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

Issue 596142 link

Starred by 1 user
Status: Fixed
Owner:
sande...@chromium.org
Closed: Apr 2016
Cc:
█ petrcermak@chromium.org
crouleau@chromium.org
mac-bugs-priority@chromium.org
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 1
Type: Bug-Regression



Sign in to add a comment

media.tough_video_cases_extra.reference flaky on Mac 10.10 Perf (4)

Project Member Reported by petrcermak@chromium.org, Mar 18 2016 
Revision range first seen: https://build.chromium.org/p/chromium.perf/builders/Mac%2010.10%20Perf%20%284%29/builds/3120 (probably as far as the bot's history goes)
Link to failing step log: https://build.chromium.org/p/chromium.perf/builders/Mac%2010.10%20Perf%20%284%29/builds/3286/steps/media.tough_video_cases_extra.reference/logs/stdio

The failures are due to native crashes:

Standard output:
********************************************************************************
	2016-03-18 11:10:14.225 Google Chrome[8400:19459] NSWindow warning: adding an unknown subview: <FullSizeContentView: 0x7f97d0782730>
	2016-03-18 11:10:14.226 Google Chrome[8400:19459] Call stack:
	(
		0   AppKit                              0x00007fff8e45b54d -[NSThemeFrame addSubview:] + 107
		1   AppKit                              0x00007fff8e45b03f -[NSView addSubview:positioned:relativeTo:] + 208
		2   Google Chrome Framework             0x0000000106031723 ChromeMain + 51699443
		3   Google Chrome Framework             0x000000010603052c ChromeMain + 51694844
		4   Google Chrome Framework             0x000000010609a3c1 ChromeMain + 52128657
		5   Google Chrome Framework             0x0000000105feeecc ChromeMain + 51426972
		6   Google Chrome Framework             0x0000000105ff95c2 ChromeMain + 51469714
		7   Google Chrome Framework             0x00000001060feebc ChromeMain + 52541068
		8   Google Chrome Framework             0x000000010613796f ChromeMain + 52773183
		9   Google Chrome Framework             0x0000000106136d98 ChromeMain + 52770152
		10  Google Chrome Framework             0x0000000106136870 ChromeMain + 52768832
		11  Google Chrome Framework             0x0000000106135cf0 ChromeMain + 52765888
		12  Google Chrome Framework             0x00000001061358ff ChromeMain + 52764879
		13  Google Chrome Framework             0x00000001061337d2 ChromeMain + 52756386
		14  Google Chrome Framework             0x00000001061335a7 ChromeMain + 52755831
		15  Google Chrome Framework             0x00000001061329c1 ChromeMain + 52752785
		16  Google Chrome Framework             0x0000000102f811d1 ChromeMain + 645537
		17  Google Chrome Framework             0x0000000102f800ca ChromeMain + 641178
		18  Google Chrome Framework             0x000000010638b1ea ChromeMain + 55212474
		19  Google Chrome Framework             0x0000000106635d27 ChromeMain + 58008823
		20  Google Chrome Framework             0x0000000106389801 ChromeMain + 55205841
		21  Google Chrome Framework             0x000000010638d9a0 ChromeMain + 55222640
		22  Google Chrome Framework             0x0000000106386df5 ChromeMain + 55195077
		23  Google Chrome Framework             0x000000010341d99c ChromeMain + 5480812
		24  Google Chrome Framework             0x000000010341cf06 ChromeMain + 5478102
		25  Google Chrome Framework             0x0000000102ee3872 ChromeMain + 66
		26  Google Chrome                       0x0000000102e7cd42 main + 530
		27  Google Chrome                       0x0000000102e7cb24 Google Chrome + 2852
		28  ???                                 0x0000000000000010 0x0 + 16
	)
********************************************************************************

There were 14 flakes in the past 200 runs. Since it's less than one in five runs, I'm not going to disable the test (https://chromium.googlesource.com/chromium/src/+/master/tools/perf/docs/perf_bot_sheriffing.md#testfailures). I'm assigning it to the test owner to decide on further action. If the test is disabled, please downgrade to Pri-2.

Comment 1 by pras...@chromium.org, Mar 22 2016 

In last 20 builds this test failed 2 times, sandersd@ can you please take a look at this

Comment 2 by sande...@chromium.org, Mar 22 2016 

First of all, this trace does not appear to be media-related.

Looking through crbug, I find many very similar traces (https://bugs.chromium.org/p/chromium/issues/list?can=1&q=NSThemeFrame+addSubview), commonly resolved by disabling tests. I don't want to continue that trend.

https://bugs.chromium.org/p/chromium/issues/detail?id=511565#c10 has a similar stack trace, but with symbols. It could be a good start.

Comment 3 by sande...@chromium.org, Mar 22 2016

Cc: crouleau@chromium.org
According to  issue 520373 , the warning and stack trace are informational and not the cause of any crash.

Adding Caleb for help investigating the test itself, as there is no information in the log about what went wrong.

Comment 4 by rnep...@chromium.org, Apr 12 2016

Status: Fixed (was: Assigned)
It appears the flakiness might have cleared up. Not counting a block where everything was failing, the last 38 runs have not shown flake. Before this it was failing ~ every 10 runs.

Closing as fixed unless someone else feels it should be reopened.
►

Sign in to add a comment