chrome-si/initializers regression on Mac and Linux |
|||
Issue descriptionBuild is broken: sizes Relevant build logs here: Linux: https://build.chromium.org/p/chromium/builders/Linux%20x64/builds/21849 Mac: https://build.chromium.org/p/chromium/builders/Mac/builds/17350 Relevant lines from the error log: Linux: FAILED linux-release-64/sizes/nacl_helper-si/initializers: actual 8, expected 7, better lower FAILED linux-release-64/sizes/chrome-si/initializers: actual 8, expected 7, better lower Mac: FAILED mac-release/sizes/chrome-si/initializers: actual 2, expected 0, better lower It seems skia roll at r403708 introduced a new static initializer in src/gpu/instanced/InstanceProcessor.cpp.
,
Jul 5 2016
Oops, reverted to wrong revision :(
,
Jul 5 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/99241c5cb75649c2642c9176183e6c016a731c15 commit 99241c5cb75649c2642c9176183e6c016a731c15 Author: ksakamoto <ksakamoto@chromium.org> Date: Tue Jul 05 03:11:07 2016 Revert of Roll src/third_party/skia/ df4f47b8f..d87a6b235 (39 commits). (patchset #1 id:1 of https://codereview.chromium.org/2121043003/ ) Reason for revert: Broke Mac and Linux builds. crbug.com/625728 Original issue's description: > Roll src/third_party/skia/ df4f47b8f..d87a6b235 (39 commits). > > https://chromium.googlesource.com/skia.git/+log/df4f47b8ff63..d87a6b235672 > > $ git log df4f47b8f..d87a6b235 --date=short --no-merges --format='%ad %ae %s' > 2016-07-04 benjaminwagner Revert of Increase batching for AA fill rects. (patchset #4 id:60001 of https://codereview.chromium.org/2116823002/ ) > 2016-07-02 benjaminwagner Exclude SkSL in public.bzl. > 2016-07-01 ethannicholas fixed SkSL memory leak GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2119903002 NOTREECHECKS=true > 2016-07-01 ethannicholas fixed non-GPU build GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2120543003 NOTREECHECKS=true > 2016-07-01 skia.buildbots Update Win toolchain > 2016-07-01 benjaminwagner Specify new Windows Toolchain isolated hash. > 2016-07-01 bsalomon Increase batching for AA fill rects. > 2016-07-01 robertphillips Split apart the GP construction for the NonAAFillRect Batches > 2016-07-01 halcanary using_skia_and_harfbuzz: use default typeface > 2016-07-01 ethannicholas initial checkin of SkSL compiler > 2016-07-01 egdaniel Use bounds of batches for render pass bounds > 2016-07-01 egdaniel Fix vulkan ImageView swizzling to match spec > 2016-07-01 bsalomon Cleanup non-AA fill rect batch construction. GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2119853002 > 2016-07-01 bsalomon Simplify AA fill rect batch construction and rename Geometry->RectInfo GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2116723003 > 2016-07-01 bsalomon Remove unneeded helpers in AA and non-AA rect GrBatch subclasses. > 2016-06-30 bsalomon Cleanup PLSPathBatch GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2116543002 > 2016-06-30 bsalomon Disable GPU dithering in Android framework builds GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2115683002 > 2016-06-30 bsalomon Remove unused batch rect append functions GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2106283003 > 2016-06-30 bsalomon Remove GrTInstanceBatch > 2016-06-30 vjiaoblack Fix bug where ovals' AA exceed bounds by .5 pixel > 2016-06-30 csmartdalton Begin instanced rendering for simple shapes > 2016-06-30 bsalomon Hide GrNonAANinePatchBatch::Geometry and rename to Patch GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111943002 > 2016-06-30 bsalomon Hide MSAAPathBatch::Geometry and rename to PathInfo GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2116533002 > 2016-06-30 halcanary SkPDF: Glyph Useage Map improvements > 2016-06-30 egdaniel Remove support for GL discards > 2016-06-30 msarett Fix png encoding in skia debugger > 2016-06-30 hstern Add cubic example and draw original path > 2016-06-30 jcgregorio skiaserve: Use snakeCase for JSON identifiers. > 2016-06-30 jvanverth Add static buffer support > 2016-06-30 egdaniel Add discard support to Vulkan > 2016-06-30 halcanary SkPDF: allow overriding Producer metadata > 2016-06-30 reed use mapRectScaleTranslate for common case in clipRect > 2016-06-30 bsalomon Hide GrDrawVerticesBatch::Geometry and rename to Mesh GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2104383003 > 2016-06-30 jcgregorio debugger: Allow for larger images. > 2016-06-30 reed inline very common intersect+rect_rgn > 2016-06-30 hcm Update skia milestone to next > 2016-06-30 bsalomon Even more hiding of Geometry structs in GrBatch subclasses. GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2110853004 > 2016-06-30 reed speed up maprect for scale+trans case > 2016-06-30 borenet Add Android SDK asset > > CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel > TBR=rmistry@google.com > > Committed: https://crrev.com/1db191f4165df78f467c37b09ee69d666fb63a8b > Cr-Commit-Position: refs/heads/master@{#403708} BUG= 625728 TBR=rmistry@google.com NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true Review-Url: https://codereview.chromium.org/2127433002 Cr-Commit-Position: refs/heads/master@{#403734} [modify] https://crrev.com/99241c5cb75649c2642c9176183e6c016a731c15/DEPS
,
Jul 5 2016
Revert of the upstream CL created (waiting for an approval): https://codereview.chromium.org/2123693002/ Assigning to csmartdalton, the author of original CL.
,
Jul 5 2016
Hmm, is there any way to prevent skia-deps-roller from rolling the bad change repeatedly?
,
Jul 5 2016
I stopped the autoroller and closed the Skia tree.
,
Jul 5 2016
Revert is in. The builds are now green: * https://build.chromium.org/p/chromium/builders/Linux%20x64 * https://build.chromium.org/p/chromium/builders/Mac Marking this as fixed. We will have to make sure that when the checks are updated before the CL relands.
,
Jul 6 2016
Re: Comment 5, I filed skia:5494. |
|||
►
Sign in to add a comment |
|||
Comment 1 by bugdroid1@chromium.org
, Jul 5 2016