Bug 216203

Summary: [Cocoa,HDR] HLS streams with HDR variants will not select HDR.
Product: WebKit Reporter: Jer Noble <jer.noble>
Component: New BugsAssignee: Jer Noble <jer.noble>
Status: RESOLVED FIXED    
Severity: Normal CC: calvaris, cdumez, changseok, darin, eric.carlson, esprehn+autocc, ews-watchlist, glenn, gyuyoung.kim, mjs, philipj, sergio, webkit-bug-importer
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch for landing
none
Patch for landing
ews-feeder: commit-queue-
Patch for Landing
none
Patch for Landing none

Jer Noble
Reported 2020-09-04 17:02:18 PDT
[Cocoa,HDR] HLS streams with HDR variants will not select HDR.
Attachments
Patch (1.97 KB, patch)
2020-09-04 17:08 PDT, Jer Noble 		no flags
DetailsFormatted DiffDiff
Patch for landing (118.48 KB, patch)
2020-09-08 17:29 PDT, Jer Noble 		no flags
DetailsFormatted DiffDiff
Patch for landing (119.16 KB, patch)
2020-09-11 11:35 PDT, Jer Noble 		ews-feeder: commit-queue-
DetailsFormatted DiffDiff
Patch for Landing (119.16 KB, patch)
2020-09-11 12:56 PDT, Jer Noble 		no flags
DetailsFormatted DiffDiff
Patch for Landing (118.85 KB, patch)
2020-09-14 15:10 PDT, Jer Noble 		no flags
DetailsFormatted DiffDiff
Show Obsolete (4) View All Add attachment proposed patch, testcase, etc.
Jer Noble
Comment 1 2020-09-04 17:02:45 PDT
<rdar://problem/67438626>
Jer Noble
Comment 2 2020-09-04 17:08:26 PDT
Created attachment 408047 [details] Patch
Darin Adler
Comment 3 2020-09-04 17:09:51 PDT
Comment on attachment 408047 [details] Patch Regression testing?
Jer Noble
Comment 4 2020-09-04 17:51:36 PDT
(In reply to Darin Adler from comment #3) > Comment on attachment 408047 [details] > Patch > > Regression testing? It’s hard; I don’t know that we can write a consistent regression test that isn’t flakey here. For one, it’s very hard to determine what variant HLS chooses; AVFoundation goes out of its way to make that difficult. For two, anything we do which _could_ be detectable will perturb HLS’s selection logic. E.g., giving the HDR and SDR variants different sizes would make the selection detectable, but might lead HLS to choose among the smaller or larger variant, rather than the HDR or SDR one. For three, HLS may only switch up to HDR _eventually_, not initially, so the test may end up taking >10s to run, which on the bots means it would be a flakey timeout. I’ll see what I can do; perhaps a stream with very short segment times, like 0.5s each, may make the switching happen quicker.
Jer Noble
Comment 5 2020-09-08 17:29:13 PDT
Created attachment 408293 [details] Patch for landing
Jer Noble
Comment 6 2020-09-11 11:35:56 PDT
Created attachment 408549 [details] Patch for landing
Jer Noble
Comment 7 2020-09-11 12:56:57 PDT
Created attachment 408555 [details] Patch for Landing
EWS
Comment 8 2020-09-14 09:20:54 PDT
Tools/Scripts/svn-apply failed to apply attachment 408555 [details] to trunk. Please resolve the conflicts and upload a new patch.
Jer Noble
Comment 9 2020-09-14 15:10:56 PDT
Created attachment 408749 [details] Patch for Landing
EWS
Comment 10 2020-09-14 16:17:58 PDT
Committed r267057: <https://trac.webkit.org/changeset/267057> All reviewed patches have been landed. Closing bug and clearing flags on attachment 408749 [details].
Note You need to log in before you can comment on or make changes to this bug.