Telemetry: Build symfs for Android VTune profiler

Telemetry: Build symfs for Android VTune profiler

Extend and use android_profiling_helper.py to set up a symfs for VTune.
Queries the required libraries from the profiler's output and pulls them from
the device. This allows VTune to resolve all symbols in the profiling output.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=276740

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=277839

[Dominik Grewe, 2014-05-30 12:49:46]

ptal

[Dominik Grewe, 2014-05-30 12:53:44]

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py (right):

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:52: #
TODO(dominikg): Ask simonhatch@ why we don't fail when exit_code = 1.
Simon, in crrev.com/23536025 you said we don't raise an exception when the
renderer got swapped out. What was the motivation behind that? We don't get any
results when that happens, do we?

[Sami, 2014-05-30 13:01:03]

Great, code reuse ftw :)

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
File
tools/telemetry/telemetry/core/platform/profiler/android_profiling_helper.py
(right):

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/android_profiling_helper.py:107:
try:
I was thinking we should use conn as context manager, but apparently that only
does transactions instead of managing the entire connection lifetime. Weird.

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py (right):

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:17: def
__init__(self, pid, output_file, browser_backend, platform_backend):
Could you pass in the device instead of entire browser backend since that's the
only thing we need?

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:51: if
exit_code:
Should we assert this doesn't happen on Android?

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:71: print 'To
view the profile, run:'
We still need to print this bit out when !self._is_android, right?

[Dominik Grewe, 2014-05-30 15:01:43]

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py (right):

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:17: def
__init__(self, pid, output_file, browser_backend, platform_backend):
On 2014/05/30 13:01:03, Sami wrote:
> Could you pass in the device instead of entire browser backend since that's
the
> only thing we need?

We only care about the device when we're on Android. So we'd have to check which
platform we're on before starting this profiler, in addition to checking here. I
don't have a strong opinion, but to me it seems cleaner to keep it the way it
is. Wdyt?

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:51: if
exit_code:
On 2014/05/30 13:01:03, Sami wrote:
> Should we assert this doesn't happen on Android?

I don't understand why we treat this differently in the first place.

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:71: print 'To
view the profile, run:'
On 2014/05/30 13:01:03, Sami wrote:
> We still need to print this bit out when !self._is_android, right?

Done.

[Sami, 2014-05-30 15:31:56]

Thanks, lgtm. Let's see what Simon says. Btw there's a similar swapped out check
in the perf profiler.

On 2014/05/30 15:01:43, Dominik Grewe wrote:
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py
(right):
> 
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:17: def
> __init__(self, pid, output_file, browser_backend, platform_backend):
> On 2014/05/30 13:01:03, Sami wrote:
> > Could you pass in the device instead of entire browser backend since that's
> the
> > only thing we need?
> 
> We only care about the device when we're on Android. So we'd have to check
which
> platform we're on before starting this profiler, in addition to checking here.
I
> don't have a strong opinion, but to me it seems cleaner to keep it the way it
> is. Wdyt?

Oh right, that's a good point. Let's keep this as is.
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:51: if
> exit_code:
> On 2014/05/30 13:01:03, Sami wrote:
> > Should we assert this doesn't happen on Android?
> 
> I don't understand why we treat this differently in the first place.
> 
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:71: print
'To
> view the profile, run:'
> On 2014/05/30 13:01:03, Sami wrote:
> > We still need to print this bit out when !self._is_android, right?
> 
> Done.

[Dominik Grewe, 2014-05-30 17:05:16]

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py (right):

https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:52: #
TODO(dominikg): Ask simonhatch@ why we don't fail when exit_code = 1.
On 2014/05/30 12:53:44, Dominik Grewe wrote:
> Simon, in crrev.com/23536025 you said we don't raise an exception when the
> renderer got swapped out. What was the motivation behind that? We don't get
any
> results when that happens, do we?

I didn't realize that we already check if the renderer has been swapped out at
the beginning of the function. So it makes sense to not print another warning
here.

[Sami, 2014-05-30 17:29:38]

That makes sense, lgtm.

[shatch, 2014-06-03 00:09:40]

On 2014/05/30 17:05:16, Dominik Grewe wrote:
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> File tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py
(right):
> 
>
https://codereview.chromium.org/308123002/diff/20001/tools/telemetry/telemetr...
> tools/telemetry/telemetry/core/platform/profiler/vtune_profiler.py:52: #
> TODO(dominikg): Ask simonhatch@ why we don't fail when exit_code = 1.
> On 2014/05/30 12:53:44, Dominik Grewe wrote:
> > Simon, in crrev.com/23536025 you said we don't raise an exception when the
> > renderer got swapped out. What was the motivation behind that? We don't get
> any
> > results when that happens, do we?
> 
> I didn't realize that we already check if the renderer has been swapped out at
> the beginning of the function. So it makes sense to not print another warning
> here.

