| Summary: | [GPUP] Convert CDMFactory away from platformStrategies() and use WebProcess settings instead | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Jer Noble <jer.noble> | ||||||||||
| Component: | New Bugs | Assignee: | Jer Noble <jer.noble> | ||||||||||
| Status: | RESOLVED FIXED | ||||||||||||
| Severity: | Normal | CC: | aakash_jain, commit-queue, eric.carlson, ews-watchlist, glenn, peng.liu6, philipj, sergio, webkit-bug-importer, youennf | ||||||||||
| Priority: | P2 | Keywords: | InRadar | ||||||||||
| Version: | WebKit Nightly Build | ||||||||||||
| Hardware: | Unspecified | ||||||||||||
| OS: | Unspecified | ||||||||||||
| Attachments: |
|
||||||||||||
|
Description
Jer Noble
2020-03-05 22:38:48 PST
Created attachment 392682 [details]
Patch
Comment on attachment 392682 [details] Patch r=me as long as bots are green. View in context: https://bugs.webkit.org/attachment.cgi?id=392682&action=review > Source/WebKit/WebProcess/WebProcess.cpp:2011 > + cdmFactories.clear(); This seems a bit odd to get the vector, clear it and then refill it and passing it as a mutable parameter. I guess this might help for extensibility. Could we just have something like : static inline Vector<CDMFactory*> computeFactories() { if (useGPUProcessForMedia) return ensureGPUProcessConnection().cdmFactory().computeFactories(); return CDMFactory::computePlatformFactories(); } ... CDMFactory::registerFactories(computeFactories()) > Source/WebKit/WebProcess/WebProcess.cpp:2013 > + if (useGPUProcessForMedia) Probably missing a ENABLE(GPU_PROCESS) (In reply to Aakash Jain from comment #3) > Please check > https://ews-build.webkit.org/#/builders/8/builds/17625/steps/8/logs/stdio Yes, that's what Youenn's comment about a missing ENABLE(GPU_PROCESS) guard addresses. (In reply to youenn fablet from comment #2) > Comment on attachment 392682 [details] > Patch > > r=me as long as bots are green. > > View in context: > https://bugs.webkit.org/attachment.cgi?id=392682&action=review > > > Source/WebKit/WebProcess/WebProcess.cpp:2011 > > + cdmFactories.clear(); > > This seems a bit odd to get the vector, clear it and then refill it and > passing it as a mutable parameter. > I guess this might help for extensibility. I took a pass at this, but it requires a lot of re-architecture of port-specific code; it should be left as a clean-up task. Created attachment 392782 [details]
Patch for landing
ChangeLog entry in Source/WebKitLegacy/mac/ChangeLog contains OOPS!. Comment on attachment 392782 [details] Patch for landing Rejecting attachment 392782 [details] from commit-queue. Failed to run "['/Volumes/Data/EWS/WebKit/Tools/Scripts/webkit-patch', '--status-host=webkit-queues.webkit.org', '--bot-id=webkit-cq-01', 'validate-changelog', '--check-oops', '--non-interactive', 392782, '--port=mac']" exit_code: 1 cwd: /Volumes/Data/EWS/WebKit ChangeLog entry in Source/WebKitLegacy/mac/ChangeLog contains OOPS!. Full output: https://webkit-queues.webkit.org/results/13335013 Created attachment 392805 [details]
Patch for landing
Created attachment 392806 [details]
Patch for landing
Comment on attachment 392806 [details] Patch for landing Clearing flags on attachment: 392806 Committed r258040: <https://trac.webkit.org/changeset/258040> |