Move matrix from VideoFrame to TextureBuffer.

Move matrix from VideoFrame to TextureBuffer.

Previously, the matrix in VideoFrame was used to crop and scale the
frame. This caused complications because webrtc::VideoFrame doesn't
include a matrix. cropAndScale method is added to VideoBuffer class for
cropping and scaling instead.

BUG=webrtc:7749, webrtc:7760
Review-Url: https://codereview.webrtc.org/2990583002
Cr-Commit-Position: refs/heads/master@{#19179}
Committed: https://chromium.googlesource.com/external/webrtc/+/836f60cda1775d65f233f17501f5fdad96e0fbc0

[sakal, 2017-07-25 15:09:23]

Description was changed from

==========
Move matrix from VideoFrame to TextureBuffer.

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

to

==========
Move matrix from VideoFrame to TextureBuffer.

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

[sakal, 2017-07-25 15:09:23]

sakal@webrtc.org changed reviewers:
+ magjed@webrtc.org

[sakal, 2017-07-25 15:10:14]

Might be a little rough on the edges because I haven't had a chance to test it
yet.

PTAL if you have time

[sakal, 2017-07-26 09:27:33]

This is now tested and working PTAL.

[sakal, 2017-07-26 09:30:17]

Description was changed from

==========
Move matrix from VideoFrame to TextureBuffer.

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

to

==========
Move matrix from VideoFrame to TextureBuffer.

Previously, the matrix in VideoFrame was used to crop and scale the
frame. This caused complications because webrtc::VideoFrame doesn't
include a matrix. cropAndScale method is added to VideoBuffer class for
cropping and scaling instead.

BUG=webrtc:7749, webrtc:7760
==========

[sakal, 2017-07-26 09:30:30]

PTAL

[magjed_webrtc, 2017-07-27 14:02:29]

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/api/or...
File webrtc/sdk/android/api/org/webrtc/VideoFrame.java (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/api/or...
webrtc/sdk/android/api/org/webrtc/VideoFrame.java:128: public static
VideoFrame.Buffer cropAndScaleI420(I420Buffer buffer, int cropX, int cropY,
Maybe we should implement an optimized path for pure cropping (i.e. when
scaleWidth==cropWidth && scaleHeight == cropHeight)? It should be possible by
just slicing the ByteBuffers.

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/TextureBufferImpl.java (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/TextureBufferImpl.java:44: return
VideoFrame.TextureBuffer.Type.OES;
Is it possible to return just Type.OES?

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/jn...
File webrtc/sdk/android/src/jni/native_handle_impl.cc (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/jn...
webrtc/sdk/android/src/jni/native_handle_impl.cc:296: JNIEnv* jni =
AttachCurrentThreadIfNeeded();
Isn't it easier to extract the width and height here directly and then pass
those as ints to AndroidVideoBuffer?

[sakal, 2017-07-28 08:01:36]

I had to modify I420BufferImpl interface to accept individual ByteBuffers to
support the optimized path for cropping.

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/api/or...
File webrtc/sdk/android/api/org/webrtc/VideoFrame.java (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/api/or...
webrtc/sdk/android/api/org/webrtc/VideoFrame.java:128: public static
VideoFrame.Buffer cropAndScaleI420(I420Buffer buffer, int cropX, int cropY,
On 2017/07/27 14:02:29, magjed_webrtc wrote:
> Maybe we should implement an optimized path for pure cropping (i.e. when
> scaleWidth==cropWidth && scaleHeight == cropHeight)? It should be possible by
> just slicing the ByteBuffers.

Done.

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/TextureBufferImpl.java (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/TextureBufferImpl.java:44: return
VideoFrame.TextureBuffer.Type.OES;
On 2017/07/27 14:02:29, magjed_webrtc wrote:
> Is it possible to return just Type.OES?

This should return type.

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/jn...
File webrtc/sdk/android/src/jni/native_handle_impl.cc (right):

https://codereview.webrtc.org/2990583002/diff/20001/webrtc/sdk/android/src/jn...
webrtc/sdk/android/src/jni/native_handle_impl.cc:296: JNIEnv* jni =
AttachCurrentThreadIfNeeded();
On 2017/07/27 14:02:29, magjed_webrtc wrote:
> Isn't it easier to extract the width and height here directly and then pass
> those as ints to AndroidVideoBuffer?

