Add initialization callback support for Video Decoder PPAPI.

Add initialization callback support for Video Decoder PPAPI.

Initializing a decoder is asynchronous, so add a callback to tell client
when the decoder is ready to use.
I confirmed that this works locally using a C plugin that I wrote on my machine.

BUG=NONE
TEST=NONE

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=86699

[vrk (LEFT CHROMIUM), 2011-05-23 21:29:12]

Small patch implementing a TODO to add an initialization callback.

brettw@ for OWNERS review.

[Ami GONE FROM CHROMIUM, 2011-05-23 21:39:02]

Basically LGTM.

http://codereview.chromium.org/7065010/diff/2001/media/video/video_decode_acc...
File media/video/video_decode_accelerator.h (right):

http://codereview.chromium.org/7065010/diff/2001/media/video/video_decode_acc...
media/video/video_decode_accelerator.h:192: // Callback to notify client that
decoder has been initialized.
Move above PPB?

http://codereview.chromium.org/7065010/diff/2001/ppapi/c/dev/ppb_video_decode...
File ppapi/c/dev/ppb_video_decoder_dev.h (right):

http://codereview.chromium.org/7065010/diff/2001/ppapi/c/dev/ppb_video_decode...
ppapi/c/dev/ppb_video_decoder_dev.h:137: PP_Resource (*Create)(PP_Instance
instance,
Update C++ API as well.

[vmr, 2011-05-24 07:16:08]

http://codereview.chromium.org/7065010/diff/2001/content/common/gpu/gpu_video...
File content/common/gpu/gpu_video_decode_accelerator.h (right):

http://codereview.chromium.org/7065010/diff/2001/content/common/gpu/gpu_video...
content/common/gpu/gpu_video_decode_accelerator.h:6: #define
CONTENT_GPU_GPU_VIDEO_DECODE_ACCELERATOR_H_
NIT: Fix header guard.

Also, it is a bit unclear to me why GpuVideoDecodeAccelerator has been moved to
content/common/gpu from content/gpu.

[Ami GONE FROM CHROMIUM, 2011-05-24 07:26:28]

>
> Also, it is a bit unclear to me why GpuVideoDecodeAccelerator has been
> moved to content/common/gpu from content/gpu.
>

From my response to vhiremath@nvidia.com in reviewlog for
http://codereview.chromium.org/7057027/:

Because of line 65
ofhttp://codereview.chromium.org/7057027/diff/1/content/common/gpu/gpu_video_se...
<http://codereview.chromium.org/7057027/diff/1/content/common/gpu/gpu_video_se...>
Namely, GpuVideoService needs to be able to set the accelerator on the decoder,
and that lives in content/common/gpu.  c/c/g may not depend on content/gpu,
hence the move.

[brettw, 2011-05-24 17:25:17]

LGTM, be sure to get to Ami's comment about updating the C++ API.

[darin (slow to review), 2011-05-31 21:40:46]

http://codereview.chromium.org/7065010/diff/3002/ppapi/c/dev/ppb_video_decode...
File ppapi/c/dev/ppb_video_decoder_dev.h (right):

http://codereview.chromium.org/7065010/diff/3002/ppapi/c/dev/ppb_video_decode...
ppapi/c/dev/ppb_video_decoder_dev.h:137: PP_Resource (*Create)(PP_Instance
instance,
This method is misusing PP_CompletionCallback.  Any function that takes a
PP_CompletionCallback MUST return int32_t, and it must return
PP_OK_COMPLETIONPENDING when it will complete its work asynchronously or PP_OK /
some error code if the completion callback will not run.

In this case, it means that you need to break out a separate "init" method, just
like PPB_URLLoader::Open.

It looks like you already have an Init method under the hood, so this change
should be trivial.

http://codereview.chromium.org/7065010/diff/3002/ppapi/c/dev/ppb_video_decode...
ppapi/c/dev/ppb_video_decoder_dev.h:161: PP_Bool (*Decode)(PP_Resource
video_decoder,
this method signature is also wrong.  it should return int32_t.

http://codereview.chromium.org/7065010/diff/3002/ppapi/c/dev/ppb_video_decode...
ppapi/c/dev/ppb_video_decoder_dev.h:232: PP_Bool (*Flush)(PP_Resource
video_decoder,
ditto

http://codereview.chromium.org/7065010/diff/3002/ppapi/c/dev/ppb_video_decode...
ppapi/c/dev/ppb_video_decoder_dev.h:248: PP_Bool (*Abort)(PP_Resource
video_decoder,
ditto