WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
108203
Allow direct compositing of background images
https://bugs.webkit.org/show_bug.cgi?id=108203
Summary
Allow direct compositing of background images
Noam Rosenthal
Reported
2013-01-29 10:33:13 PST
Allow direct compositing of background images
Attachments
Patch
(60.39 KB, patch)
2013-01-29 10:55 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(60.83 KB, patch)
2013-01-30 02:35 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(61.60 KB, patch)
2013-01-30 05:48 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(61.58 KB, patch)
2013-01-30 07:44 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(59.84 KB, patch)
2013-01-30 10:11 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(30.57 KB, patch)
2013-02-01 02:22 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(30.78 KB, patch)
2013-02-02 03:13 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(31.04 KB, patch)
2013-02-02 07:09 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(31.04 KB, patch)
2013-02-04 09:54 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(1.08 MB, patch)
2013-02-09 09:38 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(1.29 MB, patch)
2013-02-09 15:31 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(1.29 MB, patch)
2013-02-10 02:44 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(30.64 KB, patch)
2013-02-11 04:14 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(30.69 KB, patch)
2013-02-12 08:45 PST
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(137.47 KB, patch)
2013-03-18 09:57 PDT
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(34.67 KB, patch)
2013-03-22 03:48 PDT
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Archive of layout-test-results from gce-cr-linux-05
(1.26 MB, application/zip)
2013-03-22 05:16 PDT
,
WebKit Review Bot
no flags
Details
Archive of layout-test-results from gce-cr-linux-06
(1.92 MB, application/zip)
2013-03-22 05:54 PDT
,
WebKit Review Bot
no flags
Details
Patch
(34.77 KB, patch)
2013-03-22 06:13 PDT
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Patch
(34.80 KB, patch)
2013-04-10 10:38 PDT
,
Noam Rosenthal
no flags
Details
Formatted Diff
Diff
Show Obsolete
(17)
View All
Add attachment
proposed patch, testcase, etc.
Noam Rosenthal
Comment 1
2013-01-29 10:55:27 PST
Created
attachment 185267
[details]
Patch
Simon Fraser (smfr)
Comment 2
2013-01-29 11:08:21 PST
Comment on
attachment 185267
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=185267&action=review
> Source/WebCore/platform/graphics/GraphicsLayer.h:330 > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > + virtual void setContentsTileRect(const IntRect& r) { m_contentsTileRect = r; } > + IntRect contentsTileRect() const { return m_contentsTileRect; }
Are you only supporting tiled background images? What about tiling just in x or y, or untiled?
> Source/WebCore/rendering/RenderLayerBacking.cpp:1373 > + if (renderer()->isImage()) > + return;
This seems a bit out of place here. I would not expect to get into this function for an image.
Noam Rosenthal
Comment 3
2013-01-29 11:10:31 PST
(In reply to
comment #2
)
> (From update of
attachment 185267
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=185267&action=review
> > > Source/WebCore/platform/graphics/GraphicsLayer.h:330 > > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > > + virtual void setContentsTileRect(const IntRect& r) { m_contentsTileRect = r; } > > + IntRect contentsTileRect() const { return m_contentsTileRect; } > > Are you only supporting tiled background images? What about tiling just in x or y, or untiled?
> Supporting x, y and untiled as well, but those don't require a contentsTileRect.
> > Source/WebCore/rendering/RenderLayerBacking.cpp:1373 > > + if (renderer()->isImage()) > > + return; > > This seems a bit out of place here. I would not expect to get into this function for an image.
Right.
Simon Fraser (smfr)
Comment 4
2013-01-29 11:11:33 PST
(In reply to
comment #3
)
> (In reply to
comment #2
) > > (From update of
attachment 185267
[details]
[details]) > > View in context:
https://bugs.webkit.org/attachment.cgi?id=185267&action=review
> > > > > Source/WebCore/platform/graphics/GraphicsLayer.h:330 > > > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > > > + virtual void setContentsTileRect(const IntRect& r) { m_contentsTileRect = r; } > > > + IntRect contentsTileRect() const { return m_contentsTileRect; } > > > > Are you only supporting tiled background images? What about tiling just in x or y, or untiled? > > > Supporting x, y and untiled as well, but those don't require a contentsTileRect.
