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

Issue 662266 link

Starred by 2 users
Status: Duplicate
Owner:
█ alexclarke@chromium.org
Last visit > 30 days ago
Closed: Nov 2016
Cc:
█ bmcquade@chromium.org
█ kbr@chromium.org
wjmaclean@chromium.org
█ ymalik@chromium.org
nedngu...@google.com
█ glebl@chromium.org
jbudorick@chromium.org
jmad...@chromium.org
altimin@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Android
Pri: 1
Type: Bug-Regression



Sign in to add a comment

DCHECK in WebTaskRunnerImpl::postDelayedTask on Nexus 5 GPU.FYI bot

Project Member Reported by ynovikov@chromium.org, Nov 4 2016 
Many test fail in https://build.chromium.org/p/chromium.gpu.fyi/builders/Android%20Release%20%28Nexus%205%29/builds/2570
with:
[FATAL:web_task_runner_impl.cc(42)] Check failed: delayMs >= 0.0 (-5.94105e+238 vs. 0)

Assigning to alexclarke@, who wrote the DCHECK, since I can't figure out what CL in the blamelist could have caused this.

Comment 2 by alexclarke@chromium.org, Nov 4 2016 

If we could symbolize the stack we'd find out what is positing it.  I'm also going to land a patch which males the DCHECK print out the name of the task posting location.

Comment 3 by wjmaclean@chromium.org, Nov 4 2016 

The first CL in your list should only affect whether or not WebViewBrowserTests  run on *Mac* or not, and since (BrowserTag) WebView doesn't exist on Mac, I suspect it's unrelated.

Comment 4 by kbr@chromium.org, Nov 4 2016 

OK, thanks wjmaclean@ for analyzing. If your CLs definitely aren't at fault please feel free to remove yourself from the CC: list.

Comment 5 by wjmaclean@chromium.org, Nov 4 2016 

The second CL, also one of mine, also seems unlikely, but I'll stay on cc until we get a stack trace just in case it is somehow related.

Comment 6 by bmcquade@chromium.org, Nov 4 2016 

Likewise. My CL (49550fef9418c915f8d73ac8ea768d729c0b31c8) just removes code, and should not affect this codepath, but let's see what the stack trace shows.

Comment 7 by kbr@chromium.org, Nov 4 2016

Cc: nedngu...@google.com jbudorick@chromium.org
Components: Tests>Telemetry Infra>Client>Android
I don't know how the tombstone processing works; in the past I've seen symbolized stack traces on these bots. jbudorick@ do you know what code's responsible for that? nednguyen@?

Comment 8 by jbudorick@chromium.org, Nov 4 2016

Mergedinto: 662147
Status: Duplicate (was: Assigned)
Culprit was reverted last night.
►

Sign in to add a comment