remove unneeded xdisplaycheck dep in base_unittests_run

remove unneeded xdisplaycheck dep in base_unittests_run

This previous CL added an xdisplaycheck dep to base_unittests_run, but
base_unittests does not appear to depend on any X libraries as far as I
can tell:
https://codereview.chromium.org/388843004

Let's see if we can remove this dependency.

Committed: https://crrev.com/1d34e9d4dbaad8558164d3b144c42d60c95b515b
Cr-Commit-Position: refs/heads/master@{#333566}

[Mostyn Bramley-Moore, 2015-06-09 17:31:16]

mostynb@opera.com changed reviewers:
+ maruel@chromium.org, petermayo@chromium.org

[Mostyn Bramley-Moore, 2015-06-09 17:31:17]

@maruel, petermayo: it looks to me as though base_unittests has no X library
dependencies.  I'm not sure why xvfb+xdisplaycheck deps were added to
base_unittests_run (https://codereview.chromium.org/388843004) but in my testing
it doesn't seem to be needed anymore.

[Peter Mayo, 2015-06-09 18:30:28]

On 2015/06/09 17:31:17, Mostyn Bramley-Moore wrote:
> @maruel, petermayo: it looks to me as though base_unittests has no X library
> dependencies.  I'm not sure why xvfb+xdisplaycheck deps were added to
> base_unittests_run (https://codereview.chromium.org/388843004) but in my
testing
> it doesn't seem to be needed anymore.

Does your testing include running the unittest with no display?  If that work on
linux then I think it should be fine.

I am pretty sure I was reacting to a situation where there were ordering
dependencies of running the tests and there were unexpected failures
parallelizing them.  The failure rates even with the dependencies missing were
really small, but noticeable.

See correlated contemporaneous work here:
https://codereview.chromium.org/385073006/
https://codereview.chromium.org/387143004/

[Mostyn Bramley-Moore, 2015-06-09 19:20:24]

> Does your testing include running the unittest with no display?  If that work
on
> linux then I think it should be fine.

I built the base_unittests_run target with this CL, while ssh'ed into a linux
machine running X but without DISPLAY set in my terminal, and it succeeded.  I
also verified that xdisplaycheck was not built, and did not exist in the
out/Release/ directory.  

I also checked the base_unittests binary with ldd, and it has no (obvious) X
library dependencies:
$ ldd out/Release/base_unittests | awk '{ print $1 }'
linux-vdso.so.1
libicuuc.so
libbase_prefs.so
libicui18n.so
libbase.so
libmalloc_wrapper.so
libbase_i18n.so
libglib-2.0.so.0
libstdc++.so.6
libm.so.6
libpthread.so.0
libc.so.6
librt.so.1
libdl.so.2
libpcre.so.3
/lib64/ld-linux-x86-64.so.2
libgcc_s.so.1

> I am pretty sure I was reacting to a situation where there were ordering
> dependencies of running the tests and there were unexpected failures
> parallelizing them.  The failure rates even with the dependencies missing were
> really small, but noticeable.
> 
> See correlated contemporaneous work here:
> https://codereview.chromium.org/385073006/
> https://codereview.chromium.org/387143004/

I tried building the parent of the older of these two commits, and
base_unittests didn't have any X library dependencies either, and ran OK when
ssh'ed into my setup mentioned above (and without xdisplaycheck built).

Perhaps you had confused ui_base_unittests (which does depend on X libs) with
base_unittests?

[M-A Ruel, 2015-06-09 19:23:59]

I'm pretty sure base_unittests *used* to depend on X. It's fairly probable that
all UI related code was migrated to ui/base, which would totally make sense.

lgtm

[Mostyn Bramley-Moore, 2015-06-09 19:34:16]

On 2015/06/09 19:23:59, M-A Ruel wrote:
> I'm pretty sure base_unittests *used* to depend on X. It's fairly probable
that
> all UI related code was migrated to ui/base, which would totally make sense.

Looks like this CL from april 2014 removed the X11 dependency from base:
https://codereview.chromium.org/237403002

Queuing...

[Mostyn Bramley-Moore, 2015-06-09 19:34:22]

The CQ bit was checked by mostynb@opera.com

[commit-bot: I haz the power, 2015-06-09 19:35:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1171073002/1

[commit-bot: I haz the power, 2015-06-09 19:41:54]

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

[commit-bot: I haz the power, 2015-06-09 19:41:55]

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

[Mostyn Bramley-Moore, 2015-06-09 19:44:56]

mostynb@opera.com changed reviewers:
+ thakis@chromium.org

[Mostyn Bramley-Moore, 2015-06-09 19:45:34]

+Nico for base/OWNERS approval.

[Nico, 2015-06-09 19:46:15]

lgtm

[Mostyn Bramley-Moore, 2015-06-09 19:46:35]

The CQ bit was checked by mostynb@opera.com

[commit-bot: I haz the power, 2015-06-09 19:46:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1171073002/1

[commit-bot: I haz the power, 2015-06-09 20:40:38]

Committed patchset #1 (id:1)

[Peter Mayo, 2015-06-09 20:41:08]

Thanks for fixing mys over-exuberance.

LGTM

[commit-bot: I haz the power, 2015-06-09 20:41:34]

Patchset 1 (id:??) landed as
https://crrev.com/1d34e9d4dbaad8558164d3b144c42d60c95b515b
Cr-Commit-Position: refs/heads/master@{#333566}