webkit_layout_tests: Add osmesa as data_dep.

webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002, 524758
NOTRY=true

Review-Url: https://codereview.chromium.org/2702613003
Cr-Commit-Position: refs/heads/master@{#451226}
Committed: https://chromium.googlesource.com/chromium/src/+/aac85d78a79fe653d913ad9278c18da28900a25b

[mithro, 2017-02-16 23:59:19]

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

[mithro, 2017-02-17 00:00:00]

Description was changed from

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
==========

to

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
==========

[mithro, 2017-02-17 00:00:00]

tansell@chromium.org changed reviewers:
+ dpranke@google.com, jochen@chromium.org

[commit-bot: I haz the power, 2017-02-17 00:01:03]

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

[mithro, 2017-02-17 00:09:14]

Hi!

This patch should fix (or get us closer to) LayoutTests on swarming for the Mac
builders. The error they are getting is;
---------------------
15:16:52.566 5262  
[5298:771:0216/151652.444061:2246944496735:ERROR:gl_initializer_mac.cc(90)]
osmesa.so not found at /b/swarm_slave/w/iryyXQcQ/out/Release/osmesa.so
15:16:52.566 5262  
[5298:771:0216/151652.540598:2247041030049:ERROR:gl_initializer_mac.cc(90)]
osmesa.so not found at /b/swarm_slave/w/iryyXQcQ/out/Release/osmesa.so
---------------------

I'm unsure why Windows/Linux is not complaining about this file being missing
too. I wonder if the base system is providing it?

Tim 'mithro' Ansell

[mithro, 2017-02-17 00:56:54]

tansell@chromium.org changed reviewers:
+ jbudorick@chromium.org

[mithro, 2017-02-17 00:57:17]

Hi John,

Looks like Dirk is OO today. Can you give me an LGTM for this CL?

Tim 'mithro' Ansell

[jbudorick, 2017-02-17 00:59:57]

lgtm

[mithro, 2017-02-17 01:20:45]

The CQ bit was unchecked by tansell@chromium.org

[mithro, 2017-02-17 01:20:49]

The CQ bit was checked by tansell@chromium.org

[commit-bot: I haz the power, 2017-02-17 01:21:47]

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, 2017-02-17 02:39:47]

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

[commit-bot: I haz the power, 2017-02-17 02:39:48]

Try jobs failed on following builders:
  win_chromium_x64_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[mithro, 2017-02-17 02:45:35]

The CQ bit was checked by tansell@chromium.org

[commit-bot: I haz the power, 2017-02-17 02:46:48]

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, 2017-02-17 04:36:30]

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

[commit-bot: I haz the power, 2017-02-17 04:36:31]

Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_TIMED_OUT,
no build URL)

[mithro, 2017-02-17 04:40:11]

Description was changed from

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
==========

to

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
NOTRY=true
==========

[mithro, 2017-02-17 04:40:33]

The CQ bit was checked by tansell@chromium.org

[commit-bot: I haz the power, 2017-02-17 04:41:23]

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, 2017-02-17 05:13:09]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1487306433533720, "parent_rev":
"1d79a857fc39bc1e71c2e1ad523a9426a29fe14a", "commit_rev":
"aac85d78a79fe653d913ad9278c18da28900a25b"}

[commit-bot: I haz the power, 2017-02-17 05:13:56]

Description was changed from

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
NOTRY=true
==========

to

==========
webkit_layout_tests: Add osmesa as data_dep.

osmesa is a "loadable module" object and hence needs to end up in the
data_deps file.

See content/shell:content_shell_crash_test for another example of this.

BUG=693002,524758
NOTRY=true

Review-Url: https://codereview.chromium.org/2702613003
Cr-Commit-Position: refs/heads/master@{#451226}
Committed:
https://chromium.googlesource.com/chromium/src/+/aac85d78a79fe653d913ad9278c1...
==========

[commit-bot: I haz the power, 2017-02-17 05:13:59]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/aac85d78a79fe653d913ad9278c1...

[Dirk Pranke, 2017-02-23 02:59:10]

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

[Dirk Pranke, 2017-02-23 02:59:11]

https://codereview.chromium.org/2702613003/diff/1/BUILD.gn
File BUILD.gn (right):

https://codereview.chromium.org/2702613003/diff/1/BUILD.gn#newcode933
BUILD.gn:933: "//third_party/mesa:osmesa",
This doesn't seem right. Why would the layout tests themselves directly need
this? And isn't this already listed as a data dependency of content_shell?

[mithro, 2017-02-24 00:23:35]

On 2017/02/23 02:59:11, Dirk Pranke wrote:
> https://codereview.chromium.org/2702613003/diff/1/BUILD.gn
> File BUILD.gn (right):
> 
> https://codereview.chromium.org/2702613003/diff/1/BUILD.gn#newcode933
> BUILD.gn:933: "//third_party/mesa:osmesa",
> This doesn't seem right. Why would the layout tests themselves directly need
> this? And isn't this already listed as a data dependency of content_shell?

I don't quite understand /why/ this is needed but it definitely works.
content/shell:content_shell_crash_test is another example of this happening. I
think it might have to do with osmesa being a "loadable module"? Really need
someone who understands GN better than I do to figure out what is going on here.

I've logged https://crbug.com/695691 about this issue.

Tim 'mithro' Ansell