Roll googletest to 1.8.0.

Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
TBR=rkc,dpranke,agl

Review-Url: https://codereview.chromium.org/2852613002
Cr-Commit-Position: refs/heads/master@{#468860}
Committed: https://chromium.googlesource.com/chromium/src/+/dbcef9bdc1edd32d62c6de0bd382cdc78ce385da

[pwnall, 2017-04-28 03:14:28]

Description was changed from

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
==========

to

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
==========

[pwnall, 2017-04-28 03:28:52]

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

[commit-bot: I haz the power, 2017-04-28 03:29:14]

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, 2017-04-28 04:59:44]

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

[commit-bot: I haz the power, 2017-04-28 04:59:45]

Dry run: Try jobs failed on following builders:
  android_arm64_dbg_recipe on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_arm6...)

[pwnall, 2017-04-28 09:43:50]

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

[commit-bot: I haz the power, 2017-04-28 09:44:18]

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, 2017-04-28 12:55:49]

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

[commit-bot: I haz the power, 2017-04-28 12:55:51]

Dry run: This issue passed the CQ dry run.

[pwnall, 2017-04-28 13:38:58]

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

[pwnall, 2017-04-28 13:38:58]

dpranke: Can you PTAL and see if the fix for the build error makes sense? I've
been trying really hard to avoid creating an empty file, but I think I'm out of
options.

[pwnall, 2017-05-02 01:34:32]

pwnall@chromium.org changed reviewers:
+ thakis@chromium.org

[pwnall, 2017-05-02 01:35:06]

pwnall@chromium.org changed reviewers:
+ agl@chromium.org, rkc@chromium.org

[Nico, 2017-05-02 20:39:44]

lgtm

[pwnall, 2017-05-02 22:27:47]

Description was changed from

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
==========

to

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
TBR=rkc,dpranke,agl
==========

[pwnall, 2017-05-02 22:28:16]

The CQ bit was checked by pwnall@chromium.org

[commit-bot: I haz the power, 2017-05-02 22:28:55]

CQ is trying da patch.

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

[pwnall, 2017-05-02 22:29:44]

On 2017/05/02 20:39:44, Nico wrote:
> lgtm

Thank you very much! I'll monitor the bots while this lands.

[commit-bot: I haz the power, 2017-05-03 01:50:30]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1493764096574880,
"parent_rev": "5656628563d79c5e56899b0103eb35e2576e83db", "commit_rev":
"dbcef9bdc1edd32d62c6de0bd382cdc78ce385da"}

[commit-bot: I haz the power, 2017-05-03 01:50:45]

Description was changed from

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
TBR=rkc,dpranke,agl
==========

to

==========
Roll googletest to 1.8.0.

This is a re-land of http://crrev.com/2779193002 which was reverted due to
build errors on Mac.

GoogleTest (gtest) and GoogleMock (gmock) are now hosted into the same
googletest repository. In order to cope with this, the googletest
repository is now sourced at third_party/googletest.

The file/directory layout of Google Test is not yet considered stable.
To minimize disruption while Google Test stabilizes, Chromium code will
be insulated from third_party/googletest.

* testing/gtest/include/gtest/ and testing/gmock/include/gmock have
  been populated with headers that forward into the appropriate locations of
  third_party/googletest

* testing/BUILD.gn has been populated with the targets
  //testing/gtest(:gtest_main) and //testing/gmock(:gmock_main), which
  depend on the appropriate //third_party/googletest targets.

All Chromium code should keep depending on the targets and headers
in testing/{gtest,gmock} for now.

BUG=630705
TESTED=ninja -C out/Default/ ced_unittests && ninja -C out/Default
TBR=rkc,dpranke,agl

Review-Url: https://codereview.chromium.org/2852613002
Cr-Commit-Position: refs/heads/master@{#468860}
Committed:
https://chromium.googlesource.com/chromium/src/+/dbcef9bdc1edd32d62c6de0bd382...
==========

[commit-bot: I haz the power, 2017-05-03 01:50:46]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/src/+/dbcef9bdc1edd32d62c6de0bd382...

[Michael Achenbach, 2017-05-03 09:18:01]

This is suboptimal as it now blocks rolling build into V8:
https://chromium-review.googlesource.com/c/493947/

Wouldn't it be possible to make a smooth 3-way change for such things like, 1)
land first part in chromium, 2) update subprojects and roll, 3) land second part
in chromium.

If not, please fix the V8-side asap and roll build in one CL. If that's not
possible soon, please consider reverting this until
https://codereview.chromium.org/2847693002/ is ready.

