WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
136806
Web Inspector: InjectedScripts should not be profiled or displayed in Timeline
https://bugs.webkit.org/show_bug.cgi?id=136806
Summary
Web Inspector: InjectedScripts should not be profiled or displayed in Timeline
Brian Burg
Reported
2014-09-14 11:46:49 PDT
It doesn't make sense to show these in the timeline, at least for inspector^1. They only really show up when debugging, they generate a lot of profile data, and are sort of irrelevant for the purpose of profiling user content. I can think of some uses for seeing injected scripts in the timeline for inspector^2 and above. But, I'm not quite sure how to distinguish that from inside the engine (look at the PageGroup or something?)
Attachments
WIP
(8.83 KB, patch)
2014-09-24 16:15 PDT
,
Brian Burg
no flags
Details
Formatted Diff
Diff
Patch
(8.68 KB, patch)
2014-09-29 16:22 PDT
,
Brian Burg
no flags
Details
Formatted Diff
Diff
Show Obsolete
(1)
View All
Add attachment
proposed patch, testcase, etc.
Radar WebKit Bug Importer
Comment 1
2014-09-14 11:46:57 PDT
<
rdar://problem/18332123
>
Timothy Hatcher
Comment 2
2014-09-15 11:09:33 PDT
Good point. We do hide the InjectedScript from backtraces in the Debugger. We should be able to do it for the profiler too. It would make sense to show it for Inspector^2. That would be via the PageGroup identifier.
Brian Burg
Comment 3
2014-09-15 12:15:30 PDT
(In reply to
comment #2
)
> Good point. We do hide the InjectedScript from backtraces in the Debugger. We should be able to do it for the profiler too. > > It would make sense to show it for Inspector^2. That would be via the PageGroup identifier.
The profiler can distinguish different page group ids, but I don't think the inspector "level" is exposed anywhere outside of UIProcess. What's a good way to expose that? WebInspectorClient? My WIP patch has a ProfileGenerator::setSuspended flag that is set and unset under the InspectorController will/didCallInjectedScript.
Brian Burg
Comment 4
2014-09-24 16:15:43 PDT
Created
attachment 238621
[details]
WIP
Timothy Hatcher
Comment 5
2014-09-24 21:31:00 PDT
Comment on
attachment 238621
[details]
WIP View in context:
https://bugs.webkit.org/attachment.cgi?id=238621&action=review
Makes sense to me.
> Source/WebCore/inspector/InspectorController.cpp:-427 > - InspectorInstrumentationCookie cookie = InspectorInstrumentation::willCallFunction(scriptExecutionContext, scriptName, scriptLine);
Yeah I never understood why this was instrumented.
Brian Burg
Comment 6
2014-09-29 16:22:33 PDT
Created
attachment 238895
[details]
Patch
WebKit Commit Bot
Comment 7
2014-09-29 16:23:33 PDT
Attachment 238895
[details]
did not pass style-queue: ERROR: Source/JavaScriptCore/profiler/ProfileGenerator.cpp:53: Wrong number of spaces before statement. (expected: 8) [whitespace/indent] [4] Total errors found: 1 in 7 files If any of these errors are false positives, please file a bug against check-webkit-style.
WebKit Commit Bot
Comment 8
2014-09-29 20:34:16 PDT
Comment on
attachment 238895
[details]
Patch Clearing flags on attachment: 238895 Committed
r174095
: <
http://trac.webkit.org/changeset/174095
>
WebKit Commit Bot
Comment 9
2014-09-29 20:34:19 PDT
All reviewed patches have been landed. Closing bug.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug