Adding in missing tests to make chromeos build on waterfall go green....

Adding in missing tests to make chromeos build on waterfall go green.

BUG=None
TEST=None
TBR=jiesun


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=40231

[jiesun, 2010-03-01 16:37:53]

my thoughts

http://codereview.chromium.org/660248/diff/1/2
File build/all.gyp (right):

http://codereview.chromium.org/660248/diff/1/2#newcode260
build/all.gyp:260: 'temp_gyp/googleurl.gyp:googleurl_unittests',
the question are
1. Were these targets built in the past when chromeos==1? I am not sure if all
the unittests is required, but maybe some of them are helpful.
2. Were there targets built in the past when chromeos==0? Did they need these
for regression tests?
Originally I will plan to make chromeos_builder the only EXTRA target that
build-bot will build when chromeos==1 is specified.

[bradn, 2010-03-03 05:09:53]

On Mon, Mar 1, 2010 at 8:37 AM, <jiesun@chromium.org> wrote:

> my thoughts
>
>
> http://codereview.chromium.org/660248/diff/1/2
> File build/all.gyp (right):
>
> http://codereview.chromium.org/660248/diff/1/2#newcode260
> build/all.gyp:260: 'temp_gyp/googleurl.gyp:googleurl_unittests',
> the question are
> 1. Were these targets built in the past when chromeos==1? I am not sure
> if all the unittests is required, but maybe some of them are helpful.
>
This should be similar to the set of projects build before in this case.


> 2. Were there targets built in the past when chromeos==0? Did they need
> these for regression tests?
>
There were a bunch of targets called out specifically, not sure why, they
were listed manually.


> Originally I will plan to make chromeos_builder the only EXTRA target
> that build-bot will build when chromeos==1 is specified.
>
> Having chromeos_builder list all the targets is more consistent with how
the other builders are handled. That should save us having to make another
change to the buildbot configs if you ever wanted to remove something.
We could add a dependency on chromium_dbg_builder or the rel builder, but
neither one exactly matched the existing set.


> http://codereview.chromium.org/660248
>

[jiesun, 2010-03-03 16:21:59]

LGTM