Update dependencies to use new angle repository.

Update dependencies to use new angle repository.

BUG=319527
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=240595

[Geoff Lang, 2013-12-09 18:09:07]

PTAL

[Geoff Lang, 2013-12-09 18:11:14]

android_aosp failure is due to crbug.com/326504 I believe.

[Zhenyao Mo, 2013-12-09 20:18:33]

Looks good in general.

A few suggestions.

https://codereview.chromium.org/106623003/diff/20001/build/all.gyp
File build/all.gyp (right):

https://codereview.chromium.org/106623003/diff/20001/build/all.gyp#newcode166
build/all.gyp:166: '../third_party/angle/src/build_angle.gyp:*',
can you use the angle_path variable through out all the gyp files?  Just like
what you did in webkit side?  So in the future if we need to change the path
again, it would be much easier.

https://codereview.chromium.org/106623003/diff/20001/content/content_gpu.gypi
File content/content_gpu.gypi (right):

https://codereview.chromium.org/106623003/diff/20001/content/content_gpu.gypi...
content/content_gpu.gypi:33: '../third_party/angle/src/build_angle.gyp:libEGL',
here too, use angle_path

https://codereview.chromium.org/106623003/diff/20001/content/content_tests.gypi
File content/content_tests.gypi (right):

https://codereview.chromium.org/106623003/diff/20001/content/content_tests.gy...
content/content_tests.gypi:1258:
'../third_party/angle/src/build_angle.gyp:libEGL',
use angle_path here.

https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer/servi...
File gpu/command_buffer/service/shader_translator.h (right):

https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer/servi...
gpu/command_buffer/service/shader_translator.h:16: #include
"third_party/angle/include/GLSLANG/ShaderLang.h"
I would suggest define a angle_headers.h in the chromium side, which is simply a
include of ShaderLang.h.

So anywhere in chromium we include the angle_headers.h instead of the full path
of ShaderLang.h.  And in the future if we change angle path, we only need to
modify one place.

But it's up to you.  I am fine with this CL without doing this.

https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer_servi...
File gpu/command_buffer_service.gypi (right):

https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer_servi...
gpu/command_buffer_service.gypi:21:
'../third_party/angle/src/build_angle.gyp:translator',
use angle_path

https://codereview.chromium.org/106623003/diff/20001/gpu/gpu.gyp
File gpu/gpu.gyp (right):

https://codereview.chromium.org/106623003/diff/20001/gpu/gpu.gyp#newcode128
gpu/gpu.gyp:128: '../third_party/angle/src/build_angle.gyp:translator_static',
use angle_path

https://codereview.chromium.org/106623003/diff/20001/webkit/common/gpu/webkit...
File webkit/common/gpu/webkit_gpu.gyp (right):

https://codereview.chromium.org/106623003/diff/20001/webkit/common/gpu/webkit...
webkit/common/gpu/webkit_gpu.gyp:26:
'<(DEPTH)/third_party/angle/src/build_angle.gyp:translator',
use angle_path

[Geoff Lang, 2013-12-09 21:56:58]

