[Webkit-unassigned] [Bug 184261] [GTK][WPE] Memory pressure monitor doesn't reliable notify all the subprocesses

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=184261

--- Comment #1 from Carlos Alberto Lopez Perez <clopez at igalia.com> ---
Created attachment 337074

  --> https://bugs.webkit.org/attachment.cgi?id=337074&action=review

Test case to demonstrate the issue

The following patch demonstrates the issue. It patches WebKit to simulate a memory event notification each 5 seconds and then prints which process have received it.
In theory all process always should receive the notification, in practice many are missed. Example:


[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/MiniBrowser(7969)]TRIGGER MEMORY PRESSURE EVENT and sleep 5
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitNetworkProcess(8004)]Received event to trigger a memory pressure event


[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/MiniBrowser(7969)]TRIGGER MEMORY PRESSURE EVENT and sleep 5
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitWebProcess(8002)]Received event to trigger a memory pressure event


[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/MiniBrowser(7969)]TRIGGER MEMORY PRESSURE EVENT and sleep 5
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitNetworkProcess(8004)]Received event to trigger a memory pressure event
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitWebProcess(8002)]Received event to trigger a memory pressure event


[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/MiniBrowser(7969)]TRIGGER MEMORY PRESSURE EVENT and sleep 5
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitWebProcess(8002)]Received event to trigger a memory pressure event


[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/MiniBrowser(7969)]TRIGGER MEMORY PRESSURE EVENT and sleep 5
[/home/igalia/clopez/webkit/webkit/WebKitBuild/Release/bin/WebKitWebProcess(8002)]Received event to trigger a memory pressure event


As you can see only once the network and web process received it at the same time, the other times only one of this two got it.
And this is without opening multiple tabs (just only one webprocess and one network process)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180403/2be0765c/attachment-0002.html>