Fix linux_clang default try builder to only compile and not run tests.

Fix linux_clang default try builder to only compile and not run tests.

So when a user does git try -t base_unittests, base_unittests is not run on
linux_clang.

Remove an invalid line added in r125922.

R=thakis@chromium.org
BUG=
TEST=


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=135696

[Nico, 2012-04-30 03:25:43]

Is linux_clang slow? Needs too many resources? I find this useful to check that
some stuff still works after clang rolls.

[M-A Ruel, 2012-04-30 11:06:25]

On 2012/04/30 03:25:43, Nico wrote:
> Is linux_clang slow? Needs too many resources? I find this useful to check
that
> some stuff still works after clang rolls.

I added tests to linux_clang. It didn't have any up to yesterday. This CL is jut
to make sure the behavior is not affected in practice.

[M-A Ruel, 2012-05-01 01:04:55]

On 2012/04/30 11:06:25, Marc-Antoine Ruel wrote:
> On 2012/04/30 03:25:43, Nico wrote:
> > Is linux_clang slow? Needs too many resources? I find this useful to check
> that
> > some stuff still works after clang rolls.
> 
> I added tests to linux_clang. It didn't have any up to yesterday. This CL is
jut
> to make sure the behavior is not affected in practice.

And there is no point in running the tests on both linux and linux_clang.

[tfarina, 2012-05-01 01:32:19]

http://codereview.chromium.org/10263007/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

http://codereview.chromium.org/10263007/diff/1/PRESUBMIT.py#newcode331
PRESUBMIT.py:331: preferred = ['win_rel', 'linux_rel', 'mac_rel']
it would be interesting to have a win_re = '...' for files that end in _win.cc
as well, so we don't trigger linux_rel and mac_rel unnecessarily.

[M-A Ruel, 2012-05-01 01:35:41]

http://codereview.chromium.org/10263007/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

http://codereview.chromium.org/10263007/diff/1/PRESUBMIT.py#newcode331
PRESUBMIT.py:331: preferred = ['win_rel', 'linux_rel', 'mac_rel']
On 2012/05/01 01:32:19, tfarina wrote:
> it would be interesting to have a win_re = '...' for files that end in _win.cc
> as well, so we don't trigger linux_rel and mac_rel unnecessarily.

I'm fine with this idea but this is outside the scope of this CL.

[M-A Ruel, 2012-05-04 20:00:45]

ping

[Nico, 2012-05-04 20:04:51]

On 2012/05/04 20:00:45, Marc-Antoine Ruel wrote:
> ping

"Is linux_clang slow? Needs too many resources?" I maintain this is somewhat
useful for testing clang rolls, so I'm curious why you want to remove it.

[M-A Ruel, 2012-05-04 20:07:15]

On 2012/05/04 20:04:51, Nico wrote:
> On 2012/05/04 20:00:45, Marc-Antoine Ruel wrote:
> > ping
> 
> "Is linux_clang slow? Needs too many resources?" I maintain this is somewhat
> useful for testing clang rolls, so I'm curious why you want to remove it.

I'm not removing it. I'm removing a duplicate line, and only adding it when
there's a C++ file in the CL.

[Nico, 2012-05-04 20:09:06]

On 2012/05/04 20:07:15, Marc-Antoine Ruel wrote:
> On 2012/05/04 20:04:51, Nico wrote:
> > On 2012/05/04 20:00:45, Marc-Antoine Ruel wrote:
> > > ping
> > 
> > "Is linux_clang slow? Needs too many resources?" I maintain this is somewhat
> > useful for testing clang rolls, so I'm curious why you want to remove it.
> 
> I'm not removing it. I'm removing a duplicate line, and only adding it when
> there's a C++ file in the CL.

Right, and clang rolls contain no c++ files.

[M-A Ruel, 2012-05-04 20:13:17]

On 2012/05/04 20:09:06, Nico wrote:
> Right, and clang rolls contain no c++ files.

ok, fixed.

[Nico, 2012-05-04 20:20:30]

On 2012/05/04 20:13:17, Marc-Antoine Ruel wrote:
> On 2012/05/04 20:09:06, Nico wrote:
> > Right, and clang rolls contain no c++ files.
> 
> ok, fixed.

Still doesn't run tests, does it?

[M-A Ruel, 2012-05-04 20:23:18]

On 2012/05/04 20:20:30, Nico wrote:
> On 2012/05/04 20:13:17, Marc-Antoine Ruel wrote:
> > On 2012/05/04 20:09:06, Nico wrote:
> > > Right, and clang rolls contain no c++ files.
> > 
> > ok, fixed.
> 
> Still doesn't run tests, does it?

No, it's not necessary either.

[M-A Ruel, 2012-05-05 00:55:29]

On 2012/05/04 20:23:18, Marc-Antoine Ruel wrote:
> On 2012/05/04 20:20:30, Nico wrote:
> > On 2012/05/04 20:13:17, Marc-Antoine Ruel wrote:
> > > On 2012/05/04 20:09:06, Nico wrote:
> > > > Right, and clang rolls contain no c++ files.
> > > 
> > > ok, fixed.
> > 
> > Still doesn't run tests, does it?
> 
> No, it's not necessary either.

I mean, linux_clang didn't run test at all. Now it has tests registered but they
are not run by default. I just want to add that so that -t foo_tests doesn't
have any unexpected behavior.

[M-A Ruel, 2012-05-07 17:30:37]

Still unhappy about the CL?

[Nico, 2012-05-07 17:42:19]

lgtm

[commit-bot: I haz the power, 2012-05-07 17:43:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/maruel@chromium.org/10263007/11001

[commit-bot: I haz the power, 2012-05-07 19:23:15]

Change committed as 135696