Move gn _run target generation into test(). (RELAND)

Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

Consolidating the _run targets into the test template should make them
easier to maintain (and, eventually, remove).

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/3c0ef4309a77bbf172f081c99bbb33e732ea07af
Cr-Commit-Position: refs/heads/master@{#373385}

[jbudorick, 2016-02-03 19:50:18]

Description was changed from

==========
Move gn _run target generation into test(). (RELAND)

BUG=583694
==========

to

==========
Move gn _run target generation into test(). (RELAND)

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

[jbudorick, 2016-02-03 19:52:53]

Description was changed from

==========
Move gn _run target generation into test(). (RELAND)

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

to

==========
Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

[jbudorick, 2016-02-03 20:05:27]

jbudorick@chromium.org changed reviewers:
+ brettw@chromium.org, dpranke@chromium.org, kbr@chromium.org

[jbudorick, 2016-02-03 20:05:28]

Not entirely sure what happened here, but my best guess is that gn was fine with
having both //gpu:angle_end2end_tests_run and
//third_party/angle/src/tests/angle_end2end_tests_run, but because there were
two it couldn't generate an authoritative angle_end2end_tests_run target.

i.e. before:

$ grep "^build.*angle_end2end_tests_run" ./build.ninja
build gpu$:angle_end2end_tests_run: phony obj/gpu/angle_end2end_tests_run.stamp
build third_party/angle/src/tests$:angle_end2end_tests_run: phony
obj/third_party/angle/src/tests/angle_end2end_tests_run.stamp

after:

$ grep "^build.*angle_end2end_tests_run" ./build.ninja
build third_party/angle/src/tests$:angle_end2end_tests_run: phony
obj/third_party/angle/src/tests/angle_end2end_tests_run.stamp
build angle_end2end_tests_run: phony
obj/third_party/angle/src/tests/angle_end2end_tests_run.stamp

[Ken Russell (switch to Gerrit), 2016-02-03 21:14:08]

gpu/ LGTM but I'm not an owner.

[brettw, 2016-02-03 21:35:55]

On 2016/02/03 20:05:28, jbudorick wrote:
> Not entirely sure what happened here, but my best guess is that gn was fine
with
> having both //gpu:angle_end2end_tests_run and
> //third_party/angle/src/tests/angle_end2end_tests_run, but because there were
> two it couldn't generate an authoritative angle_end2end_tests_run target.

Yup, that's what it will do.

LGTM

[Dirk Pranke, 2016-02-03 22:24:42]

On 2016/02/03 21:35:55, brettw wrote:
> On 2016/02/03 20:05:28, jbudorick wrote:
> > Not entirely sure what happened here, but my best guess is that gn was fine
> with
> > having both //gpu:angle_end2end_tests_run and
> > //third_party/angle/src/tests/angle_end2end_tests_run, but because there
were
> > two it couldn't generate an authoritative angle_end2end_tests_run target.
> 
> Yup, that's what it will do.
> 
> LGTM

Right, if there were two angle_end2end_tests_run targets defined, GN will not
generate the top-level ninja target with that name (because it will be 
ambiguous) and you have to refer to the full label names instead.

So, LGTM to me as well but you should coordinate w/ cwallez@ for his patch to
angle.

[Ken Russell (switch to Gerrit), 2016-02-03 22:31:41]

On 2016/02/03 22:24:42, Dirk Pranke wrote:
> On 2016/02/03 21:35:55, brettw wrote:
> > On 2016/02/03 20:05:28, jbudorick wrote:
> > > Not entirely sure what happened here, but my best guess is that gn was
fine
> > with
> > > having both //gpu:angle_end2end_tests_run and
> > > //third_party/angle/src/tests/angle_end2end_tests_run, but because there
> were
> > > two it couldn't generate an authoritative angle_end2end_tests_run target.
> > 
> > Yup, that's what it will do.
> > 
> > LGTM
> 
> Right, if there were two angle_end2end_tests_run targets defined, GN will not
> generate the top-level ninja target with that name (because it will be 
> ambiguous) and you have to refer to the full label names instead.
> 
> So, LGTM to me as well but you should coordinate w/ cwallez@ for his patch to
> angle.

I'm quite sure cwallez@ will be happy to have this land ASAP. Let's go forward.
Thanks.

[jbudorick, 2016-02-03 22:35:17]

Description was changed from

==========
Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

to

==========
Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

Consolidating the _run targets into the test template should make them
easier to maintain (and, eventually, remove).

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

[jbudorick, 2016-02-03 22:36:30]

On 2016/02/03 22:31:41, Ken Russell wrote:
> On 2016/02/03 22:24:42, Dirk Pranke wrote:
> > On 2016/02/03 21:35:55, brettw wrote:
> > > On 2016/02/03 20:05:28, jbudorick wrote:
> > > > Not entirely sure what happened here, but my best guess is that gn was
> fine
> > > with
> > > > having both //gpu:angle_end2end_tests_run and
> > > > //third_party/angle/src/tests/angle_end2end_tests_run, but because there
> > were
> > > > two it couldn't generate an authoritative angle_end2end_tests_run
target.
> > > 
> > > Yup, that's what it will do.
> > > 
> > > LGTM
> > 
> > Right, if there were two angle_end2end_tests_run targets defined, GN will
not
> > generate the top-level ninja target with that name (because it will be 
> > ambiguous) and you have to refer to the full label names instead.
> > 
> > So, LGTM to me as well but you should coordinate w/ cwallez@ for his patch
to
> > angle.
> 
> I'm quite sure cwallez@ will be happy to have this land ASAP. Let's go
forward.
> Thanks.

Discussed with cwallez@ out-of-band.

[jbudorick, 2016-02-03 22:36:35]

The CQ bit was checked by jbudorick@chromium.org

[commit-bot: I haz the power, 2016-02-03 22:36:47]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1662053002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1662053002/20001

[commit-bot: I haz the power, 2016-02-03 23:40:27]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-03 23:41:09]

Description was changed from

==========
Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

Consolidating the _run targets into the test template should make them
easier to maintain (and, eventually, remove).

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

to

==========
Move gn _run target generation into test(). (RELAND)

This is a reland of https://codereview.chromium.org/1653003006/

Consolidating the _run targets into the test template should make them
easier to maintain (and, eventually, remove).

BUG=583694
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/3c0ef4309a77bbf172f081c99bbb33e732ea07af
Cr-Commit-Position: refs/heads/master@{#373385}
==========

[commit-bot: I haz the power, 2016-02-03 23:41:11]

Patchset 2 (id:??) landed as
https://crrev.com/3c0ef4309a77bbf172f081c99bbb33e732ea07af
Cr-Commit-Position: refs/heads/master@{#373385}

[maste, 2017-03-16 19:49:26]

t.p.taeza@gmail.com changed reviewers:
+ t.p.taeza@gmail.com