Add BUILD.gn to third_party/gRPC

Add BUILD.gn to third_party/gRPC

BUG=618889
R=gcasto@chromium.org

Committed: https://chromium.googlesource.com/external/github.com/grpc/grpc/+/26847f3fdb23459d8fbbc5fe0c054d55af0862b5

[xyzzyz, 2016-06-21 00:02:47]

Description was changed from

==========
Add BUILD.gn to third_party/gRPC

BUG=618889
==========

to

==========
Add BUILD.gn to third_party/gRPC

BUG=618889
==========

[xyzzyz, 2016-06-21 00:02:48]

xyzzyz@chromium.org changed reviewers:
+ gcasto@chromium.org

[xyzzyz, 2016-06-21 00:02:48]

xyzzyz@chromium.org changed required reviewers:
+ gcasto@chromium.org

[Garrett Casto, 2016-06-21 23:18:34]

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn
File BUILD.gn (right):

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn#newcode1
BUILD.gn:1: config("grpc_config") {
From our discussion, it seems like the general setup of this file comes from the
upstream BUILD file. Can you put a comment here pointing that out? It might make
it clearer why you have a component named gpr, why there is a multiple
intermediate targets even though Chromium code will only depend on the final
target, etc.

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn#newcode11
BUILD.gn:11: # Therefore, we need to make gRPC not use standard library
threading support.
Does this change the behavior of gRPC in any way? Also, can we file a bug for
this so that when we can know when to drop this?

[xyzzyz, 2016-06-21 23:29:22]

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn
File BUILD.gn (right):

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn#newcode1
BUILD.gn:1: config("grpc_config") {
On 2016/06/21 23:18:33, Garrett Casto wrote:
> From our discussion, it seems like the general setup of this file comes from
the
> upstream BUILD file. Can you put a comment here pointing that out? It might
make
> it clearer why you have a component named gpr, why there is a multiple
> intermediate targets even though Chromium code will only depend on the final
> target, etc.

Done.

https://codereview.chromium.org/2087693002/diff/20001/BUILD.gn#newcode11
BUILD.gn:11: # Therefore, we need to make gRPC not use standard library
threading support.
On 2016/06/21 23:18:33, Garrett Casto wrote:
> Does this change the behavior of gRPC in any way? Also, can we file a bug for
> this so that when we can know when to drop this?

As far as I know, it doesn't (or at least it shouldn't) -- if condition_variable
is not available, it implements its own using pthreads primitives.

I think this is a good bug to follow wrt this: https://crbug.com/593874. Added
as a comment.

[Garrett Casto, 2016-06-21 23:58:22]

lgtm

[xyzzyz, 2016-06-22 20:50:06]

Description was changed from

==========
Add BUILD.gn to third_party/gRPC

BUG=618889
==========

to

==========
Add BUILD.gn to third_party/gRPC

BUG=618889
R=gcasto@chromium.org

Committed:
https://chromium.googlesource.com/external/github.com/grpc/grpc/+/26847f3fdb2...
==========

[xyzzyz, 2016-06-22 20:50:07]

Committed patchset #3 (id:40001) manually as
26847f3fdb23459d8fbbc5fe0c054d55af0862b5 (presubmit successful).