Add ANGLE Platform implementation.

Add ANGLE Platform implementation.

*re-re-land with SwiftShader fix*

*re-land with fix to components_browsertests*

This logic allows us to record traces and histograms easily within
ANGLE's code. It also gives a flexible design that we can extend
with further methods later, without breaking compilation. The
design is based on blink::Platform.

Also add two new histogram descriptions to histograms.xml, and
move the GPU Process metrics init to before InitializeOneOff, so
ANGLE can record histograms during Renderer initialization.

BUG=436191
TBR=jam@chromium.org, asvitkine@chromium.org, piman@chromium.org, maruel@chromium.org, kbr@chromium.org

Committed: https://crrev.com/29fda863e960afecd39be697b54506877264a30f
Cr-Commit-Position: refs/heads/master@{#321225}

[Jamie Madill, 2015-03-18 19:12:12]

piman/kbr, PTAL:

I modified the changes in ui/gl to load the ANGLE entry points dynamically. If
they are not loaded (as in the case with SwiftShader), do nothing. Also removed
hard-linking against libGLESv2.

Manually tested with SwiftShader. Integrating a SwiftShader test is blocked on
SwiftShader decoupling the circular dependency between its libEGL and libGLESv2
- otherwise it loads ANGLE's matching DLL.

[M-A Ruel, 2015-03-18 19:15:59]

isolate still lgtm

[piman, 2015-03-18 19:27:08]

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
File ui/gl/gl_implementation_win.cc (right):

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
ui/gl/gl_implementation_win.cc:136: }
I think we'd only want to do this in the kGLImplementationEGLGLES2 path. Also,
we already load the DLL there (l. 207), so should all this move to l.213?

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
ui/gl/gl_implementation_win.cc:381: ANGLEPlatformShutdownFunc shutdownFunc =
We could just lookup this function at the same time we lookup the init one, just
keeping it in a global? That way we don't have to LoadNativeLibrary again.

[Jamie Madill, 2015-03-18 21:01:50]

PTAL

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
File ui/gl/gl_implementation_win.cc (right):

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
ui/gl/gl_implementation_win.cc:136: }
On 2015/03/18 19:27:07, piman (Very slow to review) wrote:
> I think we'd only want to do this in the kGLImplementationEGLGLES2 path. Also,
> we already load the DLL there (l. 207), so should all this move to l.213?

Done.

https://codereview.chromium.org/1007513006/diff/40001/ui/gl/gl_implementation...
ui/gl/gl_implementation_win.cc:381: ANGLEPlatformShutdownFunc shutdownFunc =
On 2015/03/18 19:27:07, piman (Very slow to review) wrote:
> We could just lookup this function at the same time we lookup the init one,
just
> keeping it in a global? That way we don't have to LoadNativeLibrary again.

Done.

[piman, 2015-03-18 22:16:57]

lgtm

[Ken Russell (switch to Gerrit), 2015-03-18 22:17:35]

lgtm

[Jamie Madill, 2015-03-18 22:37:37]

The CQ bit was checked by jmadill@chromium.org

[Jamie Madill, 2015-03-18 22:37:38]

The patchset sent to the CQ was uploaded after l-g-t-m from maruel@chromium.org
Link to the patchset: https://codereview.chromium.org/1007513006/#ps60001
(title: "Addressed pimans' review feedback")

[commit-bot: I haz the power, 2015-03-18 22:37:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1007513006/60001

[commit-bot: I haz the power, 2015-03-18 22:45:41]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-03-18 22:46:08]

Patchset 4 (id:??) landed as
https://crrev.com/29fda863e960afecd39be697b54506877264a30f
Cr-Commit-Position: refs/heads/master@{#321225}