Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=608026
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/23fa660b47e7a5fdadd6303eb966cc31a51d4b5e
Cr-Commit-Position: refs/heads/master@{#399804}

[erikchen, 2016-06-13 22:31:50]

Description was changed from

==========
IOSurfaceInUseQuery

BUG=
==========

to

==========
IOSurfaceInUseQuery

BUG=
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[erikchen, 2016-06-13 22:32:33]

The CQ bit was checked by [email protected] to run a CQ dry run

[commit-bot: I haz the power, 2016-06-13 22:33:13]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/1

[erikchen, 2016-06-13 22:43:23]

The CQ bit was checked by [email protected] to run a CQ dry run

[commit-bot: I haz the power, 2016-06-13 22:43:59]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/20001

[erikchen, 2016-06-13 23:30:30]

Description was changed from

==========
IOSurfaceInUseQuery

BUG=
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=602484
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[erikchen, 2016-06-13 23:33:33]

Description was changed from

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=602484
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=602484
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[erikchen, 2016-06-13 23:33:39]

The CQ bit was checked by [email protected] to run a CQ dry run

[erikchen, 2016-06-13 23:33:52]

[email protected] changed reviewers:
+ [email protected]

[erikchen, 2016-06-13 23:33:53]

ccameron: Please review.

[commit-bot: I haz the power, 2016-06-13 23:34:35]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/40001

[ccameron, 2016-06-14 00:08:03]

lgtm

[erikchen, 2016-06-14 00:11:08]

[email protected] changed reviewers:
+ [email protected]

[erikchen, 2016-06-14 00:11:09]

piman: Please review.

[ccameron, 2016-06-14 00:22:26]

Also, I'm very pleased to see this -- it seems much cleaner than the current
scheme (which is, IMO, cleaner than the previous one).

[piman, 2016-06-14 00:58:56]

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/buil...
File gpu/command_buffer/build_gles2_cmd_buffer.py (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/buil...
gpu/command_buffer/build_gles2_cmd_buffer.py:4427: 'type': 'Custom',
You should be able to use the PUTn type, like e.g. InvalidateFramebuffer. It
requires you to rename 'n' into 'count' in cmd_buffer_functions.txt, but then it
can autogen both client-side encoding and service side validation (as well as
tests).

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/clie...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/clie...
gpu/command_buffer/client/gles2_implementation.cc:4745: GLfloat* mem =
static_cast<GLfloat*>(buffer.address());
nit: not GLfloat... GLuint if you want to, but I think just void* is fine.

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10468: GetImmediateDataAs<const
GLuint*>(c, data_size, immediate_data_size);
On the client side, you transmits the textures into a side transfer buffer,
whereas here you read them from the command buffer, so I think one of them is
wrong :)

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10470:
std::vector<gl::GLSurface::CALayerInUseQuery> queries;
nit: queries.reserve(n); to avoid reallocations

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10492:
surface_->ScheduleCALayerInUseQuery(queries);
nit: std::move(queries)

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/common/gpu_mess...
File gpu/ipc/common/gpu_messages.h (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/common/gpu_mess...
gpu/ipc/common/gpu_messages.h:54: using SwapBuffersCompletedIOSurfaceInUseQuery
= std::pair<uint32_t, bool>;
nit: may I suggest an actual struct:

IPC_STRUCT_BEGIN(SwapBuffersCompletedIOSurfaceInUseQuery)
  IPC_STRUCT_MEMBER(uint32_t, texture)
  IPC_STRUCT_MEMBER(uint32_t, in_use)
IPC_STRUCT_END()


That way, the code that uses it can say ".texture" and ".in_use" instead of
".first" and ".second".

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
File gpu/ipc/service/image_transport_surface_overlay_mac.h (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
gpu/ipc/service/image_transport_surface_overlay_mac.h:105: explicit
IOSurfaceInUseQuery(IOSurfaceInUseQuery&);
Did you mean IOSurfaceInUseQuery(const IOSurfaceInUseQuery&) ?

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
gpu/ipc/service/image_transport_surface_overlay_mac.h:108: unsigned texture;
= 0;

https://codereview.chromium.org/2067503003/diff/40001/ui/gl/gl_surface.h
File ui/gl/gl_surface.h (right):

https://codereview.chromium.org/2067503003/diff/40001/ui/gl/gl_surface.h#newc...
ui/gl/gl_surface.h:189: GLImage* image;
I think you want to take a scoped_refptr<GLImage> here, in case the client
destroys the image before the swap.

Side note, we should probably document the semantics in that case.

