Port gles2_conform_test_windowless action to GN build.

Port gles2_conform_test_windowless action to GN build.

It requires internal test files, hence it was not tested.

Partially tested on Linux with the following command lines:

$ gn gen out-gn --args='internal_gles2_conform_tests'

Formatted with the following command line:

$ gn format --in-place gpu/gles2_conform_support/BUILD.gn

BUG=432959
TEST=see above
R=dpranke@chromium.org,sievers@chromium.org

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:android_chromium_gn_compile_dbg,android_chromium_gn_compile_rel;tryserver.chromium.win:win8_chromium_gn_rel,win8_chromium_gn_dbg;tryserver.chromium.mac:mac_chromium_gn_rel,mac_chromium_gn_dbg

Committed: https://crrev.com/719f00367465a6b940dc2804be147a0b23546cf4
Cr-Commit-Position: refs/heads/master@{#323082}

[tfarina, 2015-03-27 22:04:14]

tough (non-trivial) one.

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
gpu/gles2_conform_support/BUILD.gn:63: gles2_conform_gypi.gtf_es_sources,
Dirk, Brett, do you know if I need any rebase_path magic here? The
third_party/angle/BUILD.gn file I looked (among others), does some, hence I'm
not sure if I need one here or not.

[tfarina, 2015-03-27 22:06:45]

CL description updated.

[tfarina, 2015-03-27 22:13:33]

Not sure why I'm not getting the the following error in my machine:

ERROR at //build/config/linux/gtk/BUILD.gn:31:7: Undefined identifier
  if (internal_gles2_conform_tests) {
      ^---------------------------

I'm running:

$ gn gen out-gn --args='internal_gles2_conform_tests=true'
Done. Wrote 2030 targets from 661 files in 3627ms

[Dirk Pranke, 2015-03-30 20:38:20]

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

[Dirk Pranke, 2015-03-30 20:38:21]

https://codereview.chromium.org/1033373002/diff/1/build/config/linux/gtk/BUIL...
File build/config/linux/gtk/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/build/config/linux/gtk/BUIL...
build/config/linux/gtk/BUILD.gn:31: if (internal_gles2_conform_tests) {
this won't work because that variable isn't in scope in this file.

Just make this unconditional, i.e., add the target to the list after line 28.

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
gpu/gles2_conform_support/BUILD.gn:63: gles2_conform_gypi.gtf_es_sources,
On 2015/03/27 22:04:14, tfarina wrote:
> Dirk, Brett, do you know if I need any rebase_path magic here? The
> third_party/angle/BUILD.gn file I looked (among others), does some, hence I'm
> not sure if I need one here or not.

I'm not sure. Brett?

[brettw, 2015-03-30 21:00:26]

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
gpu/gles2_conform_support/BUILD.gn:63: gles2_conform_gypi.gtf_es_sources,
That depends on what the files in the gypi are relative to.

GN will treat them as being relative to the current directory. GYP treats them
as being relative to the .gyp file that includes the .gypi. If these don't
match, you need to rebase_path to get from the GYP path to the GN one.

[tfarina, 2015-03-30 21:03:39]

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
gpu/gles2_conform_support/BUILD.gn:63: gles2_conform_gypi.gtf_es_sources,
On 2015/03/30 21:00:26, brettw wrote:
> That depends on what the files in the gypi are relative to.
> 
> GN will treat them as being relative to the current directory. GYP treats them
> as being relative to the .gyp file that includes the .gypi. If these don't
> match, you need to rebase_path to get from the GYP path to the GN one.

So, since both gles2_conform_test.gyp and BUILD.gn are on the same directory, we
are fine?

[tfarina, 2015-03-30 21:05:30]

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/gpu/gles2_conform_support/B...
gpu/gles2_conform_support/BUILD.gn:63: gles2_conform_gypi.gtf_es_sources,
On 2015/03/30 21:03:39, tfarina wrote:
> On 2015/03/30 21:00:26, brettw wrote:
> > That depends on what the files in the gypi are relative to.
> > 
> > GN will treat them as being relative to the current directory. GYP treats
them
> > as being relative to the .gyp file that includes the .gypi. If these don't
> > match, you need to rebase_path to get from the GYP path to the GN one.
> 
> So, since both gles2_conform_test.gyp and BUILD.gn are on the same directory,
we
> are fine?