[Michael Achenbach, 2017-05-03 09:20:01]

On 2017/05/03 09:18:01, Michael Achenbach wrote:
> This is suboptimal as it now blocks rolling build into V8:
> https://chromium-review.googlesource.com/c/493947/
> 
> Wouldn't it be possible to make a smooth 3-way change for such things like, 1)
> land first part in chromium, 2) update subprojects and roll, 3) land second
part
> in chromium.
> 
> If not, please fix the V8-side asap and roll build in one CL. If that's not
> possible soon, please consider reverting this until
> https://codereview.chromium.org/2847693002/ is ready.

Or is it just the gtest roll that now breaks things because
https://codereview.chromium.org/2847693002/ didn't land yet?

[pwnall, 2017-05-03 09:22:35]

On 2017/05/03 09:20:01, Michael Achenbach wrote:
> On 2017/05/03 09:18:01, Michael Achenbach wrote:
> > This is suboptimal as it now blocks rolling build into V8:
> > https://chromium-review.googlesource.com/c/493947/
> > 
> > Wouldn't it be possible to make a smooth 3-way change for such things like,
1)
> > land first part in chromium, 2) update subprojects and roll, 3) land second
> part
> > in chromium.
> > 
> > If not, please fix the V8-side asap and roll build in one CL. If that's not
> > possible soon, please consider reverting this until
> > https://codereview.chromium.org/2847693002/ is ready.
> 
> Or is it just the gtest roll that now breaks things because
> https://codereview.chromium.org/2847693002/ didn't land yet?

The change was not intended to break V8. I avoided removing build/secondary
precisely so that V8 wouldn't break.

Sorry for the breakage. I'll investigate. Thank you very much for pinging me!

[Nico, 2017-05-03 10:06:08]

Also see the somewhat lengthy discussion about v8 in the original version
of this cl.

On May 3, 2017 5:22 AM, <pwnall@chromium.org> wrote:

On 2017/05/03 09:20:01, Michael Achenbach wrote:
> On 2017/05/03 09:18:01, Michael Achenbach wrote:
> > This is suboptimal as it now blocks rolling build into V8:
> > https://chromium-review.googlesource.com/c/493947/
> >
> > Wouldn't it be possible to make a smooth 3-way change for such things
like,
1)
> > land first part in chromium, 2) update subprojects and roll, 3) land
second
> part
> > in chromium.
> >
> > If not, please fix the V8-side asap and roll build in one CL. If that's
not
> > possible soon, please consider reverting this until
> > https://codereview.chromium.org/2847693002/ is ready.
>
> Or is it just the gtest roll that now breaks things because
> https://codereview.chromium.org/2847693002/ didn't land yet?

The change was not intended to break V8. I avoided removing build/secondary
precisely so that V8 wouldn't break.

Sorry for the breakage. I'll investigate. Thank you very much for pinging
me!

https://codereview.chromium.org/2852613002/

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[msramek, 2017-05-03 10:51:02]

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

[msramek, 2017-05-03 10:51:03]

This seems to have broken *all* content_browsertests on the ChromiumOS bot.

Stack trace:

Received signal 6
#0 0x7f04980a232b base::debug::StackTrace::StackTrace()
#1 0x7f04980a106c base::debug::StackTrace::StackTrace()
#2 0x7f04980a1e3f base::debug::(anonymous namespace)::StackDumpSignalHandler()
#3 0x7f04a0840330 <unknown>
#4 0x7f048bf0ac37 gsignal
#5 0x7f048bf0e028 abort
#6 0x0000005ba529 testing::internal::posix::Abort()
#7 0x000000d4b500 testing::internal::GTestLog::~GTestLog()
#8 0x7f048d7e6976 testing::internal::ThreadLocal<>::~ThreadLocal()
#9 0x7f048bf1053a __cxa_finalize
#10 0x7f048d6cd633 <unknown>


I'll revert again.

[msramek, 2017-05-03 10:54:30]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/2856963003/ by msramek@chromium.org.

The reason for reverting is: Seems to have caused all content_browsertests on
ChromiumOS to crash.

https://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%...

I'll run the revert through trybots so that it doesn't break anything else..