[commit-bot: I haz the power, 2016-06-14 01:25:47]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2016-06-14 01:25:48]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[erikchen, 2016-06-14 01:46:58]

The CQ bit was checked by [email protected] to run a CQ dry run

[commit-bot: I haz the power, 2016-06-14 01:47:16]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/120001

[erikchen, 2016-06-14 01:47:50]

[email protected] changed reviewers:
+ [email protected]

[erikchen, 2016-06-14 01:47:51]

piman: PTAL

dcheng: Please review gpu/ipc/common/gpu_messages.h

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/buil...
File gpu/command_buffer/build_gles2_cmd_buffer.py (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/buil...
gpu/command_buffer/build_gles2_cmd_buffer.py:4427: 'type': 'Custom',
On 2016/06/14 00:58:55, piman wrote:
> You should be able to use the PUTn type, like e.g. InvalidateFramebuffer. It
> requires you to rename 'n' into 'count' in cmd_buffer_functions.txt, but then
it
> can autogen both client-side encoding and service side validation (as well as
> tests).

Awesome. I'm a big fan of not writing boilerplate plumbing, and then getting it
wrong in the process.

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/clie...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/clie...
gpu/command_buffer/client/gles2_implementation.cc:4745: GLfloat* mem =
static_cast<GLfloat*>(buffer.address());
On 2016/06/14 00:58:55, piman wrote:
> nit: not GLfloat... GLuint if you want to, but I think just void* is fine.

No longer necessary by using PUTn type.

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10468: GetImmediateDataAs<const
GLuint*>(c, data_size, immediate_data_size);
On 2016/06/14 00:58:56, piman wrote:
> On the client side, you transmits the textures into a side transfer buffer,
> whereas here you read them from the command buffer, so I think one of them is
> wrong :)

This code is no longer necessary, because I now use PUTn type.

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10470:
std::vector<gl::GLSurface::CALayerInUseQuery> queries;
On 2016/06/14 00:58:55, piman wrote:
> nit: queries.reserve(n); to avoid reallocations

Done.

https://codereview.chromium.org/2067503003/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:10492:
surface_->ScheduleCALayerInUseQuery(queries);
On 2016/06/14 00:58:55, piman wrote:
> nit: std::move(queries)

Done.

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/common/gpu_mess...
File gpu/ipc/common/gpu_messages.h (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/common/gpu_mess...
gpu/ipc/common/gpu_messages.h:54: using SwapBuffersCompletedIOSurfaceInUseQuery
= std::pair<uint32_t, bool>;
On 2016/06/14 00:58:56, piman wrote:
> nit: may I suggest an actual struct:
> 
> IPC_STRUCT_BEGIN(SwapBuffersCompletedIOSurfaceInUseQuery)
>   IPC_STRUCT_MEMBER(uint32_t, texture)
>   IPC_STRUCT_MEMBER(uint32_t, in_use)
> IPC_STRUCT_END()
> 
> 
> That way, the code that uses it can say ".texture" and ".in_use" instead of
> ".first" and ".second".

good point. done.

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
File gpu/ipc/service/image_transport_surface_overlay_mac.h (right):

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
gpu/ipc/service/image_transport_surface_overlay_mac.h:105: explicit
IOSurfaceInUseQuery(IOSurfaceInUseQuery&);
On 2016/06/14 00:58:56, piman wrote:
> Did you mean IOSurfaceInUseQuery(const IOSurfaceInUseQuery&) ?

yes, done.

https://codereview.chromium.org/2067503003/diff/40001/gpu/ipc/service/image_t...
gpu/ipc/service/image_transport_surface_overlay_mac.h:108: unsigned texture;
On 2016/06/14 00:58:56, piman wrote:
> = 0;

Done.

https://codereview.chromium.org/2067503003/diff/40001/ui/gl/gl_surface.h
File ui/gl/gl_surface.h (right):

https://codereview.chromium.org/2067503003/diff/40001/ui/gl/gl_surface.h#newc...
ui/gl/gl_surface.h:189: GLImage* image;
On 2016/06/14 00:58:56, piman wrote:
> I think you want to take a scoped_refptr<GLImage> here, in case the client
> destroys the image before the swap.
> 
> Side note, we should probably document the semantics in that case.

ImageTransportSurfaceOverlayMac holds on to the underlying IOSurfaceRef, not the
GLImage. That being said, I prefer smart pointers to raw pointers, so I made the
change anyways.

I documented the semantics more clearly in CHROMIUM_schedule_ca_layer.txt.

"""
Clients should not destroy any textures while they are being queried, as the
swap ACK will use the original texture id to identify the queried textures.
"""
Note that there is never any confusion from the service, since |texture| is
treated as an opaque id.

I also updated the documentation in ImageTransportSurfaceOverlayMac
"""
	 109     // It's possible that the client will have destroyed the texture
before
 110     // SwapBuffersInternal is called. It's important that |texture| is only
used
 111     // as an opaque identifier, since it may no longer point to the same
 112     // texture."""

