Bug 211120 - Delay SystemStatus call from process launch to when being useful
Summary: Delay SystemStatus call from process launch to when being useful
Status: RESOLVED CONFIGURATION CHANGED
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:
Depends on:
Blocks:
 
Reported: 2020-04-28 04:55 PDT by youenn fablet
Modified: 2020-04-29 06:35 PDT (History)
13 users (show)

See Also:

Attachments
Patch (21.24 KB, patch)
2020-04-28 05:26 PDT, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (22.65 KB, patch)
2020-04-28 05:28 PDT, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (26.31 KB, patch)
2020-04-28 07:48 PDT, youenn fablet 		no flags Details | Formatted Diff | Diff
Show Obsolete (2) 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 2020-04-28 04:55:17 PDT 
Delay SystemAttribution call from process launch to when being useful
Comment 1 youenn fablet 2020-04-28 05:26:41 PDT 
Created attachment 397830 [details]
Patch
Comment 2 youenn fablet 2020-04-28 05:28:15 PDT 
Created attachment 397831 [details]
Patch
Comment 3 Geoffrey Garen 2020-04-28 06:56:08 PDT 
Comment on attachment 397831 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=397831&action=review

> Source/WebKit/Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:178
> +        (allow mach-lookup (global-name "com.apple.systemstatus.activityattribution"))

Sorry, I should have realized this earlier — I don’t think we can take this approach. Since we’re no loner calling the status function on startup, this patch has to include an unsafe privileged service in the sandbox. That’s a pretty serious sandbox regression relative to all the work we’ve done lately. I don’t think that’s ok. 

I guess we need to talk to the systemstatus and integration folks to find some way forward that allows strong linking. 😞
Comment 4 youenn fablet 2020-04-28 07:48:57 PDT 
Created attachment 397836 [details]
Patch
Comment 5 Eric Carlson 2020-04-28 09:24:33 PDT 
Comment on attachment 397836 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=397836&action=review

> Source/WebCore/platform/mediastream/RealtimeMediaSourceCenter.cpp:41
> +#include "PlatformMediaSessionManager.h"

Nit: is this necessary?

> Source/WebCore/platform/mock/MockRealtimeMediaSourceCenter.cpp:39
> +#include "PlatformMediaSessionManager.h"

Ditto
Comment 6 youenn fablet 2020-04-29 06:35:17 PDT 
We will go with another approach that is much less complex.