GN: Build NaCl tests for x64 too on x86 Windows

GN: Build NaCl tests for x64 too on x86 Windows

x86 Windows builds of Chrome run on both x86 Windows and x64 Windows.
On x64 Windows, only x64 NaCl is supported, so those tests are needed too.

BUG=609684
R=bbudge@chromium.org, dpranke@chromium.org, phosek@chromium.org

Committed: https://crrev.com/a03f119d7af32b119fb755abe3d6a661b55548e8
Cr-Commit-Position: refs/heads/master@{#393433}

[Roland McGrath, 2016-05-12 21:49:16]

PTAL

AIUI this needs to have https://codereview.chromium.org/1908733002/ folded in to
be able to do a trybot run that will actually test the fix.

Once people's eyeballs are happy with this, either I will fold that into this to
do the try run or Dirk will fold this into that to do the try run.

[bbudge, 2016-05-12 21:59:28]

Rubber stamp lgtm

[Petr Hosek, 2016-05-12 22:08:26]

lgtm

https://codereview.chromium.org/1976753002/diff/1/build/config/nacl/rules.gni
File build/config/nacl/rules.gni (right):

https://codereview.chromium.org/1976753002/diff/1/build/config/nacl/rules.gni...
build/config/nacl/rules.gni:100: "--library-path=" + rebase_path(root_out_dir),
Was this a bug?

https://codereview.chromium.org/1976753002/diff/1/ppapi/native_client/nacl_te...
File ppapi/native_client/nacl_test_data.gni (right):

https://codereview.chromium.org/1976753002/diff/1/ppapi/native_client/nacl_te...
ppapi/native_client/nacl_test_data.gni:182:
template("nacl_test_data_pretranslate_pexe") {
Since translate_cpu is a required argument, we should assert here that it's set.

You could also forward translate_cpu from the invoker to avoid having to use the
"invoker.translate_cpu" to make the code a little more readable.

[Dirk Pranke, 2016-05-12 22:12:13]

The CQ bit was checked by dpranke@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-05-12 22:12:24]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1976753002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1976753002/1

[Roland McGrath, 2016-05-12 22:29:59]

https://codereview.chromium.org/1976753002/diff/1/build/config/nacl/rules.gni
File build/config/nacl/rules.gni (right):

https://codereview.chromium.org/1976753002/diff/1/build/config/nacl/rules.gni...
build/config/nacl/rules.gni:100: "--library-path=" + rebase_path(root_out_dir),
On 2016/05/12 22:08:26, Petr Hosek wrote:
> Was this a bug?

Previously this was done above, before this 'if' block.
I made the x86/x64 case always do both unconditionally since that seemed
clearest, so the one before the 'if' would have become redundant.  Hence I moved
it into the 'else' clause where it's still needed, and rolled it into the
existing += statement.

https://codereview.chromium.org/1976753002/diff/1/ppapi/native_client/nacl_te...
File ppapi/native_client/nacl_test_data.gni (right):

https://codereview.chromium.org/1976753002/diff/1/ppapi/native_client/nacl_te...
ppapi/native_client/nacl_test_data.gni:182:
template("nacl_test_data_pretranslate_pexe") {
On 2016/05/12 22:08:26, Petr Hosek wrote:
> Since translate_cpu is a required argument, we should assert here that it's
set.
> 
> You could also forward translate_cpu from the invoker to avoid having to use
the
> "invoker.translate_cpu" to make the code a little more readable.

I thought about that but didn't bother.  It might make the uglification imposed
by 'gn format' less egregious, and the assignment can act as the assert too, so
I'll do that.

[Dirk Pranke, 2016-05-13 01:26:36]

lgtm. 

Successful builds for x86 (debug and release) and x64 can be seen in
https://codereview.chromium.org/1908733002/#ps180001 (the red
win_chromium_x64_rel_ng try job is due to an unrelated content_browsertests
failure).

[Dirk Pranke, 2016-05-13 01:26:40]

The CQ bit was checked by dpranke@chromium.org

[Dirk Pranke, 2016-05-13 01:26:41]

The patchset sent to the CQ was uploaded after l-g-t-m from bbudge@chromium.org,
phosek@chromium.org
Link to the patchset: https://codereview.chromium.org/1976753002/#ps40001
(title: "don't list x86-32 lib dirs when target_cpu==x64")

[commit-bot: I haz the power, 2016-05-13 01:26:58]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1976753002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1976753002/40001

[commit-bot: I haz the power, 2016-05-13 02:23:05]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-05-13 02:24:34]

Description was changed from

==========
GN: Build NaCl tests for x64 too on x86 Windows

x86 Windows builds of Chrome run on both x86 Windows and x64 Windows.
On x64 Windows, only x64 NaCl is supported, so those tests are needed too.

BUG=609684
R=bbudge@chromium.org, dpranke@chromium.org, phosek@chromium.org
==========

to

==========
GN: Build NaCl tests for x64 too on x86 Windows

x86 Windows builds of Chrome run on both x86 Windows and x64 Windows.
On x64 Windows, only x64 NaCl is supported, so those tests are needed too.

BUG=609684
R=bbudge@chromium.org, dpranke@chromium.org, phosek@chromium.org

Committed: https://crrev.com/a03f119d7af32b119fb755abe3d6a661b55548e8
Cr-Commit-Position: refs/heads/master@{#393433}
==========

[commit-bot: I haz the power, 2016-05-13 02:24:36]

Patchset 3 (id:??) landed as
https://crrev.com/a03f119d7af32b119fb755abe3d6a661b55548e8
Cr-Commit-Position: refs/heads/master@{#393433}