[Nico, 2017-05-03 11:02:14]

Please file a bug for the chromeos thi

On May 3, 2017 6:54 AM, <msramek@chromium.org> wrote:

> A revert of this CL (patchset #3 id:40001) has been created in
> https://codereview.chromium.org/2856963003/ by msramek@chromium.org.
>
> The reason for reverting is: Seems to have caused all content_browsertests
> on
> ChromiumOS to crash.
>
> https://build.chromium.org/p/chromium.chromiumos/builders/
> Linux%20ChromiumOS%20Tests%20%28dbg%29%281%29/builds/25582
>
> I'll run the revert through trybots so that it doesn't break anything
> else..
>
> https://codereview.chromium.org/2852613002/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[pwnall, 2017-05-03 21:56:05]

On 2017/05/03 10:54:30, msramek (recovering) wrote:
> A revert of this CL (patchset #3 id:40001) has been created in
> https://codereview.chromium.org/2856963003/ by mailto:msramek@chromium.org.
> 
> The reason for reverting is: Seems to have caused all content_browsertests on
> ChromiumOS to crash.
> 
>
https://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%...
> 
> I'll run the revert through trybots so that it doesn't break anything else..

msramek: The CL doesn't appear to have run through the trybots. Am I missing
something?

Also, did you file a bug so we can have a record of the breakage? I didn't see
anything in my inbox.

[Nico, 2017-05-03 22:02:57]

There's no bug I believe (I asked for one on the revert CL too), but
https://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%...
shows what went wrong.

On Wed, May 3, 2017 at 5:56 PM, <pwnall@chromium.org> wrote:

> On 2017/05/03 10:54:30, msramek (recovering) wrote:
> > A revert of this CL (patchset #3 id:40001) has been created in
> > https://codereview.chromium.org/2856963003/ by mailto:
> msramek@chromium.org.
> >
> > The reason for reverting is: Seems to have caused all
> content_browsertests on
> > ChromiumOS to crash.
> >
> >
> https://build.chromium.org/p/chromium.chromiumos/builders/
> Linux%20ChromiumOS%20Tests%20%28dbg%29%281%29/builds/25582
> >
> > I'll run the revert through trybots so that it doesn't break anything
> else..
>
> msramek: The CL doesn't appear to have run through the trybots. Am I
> missing
> something?
>
> Also, did you file a bug so we can have a record of the breakage? I didn't
> see
> anything in my inbox.
>
> https://codereview.chromium.org/2852613002/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[pwnall, 2017-05-03 22:46:53]

On 2017/05/03 22:02:57, Nico wrote:
> There's no bug I believe (I asked for one on the revert CL too), but
>
https://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%...
> shows what went wrong.

I'll extract the relevant snippets and write up a bug report. In my experience
(I might be holding it wrong?), the builder logs disappear after a couple of
weeks, so links to them are not good permanent records.

[Nico, 2017-05-03 22:52:17]

The buildbot urls expire, but you can hit "logdog" in the upper right to
get to
https://luci-milo.appspot.com/buildbot/chromium.chromiumos/Linux%20ChromiumOS...
and that won't expire.

Do you think another attempt will take multiple weeks? I asked someone on a
different review who was touching gtest/BUILD.gn to wait until this here is
in for good. If you think it'll take a while, I'll tell them to land first
(it's fairly easy to merge into this).

On Wed, May 3, 2017 at 6:46 PM, <pwnall@chromium.org> wrote:

> On 2017/05/03 22:02:57, Nico wrote:
> > There's no bug I believe (I asked for one on the revert CL too), but
> >
> https://build.chromium.org/p/chromium.chromiumos/builders/
> Linux%20ChromiumOS%20Tests%20%28dbg%29%281%29/builds/25582
> > shows what went wrong.
>
> I'll extract the relevant snippets and write up a bug report. In my
> experience
> (I might be holding it wrong?), the builder logs disappear after a couple
> of
> weeks, so links to them are not good permanent records.
>
> https://codereview.chromium.org/2852613002/
>

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[pwnall, 2017-05-04 00:51:29]

On 2017/05/03 22:52:17, Nico wrote:
> The buildbot urls expire, but you can hit "logdog" in the upper right to
> get to
>
https://luci-milo.appspot.com/buildbot/chromium.chromiumos/Linux%20ChromiumOS...
> and that won't expire.

Thanks for explaining! I might be lazy then.
 
> Do you think another attempt will take multiple weeks? I asked someone on a
> different review who was touching gtest/BUILD.gn to wait until this here is
> in for good. If you think it'll take a while, I'll tell them to land first
> (it's fairly easy to merge into this).

I'll try to have another attempt ready in a day or so, to keep the momentum
going. I want to take advantage of the fact that this CL's state is in our
heads.

If the other CL is ready to go today/tomorrow, I'm OK with rebasing.