I didn't see functions on GraphicsLayer to specify the tiling behavior.
Noam Rosenthal
Comment 5
2013-01-29 11:13:46 PST
(In reply to
comment #4
)
> (In reply to
comment #3
) > > (In reply to
comment #2
) > > > (From update of
attachment 185267
[details]
[details] [details]) > > > View in context:
https://bugs.webkit.org/attachment.cgi?id=185267&action=review
> > > > > > > Source/WebCore/platform/graphics/GraphicsLayer.h:330 > > > > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > > > > + virtual void setContentsTileRect(const IntRect& r) { m_contentsTileRect = r; } > > > > + IntRect contentsTileRect() const { return m_contentsTileRect; } > > > > > > Are you only supporting tiled background images? What about tiling just in x or y, or untiled? > > > > > Supporting x, y and untiled as well, but those don't require a contentsTileRect. > > I didn't see functions on GraphicsLayer to specify the tiling behavior.
It's not necessary. RenderBoxObjectModel::getGeometryForBackgroundImage takes care of that. For example if the background is tiled only horizontally, the contentsRect would be cropped to include only the tiled row.
Noam Rosenthal
Comment 6
2013-01-29 11:15:23 PST
(In reply to
comment #5
)
> (In reply to
comment #4
) > > (In reply to
comment #3
) > > > (In reply to
comment #2
) > > > > (From update of
attachment 185267
[details]
[details] [details] [details]) > > > > View in context:
https://bugs.webkit.org/attachment.cgi?id=185267&action=review
> > > > > > > > > Source/WebCore/platform/graphics/GraphicsLayer.h:330 > > > > > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > > > > > + virtual void setContentsTileRect(const IntRect& r) { m_contentsTileRect = r; } > > > > > + IntRect contentsTileRect() const { return m_contentsTileRect; } > > > > > > > > Are you only supporting tiled background images? What about tiling just in x or y, or untiled? > > > > > > > Supporting x, y and untiled as well, but those don't require a contentsTileRect. > > > > I didn't see functions on GraphicsLayer to specify the tiling behavior. > It's not necessary. RenderBoxObjectModel::getGeometryForBackgroundImage takes care of that. > For example if the background is tiled only horizontally, the contentsRect would be cropped to include only the tiled row.
I should probably add that to the Changelog :)
Build Bot
Comment 7
2013-01-29 12:02:35 PST
Comment on
attachment 185267
[details]
Patch
Attachment 185267
[details]
did not pass mac-wk2-ews (mac-wk2): Output:
http://queues.webkit.org/results/16218108
Build Bot
Comment 8
2013-01-29 13:04:59 PST
Comment on
attachment 185267
[details]
Patch
Attachment 185267
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/16195270
Noam Rosenthal
Comment 9
2013-01-30 02:35:27 PST
Created
attachment 185443
[details]
Patch
Noam Rosenthal
Comment 10
2013-01-30 05:48:04 PST
Created
attachment 185478
[details]
Patch
Build Bot
Comment 11
2013-01-30 06:44:32 PST
Comment on
attachment 185478
[details]
Patch
Attachment 185478
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/16195694
New failing tests: compositing/patterns/direct-pattern-compositing-rotation.html
Noam Rosenthal
Comment 12
2013-01-30 07:44:28 PST
Created
attachment 185505
[details]
Patch
Build Bot
Comment 13
2013-01-30 10:05:20 PST
Comment on
attachment 185505
[details]
Patch
Attachment 185505
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/16218609
New failing tests: compositing/patterns/direct-pattern-compositing-rotation.html
Noam Rosenthal
Comment 14
2013-01-30 10:11:33 PST
Created
attachment 185513
[details]
Patch
Noam Rosenthal
Comment 15
2013-01-30 13:44:00 PST
Removed the rotation test, it doesn't make sense as a ref test.
