WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
Bug 81785
[GTK] Build fix for Accelerated Compositing with Clutter
https://bugs.webkit.org/show_bug.cgi?id=81785
Summary
[GTK] Build fix for Accelerated Compositing with Clutter
Joone Hur
Reported
2012-03-21 08:37:46 PDT
AcceleratedCompositingContext was introduced to isolate different accelerated compositing implementations:
r104194
(
https://bugs.webkit.org/show_bug.cgi?id=75519
). However, Clutter implementation doesn't build with it.
Attachments
Proposed Patch
(6.21 KB, patch)
2012-03-21 09:27 PDT
,
Joone Hur
no flags
Details
Formatted Diff
Diff
Proposed Patch2
(19.31 KB, patch)
2012-03-25 23:54 PDT
,
Joone Hur
no flags
Details
Formatted Diff
Diff
Proposed Patch3
(19.31 KB, patch)
2012-03-26 00:15 PDT
,
Joone Hur
no flags
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
Joone Hur
Comment 1
2012-03-21 09:27:31 PDT
Created
attachment 133060
[details]
Proposed Patch
Martin Robinson
Comment 2
2012-03-21 10:26:43 PDT
Comment on
attachment 133060
[details]
Proposed Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=133060&action=review
> Source/WebCore/platform/graphics/cairo/GraphicsContext3DCairo.cpp:39 > #include "Extensions3DOpenGL.h" > #include "GraphicsContext3DPrivate.h" > #include "Image.h" > +#include "NotImplemented.h" > #include "OpenGLShims.h" > #include "PlatformContextCairo.h" > #include "RefPtrCairo.h"
Clutter is going to need its own GraphicsContext3D right? Why not just add a stub?
> Source/WebKit/gtk/WebCoreSupport/AcceleratedCompositingContext.h:75 > - OwnPtr<WebCore::GraphicsLayer> m_rootGraphicsLayer; > unsigned int m_syncTimerCallbackId; > > #if USE(CLUTTER) > + WebCore::GraphicsLayer* m_rootGraphicsLayer; > GtkWidget* m_rootLayerEmbedder; > #elif USE(TEXTURE_MAPPER_GL) > WebCore::GLContext* glContext(); > WebCore::TextureMapperLayer* m_rootTextureMapperLayer; > + OwnPtr<WebCore::GraphicsLayer> m_rootGraphicsLayer; > OwnPtr<WebCore::TextureMapper> m_textureMapper;
Why can't this be shared?
Joone Hur
Comment 3
2012-03-21 22:34:21 PDT
Comment on
attachment 133060
[details]
Proposed Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=133060&action=review
>> Source/WebCore/platform/graphics/cairo/GraphicsContext3DCairo.cpp:39 >> #include "RefPtrCairo.h" > > Clutter is going to need its own GraphicsContext3D right? Why not just add a stub?
I will add GraphicsContext3DClutter later. This patch just fixes the build error.
>> Source/WebKit/gtk/WebCoreSupport/AcceleratedCompositingContext.h:75 >> OwnPtr<WebCore::TextureMapper> m_textureMapper; > > Why can't this be shared?
Currently, we just use the given layer for the root layer, but I will create a root layer to add the given layer as a child. Thanks you for your comments.
Martin Robinson
Comment 4
2012-03-22 07:49:09 PDT
(In reply to
comment #3
)
> (From update of
attachment 133060
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=133060&action=review
> > >> Source/WebCore/platform/graphics/cairo/GraphicsContext3DCairo.cpp:39 > >> #include "RefPtrCairo.h" > > > > Clutter is going to need its own GraphicsContext3D right? Why not just add a stub? > > I will add GraphicsContext3DClutter later. This patch just fixes the build error.
I think it makes sense to add the stub now, since in the end the file will never be shared between the two.
> > >> Source/WebKit/gtk/WebCoreSupport/AcceleratedCompositingContext.h:75 > >> OwnPtr<WebCore::TextureMapper> m_textureMapper; > > > > Why can't this be shared? > > Currently, we just use the given layer for the root layer, but I will create a root layer to add the given layer as a child.
There's no need. Now that I understand why you are doing this, I don't think it needs to be changed.
Joone Hur
Comment 5
2012-03-25 23:53:19 PDT
Comment on
attachment 133060
[details]
Proposed Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=133060&action=review
>>>> Source/WebCore/platform/graphics/cairo/GraphicsContext3DCairo.cpp:39 >>>> #include "RefPtrCairo.h" >>> >>> Clutter is going to need its own GraphicsContext3D right? Why not just add a stub? >> >> I will add GraphicsContext3DClutter later. This patch just fixes the build error. > > I think it makes sense to add the stub now, since in the end the file will never be shared between the two.
Okay, I will add GraphicsContext3D stub to be built with Clutter.
Joone Hur
Comment 6
2012-03-25 23:54:28 PDT
Created
attachment 133737
[details]
Proposed Patch2
WebKit Review Bot
Comment 7
2012-03-25 23:58:17 PDT
Attachment 133737
[details]
did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebCore/ChangeLog', u'Source/WebCor..." exit_code: 1 Source/WebCore/platform/graphics/clutter/GraphicsContext3DPrivate.h:34: This { should be at the end of the previous line [whitespace/braces] [4] Source/WebCore/platform/graphics/clutter/GraphicsContext3DPrivate.h:40: Missing space before { [whitespace/braces] [5] Source/WebCore/platform/graphics/clutter/GraphicsContext3DPrivate.h:43: The parameter name "matrix" adds no information, so it should be removed. [readability/parameter_name] [5] Total errors found: 3 in 9 files If any of these errors are false positives, please file a bug against check-webkit-style.
Joone Hur
Comment 8
2012-03-26 00:15:12 PDT
Created
attachment 133740
[details]
Proposed Patch3 Fixed style errors.
Martin Robinson
Comment 9
2012-03-26 08:29:22 PDT
Comment on
attachment 133740
[details]
Proposed Patch3 Thanks for cleaning up my mess!
Joone Hur
Comment 10
2012-03-26 12:16:18 PDT
Comment on
attachment 133740
[details]
Proposed Patch3 Thank you! :-)
WebKit Review Bot
Comment 11
2012-03-26 12:41:58 PDT
Comment on
attachment 133740
[details]
Proposed Patch3 Clearing flags on attachment: 133740 Committed
r112141
: <
http://trac.webkit.org/changeset/112141
>
WebKit Review Bot
Comment 12
2012-03-26 12:42:03 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