+tonyg

Hi sorry, missed my name in here. I forget the exact reasons, had something to
do with the first renderer getting swapped out.

[Dominik Grewe, 2014-06-12 13:20:52]

The CQ bit was checked by dominikg@chromium.org

[commit-bot: I haz the power, 2014-06-12 13:22:15]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dominikg@chromium.org/308123002/80001

[commit-bot: I haz the power, 2014-06-12 15:15:01]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/win_gpu_triggered...)

[commit-bot: I haz the power, 2014-06-12 15:17:45]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-06-12 15:17:46]

Try jobs failed on following builders:
  win_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/win_gpu_triggered...)

[Dominik Grewe, 2014-06-12 17:23:44]

The CQ bit was checked by dominikg@chromium.org

[commit-bot: I haz the power, 2014-06-12 17:26:07]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dominikg@chromium.org/308123002/80001

[commit-bot: I haz the power, 2014-06-12 18:09:41]

Change committed as 276740

[xiyuan, 2014-06-12 22:25:09]

This breaks ChromeOS build at
http://build.chromium.org/p/chromiumos.chromium/builders/X86%20%28chromium%29...

16:09:25 INFO | autoserv|
FAIL	login_Cryptohome	login_Cryptohome	timestamp=1402607364	localtime=Jun 12
14:09:24	Unhandled ImportError: No module named sqlite3
16:09:25 INFO | autoserv| AUTOTEST_STATUS::  Traceback (most recent call last):
16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
"/usr/local/autotest/common_lib/test.py", line 725, in _call_test_function
16:09:25 INFO | autoserv| AUTOTEST_STATUS::      return func(*args, **dargs)
...
16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
"/usr/local/telemetry/src/tools/telemetry/telemetry/core/platform/profiler/android_profiling_helper.py",
line 12, in <module>
16:09:25 INFO | autoserv| AUTOTEST_STATUS::      import sqlite3
16:09:25 INFO | autoserv| AUTOTEST_STATUS::  ImportError: No module named
sqlite3

ChromeOS image does not have sqlite3. Please either file a bug (cc vapier@) to
add sqlite3. Or skip the sqlite3 logic for chromeos.

I suggest to revert this for now so that it does not shadow other failures.

Please use 

cbuildbot --remote -G <rietveld_id> x86-generic-tot-chrome-pfq-informational
amd64-generic-tot-chrome-pfq-informational daisy-tot-chrome-pfq-informational

to ensure the CL runs fine on ChromeOS when landing the new version.

[xiyuan, 2014-06-12 22:39:04]

On 2014/06/12 22:25:09, xiyuan wrote:
> This breaks ChromeOS build at
>
http://build.chromium.org/p/chromiumos.chromium/builders/X86%20%28chromium%29...
> 
> 16:09:25 INFO | autoserv|
> FAIL	login_Cryptohome	login_Cryptohome	timestamp=1402607364	localtime=Jun 12
> 14:09:24	Unhandled ImportError: No module named sqlite3
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::  Traceback (most recent call
last):
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
> "/usr/local/autotest/common_lib/test.py", line 725, in _call_test_function
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::      return func(*args, **dargs)
> ...
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
>
"/usr/local/telemetry/src/tools/telemetry/telemetry/core/platform/profiler/android_profiling_helper.py",
> line 12, in <module>
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::      import sqlite3
> 16:09:25 INFO | autoserv| AUTOTEST_STATUS::  ImportError: No module named
> sqlite3
> 
> ChromeOS image does not have sqlite3. Please either file a bug (cc vapier@) to
> add sqlite3. Or skip the sqlite3 logic for chromeos.
> 
> I suggest to revert this for now so that it does not shadow other failures.
> 
> Please use 
> 
> cbuildbot --remote -G <rietveld_id> x86-generic-tot-chrome-pfq-informational
> amd64-generic-tot-chrome-pfq-informational daisy-tot-chrome-pfq-informational
> 
> to ensure the CL runs fine on ChromeOS when landing the new version.

This might prevent PFQ uprev chrome tonight. We should really decide to revert
or fix asap.

[achuithb, 2014-06-12 22:41:52]

A revert of this CL has been created in
https://codereview.chromium.org/337603002/ by achuith@chromium.org.

