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

Issue 686080 link

Starred by 1 user

Issue metadata

Status: Started
Owner:
Last visit > 30 days ago
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Bug

Blocking:
issue 685764
issue 685804
issue 685810



Sign in to add a comment

Bisect - Repro'ing test failures but not continuing

Project Member Reported by simonhatch@chromium.org, Jan 27 2017

Issue description

AFAIK it's actually reproducing the failures but we don't seem to be continuing bisecting after that.
 
Owner: mikec...@chromium.org
+mikecase

I think what's happening is that everything now passes through tee.py, which swallows the original return code. Bisect gets the return code of running tee.py, not of the benchmark, which is always successful and so we never end up moving ahead.
Status: Started (was: Untriaged)
Wow, didn't think of this at all. Will work on fix.
I think this simple fix should do the trick.

https://chromium-review.googlesource.com/c/433330/
Project Member

Comment 4 by bugdroid1@chromium.org, Jan 27 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/tools/build.git/+/2933ae7994ff7089034c13e106c1ba94807cad2d

commit 2933ae7994ff7089034c13e106c1ba94807cad2d
Author: Mike Case <mikecase@google.com>
Date: Fri Jan 27 20:30:44 2017

Fix tee.py eating return code for commands it wraps.

The fact that tee.py always return 0 exit code breaks bisect.

BUG=686080

Change-Id: I2b1904ffafcb2f761c420e211ac597fa6fbec5aa
Reviewed-on: https://chromium-review.googlesource.com/433330
Reviewed-by: John Budorick <jbudorick@chromium.org>
Reviewed-by: Michael Case <mikecase@chromium.org>
Commit-Queue: Michael Case <mikecase@chromium.org>

[modify] https://crrev.com/2933ae7994ff7089034c13e106c1ba94807cad2d/scripts/slave/tee.py

Components: Speed>Bisection

Sign in to add a comment