Roll clang 277962:278861

Roll clang 277962:278861

BUG=636558, 637866
R=dcheng@chromium.org, dpranke@chromium.org, inferno@chromium.org, rnk@chromium.org

Committed: https://chromium.googlesource.com/chromium/src/+/3829c6ad5aeef24992be742340df28c324da9e61

[hans, 2016-08-16 03:52:30]

hans@chromium.org changed reviewers:
+ rnk@chromium.org

[hans, 2016-08-16 03:52:31]

Trying r278762 which has rnk's fix for a failing AMDGPU test.

[hans, 2016-08-16 03:56:32]

The CQ bit was checked by hans@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-16 03:56:45]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-16 04:14:05]

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

[commit-bot: I haz the power, 2016-08-16 04:14:06]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[hans, 2016-08-16 15:31:48]

The CQ bit was checked by hans@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-16 15:32:07]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-16 16:49:41]

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

[commit-bot: I haz the power, 2016-08-16 16:49:42]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[hans, 2016-08-17 00:37:00]

try 278861 instead

[hans, 2016-08-17 00:38:54]

Description was changed from

==========
Roll clang 277962:278762.

Ran `tools/clang/scripts/upload_revision.py 278762`.

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
==========

[hans, 2016-08-17 03:10:21]

The CQ bit was checked by hans@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-17 03:10:44]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-17 04:53:44]

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

[commit-bot: I haz the power, 2016-08-17 04:53:45]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[hans, 2016-08-17 17:24:01]

Rerunning made mac_chromium_rel_ng green, but the ios-simulator failure is
persistent. We don't run ios tests on the ToT bots, so there's not much to go on
:-/

I wonder if I can run this locally..

[Reid Kleckner, 2016-08-17 17:31:37]

Yep, looks like we have to debug manually. The normal builder is green:
https://build.chromium.org/p/chromium.mac/builders/ios-simulator/

[hans, 2016-08-17 20:11:57]

hans@chromium.org changed reviewers:
+ dcheng@chromium.org

[hans, 2016-08-17 20:11:58]

rnk figured out that the test failure is just due to the stack growing with the
new clang. We already have an #ifdef for ASan, so let's expand that for this
case too.

rnk, I think we're good to roll

dcheng: please take an OWNER's look for base/

[Reid Kleckner, 2016-08-17 20:12:59]

lgtm

[dcheng, 2016-08-17 21:51:27]

//base LGTM

[hans, 2016-08-17 23:02:01]

Thanks!

For those following along, I'm doing a double check on that base_unittests test
just to make sure we're not growing the stack in some pathological way.

[hans, 2016-08-17 23:59:45]

I think the increased stack size might be due to
http://llvm.org/viewvc/llvm-project?rev=278356&view=rev

[hans, 2016-08-17 23:59:55]

The CQ bit was checked by hans@chromium.org

[commit-bot: I haz the power, 2016-08-18 00:01:05]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-18 00:07:49]

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

[commit-bot: I haz the power, 2016-08-18 00:07:51]

Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[hans, 2016-08-18 00:14:51]

hans@chromium.org changed reviewers:
+ dpranke@chromium.org

[hans, 2016-08-18 00:14:52]

dpranke: can you take an OWNER's look at the .gn change?

[inferno, 2016-08-18 02:12:46]

inferno@chromium.org changed reviewers:
+ inferno@chromium.org

[Dirk Pranke, 2016-08-18 02:14:55]

lgtm

[hans, 2016-08-18 04:28:07]

The CQ bit was checked by hans@chromium.org

[commit-bot: I haz the power, 2016-08-18 04:28:18]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-18 04:32:38]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
==========

[commit-bot: I haz the power, 2016-08-18 04:32:40]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-08-18 04:34:37]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866

Committed: https://crrev.com/29ecb32ba7b5ca183782c5f5d2b7a1ae23309edb
Cr-Commit-Position: refs/heads/master@{#412745}
==========

[commit-bot: I haz the power, 2016-08-18 04:34:39]

Patchset 5 (id:??) landed as
https://crrev.com/29ecb32ba7b5ca183782c5f5d2b7a1ae23309edb
Cr-Commit-Position: refs/heads/master@{#412745}

[Guido Urdaneta, 2016-08-18 10:13:56]

A revert of this CL (patchset #5 id:80001) has been created in
https://codereview.chromium.org/2249163007/ by guidou@chromium.org.

The reason for reverting is: This is suspect of breaking Mac 10.9 builder.

