Rework how the gn migration targets are included from all.gyp.

Rework how the gn migration targets are included from all.gyp.

This change splits out the targets being used to track the gn
migration from  all.gyp into a separate, new file
(gn_migration.gypi), and reworks their structure to

  1) better match the gn_all target in src/BUILD.gn,
  2) identify the targets built by 'all' in gyp that
     are not yet explicitly listed in the gn_all target
     (the new 'add_to_gn_all' target),
  3) identify the targets left to be converted.

The 'add_to_gn_all' targets will simply be moved to gn_all in
a follow-up CL.

Right now these targets will be linux-only, but follow-up CLs
will expand them to the other configurations.

This CL should produce no actual changes in the builds.

R=brettw@chromium.org
BUG=461019
CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:android_chromium_gn_compile_dbg,android_chromium_gn_compile_rel;tryserver.chromium.win:win8_chromium_gn_rel,win8_chromium_gn_dbg;tryserver.chromium.mac:mac_chromium_gn_rel,mac_chromium_gn_dbg

Committed: https://crrev.com/91cb1747a84d80acc9fc5045c971986f1799e02b
Cr-Commit-Position: refs/heads/master@{#318599}

[brettw, 2015-02-27 21:48:26]

lgtm

https://codereview.chromium.org/962343002/diff/1/build/gn_migration.gypi
File build/gn_migration.gypi (right):

https://codereview.chromium.org/962343002/diff/1/build/gn_migration.gypi#newc...
build/gn_migration.gypi:498: ]
indenting.

[Dirk Pranke, 2015-02-27 22:14:41]

https://codereview.chromium.org/962343002/diff/1/build/gn_migration.gypi
File build/gn_migration.gypi (right):

https://codereview.chromium.org/962343002/diff/1/build/gn_migration.gypi#newc...
build/gn_migration.gypi:498: ]
On 2015/02/27 21:48:26, brettw wrote:
> indenting.

Acknowledged.

[Dirk Pranke, 2015-02-28 00:33:24]

Note that this patch depends on https://codereview.chromium.org/967683003/
landing first, and getting brett's changes to chromedriver, remoting, and ppapi
landed first (merged into this patch if needed).

[Dirk Pranke, 2015-02-28 03:16:46]

The CQ bit was checked by dpranke@chromium.org

[Dirk Pranke, 2015-02-28 03:16:51]

The CQ bit was unchecked by dpranke@chromium.org

[Dirk Pranke, 2015-02-28 03:21:09]

Patchset #2 (id:20001) has been deleted

[Dirk Pranke, 2015-02-28 03:27:10]

The CQ bit was checked by dpranke@chromium.org

[Dirk Pranke, 2015-02-28 03:27:10]

The patchset sent to the CQ was uploaded after l-g-t-m from brettw@chromium.org
Link to the patchset: https://codereview.chromium.org/962343002/#ps40001 (title:
"merge to #318587")

[commit-bot: I haz the power, 2015-02-28 03:27:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/962343002/40001

[commit-bot: I haz the power, 2015-02-28 05:01:44]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-02-28 05:02:13]

Patchset 2 (id:??) landed as
https://crrev.com/91cb1747a84d80acc9fc5045c971986f1799e02b
Cr-Commit-Position: refs/heads/master@{#318599}

[gunsch, 2015-03-02 00:23:00]

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

[gunsch, 2015-03-02 00:23:00]

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi
File build/gn_migration.gypi (right):

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:370:
'../components/components.gyp:session_manager_component',
This target isn't defined for all Linux builds:

https://code.google.com/p/chromium/codesearch#chromium/src/components/compone...

[gunsch, 2015-03-02 00:31:03]

Found a handful more, even after the first.

Then something is still including ash.gyp:ash (not defined for Chromecast).

We have a pair of trybots up, we're still trying to get them to advisory and
then blocking, but they can be run with:

git cl try -m tryserver.chromium.linux -b cast_shell -b cast_shell_apk

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi
File build/gn_migration.gypi (right):

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:386: '../gpu/tools/tools.gyp:compositor_model_bench',
this is also not found in Chromecast build

https://code.google.com/p/chromium/codesearch#chromium/src/gpu/tools/tools.gy...

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:394: '../media/media.gyp:player_x11',
neither is this

https://code.google.com/p/chromium/codesearch#chromium/src/media/media.gyp&q=...

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:404: '../remoting/remoting.gyp:remoting_host',
neither is this

https://code.google.com/p/chromium/codesearch#chromium/src/remoting/remoting_...