Do'h, no, actually you were talking about the source files. They are in
third_party/gles2_conform/GTF_ES/...

[tfarina, 2015-03-30 23:55:44]

Brett, could you answer what rebase_path magic we need? We want to get this
until Wednesday.

Thanks,

[Dirk Pranke, 2015-03-31 00:14:22]

On 2015/03/30 23:55:44, tfarina wrote:
> Brett, could you answer what rebase_path magic we need? We want to get this
> until Wednesday.
> 
> Thanks,

I think that since the files are relative to <(DEPTH), you don't need to rebase
things? 

At any rate, after talking to kbr@, we should check in a stub of this target and
let him fill in the rest as he has access to the internal files to verify
things.

(see https://codereview.chromium.org/1041363002/ where I just did something
similar for the khronos_glcts_test target)

[tfarina, 2015-03-31 01:00:08]

Dirk, could you take another look?

https://codereview.chromium.org/1033373002/diff/1/build/config/linux/gtk/BUIL...
File build/config/linux/gtk/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/1/build/config/linux/gtk/BUIL...
build/config/linux/gtk/BUILD.gn:31: if (internal_gles2_conform_tests) {
On 2015/03/30 20:38:21, Dirk Pranke wrote:
> this won't work because that variable isn't in scope in this file.
> 
> Just make this unconditional, i.e., add the target to the list after line 28.

Done.

https://codereview.chromium.org/1033373002/diff/1/build/config/linux/gtk/BUIL...
build/config/linux/gtk/BUILD.gn:31: if (internal_gles2_conform_tests) {
On 2015/03/30 20:38:21, Dirk Pranke wrote:
> this won't work because that variable isn't in scope in this file.
> 
> Just make this unconditional, i.e., add the target to the list after line 28.

Done.

[Dirk Pranke, 2015-03-31 05:10:29]

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
gpu/gles2_conform_support/BUILD.gn:41: if (internal_gles2_conform_tests) {
where is 'internal_gles2_conform_tests' declared?

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
gpu/gles2_conform_support/BUILD.gn:164: ]
this should have an

  if (internal_gles2_conform_test) {
    deps += [ ":gles2_conform_test_windowless" ]
  }

or some such thing, right?

[tfarina, 2015-03-31 13:00:30]

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
File gpu/gles2_conform_support/BUILD.gn (right):

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
gpu/gles2_conform_support/BUILD.gn:41: if (internal_gles2_conform_tests) {
On 2015/03/31 05:10:29, Dirk Pranke wrote:
> where is 'internal_gles2_conform_tests' declared?

from
https://chromium.googlesource.com/chromium/src/+/master/build/config/BUILD.gn#31

https://codereview.chromium.org/1033373002/diff/20001/gpu/gles2_conform_suppo...
gpu/gles2_conform_support/BUILD.gn:164: ]
On 2015/03/31 05:10:28, Dirk Pranke wrote:
> this should have an
> 
>   if (internal_gles2_conform_test) {
>     deps += [ ":gles2_conform_test_windowless" ]
>   }
> 
> or some such thing, right?

Done.

[Dirk Pranke, 2015-03-31 17:34:30]

ok, lgtm.

[no sievers, 2015-03-31 18:50:03]

lgtm stamp

[tfarina, 2015-03-31 18:56:01]

The CQ bit was checked by tfarina@chromium.org

[commit-bot: I haz the power, 2015-03-31 18:56:33]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1033373002/40001

[commit-bot: I haz the power, 2015-03-31 19:00:59]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-03-31 19:01:35]

Patchset 3 (id:??) landed as
https://crrev.com/719f00367465a6b940dc2804be147a0b23546cf4
Cr-Commit-Position: refs/heads/master@{#323082}