Use Git hash to post try job http request to bisect builders.

Use Git hash to post try job http request to bisect builders.

Always use Git hash to post build request since Commit positions are
not supported by builders to build.

BUG=407898
NOTRY=true

Committed: https://crrev.com/a07262e0a10ce16bfc41765e5269e531a098a94e
Cr-Commit-Position: refs/heads/master@{#292188}

[prasadv, 2014-08-27 00:05:57]

prasadv@chromium.org changed reviewers:
+ qyearsley@chromium.org

[prasadv, 2014-08-27 00:41:17]

prasadv@chromium.org changed reviewers:
+ stip@chromium.org

[qyearsley, 2014-08-27 16:41:55]

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression.py
File tools/bisect-perf-regression.py (right):

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1319: revision: A Git hash revision.
I think it would be less confusing if this were called git_hash or at least
git_revision.

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1331: 'Failed to determine SVN revision for %s'
% revision)
We don't need to do this now, right?

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1358: '%s-%s-%s' % (svn_revision, patch,
time.time()))
Perhaps the build ID request should use the git hash instead of svn revision
now?

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1377: print '%s [revision: %s]' % (error_msg,
svn_revision)
Git hash can be printed here.

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1380: print 'Failed to post build request for
revision: [%s]' % svn_revision
And here.

[prasadv, 2014-08-27 17:15:40]

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression.py
File tools/bisect-perf-regression.py (right):

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1319: revision: A Git hash revision.
On 2014/08/27 16:41:55, qyearsley wrote:
> I think it would be less confusing if this were called git_hash or at least
> git_revision.

Done.

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1331: 'Failed to determine SVN revision for %s'
% revision)
On 2014/08/27 16:41:55, qyearsley wrote:
> We don't need to do this now, right?
I felt SVN or Commit position are more readable rather than git hash,therefore I
kept this as is so that we can use this value in error msgs.
Let me know if this should still be removed.

[qyearsley, 2014-08-27 17:58:24]

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression.py
File tools/bisect-perf-regression.py (right):

https://codereview.chromium.org/508723004/diff/1/tools/bisect-perf-regression...
tools/bisect-perf-regression.py:1331: 'Failed to determine SVN revision for %s'
% revision)
On 2014/08/27 17:15:39, prasadv wrote:
> On 2014/08/27 16:41:55, qyearsley wrote:
> > We don't need to do this now, right?
> I felt SVN or Commit position are more readable rather than git hash,therefore
I
> kept this as is so that we can use this value in error msgs.
> Let me know if this should still be removed.

Now, the function SVNFindRev might also find git commit position as well, right?
Saying that we're finding SVN revision (and naming that function SVNFindRev)
might be a bit misleading.

I think ideally, we should either:
  (1) refer to the revision in error messages by showing the first 8-10
characters of the git hash -- this is shorter and can also be used to find a
particular commit on all of our tools, e.g. http://crrev.com/459825d2 or "git
log -1 459825d2".
  (2) If you want to use an integer, then the function to fetch it should be
called FindSVNRevisionOrGitNumber.

I'm in favor of option 1 because it's simpler.

[qyearsley, 2014-08-27 18:19:42]

lgtm

[prasadv, 2014-08-27 18:20:08]

The CQ bit was checked by prasadv@chromium.org

[commit-bot: I haz the power, 2014-08-27 18:21:19]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/prasadv@chromium.org/508723004/40001

[commit-bot: I haz the power, 2014-08-27 18:22:14]

Committed patchset #3 (id:40001) as 8cd734317b746142beeb36b400d752ca0519349b

[commit-bot: I haz the power, 2014-09-10 02:52:52]

Patchset 3 (id:??) landed as
https://crrev.com/a07262e0a10ce16bfc41765e5269e531a098a94e
Cr-Commit-Position: refs/heads/master@{#292188}