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:405:
'../remoting/remoting.gyp:remoting_it2me_native_messaging_host',
or this

https://code.google.com/p/chromium/codesearch#chromium/src/remoting/remoting_...

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:407: '../remoting/remoting.gyp:remoting_me2me_host',
or the next two deps:
https://code.google.com/p/chromium/codesearch#chromium/src/remoting/remoting_...

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:409:
'../remoting/remoting.gyp:remoting_native_messaging_manifests',
or this, remoting_perftests, and remoting_start_host (all guarded with
use_remoting_host)

https://codereview.chromium.org/962343002/diff/40001/build/gn_migration.gypi#...
build/gn_migration.gypi:429: '../ui/app_list/app_list.gyp:app_list_demo',
or this (toolkit_views==1):

https://code.google.com/p/chromium/codesearch#chromium/src/ui/app_list/app_li...

[gunsch, 2015-03-02 00:33:52]

https://codereview.chromium.org/962343002/diff/40001/build/all.gyp
File build/all.gyp (right):

https://codereview.chromium.org/962343002/diff/40001/build/all.gyp#newcode471
build/all.gyp:471: ['OS=="linux" and chromeos==0', {
Alternately, you could put "chromecast==0" here and probably just call it a day
so we don't step on each other's toes while you're working on this :)

[Dirk Pranke, 2015-03-02 01:31:53]

On 2015/03/02 00:33:52, gunsch wrote:
> https://codereview.chromium.org/962343002/diff/40001/build/all.gyp
> File build/all.gyp (right):
> 
> https://codereview.chromium.org/962343002/diff/40001/build/all.gyp#newcode471
> build/all.gyp:471: ['OS=="linux" and chromeos==0', {
> Alternately, you could put "chromecast==0" here and probably just call it a
day
> so we don't step on each other's toes while you're working on this :)

Well, I actually ended up reverting this change for unrelated reasons, but all
this is good feedback.

This patch was only intended to work on the normal linux release build, but it's
been
 interesting to see which things don't quite meet that criteria and aren't on
the waterfall.

I thought chromecast was a kind of chromeos build (and so chromeos==0 would
exclude
it); I guess that's not the case?

Is chromecast using ozone? I'll take a look at the trybots you linked to and see
what all flags are being set ...

[gunsch, 2015-03-02 05:21:38]

On 2015/03/02 01:31:53, Dirk Pranke wrote:
> On 2015/03/02 00:33:52, gunsch wrote:
> > https://codereview.chromium.org/962343002/diff/40001/build/all.gyp
> > File build/all.gyp (right):
> > 
> >
https://codereview.chromium.org/962343002/diff/40001/build/all.gyp#newcode471
> > build/all.gyp:471: ['OS=="linux" and chromeos==0', {
> > Alternately, you could put "chromecast==0" here and probably just call it a
> day
> > so we don't step on each other's toes while you're working on this :)
> 
> Well, I actually ended up reverting this change for unrelated reasons, but all
> this is good feedback.
> 
> This patch was only intended to work on the normal linux release build, but
it's
> been
>  interesting to see which things don't quite meet that criteria and aren't on
> the waterfall.
> 
> I thought chromecast was a kind of chromeos build (and so chromeos==0 would
> exclude
> it); I guess that's not the case?
> 
> Is chromecast using ozone? I'll take a look at the trybots you linked to and
see
> what all flags are being set ...

Couple quick comments:

There are two top-level Chromecast builds: cast_shell on Linux (what runs on the
Chromecast), and cast_shell_apk on Android (for Android TV). The "cast_shell"
you can build from a public Chromium checkout is analogous to Clank's
"content_shell_apk", in the sense you can build a single-page embedder with
similar behavior, but we also have a private repo that tacks on much of the
Chromecast functionality. We're also still partly building from a downstream
fork, but are getting very close to building using [Chromium TOT + our internal
repo] rather than only [downstream fork of Chromium + our internal repo].

Chromecast is an embedded Linux system, no ChromeOS.

The cast_shell Linux build sets the GYP variable "embedded=1" (for which, last I
checked, no GN equivalent has been added), which:
* enables Ozone/Aura
* disables toolkit_views, dbus, udev, printing, plugins, remoting, etc.

The cast_shell_apk Chromecast build also sets GYP variable chromecast==1, but is
generally pretty close to the content_shell_apk build with Chromecast's
embedder.

Thanks for taking a look.