New issue
Advanced search Search tips

Issue 767106 link

Starred by 2 users

Issue metadata

Status: Untriaged
Owner: ----
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Feature

Blocked on:
issue 774179



Sign in to add a comment

rebaseline-cl should not re-trigger try jobs for commit message updates

Project Member Reported by fmalita@chromium.org, Sep 20 2017

Issue description

This has bitten me a few times, and couldn't figure out what was going on:

1) rebaseline-cl to trigger trybots
2) wait for all bots to finish
3) rebaseline-cl again to pull the results => No finished try jobs. Triggering try jobs: .....  =>  arrgh :)

It just occurred again for https://chromium-review.googlesource.com/c/chromium/src/+/675184, but this time I think I see what is going on: between #1 and #3 I updated the commit message, which shows as a new patch set in Gerrit.

Presumably the patch set number increment confuses rebaseline-cl into triggering a new try batch?

Since Gerrit is smart enough to recognize commit message updates as special events (doesn't discard the last trybot results for example), it would be awesome if rebaseline-cl could do the same!
 
Summary: rebaseline-cl should not re-trigger try jobs for commit message updates (was: rebasline-cl should not re-trigger try jobs for commit message updates)
Ah, good finding.

rebaseline-cl uses git-cl to get current try jobs and try job status.

This seems like something that perhaps should be changed in depot_tools/git_cl.py; I'll file a bug after investigating how git-cl gets try job status.
Blockedon: 774179
Labels: -OS-Linux -OS-Windows -OS-Mac
Owner: ----
Status: Available (was: Assigned)
So, git cl try-results lists results for the latest patch set (called a "revision" in Gerrit), whereas the Gerrit plugin for showing the results has logic for showing the latest non-trivial patch set.

Filed bug 774179 to suggest change to git cl try-results.
Cc: robertma@chromium.org qyears...@chromium.org
 Issue 796669  has been merged into this issue.

Comment 5 Deleted

It'll be also good if I can specify a patch set number of URL. This is helpful not only to workaround this bug, but also to unblock other tasks from rebaseline-cl.

For example, for the following timeline:
1. I execute rebaseline-cl at for patch-set a;
2. The reviewer submits some comments;
3. I upload a new patch set (which doesn't change layout test results) to address the comments;
4. The reviewer reviews the new patch set and LGTM.

For now, I must wait for the finish of 1 before doing 3. If we can specify patch set to rebaseline, we can parallelize 1 and 3/4 which will save much time in the ciritical path.
That's a good suggestion as well.

Since git cl try-results already supports specifying a patchset, I believe that in order to support specifying a patchset we would need to pass the --patchset argument through to git cl try-results.

(Unfortunately passing it through requires passing it through several intermediary functions, so the change isn't as simple as it could perhaps have been. CL for this idea: https://chromium-review.googlesource.com/c/chromium/src/+/858228)


Project Member

Comment 8 by bugdroid1@chromium.org, Jan 10 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6518a943f3297239e617d9816aae1e3e5d538854

commit 6518a943f3297239e617d9816aae1e3e5d538854
Author: Quinten Yearsley <qyearsley@chromium.org>
Date: Wed Jan 10 17:37:39 2018

Add a --patchset argument to webkit-patch rebaseline-cl

Purpose: Sometimes users of rebaseline-cl may want to fetch
baselines generated in try jobs for a patchset other than the
latest patchset.

This would add an argument --patchset to webkit-patch
rebaseline-cl.

Bug: 767106
Change-Id: I53d8ce29554b4dc70981d73dd09b8799edce0925
Reviewed-on: https://chromium-review.googlesource.com/858228
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Quinten Yearsley <qyearsley@chromium.org>
Cr-Commit-Position: refs/heads/master@{#528347}
[modify] https://crrev.com/6518a943f3297239e617d9816aae1e3e5d538854/third_party/WebKit/Tools/Scripts/webkitpy/common/net/git_cl.py
[modify] https://crrev.com/6518a943f3297239e617d9816aae1e3e5d538854/third_party/WebKit/Tools/Scripts/webkitpy/common/net/git_cl_mock.py
[modify] https://crrev.com/6518a943f3297239e617d9816aae1e3e5d538854/third_party/WebKit/Tools/Scripts/webkitpy/common/net/git_cl_unittest.py
[modify] https://crrev.com/6518a943f3297239e617d9816aae1e3e5d538854/third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_cl.py
[modify] https://crrev.com/6518a943f3297239e617d9816aae1e3e5d538854/third_party/WebKit/Tools/Scripts/webkitpy/tool/commands/rebaseline_cl_unittest.py

I just confirmed the above workaround -- this doesn't exactly solve the original problem, which I still think will be solved if bug 774179 is done.
Project Member

Comment 10 by sheriffbot@chromium.org, Jan 11

Labels: Hotlist-Recharge-Cold
Status: Untriaged (was: Available)
This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Sign in to add a comment