mus: Modify mojo command buffer to match current chrome gpu ipc.

mus: Modify mojo command buffer to match current chrome gpu ipc.

We are implementing an ipc layer for GPU command buffer, so the command
buffer should not know the underneath ipc (see CL). To achive this goal,
we need modify mojo command buffer interface to match the current chrome
gpu ipc.

CL https://codereview.chromium.org/1656433002/

BUG=None

Committed: https://crrev.com/dbac2c6af623454ae7481c683e4348cf8ca8e6fa
Cr-Commit-Position: refs/heads/master@{#374714}

[Peng, 2016-02-09 18:30:13]

penghuang@chromium.org changed reviewers:
+ fsamuel@chromium.org

[Peng, 2016-02-09 18:30:14]

Hi Fady, PTAL. Thanks.

[Fady Samuel, 2016-02-09 22:45:16]

fsamuel@chromium.org changed reviewers:
+ yzshen@chromium.org

[Fady Samuel, 2016-02-09 22:45:17]

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
File components/mus/public/interfaces/command_buffer.mojom (right):

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
components/mus/public/interfaces/command_buffer.mojom:52:
CreateStreamTexture(uint32 client_texture_id)
Yuzhu: These IPCs with callbacks need to be sync IPCs? How are we going to do
sync IPCs here? Would adding the [Sync] attribute give us sync? Is that work
ready yet?

[yzshen1, 2016-02-09 22:51:11]

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
File components/mus/public/interfaces/command_buffer.mojom (right):

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
components/mus/public/interfaces/command_buffer.mojom:52:
CreateStreamTexture(uint32 client_texture_id)
On 2016/02/09 22:45:16, Fady Samuel wrote:
> Yuzhu: These IPCs with callbacks need to be sync IPCs?
I don't think I understand this part well enough to tell whether these messages
need to be sync or not.

Async calls can have callback, too.

> How are we going to do
> sync IPCs here? Would adding the [Sync] attribute give us sync? Is that work
> ready yet?

The work is not ready yet. It will be ready in a few weeks, I think.
I am currently working on another issue (WTF-friendly generated code), but I
will give it higher priority if you or others need it soon.

When it is ready, yes, you add [Sync] and the call will become synchronous.

[Fady Samuel, 2016-02-09 22:53:39]

lgtm + nit.

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
File components/mus/public/interfaces/command_buffer.mojom (right):

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
components/mus/public/interfaces/command_buffer.mojom:19: interface
CommandBufferObserver {
Could we call this CommandBufferClient instead?

[Fady Samuel, 2016-02-09 22:56:00]

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
File components/mus/public/interfaces/command_buffer.mojom (right):

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
components/mus/public/interfaces/command_buffer.mojom:52:
CreateStreamTexture(uint32 client_texture_id)
On 2016/02/09 22:51:11, yzshen1 wrote:
> On 2016/02/09 22:45:16, Fady Samuel wrote:
> > Yuzhu: These IPCs with callbacks need to be sync IPCs?
> I don't think I understand this part well enough to tell whether these
messages
> need to be sync or not.
> 
> Async calls can have callback, too.
> 
> > How are we going to do
> > sync IPCs here? Would adding the [Sync] attribute give us sync? Is that work
> > ready yet?
> 
> The work is not ready yet. It will be ready in a few weeks, I think.
> I am currently working on another issue (WTF-friendly generated code), but I
> will give it higher priority if you or others need it soon.
> 
> When it is ready, yes, you add [Sync] and the call will become synchronous.
> 

Actually, not lgtm Chatting with Yuzhu offline, we won't have [Sync] for a few
weeks. Could you please follow my porting to mojo doc's pattern for sync IPC?
Thanks!

[Fady Samuel, 2016-02-10 15:59:42]

On 2016/02/09 22:56:00, Fady Samuel wrote:
>
https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
> File components/mus/public/interfaces/command_buffer.mojom (right):
> 
>
https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
> components/mus/public/interfaces/command_buffer.mojom:52:
> CreateStreamTexture(uint32 client_texture_id)
> On 2016/02/09 22:51:11, yzshen1 wrote:
> > On 2016/02/09 22:45:16, Fady Samuel wrote:
> > > Yuzhu: These IPCs with callbacks need to be sync IPCs?
> > I don't think I understand this part well enough to tell whether these
> messages
> > need to be sync or not.
> > 
> > Async calls can have callback, too.
> > 
> > > How are we going to do
> > > sync IPCs here? Would adding the [Sync] attribute give us sync? Is that
work
> > > ready yet?
> > 
> > The work is not ready yet. It will be ready in a few weeks, I think.
> > I am currently working on another issue (WTF-friendly generated code), but I
> > will give it higher priority if you or others need it soon.
> > 
> > When it is ready, yes, you add [Sync] and the call will become synchronous.
> > 
> 
> Actually, not lgtm Chatting with Yuzhu offline, we won't have [Sync] for a few
> weeks. Could you please follow my porting to mojo doc's pattern for sync IPC?
> Thanks!

Ohh I forgot about waitforincomingresponse. LGTM

[Peng, 2016-02-10 18:56:55]

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
File components/mus/public/interfaces/command_buffer.mojom (right):

https://codereview.chromium.org/1686543004/diff/1/components/mus/public/inter...
components/mus/public/interfaces/command_buffer.mojom:19: interface
CommandBufferObserver {
On 2016/02/09 22:53:39, Fady Samuel wrote:
> Could we call this CommandBufferClient instead?

Done.

[Peng, 2016-02-10 18:58:29]

penghuang@chromium.org changed reviewers:
+ sky@chromium.org

[Peng, 2016-02-10 18:58:30]

sky, PTAL. Thanks.

[sky, 2016-02-10 20:04:23]

Rubber stamp LGTM

[Peng, 2016-02-10 20:30:14]

The CQ bit was checked by penghuang@chromium.org

[Peng, 2016-02-10 20:30:14]

The patchset sent to the CQ was uploaded after l-g-t-m from fsamuel@chromium.org
Link to the patchset: https://codereview.chromium.org/1686543004/#ps20001
(title: "Address review issues")

[commit-bot: I haz the power, 2016-02-10 20:30:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1686543004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1686543004/20001

[commit-bot: I haz the power, 2016-02-10 20:43:23]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-16 22:30:48]

Description was changed from

==========
mus: Modify mojo command buffer to match current chrome gpu ipc.

We are implementing an ipc layer for GPU command buffer, so the command
buffer should not know the underneath ipc (see CL). To achive this goal,
we need modify mojo command buffer interface to match the current chrome
gpu ipc.

CL https://codereview.chromium.org/1656433002/

BUG=None
==========

to

==========
mus: Modify mojo command buffer to match current chrome gpu ipc.

We are implementing an ipc layer for GPU command buffer, so the command
buffer should not know the underneath ipc (see CL). To achive this goal,
we need modify mojo command buffer interface to match the current chrome
gpu ipc.

CL https://codereview.chromium.org/1656433002/

BUG=None

Committed: https://crrev.com/dbac2c6af623454ae7481c683e4348cf8ca8e6fa
Cr-Commit-Position: refs/heads/master@{#374714}
==========

[commit-bot: I haz the power, 2016-02-16 22:30:49]

Patchset 2 (id:??) landed as
https://crrev.com/dbac2c6af623454ae7481c683e4348cf8ca8e6fa
Cr-Commit-Position: refs/heads/master@{#374714}