On 2013/12/09 20:18:33, Zhenyao Mo wrote:
> Looks good in general.
> 
> A few suggestions.
> 
> https://codereview.chromium.org/106623003/diff/20001/build/all.gyp
> File build/all.gyp (right):
> 
> https://codereview.chromium.org/106623003/diff/20001/build/all.gyp#newcode166
> build/all.gyp:166: '../third_party/angle/src/build_angle.gyp:*',
> can you use the angle_path variable through out all the gyp files?  Just like
> what you did in webkit side?  So in the future if we need to change the path
> again, it would be much easier.
> 
> https://codereview.chromium.org/106623003/diff/20001/content/content_gpu.gypi
> File content/content_gpu.gypi (right):
> 
>
https://codereview.chromium.org/106623003/diff/20001/content/content_gpu.gypi...
> content/content_gpu.gypi:33:
'../third_party/angle/src/build_angle.gyp:libEGL',
> here too, use angle_path
> 
>
https://codereview.chromium.org/106623003/diff/20001/content/content_tests.gypi
> File content/content_tests.gypi (right):
> 
>
https://codereview.chromium.org/106623003/diff/20001/content/content_tests.gy...
> content/content_tests.gypi:1258:
> '../third_party/angle/src/build_angle.gyp:libEGL',
> use angle_path here.
> 
>
https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer/servi...
> File gpu/command_buffer/service/shader_translator.h (right):
> 
>
https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer/servi...
> gpu/command_buffer/service/shader_translator.h:16: #include
> "third_party/angle/include/GLSLANG/ShaderLang.h"
> I would suggest define a angle_headers.h in the chromium side, which is simply
a
> include of ShaderLang.h.
> 
> So anywhere in chromium we include the angle_headers.h instead of the full
path
> of ShaderLang.h.  And in the future if we change angle path, we only need to
> modify one place.
> 
> But it's up to you.  I am fine with this CL without doing this.
> 
>
https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer_servi...
> File gpu/command_buffer_service.gypi (right):
> 
>
https://codereview.chromium.org/106623003/diff/20001/gpu/command_buffer_servi...
> gpu/command_buffer_service.gypi:21:
> '../third_party/angle/src/build_angle.gyp:translator',
> use angle_path
> 
> https://codereview.chromium.org/106623003/diff/20001/gpu/gpu.gyp
> File gpu/gpu.gyp (right):
> 
> https://codereview.chromium.org/106623003/diff/20001/gpu/gpu.gyp#newcode128
> gpu/gpu.gyp:128: '../third_party/angle/src/build_angle.gyp:translator_static',
> use angle_path
> 
>
https://codereview.chromium.org/106623003/diff/20001/webkit/common/gpu/webkit...
> File webkit/common/gpu/webkit_gpu.gyp (right):
> 
>
https://codereview.chromium.org/106623003/diff/20001/webkit/common/gpu/webkit...
> webkit/common/gpu/webkit_gpu.gyp:26:
> '<(DEPTH)/third_party/angle/src/build_angle.gyp:translator',
> use angle_path

Updated to use angle_path whereever possible.

Putting ShaderLang.h behind an angle header is a good idea. I'll do that in a
separate CL.

[piman, 2013-12-10 00:39:45]

lgtm

[Zhenyao Mo, 2013-12-10 00:43:18]

On 2013/12/10 00:39:45, piman wrote:
> lgtm

lgtm if win_gpu,mac_gpu,linux_gpu try bots all return green

[Zhenyao Mo, 2013-12-10 19:04:52]

On 2013/12/10 00:43:18, Zhenyao Mo wrote:
> On 2013/12/10 00:39:45, piman wrote:
> > lgtm
> 
> lgtm if win_gpu,mac_gpu,linux_gpu try bots all return green

OK, all four gpu try bots are green, so unconditional LGTM

[Geoff Lang, 2013-12-10 21:07:18]

PTAL

[Geoff Lang, 2013-12-10 21:13:24]

On 2013/12/10 21:07:18, Geoff Lang wrote:
> PTAL

jamesr: webkit/DEPS, webkit/common/gpu/webkit_gpu.gyp
mkosiba: android_webview/buildbot/deps_whitelist.py

[jamesr, 2013-12-10 23:47:30]

Hm, didn't angle used to live in this directory? could this cause issues if a
bot/checkout hasn't removed the old directory and then tries to update on this
new one?

webkit/ lgtm

[Zhenyao Mo, 2013-12-10 23:51:03]

On 2013/12/10 23:47:30, jamesr wrote:
> Hm, didn't angle used to live in this directory? could this cause issues if a
> bot/checkout hasn't removed the old directory and then tries to update on this
> new one?
> 
> webkit/ lgtm

The angle is already checked out in third_party/angle (so we actually have two
versions checked out at the moment), now we just switch from using the one in
third_party/angle_dx11 to the one in third_party/angle.

