Bug 212970 - BaseAudioSharedUnit should unmute its clients in case of suspension even if not having any audio unit
Summary: BaseAudioSharedUnit should unmute its clients in case of suspension even if n...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebRTC (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: youenn fablet
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2020-06-09 06:30 PDT by youenn fablet
Modified: 2020-06-09 11:16 PDT (History)
10 users (show)

See Also:

Attachments
Patch (13.81 KB, patch)
2020-06-09 06:41 PDT, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (13.81 KB, patch)
2020-06-09 08:37 PDT, youenn fablet 		no flags Details | Formatted Diff | Diff
Show Obsolete (1) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description youenn fablet 2020-06-09 06:30:35 PDT 
BaseAudioSharedUnit should unmute its clients in case of suspension even if not having any audio unit
Comment 1 Radar WebKit Bug Importer 2020-06-09 06:31:30 PDT 
<rdar://problem/64161891>
Comment 2 youenn fablet 2020-06-09 06:41:52 PDT 
Created attachment 401437 [details]
Patch
Comment 3 Eric Carlson 2020-06-09 07:15:35 PDT 
Comment on attachment 401437 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=401437&action=review

> Source/WebCore/ChangeLog:9
> +        This will, in turn, make the BaseAudioSharedUnit to stop and no longer have any audio unit.

s/the BaseAudioSharedUnit to stop/the BaseAudioSharedUnit stop/

> Source/WebCore/platform/mediastream/mac/BaseAudioSharedUnit.cpp:179
> +    ASSERT(!m_producingCount);

Is this backwards?
Comment 4 youenn fablet 2020-06-09 08:36:57 PDT 
Comment on attachment 401437 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=401437&action=review

>> Source/WebCore/platform/mediastream/mac/BaseAudioSharedUnit.cpp:179
>> +    ASSERT(!m_producingCount);
> 
> Is this backwards?

Given the current model, m_producingCount should be zero.
I guess I could remove the if(m_producingCount) check.
Comment 5 youenn fablet 2020-06-09 08:37:21 PDT 
Created attachment 401442 [details]
Patch
Comment 6 Eric Carlson 2020-06-09 08:51:14 PDT 
Comment on attachment 401442 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=401442&action=review

> Source/WebCore/platform/mediastream/mac/BaseAudioSharedUnit.cpp:180
>      if (m_producingCount) {

This is confusing given the assert.
Comment 7 youenn fablet 2020-06-09 11:03:51 PDT 
(In reply to Eric Carlson from comment #6)
> Comment on attachment 401442 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=401442&action=review
> 
> > Source/WebCore/platform/mediastream/mac/BaseAudioSharedUnit.cpp:180
> >      if (m_producingCount) {
> 
> This is confusing given the assert.

Will remove it in a follow-up
Comment 8 EWS 2020-06-09 11:16:13 PDT 
Committed r262798: <https://trac.webkit.org/changeset/262798>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 401442 [details].