New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 648464 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner: ----
Closed: Sep 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 1
Type: Bug



Sign in to add a comment

Build failure - check on ColorSpace enum boundary

Project Member Reported by dim...@chromium.org, Sep 20 2016

Issue description

Caused by https://codereview.chromium.org/2338213009/
which was reverted here: https://codereview.chromium.org/2352903002

Link to build or pfq page.

https://build.chromium.org/p/chromium.gpu/builders/Mac%2010.10%20Retina%20Release%20%28AMD%29/builds/17323


Snippet of log that contains the failure.

[4590:14851:0919/144840:FATAL:gfx_param_traits_macros.h(19)] Check failed: (static_cast<int>(value) >= static_cast<int>(0) && static_cast<int>(value) <= static_cast<int>(gfx::ColorSpace::PrimaryID::LAST)). 
0   libbase.dylib                       0x000000011e3cc19e _ZN4base5debug10StackTraceC2Ev + 30
1   libbase.dylib                       0x000000011e3cc205 _ZN4base5debug10StackTraceC1Ev + 21
2   libbase.dylib                       0x000000011e4636e0 _ZN7logging10LogMessageD2Ev + 80
3   libbase.dylib                       0x000000011e461295 _ZN7logging10LogMessageD1Ev + 21
4   libgfx_ipc_color.dylib              0x00000001382053ca _ZN3IPC11ParamTraitsIN3gfx10ColorSpace9PrimaryIDEE5WriteEPN4base6PickleERKS3_ + 250
5   libgfx_ipc_color.dylib              0x0000000138204dcd _ZN3IPCL10WriteParamIN3gfx10ColorSpace9PrimaryIDEEEvPN4base6PickleERKT_ + 29
6   libgfx_ipc_color.dylib              0x0000000138204d0d _ZN3IPC11ParamTraitsIN3gfx10ColorSpaceEE5WriteEPN4base6PickleERKS2_ + 29
 

Comment 1 by dim...@chromium.org, Sep 20 2016

Revert landed at 15:52 PST, many bots affected but should clear in couple hours.

Comment 2 by jam@chromium.org, Sep 20 2016

Do we understand how this made it through the CQ?

Comment 3 by jam@chromium.org, Sep 20 2016

Cc: ccameron@chromium.org thakis@chromium.org hubbe@chromium.org sande...@chromium.org

Comment 4 by kbr@chromium.org, Sep 20 2016

Cc: kbr@chromium.org
Components: Infra>CQ
Labels: Build-CommitQueue
I don't think it's understood how this made it through the CQ.

Note that Patch Set 3 was the one that was CQ'd. The mac_chromium_rel_ng tryjob for that patch went through successfully.

Comment 5 by dim...@chromium.org, Sep 20 2016

Interesting indeed. Perhaps some other CL went through around same time that actually caused some specific ColorSpace enum value to be used on Mac?
I landed https://codereview.chromium.org/2336853002/ around the same time, but all of that should be behind a flag.
Indeed, mac_chromium_rel_ng was ran on PS#3 of https://codereview.chromium.org/2338213009/ at ~12 PM and landed 14:18 MTV time.

https://codereview.chromium.org/2336853002 landed ~14:02 MTV time.
Status: Fixed (was: Untriaged)

Sign in to add a comment