Bug 76751
| Summary: | Everyone now implements RunLoop | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Sam Weinig <sam> |
| Component: | Platform | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED CONFIGURATION CHANGED | ||
| Severity: | Normal | CC: | fishd, gram, jamesr, levin, schenney, skyul |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | All | ||
| OS: | All | ||
Sam Weinig
We have RunLoop (WebCore/platform/RunLoop.h) implementations for Mac, Windows, Qt, and Gtk. To make it usable in WebCore, we need a Chromium implementation as well.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
James Robinson
I think in Chromium this could be implemented on top of WebThread once https://bugs.webkit.org/show_bug.cgi?id=76012 lands.
Sam Weinig
(In reply to comment #1)
> I think in Chromium this could be implemented on top of WebThread once https://bugs.webkit.org/show_bug.cgi?id=76012 lands.
That seems like it would be wrong from a layering perspective. WebCore should not depend on WebKit. Perhaps WebThread could be implemented in terms of RunLoop, or whatever WebThread is implemented in terms of could be the basis for both.
James Robinson
(In reply to comment #2)
> (In reply to comment #1)
> > I think in Chromium this could be implemented on top of WebThread once https://bugs.webkit.org/show_bug.cgi?id=76012 lands.
>
> That seems like it would be wrong from a layering perspective. WebCore should not depend on WebKit. Perhaps WebThread could be implemented in terms of RunLoop, or whatever WebThread is implemented in terms of could be the basis for both.
We're planning to move WebThread to Platform/ to resolve this.
Sam Weinig
(In reply to comment #3)
> (In reply to comment #2)
> > (In reply to comment #1)
> > > I think in Chromium this could be implemented on top of WebThread once https://bugs.webkit.org/show_bug.cgi?id=76012 lands.
> >
> > That seems like it would be wrong from a layering perspective. WebCore should not depend on WebKit. Perhaps WebThread could be implemented in terms of RunLoop, or whatever WebThread is implemented in terms of could be the basis for both.
>
> We're planning to move WebThread to Platform/ to resolve this.
Hm. The Source/Platform directory is very confusing to me at the moment. It was supposed to be where we moved Source/WebCore/platform initially, but it seems like it is holding some Chrome API right now, and nothing that should be used from cross platform code.
James Robinson
(In reply to comment #4)
> (In reply to comment #3)
> > (In reply to comment #2)
> > > (In reply to comment #1)
> > > > I think in Chromium this could be implemented on top of WebThread once https://bugs.webkit.org/show_bug.cgi?id=76012 lands.
> > >
> > > That seems like it would be wrong from a layering perspective. WebCore should not depend on WebKit. Perhaps WebThread could be implemented in terms of RunLoop, or whatever WebThread is implemented in terms of could be the basis for both.
> >
> > We're planning to move WebThread to Platform/ to resolve this.
>
> Hm. The Source/Platform directory is very confusing to me at the moment. It was supposed to be where we moved Source/WebCore/platform initially, but it seems like it is holding some Chrome API right now, and nothing that should be used from cross platform code.
That's just because we haven't figured out how to move WebCore/platform code into there in a way that compiles on all platforms yet. I've been planning to wait for the Source/WTF move to happen successfully, then do whatever was necessary to make that happen and start moving WebCore/platform code there.
Stephen Chenney
There is no more Chromium in WebKit, so whatever depends on this can now move forward.