RESOLVED FIXED 65736
[chromium] Rebaseline following r92255 
https://bugs.webkit.org/show_bug.cgi?id=65736
Summary [chromium] Rebaseline following r92255
David Reveman
Reported 2011-08-04 18:15:36 PDT
Rebaseline tests affected by new layer anti-aliasing method.
Attachments
Patch (deleted)
2011-08-04 20:47 PDT, David Reveman 		no flags
DetailsFormatted DiffDiff
Patch (1.50 MB, patch)
2011-08-08 07:37 PDT, David Reveman 		jamesr: review-
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
David Reveman
Comment 1 2011-08-04 20:47:28 PDT
Created attachment 103025 [details] Patch
David Reveman
Comment 2 2011-08-04 22:19:03 PDT
Rebaselines most of the tests affected by the new anti-aliasing method. The patch is missing a few win/mac test that don't seem to have results. Not sure how to handle that.
James Robinson
Comment 3 2011-08-05 00:46:02 PDT
Comment on attachment 103025 [details] Patch Thanks, these look OK. The banding on platform/chromium/compositing/huge-layer-rotated.html and platform/chromium/compositing/tiny-layer-rotated.html is really weird, but it seems that it's not new.
WebKit Review Bot
Comment 4 2011-08-05 01:36:45 PDT
Comment on attachment 103025 [details] Patch Clearing flags on attachment: 103025 Committed r92466: <http://trac.webkit.org/changeset/92466>
WebKit Review Bot
Comment 5 2011-08-05 01:36:54 PDT
All reviewed patches have been landed. Closing bug.
David Reveman
Comment 6 2011-08-05 12:58:33 PDT
Reopen this bug since the following tests are still failing: BUGWK64942 WIN GPU : compositing/color-matching/image-color-matching.html = IMAGE BUGWK64942 WIN GPU : compositing/direct-image-compositing.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/ancestor-overflow-change.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/fixed-position.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/horizontal-scroll-composited.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/layer-due-to-layer-children-deep.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/layer-due-to-layer-children.html = IMAGE BUGWK64942 WIN GPU : compositing/geometry/vertical-scroll-composited.html = IMAGE BUGWK64942 WIN GPU : compositing/iframes/composited-iframe-alignment.html = IMAGE BUGWK64942 WIN GPU : compositing/masks/masked-ancestor.html = IMAGE BUGWK64942 WIN GPU : compositing/overflow/fixed-position-ancestor-clip.html = IMAGE BUGWK64942 WIN GPU : compositing/reflections/nested-reflection-transition.html = IMAGE BUGWK64942 WIN GPU : compositing/reflections/transform-inside-reflection.html = IMAGE BUGWK64942 WIN GPU : compositing/shadows/shadow-drawing.html = IMAGE BUGWK64942 WIN MAC GPU : compositing/transitions/scale-transition-no-start.html = IMAGE BUGWK64942 MAC GPU : media/video-transformed.html = IMAGE BUGWK64942 MAC GPU : media/video-zoom-controls.html = IMAGE BUGWK64942 MAC GPU : platform/chromium/compositing/backface-visibility-transformed.html = IMAGE BUGWK64942 MAC GPU : platform/chromium/compositing/huge-layer-rotated.html = IMAGE BUGWK64942 MAC GPU : platform/chromium/compositing/layout-width-change.html = IMAGE BUGWK64942 MAC GPU : platform/chromium/compositing/perpendicular-layer-sorting.html = IMAGE BUGWK64942 MAC GPU : platform/chromium/compositing/tiny-layer-rotated.html = IMAGE
David Reveman
Comment 7 2011-08-08 07:37:40 PDT
Created attachment 103244 [details] Patch
David Reveman
Comment 8 2011-08-08 08:09:45 PDT
Some of the expected images were not included in the previous patch. I suspect that it is related to the size of the patch. webkit-patch upload seem to be having issues with large patches. This new patch rebaselines the tests that were somehow left out in previous patch.
James Robinson
Comment 9 2011-08-10 12:57:27 PDT
Comment on attachment 103244 [details] Patch rebaselines look OK, but this patch doesn't apply to ToT. It looks like there's some garbage instead of LayoutTests/platform/chromium-gpu-win/compositing/transitions/scale-transition-no-start-expected.png. I don't think our bugzilla scripts really work for patches this big :/
James Robinson
Comment 10 2011-08-10 13:40:22 PDT
Landed rebaselines http://trac.webkit.org/changeset/92793. Sadly rebaselining is something that our tooling doesn't really handle for non-committers. Oh well :(
Note You need to log in before you can comment on or make changes to this bug.