See
https://build.chromium.org/p/chromium.mac/builders/Mac10.9%20Tests%20%28dbg%2...

Sample logs:
Retrying 1 test (retry #1)
[0817/224954:ERROR:kill_posix.cc(84)] Unable to terminate process group 8687: No
such process
[ RUN      ] ThreadTest.StartWithOptions_StackSize
[2060/2060] ThreadTest.StartWithOptions_StackSize (CRASHED)
Retrying 1 test (retry #2)
[0817/224955:ERROR:kill_posix.cc(84)] Unable to terminate process group 8688: No
such process
[ RUN      ] ThreadTest.StartWithOptions_StackSize
[2061/2061] ThreadTest.StartWithOptions_StackSize (CRASHED)
Retrying 1 test (retry #3)
[0817/224955:ERROR:kill_posix.cc(84)] Unable to terminate process group 8689: No
such process
[ RUN      ] ThreadTest.StartWithOptions_StackSize
[2062/2062] ThreadTest.StartWithOptions_StackSize (CRASHED)
1 test crashed:
    ThreadTest.StartWithOptions_StackSize
(../../base/threading/thread_unittest.cc:135)
Tests took 21 seconds.
Additional test environment:
    CHROME_DEVEL_SANDBOX=/opt/chromium/chrome_sandbox
    LANG=en_US.UTF-8
Command: ./base_unittests --brave-new-test-launcher --test-launcher-bot-mode
--test-launcher-summary-output=/b/swarm_slave/w/ionk6XHB/output.json.

[Reid Kleckner, 2016-08-18 15:33:43]

Can we just use 24K as the stack size for all platforms and cut the ifdef? We're
pretty sure at this point that this stack size change was caused by a libc++
change, and affects all mac targets.

[hans, 2016-08-18 17:30:31]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866

Committed: https://crrev.com/29ecb32ba7b5ca183782c5f5d2b7a1ae23309edb
Cr-Commit-Position: refs/heads/master@{#412745}
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
==========

[hans, 2016-08-18 17:31:59]

On 2016/08/18 15:33:43, Reid Kleckner wrote:
> Can we just use 24K as the stack size for all platforms and cut the ifdef?
We're
> pretty sure at this point that this stack size change was caused by a libc++
> change, and affects all mac targets.

I'm widening the #ifdef to asan + non-release builds.

[hans, 2016-08-18 18:21:32]

The CQ bit was checked by hans@chromium.org

[hans, 2016-08-18 18:21:33]

The patchset sent to the CQ was uploaded after l-g-t-m from rnk@chromium.org,
dpranke@chromium.org, dcheng@chromium.org
Link to the patchset: https://codereview.chromium.org/2241413003/#ps100001
(title: "Widening the #ifdef")

[commit-bot: I haz the power, 2016-08-18 18:22:11]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Reid Kleckner, 2016-08-18 18:26:50]

lgtm x2

[dcheng, 2016-08-18 19:20:52]

still lgtm

[commit-bot: I haz the power, 2016-08-18 21:50:02]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
==========

[commit-bot: I haz the power, 2016-08-18 21:50:03]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-08-18 21:54:10]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866

Committed: https://crrev.com/df7510f96322254211068a52c2b459c0fe0b2090
Cr-Commit-Position: refs/heads/master@{#412943}
==========

[commit-bot: I haz the power, 2016-08-18 21:54:11]

Patchset 6 (id:??) landed as
https://crrev.com/df7510f96322254211068a52c2b459c0fe0b2090
Cr-Commit-Position: refs/heads/master@{#412943}

[hans, 2016-08-19 01:22:50]

A revert of this CL (patchset #6 id:100001) has been created in
https://codereview.chromium.org/2257333002/ by hans@chromium.org.

The reason for reverting is: This broke the build on a certain Mac builder: 
https://build.chromium.org/p/chromium/builders/Mac/builds/18669

It looks like the build is timing out maybe? All other Mac
builders seem happy though, which makes this confusing, but
the buildbot logs very clearly implicate the roll: the same
kind of error happened last time it landed, and it went away
on the previous revert..