WebKit Review Bot
Comment 16
2013-01-31 02:17:06 PST
Comment on
attachment 185513
[details]
Patch
Attachment 185513
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16272096
New failing tests: platform/chromium/virtual/gpu/compositedscrolling/overflow/scroll-ancestor-update.html platform/chromium/virtual/gpu/fast/canvas/fillrect_gradient.html platform/chromium/virtual/softwarecompositing/overflow/scroll-ancestor-update.html platform/chromium/virtual/gpu/fast/canvas/patternfill-repeat.html platform/chromium/virtual/gpu/fast/canvas/canvas-composite.html platform/chromium/virtual/softwarecompositing/geometry/video-fixed-scrolling.html compositing/patterns/direct-pattern-compositing-padding.html compositing/overflow/overflow-compositing-descendant.html compositing/layers-inside-overflow-scroll.html platform/chromium/virtual/softwarecompositing/overflow/overflow-compositing-descendant.html fast/loader/text-document-wrapping.html platform/chromium/virtual/softwarecompositing/tile-cache-must-flatten.html compositing/self-painting-layers.html compositing/overflow/scroll-ancestor-update.html platform/chromium/virtual/softwarecompositing/self-painting-layers.html platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment.html compositing/visibility/visibility-simple-video-layer.html platform/chromium/virtual/gpu/fast/canvas/canvas-text-baseline.html compositing/geometry/video-opacity-overlay.html platform/chromium/virtual/softwarecompositing/repaint/clipped-layer-size-change.html compositing/geometry/video-fixed-scrolling.html platform/chromium/virtual/gpu/compositedscrolling/overflow/overflow-compositing-descendant.html platform/chromium/virtual/softwarecompositing/geometry/video-opacity-overlay.html platform/chromium/virtual/softwarecompositing/layers-inside-overflow-scroll.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-padding.html compositing/webgl/webgl-no-alpha.html compositing/webgl/webgl-reflection.html fast/loader/javascript-url-in-object.html
Noam Rosenthal
Comment 17
2013-02-01 02:22:47 PST
Created
attachment 185990
[details]
Patch
Build Bot
Comment 18
2013-02-01 07:36:14 PST
Comment on
attachment 185990
[details]
Patch
Attachment 185990
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/16293679
New failing tests: compositing/background-color/background-color-change-to-text.html css3/filters/huge-region-composited.html compositing/background-color/background-color-padding-change.html
Build Bot
Comment 19
2013-02-01 08:34:53 PST
Comment on
attachment 185990
[details]
Patch
Attachment 185990
[details]
did not pass mac-ews (mac): Output:
http://queues.webkit.org/results/16306732
New failing tests: compositing/background-color/background-color-change-to-text.html css3/filters/huge-region-composited.html compositing/background-color/background-color-padding-change.html
Build Bot
Comment 20
2013-02-01 09:22:54 PST
Comment on
attachment 185990
[details]
Patch
Attachment 185990
[details]
did not pass mac-wk2-ews (mac-wk2): Output:
http://queues.webkit.org/results/16308768
New failing tests: fast/css/sticky/sticky-as-positioning-container.html compositing/background-color/background-color-change-to-text.html css3/filters/huge-region-composited.html compositing/background-color/background-color-padding-change.html compositing/patterns/direct-pattern-compositing-add-text.html
WebKit Review Bot
Comment 21
2013-02-01 15:28:10 PST
Comment on
attachment 185990
[details]
Patch
Attachment 185990
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16331211
New failing tests: compositing/patterns/direct-pattern-compositing-padding.html css3/filters/huge-region-composited.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-padding.html
Noam Rosenthal
Comment 22
2013-02-02 03:13:38 PST
Created
attachment 186221
[details]
Patch
WebKit Review Bot
Comment 23
2013-02-02 04:24:03 PST
Comment on
attachment 186221
[details]
Patch
Attachment 186221
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16343677
New failing tests: compositing/patterns/direct-pattern-compositing-padding.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-padding.html
Build Bot
Comment 24
2013-02-02 05:20:15 PST
Comment on
attachment 186221
[details]
Patch
Attachment 186221
[details]
did not pass win-ews (win): Output:
http://queues.webkit.org/results/16337768
WebKit Review Bot
Comment 25
2013-02-02 05:30:23 PST
Comment on
attachment 186221
[details]
Patch
Attachment 186221
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16342710