[piman, 2016-06-14 02:00:20]

lgtm

[commit-bot: I haz the power, 2016-06-14 02:03:01]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2016-06-14 02:03:02]

Dry run: Try jobs failed on following builders:
  android_compile_dbg on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_comp...)

[dcheng, 2016-06-14 04:37:27]

Where's the code that will consume the new IPC param?

[erikchen, 2016-06-14 07:33:25]

On 2016/06/14 04:37:27, dcheng wrote:
> Where's the code that will consume the new IPC param?

That code has not yet been written. This is the service side implementation of
the command buffer function.

[dcheng, 2016-06-14 08:13:13]

On 2016/06/14 07:33:25, erikchen wrote:
> On 2016/06/14 04:37:27, dcheng wrote:
> > Where's the code that will consume the new IPC param?
> 
> That code has not yet been written. This is the service side implementation of
> the command buffer function.

If I'm correct in my understanding that this is going GPU -> browser, would it
be possible to include the IPC changes with that patch instead?

[erikchen, 2016-06-14 16:41:17]

Description was changed from

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=602484
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=608026
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[erikchen, 2016-06-14 16:42:29]

On 2016/06/14 08:13:13, dcheng wrote:
> On 2016/06/14 07:33:25, erikchen wrote:
> > On 2016/06/14 04:37:27, dcheng wrote:
> > > Where's the code that will consume the new IPC param?
> > 
> > That code has not yet been written. This is the service side implementation
of
> > the command buffer function.
> 
> If I'm correct in my understanding that this is going GPU -> browser, would it
> be possible to include the IPC changes with that patch instead?

Yes, I removed the changes to gpu/ipc/common/gpu_messages.h.

[erikchen, 2016-06-14 16:42:49]

The CQ bit was checked by [email protected]

[erikchen, 2016-06-14 16:42:50]

The patchset sent to the CQ was uploaded after l-g-t-m from
[email protected], [email protected]
Link to the patchset: https://codereview.chromium.org/2067503003/#ps160001
(title: "Comments from dcheng.")

[commit-bot: I haz the power, 2016-06-14 16:43:13]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/160001

[commit-bot: I haz the power, 2016-06-14 17:37:22]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2016-06-14 17:37:23]

Try jobs failed on following builders:
  win_chromium_compile_dbg_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_comp...)

[erikchen, 2016-06-14 19:06:54]

The CQ bit was checked by [email protected]

[erikchen, 2016-06-14 19:06:54]

The patchset sent to the CQ was uploaded after l-g-t-m from
[email protected], [email protected]
Link to the patchset: https://codereview.chromium.org/2067503003/#ps180001
(title: "GL_EXPORT.")

[commit-bot: I haz the power, 2016-06-14 19:07:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/180001

[commit-bot: I haz the power, 2016-06-14 20:03:52]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2016-06-14 20:03:53]

Try jobs failed on following builders:
  win_clang on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_clang/builds/...)

[erikchen, 2016-06-14 20:26:11]

The CQ bit was checked by [email protected]

[erikchen, 2016-06-14 20:26:12]

The patchset sent to the CQ was uploaded after l-g-t-m from
[email protected], [email protected]
Link to the patchset: https://codereview.chromium.org/2067503003/#ps200001
(title: "Include gl_image.h")

[commit-bot: I haz the power, 2016-06-14 20:26:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/200001

[commit-bot: I haz the power, 2016-06-14 22:28:03]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2016-06-14 22:28:04]

Try jobs failed on following builders:
  ios-device on tryserver.chromium.mac (JOB_TIMED_OUT, no build URL)
  ios-device-gn on tryserver.chromium.mac (JOB_TIMED_OUT, no build URL)

[erikchen, 2016-06-14 22:40:22]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2016-06-14 22:41:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2067503003/200001

[commit-bot: I haz the power, 2016-06-14 23:50:25]

Committed patchset #11 (id:200001)

[commit-bot: I haz the power, 2016-06-14 23:50:29]

CQ bit was unchecked

[commit-bot: I haz the power, 2016-06-14 23:55:45]

Description was changed from

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=608026
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Add a new command buffer function glScheduleCALayerInUseQueryCHROMIUM.

The function allows the client to asynchronously query whether a texture is
still in use by the MacOS Window Server.

BUG=608026
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/23fa660b47e7a5fdadd6303eb966cc31a51d4b5e
Cr-Commit-Position: refs/heads/master@{#399804}
==========

[commit-bot: I haz the power, 2016-06-14 23:55:46]

Patchset 11 (id:??) landed as
https://crrev.com/23fa660b47e7a5fdadd6303eb966cc31a51d4b5e
Cr-Commit-Position: refs/heads/master@{#399804}