Bug 205541 – [Media in GPU process] Add remote media player configuration

• Component: Media
• Status: RESOLVED FIXED

[Eric Carlson]

Add remote media player configuration for properties that won't change

[Radar WebKit Bug Importer]

<rdar://problem/58137418>

[Eric Carlson]

Created attachment 386303 [details]
Patch

[Eric Carlson]

Created attachment 386304 [details]
Patch

[Simon Fraser (smfr)]

Comment on attachment 386304 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=386304&action=review

> Source/WebKit/GPUProcess/media/RemoteMediaPlayerProxyConfiguration.h:46
> +    uint64_t logIdentifier;
> +    bool shouldUsePersistentCache;
> +    bool isVideo;

Maybe initialize these to be safe.

> Source/WebKit/GPUProcess/media/RemoteMediaPlayerProxyConfiguration.h:122
> +        return {{ WTFMove(*mediaKeysStorageDirectory), WTFMove(*referrer), WTFMove(*userAgent), WTFMove(*sourceApplicationIdentifier), WTFMove(*networkInterfaceName), WTFMove(*mediaCacheDirectory), WTFMove(*mediaContentTypesRequiringHardwareSupport), WTFMove(*preferredAudioCharacteristics), WTFMove(*logIdentifier), WTFMove(*shouldUsePersistentCache), WTFMove(*isVideo), }};

I'd put each WTFMove on its own line.

> Source/WebKit/WebProcess/GPU/media/RemoteMediaPlayerManager.cpp:130
> +    configuration.logIdentifier = reinterpret_cast<uint64_t>(player->mediaPlayerLogIdentifier());

Are these identifiers unique across all processes that share a GPU process?

[Eric Carlson]

Created attachment 386307 [details]
Patch for landing

[Eric Carlson]

Comment on attachment 386304 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=386304&action=review

>> Source/WebKit/GPUProcess/media/RemoteMediaPlayerProxyConfiguration.h:46
>> +    bool isVideo;
> 
> Maybe initialize these to be safe.

Good point, fixed.

>> Source/WebKit/GPUProcess/media/RemoteMediaPlayerProxyConfiguration.h:122
>> +        return {{ WTFMove(*mediaKeysStorageDirectory), WTFMove(*referrer), WTFMove(*userAgent), WTFMove(*sourceApplicationIdentifier), WTFMove(*networkInterfaceName), WTFMove(*mediaCacheDirectory), WTFMove(*mediaContentTypesRequiringHardwareSupport), WTFMove(*preferredAudioCharacteristics), WTFMove(*logIdentifier), WTFMove(*shouldUsePersistentCache), WTFMove(*isVideo), }};
> 
> I'd put each WTFMove on its own line.

Done.

>> Source/WebKit/WebProcess/GPU/media/RemoteMediaPlayerManager.cpp:130
>> +    configuration.logIdentifier = reinterpret_cast<uint64_t>(player->mediaPlayerLogIdentifier());
> 
> Are these identifiers unique across all processes that share a GPU process?

A media player's unique identifier is the same as the HTMLMediaElement that owns it, but each HTMLMediaElement has a unique ID.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 386307 [details]:

imported/w3c/web-platform-tests/IndexedDB/key-generators/reading-autoincrement-indexes-cursors.any.html bug 205545 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

Comment on attachment 386307 [details]
Patch for landing

Clearing flags on attachment: 386307

Committed r253866: <https://trac.webkit.org/changeset/253866>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.