New failing tests: compositing/patterns/direct-pattern-compositing-padding.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-padding.html
Noam Rosenthal
Comment 26
2013-02-02 07:09:42 PST
Created
attachment 186229
[details]
Patch
Build Bot
Comment 27
2013-02-02 09:13:27 PST
Comment on
attachment 186229
[details]
Patch
Attachment 186229
[details]
did not pass win-ews (win): Output:
http://queues.webkit.org/results/16335922
Build Bot
Comment 28
2013-02-02 10:15:35 PST
Comment on
attachment 186229
[details]
Patch
Attachment 186229
[details]
did not pass win-ews (win): Output:
http://queues.webkit.org/results/16346026
Noam Rosenthal
Comment 29
2013-02-04 09:54:37 PST
Created
attachment 186405
[details]
Patch
Build Bot
Comment 30
2013-02-04 12:29:36 PST
Comment on
attachment 186405
[details]
Patch
Attachment 186405
[details]
did not pass mac-wk2-ews (mac-wk2): Output:
http://queues.webkit.org/results/16374289
New failing tests: compositing/patterns/direct-pattern-compositing-change.html
Build Bot
Comment 31
2013-02-04 14:26:36 PST
Comment on
attachment 186405
[details]
Patch
Attachment 186405
[details]
did not pass win-ews (win): Output:
http://queues.webkit.org/results/16371408
Noam Rosenthal
Comment 32
2013-02-07 10:04:48 PST
Those tests work perfectly on my machine; Is there some flakiness with running ref-tests with CA?
Noam Rosenthal
Comment 33
2013-02-09 09:38:14 PST
Created
attachment 187434
[details]
Patch
WebKit Review Bot
Comment 34
2013-02-09 15:04:22 PST
Comment on
attachment 187434
[details]
Patch
Attachment 187434
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16467620
New failing tests: platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-add-text.html compositing/patterns/direct-pattern-compositing-padding.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-position.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-cover.html compositing/patterns/direct-pattern-compositing.html compositing/patterns/direct-pattern-compositing-cover.html compositing/patterns/direct-pattern-compositing-position.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing.html compositing/patterns/direct-pattern-compositing-contain.html compositing/patterns/direct-pattern-compositing-change.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-change.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-contain.html compositing/patterns/direct-pattern-compositing-size.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-size.html compositing/patterns/direct-pattern-compositing-add-text.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-padding.html
Noam Rosenthal
Comment 35
2013-02-09 15:31:03 PST
Created
attachment 187449
[details]
Patch
WebKit Review Bot
Comment 36
2013-02-10 00:40:21 PST
Comment on
attachment 187449
[details]
Patch
Attachment 187449
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16473699
New failing tests: compositing/patterns/direct-pattern-compositing-change.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-change.html
Noam Rosenthal
Comment 37
2013-02-10 02:44:51 PST
Created
attachment 187473
[details]
Patch
WebKit Review Bot
Comment 38
2013-02-10 06:53:33 PST
Comment on
attachment 187473
[details]
Patch
Attachment 187473
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://queues.webkit.org/results/16479703
New failing tests: platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-add-text.html compositing/patterns/direct-pattern-compositing-change.html platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-change.html compositing/patterns/direct-pattern-compositing-add-text.html
Noam Rosenthal
Comment 39
2013-02-11 04:14:08 PST
Created
attachment 187543
[details]
Patch
Noam Rosenthal
Comment 40
2013-02-12 08:45:14 PST
Created
attachment 187875
[details]
Patch
Noam Rosenthal
Comment 41
2013-02-21 12:37:26 PST
Ping?
Noam Rosenthal
Comment 42
2013-03-15 11:11:28 PDT
(In reply to
comment #41
)
> Ping?
Simon, any chance for review? This has been sitting here for a while.
Simon Fraser (smfr)
Comment 43
2013-03-15 11:41:27 PDT
Comment on
attachment 187875
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=187875&action=review
> Source/WebCore/platform/graphics/GraphicsLayer.h:328 > + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated.
This comment and name are a bit ambiguous. For repeated contents, does this represent just a single tile (in order to specify the offset and size), not the area covered by tiles? Might it be clearer to use phase and tileSize instead?
