Added virtual test suite with 1MB MSE buffers

Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
Committed: https://crrev.com/2f93caffe7a63bdcb391e23f67c67ceb69718a0f
Cr-Commit-Position: refs/heads/master@{#409518}

[servolk, 2016-07-29 23:14:25]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-07-29 23:14:43]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[servolk, 2016-07-29 23:15:10]

Description was changed from

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
==========

to

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
==========

[servolk, 2016-07-29 23:15:10]

servolk@chromium.org changed reviewers:
+ wolenetz@chromium.org

[servolk, 2016-07-29 23:17:24]

servolk@chromium.org changed reviewers:
+ chcunningham@chromium.org

[commit-bot: I haz the power, 2016-07-30 00:59:12]

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

[commit-bot: I haz the power, 2016-07-30 00:59:13]

Dry run: This issue passed the CQ dry run.

[chcunningham, 2016-08-01 17:44:42]

LGTM, but can you change the name of the subfolder under mediasource? 

I think "mse-1mb-buffers" is a good name for the virtual test suite, but its not
a great name for the folder where the html files live since they are only
appending 1mb buffers when the 1mb flag is set. A better name for this folder
might be "append-quota-exceeded".

[wolenetz, 2016-08-01 18:26:49]

On 2016/08/01 17:44:42, chcunningham wrote:
> LGTM, but can you change the name of the subfolder under mediasource? 
> 
> I think "mse-1mb-buffers" is a good name for the virtual test suite, but its
not
> a great name for the folder where the html files live since they are only
> appending 1mb buffers when the 1mb flag is set. A better name for this folder
> might be "append-quota-exceeded".

Based on the direction I believe this CL is paving the way for
(https://codereview.chromium.org/2170303002/), "append-quota-exceeded" isn't the
only test targeted.

Perhaps use "stream_memory_tests" instead of "mse-1mb-buffers" as the folder
name?

[wolenetz, 2016-08-01 20:21:25]

LGTM % rename or rename + more precise limit verification. I prefer the latter,
with [ Fail ] for the non-virtual-suite version (with additional logic added to
expect (and fail if it doesn't happen) QuotaExceededError on attempted append of
>1MB total coded frame sizes. Note that the initialization segment sizes are
*not* included in the buffer limit thresholding, just the coded frame sizes.

https://codereview.chromium.org/2195933002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/mse-1mb-buffers/mediasource-appendbuffer-quota-exceeded.html
(right):

https://codereview.chromium.org/2195933002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/media/media-source/mse-1mb-buffers/mediasource-appendbuffer-quota-exceeded.html:38:
assert_equals(ex.name, 'QuotaExceededError');
I wonder if we should verify that the attempted append that failed would exceed
the virtual test suite buffer limit, in combination with either marking these [
Fail Timeout ] or just [ Skip ] for non-virtual-test-suite expectation in
TestExpectations? If so, rename the subfolder to "small-sourcebuffer-limit" or
something like that; otherwise, follow chcunningham@'s guidance and rename to
something like "append-quota-tests"

[wolenetz, 2016-08-01 20:33:14]

I chatted w/servolk@ and suggested:

1) rename the folder to "stream_limit_tests"
2) refactor out the guts of the test into a common .js for inclusion
3) have 2 versions of the test HTML: desktop and VTS
Skip the desktop version in expectations for the VTS. Skip the VTS version in
the expectations for the desktop.
Make VTS use and verify >1MB, <2MB or so triggers the error.
Make desktop use and verify the appropriate desktop limit(s).
Then we have all the coverage

[servolk, 2016-08-02 01:46:13]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 01:46:36]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-02 02:54:18]

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

[commit-bot: I haz the power, 2016-08-02 02:54:19]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[servolk, 2016-08-02 16:39:52]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 16:40:23]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-02 17:56:38]

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

[commit-bot: I haz the power, 2016-08-02 17:56:38]

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

[servolk, 2016-08-02 20:27:03]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 20:27:38]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[servolk, 2016-08-02 20:51:37]

On 2016/08/01 20:33:14, wolenetz wrote:
> I chatted w/servolk@ and suggested:
> 
> 1) rename the folder to "stream_limit_tests"
> 2) refactor out the guts of the test into a common .js for inclusion
> 3) have 2 versions of the test HTML: desktop and VTS
> Skip the desktop version in expectations for the VTS. Skip the VTS version in
> the expectations for the desktop.
> Make VTS use and verify >1MB, <2MB or so triggers the error.
> Make desktop use and verify the appropriate desktop limit(s).
> Then we have all the coverage

This is all done in patchset #4, PTAL

[commit-bot: I haz the power, 2016-08-02 21:43:32]

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

[commit-bot: I haz the power, 2016-08-02 21:43:33]

Dry run: This issue passed the CQ dry run.

[wolenetz, 2016-08-02 21:56:36]

LGTM (my nit doesn't need fixing here or now. just if easy :) )

Also note that we'd like to have a more general content-embedder interface for
setting/querying the memory limits in the longer run. Any ideas for accelerating
such?

https://codereview.chromium.org/2195933002/diff/60001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/mediasource-util.js
(right):

https://codereview.chromium.org/2195933002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/mediasource-util.js:248:
MediaSourceUtil.fetchManifestAndData(test,
'webm/test-a-5min-44100Hz-1ch-manifest.json', function(type, mediaData)
nit: (not a regression, and certainly not a strong feeling, just a note if you
could fix this easily here and now, please try to:) Allow for webm-lacking
implementations to also pass this? It might ease interop test suite merger of
this. (Ditto in the stream version, below, though the interop suite will likely
drop appendStream since that's the current MSE v1 consensus).