[hans, 2016-08-19 20:50:47]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866

Committed: https://crrev.com/df7510f96322254211068a52c2b459c0fe0b2090
Cr-Commit-Position: refs/heads/master@{#412943}
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
==========

[hans, 2016-08-19 20:51:28]

Third time's the charm :-)

inferno: can you take an OWNER's look at the ipc_fuzzer change

[inferno, 2016-08-19 20:55:18]

lgtm

[hans, 2016-08-19 20:55:48]

The CQ bit was checked by hans@chromium.org

[hans, 2016-08-19 20:55:49]

The patchset sent to the CQ was uploaded after l-g-t-m from rnk@chromium.org,
dpranke@chromium.org, dcheng@chromium.org
Link to the patchset: https://codereview.chromium.org/2241413003/#ps120001
(title: "Work around llvm.org/PR29064: compile times out compiling
message_names.cc")

[commit-bot: I haz the power, 2016-08-19 20:56:46]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[hans, 2016-08-19 20:59:47]

More explanation for anyone looking at this:

PopulateIpcMessageNames() calls ipc_fuzzer::MessageNames::Add() with a very
large number of constant strings.

In r278356, libc++ marked std::string::__init always_inline, which caused
PopulateIpcMessageNames() to become very large and become pathologically slow to
optimize.

The change makes MessageNames::Add() take a "const char*" instead, which avoids
constructing all the std::string objects in PopulateIpcMessageNames().

[commit-bot: I haz the power, 2016-08-19 22:55:22]

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

[commit-bot: I haz the power, 2016-08-19 22:55:23]

Try jobs failed on following builders:
  android_clang_dbg_recipe on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_clan...)
  linux_chromium_chromeos_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[hans, 2016-08-20 00:11:10]

The CQ bit was checked by hans@chromium.org

[hans, 2016-08-20 00:11:10]

The patchset sent to the CQ was uploaded after l-g-t-m from rnk@chromium.org,
dpranke@chromium.org, dcheng@chromium.org, inferno@chromium.org
Link to the patchset: https://codereview.chromium.org/2241413003/#ps140001
(title: "Build fix")

[commit-bot: I haz the power, 2016-08-20 00:11:45]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[inferno, 2016-08-20 00:27:09]

lgtm

[commit-bot: I haz the power, 2016-08-20 04:14:31]

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

[commit-bot: I haz the power, 2016-08-20 04:14:33]

Try jobs failed on following builders:
  linux_android_rel_ng on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

[inferno, 2016-08-20 05:03:13]

The CQ bit was checked by inferno@chromium.org

[commit-bot: I haz the power, 2016-08-20 05:03:20]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-20 06:34:11]

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

[commit-bot: I haz the power, 2016-08-20 06:34:13]

Try jobs failed on following builders:
  linux_android_rel_ng on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

[hans, 2016-08-20 14:29:56]

On 2016/08/20 06:34:13, commit-bot: I haz the power wrote:
> Try jobs failed on following builders:
>   linux_android_rel_ng on master.tryserver.chromium.android (JOB_FAILED,
>
https://build.chromium.org/p/tryserver.chromium.android/builders/linux_androi...)

That test seems to be failing a lot on that bot, for other changes than this.

Given that the roll has been landed twice before without any problems with
unit_tests on Android, I'll land manually.

[hans, 2016-08-20 14:39:10]

(The bad trybot is http://crbug.com/639538)

[commit-bot: I haz the power, 2016-08-20 14:43:18]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
R=dcheng@chromium.org, dpranke@chromium.org, inferno@chromium.org,
rnk@chromium.org

Committed: https://crrev.com/3829c6ad5aeef24992be742340df28c324da9e61
Cr-Commit-Position: refs/heads/master@{#413337}
==========

[commit-bot: I haz the power, 2016-08-20 14:43:20]

Patchset 8 (id:??) landed as
https://crrev.com/3829c6ad5aeef24992be742340df28c324da9e61
Cr-Commit-Position: refs/heads/master@{#413337}

[hans, 2016-08-20 14:44:36]

Description was changed from

==========
Roll clang 277962:278861

BUG=636558,637866
R=dcheng@chromium.org, dpranke@chromium.org, inferno@chromium.org,
rnk@chromium.org

Committed: https://crrev.com/3829c6ad5aeef24992be742340df28c324da9e61
Cr-Commit-Position: refs/heads/master@{#413337}
==========

to

==========
Roll clang 277962:278861

BUG=636558,637866
R=dcheng@chromium.org, dpranke@chromium.org, inferno@chromium.org,
rnk@chromium.org

Committed:
https://chromium.googlesource.com/chromium/src/+/3829c6ad5aeef24992be742340df...
==========

[hans, 2016-08-20 14:44:38]

Committed patchset #8 (id:140001) manually as
3829c6ad5aeef24992be742340df28c324da9e61 (presubmit successful).