> Source/WebCore/rendering/RenderLayerBacking.cpp:1344 > + Color color = style->visitedDependentColor(CSSPropertyBackgroundColor); > + if (color.isValid() && color.alpha()) > + return false;
Why does an alpha color bail here? Can't you still set the layer's background color and also do tiling?
> Source/WebCore/rendering/RenderLayerBacking.cpp:1353 > + if (!styleImage->isLoaded()) > + return false;
Does the image loading cause us to re-evaluate compositing?
> Source/WebCore/rendering/RenderLayerBacking.cpp:1363 > + enum { MaxSizeForPatten = 1024 }; > + if (image->width() > MaxSizeForPatten || image->height() > MaxSizeForPatten) > + return false;
This looks pretty platform-specific.
> Source/WebCore/rendering/RenderLayerBacking.cpp:1377 > + if (!isSimpleContainer || !style->hasBackgroundImage()) { > + m_graphicsLayer->setContentsToImage(0); > + return;
It's not clear from the function name that you'll never get here for composited <img>, so setting contents to image unconditionally looks a bit weird.
> LayoutTests/compositing/patterns/direct-pattern-compositing-add-text-expected.html:2 > +<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" > + "
http://www.w3.org/TR/html4/loose.dtd
">
Yuck.
> LayoutTests/compositing/patterns/direct-pattern-compositing-add-text.html:2 > +<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" > + "
http://www.w3.org/TR/html4/loose.dtd
">
Ditto.
Noam Rosenthal
Comment 44
2013-03-15 12:02:49 PDT
Comment on
attachment 187875
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=187875&action=review
>> Source/WebCore/platform/graphics/GraphicsLayer.h:328 >> + // contentsTileRect is a single tile of contents in layer coordinates, if contents is to be repeated. > > This comment and name are a bit ambiguous. For repeated contents, does this represent just a single tile (in order to specify the offset and size), not the area covered by tiles? > > Might it be clearer to use phase and tileSize instead?
I was contemplating this. Yes, maybe setContentsTilePhase and setContentsTileSize would work better.
>> Source/WebCore/rendering/RenderLayerBacking.cpp:1344 >> + return false; > > Why does an alpha color bail here? Can't you still set the layer's background color and also do tiling?
It's currently not supported. We can support this in the future, but need to take care of background-compositing and other issues. Trying to err on the careful side for this patch. also, the more elements added to the layer, the less it benefits from direct compositing.
>> Source/WebCore/rendering/RenderLayerBacking.cpp:1353 >> + return false; > > Does the image loading cause us to re-evaluate compositing?
Actually, no. I'll investigate what's the right thing to do with contentChanged.
>> Source/WebCore/rendering/RenderLayerBacking.cpp:1363 >> + return false; > > This looks pretty platform-specific.
OK, I'll move to GraphicsLayer::shouldDirectlyCompositeTiledImage
>> Source/WebCore/rendering/RenderLayerBacking.cpp:1377 >> + return; > > It's not clear from the function name that you'll never get here for composited <img>, so setting contents to image unconditionally looks a bit weird.
Would a comment suffice for this? or maybe change the function name to updateDirectlyCompositedBackgroundImage?
>> LayoutTests/compositing/patterns/direct-pattern-compositing-add-text-expected.html:2 >> + "
http://www.w3.org/TR/html4/loose.dtd
"> > > Yuck.
Oops, copy paste from somewhere.
Simon Fraser (smfr)
Comment 45
2013-03-15 12:07:05 PDT
(In reply to
comment #44
)
> (From update of
attachment 187875
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=187875&action=review
> > Why does an alpha color bail here? Can't you still set the layer's background color and also do tiling? > > It's currently not supported. We can support this in the future, but need to take care of background-compositing and other issues. > Trying to err on the careful side for this patch. also, the more elements added to the layer, the less it benefits from direct compositing.
OK.
> > It's not clear from the function name that you'll never get here for composited <img>, so setting contents to image unconditionally looks a bit weird. > > Would a comment suffice for this? or maybe change the function name to updateDirectlyCompositedBackgroundImage?
That name is better.