Done.

[magjed_webrtc, 2017-07-28 09:51:53]

lgtm

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java (right):

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java:619:
buffer.limit(yPos + stride * sliceHeight);
Maybe use uPos here?

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java:623:
buffer.limit(uPos + uvStride * sliceHeight / 2);
Maybe use vPos here?

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java (right):

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java:57: buffer.limit(yPos
+ width * height);
Maybe use uPos here?

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java:61: buffer.limit(uPos
+ strideUV * chromaHeight);
Maybe use vPos here?

[sakal, 2017-07-28 11:05:39]

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java (right):

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java:619:
buffer.limit(yPos + stride * sliceHeight);
On 2017/07/28 09:51:53, magjed_webrtc wrote:
> Maybe use uPos here?

Done.

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/HardwareVideoDecoder.java:623:
buffer.limit(uPos + uvStride * sliceHeight / 2);
On 2017/07/28 09:51:53, magjed_webrtc wrote:
> Maybe use vPos here?

Done.

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
File webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java (right):

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java:57: buffer.limit(yPos
+ width * height);
On 2017/07/28 09:51:53, magjed_webrtc wrote:
> Maybe use uPos here?

Done.

https://codereview.webrtc.org/2990583002/diff/60001/webrtc/sdk/android/src/ja...
webrtc/sdk/android/src/java/org/webrtc/I420BufferImpl.java:61: buffer.limit(uPos
+ strideUV * chromaHeight);
On 2017/07/28 09:51:53, magjed_webrtc wrote:
> Maybe use vPos here?

Done.

[sakal, 2017-07-28 11:05:44]

The CQ bit was checked by sakal@webrtc.org

[sakal, 2017-07-28 11:05:44]

The patchset sent to the CQ was uploaded after l-g-t-m from magjed@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2990583002/#ps80001 (title:
"Address comments.")

[commit-bot: I haz the power, 2017-07-28 11:05:49]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-07-28 11:17:39]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-07-28 11:17:40]

Try jobs failed on following builders:
  android_compile_mips_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/android_compile_mips_db...)

[sakal, 2017-07-28 11:44:29]

The CQ bit was checked by sakal@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2017-07-28 11:44:38]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-07-28 12:20:29]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-07-28 12:20:30]

Dry run: This issue passed the CQ dry run.

[sakal, 2017-07-28 14:09:29]

The CQ bit was checked by sakal@webrtc.org

[sakal, 2017-07-28 14:09:30]

The patchset sent to the CQ was uploaded after l-g-t-m from magjed@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2990583002/#ps100001 (title:
"Rebase & fix tests.")

[commit-bot: I haz the power, 2017-07-28 14:09:41]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-07-28 14:12:24]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1501250969998280,
"parent_rev": "5ba9730265c3bb4069fdda557805d594b1cff789", "commit_rev":
"836f60cda1775d65f233f17501f5fdad96e0fbc0"}

[commit-bot: I haz the power, 2017-07-28 14:12:31]

Description was changed from

==========
Move matrix from VideoFrame to TextureBuffer.

Previously, the matrix in VideoFrame was used to crop and scale the
frame. This caused complications because webrtc::VideoFrame doesn't
include a matrix. cropAndScale method is added to VideoBuffer class for
cropping and scaling instead.

BUG=webrtc:7749, webrtc:7760
==========

to

==========
Move matrix from VideoFrame to TextureBuffer.

Previously, the matrix in VideoFrame was used to crop and scale the
frame. This caused complications because webrtc::VideoFrame doesn't
include a matrix. cropAndScale method is added to VideoBuffer class for
cropping and scaling instead.

BUG=webrtc:7749, webrtc:7760

Review-Url: https://codereview.webrtc.org/2990583002
Cr-Commit-Position: refs/heads/master@{#19179}
Committed:
https://chromium.googlesource.com/external/webrtc/+/836f60cda1775d65f233f1750...
==========

[commit-bot: I haz the power, 2017-07-28 14:12:33]

Committed patchset #6 (id:100001) as
https://chromium.googlesource.com/external/webrtc/+/836f60cda1775d65f233f1750...