Bug 95275 – [Chromium] Update some compositor trace events

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[John Bates]

[Chromium] Update some compositor trace events

[John Bates]

Created attachment 161107 [details]
Patch

[John Bates]

ptal

[James Robinson]

Comment on attachment 161107 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=161107&action=review

> Source/WebCore/platform/graphics/chromium/cc/CCFrameRateController.cpp:84
> -    TRACE_EVENT1("cc", "CCFrameRateController::setActive", "active", active);
>      if (m_active == active)
>          return;
> +    TRACE_EVENT1("cc", "CCFrameRateController::setActive", "active", active);

so is the thread code now working (https://bugs.webkit.org/show_bug.cgi?id=93918#c3) ?

> Source/WebCore/platform/graphics/chromium/cc/CCThreadProxy.cpp:331
> +    TRACE_EVENT2("cc", "CCThreadProxy::onVSyncParametersChanged", "monotonicTimebase", monotonicTimebase, "intervalInSeconds", intervalInSeconds);

Hm, the tracing code can take doubles? That's nice - but the type of the value parameter looks like unsigned long long to me. Do we need any casts?

[John Bates]

(In reply to comment #3)
> (From update of attachment 161107 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=161107&action=review
> 
> > Source/WebCore/platform/graphics/chromium/cc/CCFrameRateController.cpp:84
> > -    TRACE_EVENT1("cc", "CCFrameRateController::setActive", "active", active);
> >      if (m_active == active)
> >          return;
> > +    TRACE_EVENT1("cc", "CCFrameRateController::setActive", "active", active);
> 
> so is the thread code now working (https://bugs.webkit.org/show_bug.cgi?id=93918#c3) ?

yeah, I found the issue a while ago.

> 
> > Source/WebCore/platform/graphics/chromium/cc/CCThreadProxy.cpp:331
> > +    TRACE_EVENT2("cc", "CCThreadProxy::onVSyncParametersChanged", "monotonicTimebase", monotonicTimebase, "intervalInSeconds", intervalInSeconds);
> 
> Hm, the tracing code can take doubles? That's nice - but the type of the value parameter looks like unsigned long long to me. Do we need any casts?

double should go through - the macros reinterpret it to/from unsigned long long.

[James Robinson]

Comment on attachment 161107 [details]
Patch

OK

[WebKit Review Bot]

Comment on attachment 161107 [details]
Patch

Clearing flags on attachment: 161107

Committed r126954: <http://trac.webkit.org/changeset/126954>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.