Noam Rosenthal
Comment 46
2013-03-18 09:57:57 PDT
Created
attachment 193589
[details]
Patch
Build Bot
Comment 47
2013-03-18 15:01:16 PDT
Comment on
attachment 193589
[details]
Patch
Attachment 193589
[details]
did not pass mac-ews (mac): Output:
http://webkit-commit-queue.appspot.com/results/17187500
New failing tests: css3/filters/custom/effect-custom-transform-parameters.html compositing/background-color/background-color-alpha.html fast/repaint/region-painting-in-composited-view.html compositing/background-color/background-color-text-change.html compositing/background-color/background-color-composite.html compositing/background-color/background-color-text-clip.html compositing/background-color/background-color-content-clip.html compositing/geometry/fixed-position-composited-page-scale-smaller-than-viewport.html compositing/background-color/background-color-padding-clip.html fast/layers/no-clipping-overflow-hidden-added-after-transition.html compositing/background-color/background-color-container.html compositing/background-color/background-color-simple.html css3/filters/huge-region-composited.html compositing/overlap-blending/children-opacity-no-overlap.html
WebKit Review Bot
Comment 48
2013-03-18 16:56:50 PDT
Comment on
attachment 193589
[details]
Patch
Attachment 193589
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://webkit-commit-queue.appspot.com/results/17132280
New failing tests: css3/filters/effect-blur-hw.html compositing/webgl/webgl-nonpremultiplied-blend.html compositing/nested-direct-image-compositing.html compositing/direct-image-compositing.html http/tests/cache/subresource-failover-to-network.html fast/css/font-family-pictograph.html css3/filters/huge-region-composited.html compositing/visibility/visibility-simple-canvas2d-layer.html fast/canvas/webgl/webgl-layer-update.html compositing/webgl/webgl-repaint.html compositing/images/content-image.html compositing/images/content-image-change.html fast/canvas/webgl/webgl-composite-modes-repaint.html compositing/visibility/visibility-simple-webgl-layer.html compositing/patterns/direct-pattern-compositing.html compositing/culling/tile-occlusion-boundaries.html compositing/overflow/overflow-hidden-canvas-layer.html compositing/visibility/visibility-image-layers.html compositing/webgl/webgl-no-alpha.html css3/filters/effect-reference-hw.html compositing/webgl/webgl-reflection.html compositing/visibility/visibility-image-layers-dynamic.html compositing/backface-visibility/backface-visibility-image.html compositing/masks/direct-image-mask.html fast/canvas/webgl/webgl-composite-modes.html fast/loader/javascript-url-in-object.html compositing/color-matching/image-color-matching.html compositing/overflow/overflow-compositing-descendant.html compositing/backface-visibility/backface-visibility-webgl.html compositing/reflections/simple-composited-reflections.html
Build Bot
Comment 49
2013-03-19 03:31:47 PDT
Comment on
attachment 193589
[details]
Patch
Attachment 193589
[details]
did not pass mac-wk2-ews (mac-wk2): Output:
http://webkit-commit-queue.appspot.com/results/17121799
New failing tests: fast/css/sticky/sticky-margins.html fast/css/sticky/sticky-top-margins.html compositing/background-color/background-color-composite.html fast/css/sticky/sticky-left-percentage.html fast/css/sticky/sticky-stacking-context.html fast/css/sticky/sticky-writing-mode-vertical-rl.html fast/css/sticky/sticky-both-sides.html css3/filters/huge-region-composited.html fast/css/sticky/sticky-side-margins.html fast/css/sticky/sticky-left.html compositing/background-color/background-color-text-change.html compositing/background-color/background-color-simple.html css3/filters/custom/effect-custom-transform-parameters.html fast/css/sticky/sticky-top.html fast/layers/no-clipping-overflow-hidden-added-after-transition.html fast/css/sticky/inflow-sticky.html compositing/background-color/background-color-content-clip.html editing/pasteboard/smart-paste-001.html compositing/overlap-blending/children-opacity-no-overlap.html fast/forms/number/number-spinbutton-in-multi-column.html compositing/geometry/fixed-position-composited-page-scale-smaller-than-viewport.html compositing/background-color/background-color-padding-clip.html compositing/background-color/background-color-container.html fast/css/sticky/sticky-writing-mode-vertical-lr.html fast/css/sticky/sticky-writing-mode-horizontal-bt.html http/tests/security/cross-origin-appcache-allowed.html canvas/philip/tests/2d.text.draw.fontface.notinpage.html compositing/background-color/background-color-text-clip.html fast/css/sticky/sticky-overflowing.html compositing/background-color/background-color-alpha.html