[servolk, 2016-08-02 22:59:14]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[servolk, 2016-08-02 22:59:15]

https://codereview.chromium.org/2195933002/diff/60001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/mediasource-util.js
(right):

https://codereview.chromium.org/2195933002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/mediasource-util.js:248:
MediaSourceUtil.fetchManifestAndData(test,
'webm/test-a-5min-44100Hz-1ch-manifest.json', function(type, mediaData)
On 2016/08/02 21:56:36, wolenetz wrote:
> nit: (not a regression, and certainly not a strong feeling, just a note if you
> could fix this easily here and now, please try to:) Allow for webm-lacking
> implementations to also pass this? It might ease interop test suite merger of
> this. (Ditto in the stream version, below, though the interop suite will
likely
> drop appendStream since that's the current MSE v1 consensus).

Sure we can easily pass in the manifest/stream URLs as input params in here, if
that's what you mean. Done in the latest patchset.

[commit-bot: I haz the power, 2016-08-02 22:59:45]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[wolenetz, 2016-08-02 23:32:27]

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendbuffer-quota-exceeded-1mb-buffers.html
(right):

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendbuffer-quota-exceeded-1mb-buffers.html:12:
var mediaDataManifest = 'webm/test-a-5min-44100Hz-1ch-manifest.json';
This is partially there. Note that the util's usage of "AUDIO_ONLY_TYPE" will be
mp4 for a browser that doesn't support MSE webm, but does support MSE mp4 (for
the specific codec strings in the utils). Please see
https://cs.chromium.org/chromium/src/third_party/WebKit/LayoutTests/http/test...
for an example of how to make this manifestURL covariant with
MediaSourceUtil.AUDIO_ONLY_TYPE

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendstream-quota-exceeded-1mb-buffers.html
(right):

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendstream-quota-exceeded-1mb-buffers.html:10:
var mediaURL =
"/media/resources/media-source/webm/test-a-5min-44100Hz-1ch.webm";
if we ended up *not* dropping appendStream from MSE v1, then this would need to
be variant based on the supported types in the browser (and the appendSize
551*1024 would need to change, too). No worries though.

[servolk, 2016-08-02 23:53:20]

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendbuffer-quota-exceeded-1mb-buffers.html
(right):

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendbuffer-quota-exceeded-1mb-buffers.html:12:
var mediaDataManifest = 'webm/test-a-5min-44100Hz-1ch-manifest.json';
On 2016/08/02 23:32:26, wolenetz wrote:
> This is partially there. Note that the util's usage of "AUDIO_ONLY_TYPE" will
be
> mp4 for a browser that doesn't support MSE webm, but does support MSE mp4 (for
> the specific codec strings in the utils). Please see
>
https://cs.chromium.org/chromium/src/third_party/WebKit/LayoutTests/http/test...
> for an example of how to make this manifestURL covariant with
> MediaSourceUtil.AUDIO_ONLY_TYPE

Done.

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendstream-quota-exceeded-1mb-buffers.html
(right):

https://codereview.chromium.org/2195933002/diff/80001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/http/tests/media/media-source/stream_memory_tests/mediasource-appendstream-quota-exceeded-1mb-buffers.html:10:
var mediaURL =
"/media/resources/media-source/webm/test-a-5min-44100Hz-1ch.webm";
On 2016/08/02 23:32:27, wolenetz wrote:
> if we ended up *not* dropping appendStream from MSE v1, then this would need
to
> be variant based on the supported types in the browser (and the appendSize
> 551*1024 would need to change, too). No worries though.

Acknowledged.

[servolk, 2016-08-02 23:53:42]

The CQ bit was checked by servolk@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 23:53:55]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-03 00:59:49]

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

[commit-bot: I haz the power, 2016-08-03 00:59:50]

Dry run: This issue passed the CQ dry run.

[servolk, 2016-08-03 15:44:01]

The CQ bit was checked by servolk@chromium.org

[servolk, 2016-08-03 15:44:02]

The patchset sent to the CQ was uploaded after l-g-t-m from
chcunningham@chromium.org, wolenetz@chromium.org
Link to the patchset: https://codereview.chromium.org/2195933002/#ps100001
(title: "Use GetSubType for building manifest URL")

[commit-bot: I haz the power, 2016-08-03 15:44:29]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-03 15:48:56]

Description was changed from

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
==========

to

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
==========

[commit-bot: I haz the power, 2016-08-03 15:48:59]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-08-03 15:50:56]

Description was changed from

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342
==========

to

==========
Added virtual test suite with 1MB MSE buffers

This will run media-source*quota-exceeded tests in a virtual test suite
with MSE audio and video buffer sizes set to 1MB. Normally those tests
time out in Debug mode, since the default MSE audio buffer size is 12MB
and that takes to long to fill up in Debug build.

BUG=630342

Committed: https://crrev.com/2f93caffe7a63bdcb391e23f67c67ceb69718a0f
Cr-Commit-Position: refs/heads/master@{#409518}
==========

[commit-bot: I haz the power, 2016-08-03 15:51:00]

Patchset 6 (id:??) landed as
https://crrev.com/2f93caffe7a63bdcb391e23f67c67ceb69718a0f
Cr-Commit-Position: refs/heads/master@{#409518}