[Geoff Lang, 2013-12-10 23:53:36]

On 2013/12/10 23:47:30, jamesr wrote:
> Hm, didn't angle used to live in this directory? could this cause issues if a
> bot/checkout hasn't removed the old directory and then tries to update on this
> new one?
> 
> webkit/ lgtm

Yep, angle used to live in third_party/angle (~6 months ago) but chromium has
been pulling to both third_party/angle and third_party/angle_dx11 for a couple
weeks in preparation for this change.

[jamesr, 2013-12-11 00:25:38]

Ah, I see.

[mkosiba (inactive), 2013-12-11 11:56:12]

android_webview/ LGTM.

Unfortunately the trybot won't be happy since it doesn't re-run the DEPS
whitelist after apply_patch. I'm going to land
https://codereview.chromium.org/111343004 which adds angle to the whitelist.
Then you should change this CL to remove angle_dx11 from the whitelist.

Sorry for making this a bit more complex.

[Geoff Lang, 2013-12-11 14:53:28]

On 2013/12/11 11:56:12, mkosiba wrote:
> android_webview/ LGTM.
> 
> Unfortunately the trybot won't be happy since it doesn't re-run the DEPS
> whitelist after apply_patch. I'm going to land
> https://codereview.chromium.org/111343004 which adds angle to the whitelist.
> Then you should change this CL to remove angle_dx11 from the whitelist.
> 
> Sorry for making this a bit more complex.

Thanks for this, I'll update this CL as soon as that one goes in.

[mkosiba (inactive), 2013-12-11 16:47:32]

On 2013/12/11 14:53:28, Geoff Lang wrote:
> On 2013/12/11 11:56:12, mkosiba wrote:
> > android_webview/ LGTM.
> > 
> > Unfortunately the trybot won't be happy since it doesn't re-run the DEPS
> > whitelist after apply_patch. I'm going to land
> > https://codereview.chromium.org/111343004 which adds angle to the whitelist.
> > Then you should change this CL to remove angle_dx11 from the whitelist.
> > 
> > Sorry for making this a bit more complex.
> 
> Thanks for this, I'll update this CL as soon as that one goes in.

It's already in

[Geoff Lang, 2013-12-11 19:51:29]

On 2013/12/11 16:47:32, mkosiba wrote:
> On 2013/12/11 14:53:28, Geoff Lang wrote:
> > On 2013/12/11 11:56:12, mkosiba wrote:
> > > android_webview/ LGTM.
> > > 
> > > Unfortunately the trybot won't be happy since it doesn't re-run the DEPS
> > > whitelist after apply_patch. I'm going to land
> > > https://codereview.chromium.org/111343004 which adds angle to the
whitelist.
> > > Then you should change this CL to remove angle_dx11 from the whitelist.
> > > 
> > > Sorry for making this a bit more complex.
> > 
> > Thanks for this, I'll update this CL as soon as that one goes in.
> 
> It's already in

Updated the CL, getting some issues on the trybots but I don't think they're
related.

[commit-bot: I haz the power, 2013-12-11 22:57:57]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/geofflang@chromium.org/106623003/60001

[commit-bot: I haz the power, 2013-12-11 23:18:01]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[Geoff Lang, 2013-12-12 00:08:57]

PTAL

brettw:third_party owners review

[brettw, 2013-12-12 22:27:41]

lgtm

[commit-bot: I haz the power, 2013-12-12 22:30:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/geofflang@chromium.org/106623003/60001

[commit-bot: I haz the power, 2013-12-13 11:49:11]

Change committed as 240595

[alokp, 2013-12-16 19:06:39]

On 2013/12/13 11:49:11, I haz the power (commit-bot) wrote:
> Change committed as 240595

There a still a few places where we do not reference angle path via angle_path
variable. We should either replace all of them or not bother. I would personally
not use a variable for path unless you anticipate changing it in the future.

Are you also planning to delete angle_dx11?