The reason for reverting is: This breaks ChromeOS build at
http://build.chromium.org/p/chromiumos.chromium/builders/X86%20%28chromium%29...

16:09:25 INFO | autoserv|
FAIL	login_Cryptohome	login_Cryptohome	timestamp=1402607364	localtime=Jun 12
14:09:24	Unhandled ImportError: No module named sqlite3
16:09:25 INFO | autoserv| AUTOTEST_STATUS::  Traceback (most recent call last):
16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
"/usr/local/autotest/common_lib/test.py", line 725, in _call_test_function
16:09:25 INFO | autoserv| AUTOTEST_STATUS::      return func(*args, **dargs)
...
16:09:25 INFO | autoserv| AUTOTEST_STATUS::    File
"/usr/local/telemetry/src/tools/telemetry/telemetry/core/platform/profiler/android_profiling_helper.py",
line 12, in <module>
16:09:25 INFO | autoserv| AUTOTEST_STATUS::      import sqlite3
16:09:25 INFO | autoserv| AUTOTEST_STATUS::  ImportError: No module named
sqlite3

ChromeOS image does not have sqlite3. Please either file a bug (cc vapier@) to
add sqlite3. Or skip the sqlite3 logic for chromeos.

I suggest to revert this for now so that it does not shadow other failures.

Please use 

cbuildbot --remote -G <rietveld_id> x86-generic-tot-chrome-pfq-informational
amd64-generic-tot-chrome-pfq-informational daisy-tot-chrome-pfq-informational

to ensure the CL runs fine on ChromeOS when landing the new version..

[achuithb, 2014-06-12 22:50:46]

Sorry about the quick revert but the cros builders were all red increasing the
risk of other breaking CLs landing on top of this one.

Perhaps the best solution here is to conditionally import sqlite3?

[Dominik Grewe, 2014-06-13 10:57:12]

On 2014/06/12 22:50:46, achuithb wrote:
> Sorry about the quick revert but the cros builders were all red increasing the
> risk of other breaking CLs landing on top of this one.
> 
> Perhaps the best solution here is to conditionally import sqlite3?

Sorry about causing the bot to break. I've made the import conditional, so it
should work now.

I've tried to start a trybot using cbuildbot but I keep getting an error
message. Here's the output:


Verifying patches...
Submitting tryjob...
10:29:12: ERROR: Failed to submit tryjob.  This could be due to too many
submission requests by users.  Please try again.
cbuildbot: Unhandled exception:
Traceback (most recent call last):
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/bin/cbuildbot",
line 130, in <module>
    commandline.ScriptWrapperMain(FindTarget)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/commandline.py",
line 557, in ScriptWrapperMain
    ret = target(argv[1:])
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/scripts/cbuildbot.py",
line 1619, in main
    tryjob.Submit(testjob=options.test_tryjob, dryrun=False)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
line 242, in Submit
    self._Submit(ref.path, testjob, dryrun)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
line 222, in _Submit
    git.PushWithRetry(push_branch, workdir, retries=3, dryrun=dryrun)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py", line
1192, in PushWithRetry
    RunGit(git_repo, push_command)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py", line
834, in RunGit
    ['git'] + cmd, **kwargs)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/retry_util.py",
line 51, in GenericRetry
    return functor(*args, **kwargs)
  File
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/cros_build_lib.py",
line 561, in RunCommand
    raise RunCommandError(msg, cmd_result)
chromite.lib.cros_build_lib.RunCommandError: return code: 1
remote: Resolving deltas: 100% (1/1)
remote: Processing changes: refs: 1, done    
To https://chromium.googlesource.com/chromiumos/tryjobs
 ! [remote rejected] temp_auto_checkin_branch -> master (prohibited by Gerrit)
error: failed to push some refs to
'https://chromium.googlesource.com/chromiumos/tryjobs'

Failed command "git push origin temp_auto_checkin_branch:refs/heads/master",
cwd=/usr/local/google/code/sdb1/clankium/src/third_party/.cache/trybot, extra
env=None

[xiyuan, 2014-06-13 16:14:37]

