Bug 222465 – Allow access to more cmio services from Web/GPU processes

• Status: RESOLVED
• Resolution: FIXED
• Component: WebRTC

[youenn fablet]

Allow access to more cmio services from Web/GPU processes

[Radar WebKit Bug Importer]

<rdar://problem/74783208>

[youenn fablet]

<rdar://problem/74782404>

[youenn fablet]

Created attachment 421626 [details]
Patch

[EWS]

[e-mail] does not have committer permissions according to https://raw.githubusercontent.com/WebKit/WebKit/main/Tools/Scripts/webkitpy/common/config/contributors.json.

Rejecting attachment 421626 [details] from commit queue.

[youenn fablet]

Comment on attachment 421626 [details]
Patch

Let's wait for PerArne to validate this.

[Per Arne Vollan]

Comment on attachment 421626 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=421626&action=review

R=me.

> Source/WebKit/WebProcess/com.apple.WebProcess.sb.in:1401
> +        (global-name "com.apple.cmio.registerassistantservice")

Are these needed in the WebContent process when they are present in the GPU process' sandbox?

[youenn fablet]

(In reply to Per Arne Vollan from comment #6)
> Comment on attachment 421626 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=421626&action=review
> 
> R=me.
> 
> > Source/WebKit/WebProcess/com.apple.WebProcess.sb.in:1401
> > +        (global-name "com.apple.cmio.registerassistantservice")
> 
> Are these needed in the WebContent process when they are present in the GPU
> process' sandbox?

They are only needed in WebContent process when capturing in WebProcess.

[Per Arne Vollan]

Comment on attachment 421626 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=421626&action=review

>>> Source/WebKit/WebProcess/com.apple.WebProcess.sb.in:1401
>>> +        (global-name "com.apple.cmio.registerassistantservice")
>> 
>> Are these needed in the WebContent process when they are present in the GPU process' sandbox?
> 
> They are only needed in WebContent process when capturing in WebProcess.

Are they currently strictly needed in WebProcess, or can we rely on capturing being done in GPU process going forward?

[youenn fablet]

(In reply to Per Arne Vollan from comment #8)
> Comment on attachment 421626 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=421626&action=review
> 
> >>> Source/WebKit/WebProcess/com.apple.WebProcess.sb.in:1401
> >>> +        (global-name "com.apple.cmio.registerassistantservice")
> >> 
> >> Are these needed in the WebContent process when they are present in the GPU process' sandbox?
> > 
> > They are only needed in WebContent process when capturing in WebProcess.
> 
> Are they currently strictly needed in WebProcess, or can we rely on
> capturing being done in GPU process going forward?

The idea is indeed to do capture in GPUProcess going forward.
These services are allowed with a camera sandbox extension.
We should not send it to WebProcess it if capturing happens in GPUProcess.

[EWS]

Committed r274138: <https://commits.webkit.org/r274138>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 421626 [details].