one for chrome and one for testing) and people will easily forget to update one or the other.

As for this patch, I didn't change the logic, and try to minimize the refactoring.  All improvements enabled by this relocation will be done in follow-up CLs.

BUG=230477
TEST=tree
TBR=gman@chromium.org, joi@chromium.org, kbr@chromium.org, piman@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=201380

[Zhenyao Mo, 2013-05-20 18:34:24]

This CL touches so many folders, but most of them are just include path rebase,
so I think I will do TBR for these folders.

Please review the following:

1) gyp changes
2) whether you agree I move these files (I think what I moved is a minimum and I
probably would move a few more in another CL).

gman: gpu/ owner.
piman: content/ owner.
joi: content/public owner review.

Please review.

[Zhenyao Mo, 2013-05-20 18:35:54]

As for testing: I build all in linux and run chrome, content_unittests,
gpu_unittests in Linux.

Will depend on try bots for other platforms.

[piman, 2013-05-20 19:16:35]

lgtm

[greggman, 2013-05-20 19:37:24]

lgtm

[Jói, 2013-05-20 20:02:05]

LGTM for //content/public.

[Zhenyao Mo, 2013-05-20 20:07:36]

On 2013/05/20 20:02:05, Jói wrote:
> LGTM for //content/public.

Thanks guys for the quick turnaround of reviewing this huge CL! I really
appreciate it.

I'll land it once the bots turn green.

[Ken Russell (switch to Gerrit), 2013-05-20 21:32:48]

Awesome. LGTM

https://codereview.chromium.org/15385003/diff/12041/gpu/config/gpu_driver_bug...
File gpu/config/gpu_driver_bug_workaround_type.h (right):

https://codereview.chromium.org/15385003/diff/12041/gpu/config/gpu_driver_bug...
gpu/config/gpu_driver_bug_workaround_type.h:5: #ifndef
GPU_CONFIG_GPU_DRIVER_BUG_WORDAROUND_TYPE_H_
Happened to notice a typo ("wordaround") in the include guards here.

[Zhenyao Mo, 2013-05-20 21:41:59]

https://codereview.chromium.org/15385003/diff/12041/gpu/config/gpu_driver_bug...
File gpu/config/gpu_driver_bug_workaround_type.h (right):

https://codereview.chromium.org/15385003/diff/12041/gpu/config/gpu_driver_bug...
gpu/config/gpu_driver_bug_workaround_type.h:5: #ifndef
GPU_CONFIG_GPU_DRIVER_BUG_WORDAROUND_TYPE_H_
On 2013/05/20 21:32:48, kbr wrote:
> Happened to notice a typo ("wordaround") in the include guards here.

Done.

[Zhenyao Mo, 2013-05-21 01:03:37]

Al, could you please have a look at the win try bot link failure?

I fail to see how my CL may cause this failure.

[apatrick_chromium, 2013-05-21 20:41:45]

Summary of offline conversation. I think that to make patch set #6 build on
Windows:

1) content/content_browser.gypi needs to depend on dxguid.lib because there is
still code in content/browser/gamepad/gamepad_platform_data_fetcher_win.cc that
depends on it. The dependency was previously satisfied by a #pragma comment(...)
in gpu/config/gpu_dx_diagnostics_win.cc that has now moved to content/config.

2) Optionally, do the same thing in gpu/config because
gpu/config/gpu_dx_diagnostics_win.cc still depends on dxuid.lib and has moved,
although I think it is now satisfied by the #pragma comment(...).

3) Optionally, remove the #pragma comment(...) line from
gpu/config/gpu_dx_diagnostics_win.cc because it is really confusing.

Since this patch is big, maybe only do 1) for now and follow up.

I think this is the syntax for the Windows specific part of the target:

    ['OS=="win"', {
      'link_settings': {
        'libraries': [
          '-ldxguid.lib',
        ],
      },
    }],

[Zhenyao Mo, 2013-05-21 20:47:03]

On 2013/05/21 20:41:45, apatrick_chromium wrote:
> Summary of offline conversation. I think that to make patch set #6 build on
> Windows:
> 
> 1) content/content_browser.gypi needs to depend on dxguid.lib because there is
> still code in content/browser/gamepad/gamepad_platform_data_fetcher_win.cc
that
> depends on it. The dependency was previously satisfied by a #pragma
comment(...)
> in gpu/config/gpu_dx_diagnostics_win.cc that has now moved to content/config.
> 
> 2) Optionally, do the same thing in gpu/config because
> gpu/config/gpu_dx_diagnostics_win.cc still depends on dxuid.lib and has moved,
> although I think it is now satisfied by the #pragma comment(...).
> 
> 3) Optionally, remove the #pragma comment(...) line from
> gpu/config/gpu_dx_diagnostics_win.cc because it is really confusing.
> 
> Since this patch is big, maybe only do 1) for now and follow up.
> 
> I think this is the syntax for the Windows specific part of the target:
> 
>     ['OS=="win"', {
>       'link_settings': {
>         'libraries': [
>           '-ldxguid.lib',
>         ],
>       },
>     }],


Thanks Al.  I'll split the CL and land the part that's not affected by this
first.  Then I'll get the remaining part with the suggested change up for
review.

[Zhenyao Mo, 2013-05-21 22:59:47]

Committed patchset #8 manually as r201380 (presubmit successful).