On 2014/06/13 10:57:12, Dominik Grewe wrote:
> On 2014/06/12 22:50:46, achuithb wrote:
> > Sorry about the quick revert but the cros builders were all red increasing
the
> > risk of other breaking CLs landing on top of this one.
> > 
> > Perhaps the best solution here is to conditionally import sqlite3?
> 
> Sorry about causing the bot to break. I've made the import conditional, so it
> should work now.
> 
> I've tried to start a trybot using cbuildbot but I keep getting an error
> message. Here's the output:
> 
> 
> Verifying patches...
> Submitting tryjob...
> 10:29:12: ERROR: Failed to submit tryjob.  This could be due to too many
> submission requests by users.  Please try again.
> cbuildbot: Unhandled exception:
> Traceback (most recent call last):
>   File
> "/usr/local/google/code/sdb1/clankium/src/third_party/chromite/bin/cbuildbot",
> line 130, in <module>
>     commandline.ScriptWrapperMain(FindTarget)
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/commandline.py",
> line 557, in ScriptWrapperMain
>     ret = target(argv[1:])
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/scripts/cbuildbot.py",
> line 1619, in main
>     tryjob.Submit(testjob=options.test_tryjob, dryrun=False)
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
> line 242, in Submit
>     self._Submit(ref.path, testjob, dryrun)
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
> line 222, in _Submit
>     git.PushWithRetry(push_branch, workdir, retries=3, dryrun=dryrun)
>   File
> "/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py",
line
> 1192, in PushWithRetry
>     RunGit(git_repo, push_command)
>   File
> "/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py",
line
> 834, in RunGit
>     ['git'] + cmd, **kwargs)
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/retry_util.py",
> line 51, in GenericRetry
>     return functor(*args, **kwargs)
>   File
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/cros_build_lib.py",
> line 561, in RunCommand
>     raise RunCommandError(msg, cmd_result)
> chromite.lib.cros_build_lib.RunCommandError: return code: 1
> remote: Resolving deltas: 100% (1/1)
> remote: Processing changes: refs: 1, done    
> To https://chromium.googlesource.com/chromiumos/tryjobs
>  ! [remote rejected] temp_auto_checkin_branch -> master (prohibited by Gerrit)
> error: failed to push some refs to
> 'https://chromium.googlesource.com/chromiumos/tryjobs'
> 
> Failed command "git push origin temp_auto_checkin_branch:refs/heads/master",
> cwd=/usr/local/google/code/sdb1/clankium/src/third_party/.cache/trybot, extra
> env=None

I suspect you have not setup credentials to chromium.googlesource.com. Try
follow the "(Googler) To get access to the internal Chrome OS gerrit instance"
section at
http://dev.chromium.org/chromium-os/developer-guide/gerrit-guide

After that, you should be able to 
git ls-remote https://chromium.googlesource.com/chromiumos/tryjobs

If that works, you should be able to push your tryjobs.

[Dominik Grewe, 2014-06-13 16:47:08]

On 2014/06/13 16:14:37, xiyuan wrote:
> On 2014/06/13 10:57:12, Dominik Grewe wrote:
> > On 2014/06/12 22:50:46, achuithb wrote:
> > > Sorry about the quick revert but the cros builders were all red increasing
> the
> > > risk of other breaking CLs landing on top of this one.
> > > 
> > > Perhaps the best solution here is to conditionally import sqlite3?
> > 
> > Sorry about causing the bot to break. I've made the import conditional, so
it
> > should work now.
> > 
> > I've tried to start a trybot using cbuildbot but I keep getting an error
> > message. Here's the output:
> > 
> > 
> > Verifying patches...
> > Submitting tryjob...
> > 10:29:12: ERROR: Failed to submit tryjob.  This could be due to too many
> > submission requests by users.  Please try again.
> > cbuildbot: Unhandled exception:
> > Traceback (most recent call last):
> >   File
> >
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/bin/cbuildbot",
> > line 130, in <module>
> >     commandline.ScriptWrapperMain(FindTarget)
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/commandline.py",
> > line 557, in ScriptWrapperMain
> >     ret = target(argv[1:])
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/scripts/cbuildbot.py",
> > line 1619, in main
> >     tryjob.Submit(testjob=options.test_tryjob, dryrun=False)
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
> > line 242, in Submit
> >     self._Submit(ref.path, testjob, dryrun)
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/cbuildbot/remote_try.py",
> > line 222, in _Submit
> >     git.PushWithRetry(push_branch, workdir, retries=3, dryrun=dryrun)
> >   File
> > "/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py",
> line
> > 1192, in PushWithRetry
> >     RunGit(git_repo, push_command)
> >   File
> > "/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/git.py",
> line
> > 834, in RunGit
> >     ['git'] + cmd, **kwargs)
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/retry_util.py",
> > line 51, in GenericRetry
> >     return functor(*args, **kwargs)
> >   File
> >
>
"/usr/local/google/code/sdb1/clankium/src/third_party/chromite/lib/cros_build_lib.py",
> > line 561, in RunCommand
> >     raise RunCommandError(msg, cmd_result)
> > chromite.lib.cros_build_lib.RunCommandError: return code: 1
> > remote: Resolving deltas: 100% (1/1)
> > remote: Processing changes: refs: 1, done    
> > To https://chromium.googlesource.com/chromiumos/tryjobs
> >  ! [remote rejected] temp_auto_checkin_branch -> master (prohibited by
Gerrit)
> > error: failed to push some refs to
> > 'https://chromium.googlesource.com/chromiumos/tryjobs'
> > 
> > Failed command "git push origin temp_auto_checkin_branch:refs/heads/master",
> > cwd=/usr/local/google/code/sdb1/clankium/src/third_party/.cache/trybot,
extra
> > env=None
> 
> I suspect you have not setup credentials to http://chromium.googlesource.com.
Try
> follow the "(Googler) To get access to the internal Chrome OS gerrit instance"
> section at
> http://dev.chromium.org/chromium-os/developer-guide/gerrit-guide
> 
> After that, you should be able to 
> git ls-remote https://chromium.googlesource.com/chromiumos/tryjobs
> 
> If that works, you should be able to push your tryjobs.