Noam Rosenthal
Comment 50
2013-03-22 03:48:01 PDT
Created
attachment 194501
[details]
Patch
WebKit Review Bot
Comment 51
2013-03-22 05:16:02 PDT
Comment on
attachment 194501
[details]
Patch
Attachment 194501
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://webkit-commit-queue.appspot.com/results/17140618
New failing tests: platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-load.html compositing/patterns/direct-pattern-compositing-load.html
WebKit Review Bot
Comment 52
2013-03-22 05:16:07 PDT
Created
attachment 194517
[details]
Archive of layout-test-results from gce-cr-linux-05 The attached test failures were seen while running run-webkit-tests on the chromium-ews. Bot: gce-cr-linux-05 Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'> Platform: Linux-3.3.8-gcg-201212281604-x86_64-with-GCEL-10.04-gcel_10.04
WebKit Review Bot
Comment 53
2013-03-22 05:53:59 PDT
Comment on
attachment 194501
[details]
Patch
Attachment 194501
[details]
did not pass chromium-ews (chromium-xvfb): Output:
http://webkit-commit-queue.appspot.com/results/17247356
New failing tests: platform/chromium/virtual/softwarecompositing/patterns/direct-pattern-compositing-load.html compositing/patterns/direct-pattern-compositing-load.html
WebKit Review Bot
Comment 54
2013-03-22 05:54:05 PDT
Created
attachment 194522
[details]
Archive of layout-test-results from gce-cr-linux-06 The attached test failures were seen while running run-webkit-tests on the chromium-ews. Bot: gce-cr-linux-06 Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'> Platform: Linux-3.3.8-gcg-201212281604-x86_64-with-GCEL-10.04-gcel_10.04
Noam Rosenthal
Comment 55
2013-03-22 06:13:18 PDT
Created
attachment 194529
[details]
Patch
Noam Rosenthal
Comment 56
2013-03-25 14:59:02 PDT
This should now be cleaned up and ready for review. I've added a test for the case where the background image is loaded after the layer is already composited.
Simon Fraser (smfr)
Comment 57
2013-04-10 09:31:19 PDT
Comment on
attachment 194529
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=194529&action=review
> Source/WebCore/platform/graphics/GraphicsLayer.h:330 > + virtual void setContentsTilePhase(const IntPoint& p) { m_contentsTilePhase = p; } > + IntPoint contentsTilePhase() const { return m_contentsTilePhase; }
Would be nice to add a comment saying what this represents. What is it relative to; the GrpahicsLayer bounds? The content layer bounds? The offsetFromRenderer?
Noam Rosenthal
Comment 58
2013-04-10 10:38:53 PDT
Created
attachment 197309
[details]
Patch
Noam Rosenthal
Comment 59
2013-04-10 10:39:48 PDT
(In reply to
comment #57
)
> (From update of
attachment 194529
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=194529&action=review
> > > Source/WebCore/platform/graphics/GraphicsLayer.h:330 > > + virtual void setContentsTilePhase(const IntPoint& p) { m_contentsTilePhase = p; } > > + IntPoint contentsTilePhase() const { return m_contentsTilePhase; } > > Would be nice to add a comment saying what this represents. What is it relative to; the GrpahicsLayer bounds? The content layer bounds? The offsetFromRenderer?
Thanks for the review Simon, I've added a comment and will land after it gets another EWS run.
WebKit Commit Bot
Comment 60
2013-04-10 22:46:16 PDT
Comment on
attachment 197309
[details]
Patch Clearing flags on attachment: 197309 Committed
r148172
: <
http://trac.webkit.org/changeset/148172
>
WebKit Commit Bot
Comment 61
2013-04-10 22:46:23 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