I've got the right credentials, I think. The 'git ls-remote ...' works but I
still get the error when running cbuildbot. The command I'm using is

cbuildbot --remote -G 308123002 x86-generic-tot-chrome-pfq-informational
amd64-generic-tot-chrome-pfq-informational daisy-tot-chrome-pfq-informational

[xiyuan, 2014-06-13 16:51:41]

+davidjames,

David, could you help to see why Dominik is not able to run cbuildbot to submit
tryjobs?

His command line looks correct. Initially I thought he does not have netrc setup
but that seems not the case.

[Dominik Grewe, 2014-06-17 10:26:42]

On 2014/06/13 16:51:41, xiyuan wrote:
> +davidjames,
> 
> David, could you help to see why Dominik is not able to run cbuildbot to
submit
> tryjobs?
> 
> His command line looks correct. Initially I thought he does not have netrc
setup
> but that seems not the case.

Anything I could try here? Otherwise I'll just commit again and we'll see if it
breaks ;)

[xiyuan, 2014-06-17 16:12:37]

On 2014/06/17 10:26:42, Dominik Grewe wrote:
> On 2014/06/13 16:51:41, xiyuan wrote:
> > +davidjames,
> > 
> > David, could you help to see why Dominik is not able to run cbuildbot to
> submit
> > tryjobs?
> > 
> > His command line looks correct. Initially I thought he does not have netrc
> setup
> > but that seems not the case.
> 
> Anything I could try here? Otherwise I'll just commit again and we'll see if
it
> breaks ;)

Still not sure what the problem is. I have tried to submit the job for you. But
the bots is complaining missing package.

https://uberchromegw.corp.google.com/i/chromiumos.tryserver/builders/amd64-ge...

cbuildbot: error: Option --buildbot/--remote-trybot requires the following
binaries which couldn't be found in $PATH: pbzip2

I'd say just commit again and see how it goes this time.

[Dominik Grewe, 2014-06-17 16:15:21]

On 2014/06/17 16:12:37, xiyuan wrote:
> On 2014/06/17 10:26:42, Dominik Grewe wrote:
> > On 2014/06/13 16:51:41, xiyuan wrote:
> > > +davidjames,
> > > 
> > > David, could you help to see why Dominik is not able to run cbuildbot to
> > submit
> > > tryjobs?
> > > 
> > > His command line looks correct. Initially I thought he does not have netrc
> > setup
> > > but that seems not the case.
> > 
> > Anything I could try here? Otherwise I'll just commit again and we'll see if
> it
> > breaks ;)
> 
> Still not sure what the problem is. I have tried to submit the job for you.
But
> the bots is complaining missing package.
> 
>
https://uberchromegw.corp.google.com/i/chromiumos.tryserver/builders/amd64-ge...
> 
> cbuildbot: error: Option --buildbot/--remote-trybot requires the following
> binaries which couldn't be found in $PATH: pbzip2
> 
> I'd say just commit again and see how it goes this time.

Thanks. I'll commit then.

[Dominik Grewe, 2014-06-17 16:15:25]

The CQ bit was checked by dominikg@chromium.org

[commit-bot: I haz the power, 2014-06-17 16:16:02]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dominikg@chromium.org/308123002/100001

[commit-bot: I haz the power, 2014-06-17 20:06:00]

Change committed as 277839