win: Decide how to include system dirs at toolchain time instead of globally.

win: Decide how to include system dirs at toolchain time instead of globally.

Before this patch, we used the global is_clang setting to decide if
we include system libraries via /I or -imsvc in the clang toolchains.
Now, always use -imsvc in the clang toolchains.

The motivation is that we (maybe) want to build libyuv with a clang
toolchain even when is_clang is globally set to false.

BUG=692600
Review-Url: https://codereview.chromium.org/2728233002
Cr-Commit-Position: refs/heads/master@{#454704}
Committed: https://chromium.googlesource.com/chromium/src/+/cbdbc2fd19dcde3cf06296bb9bad8f41978214b5

[Nico, 2017-03-03 22:15:47]

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

[commit-bot: I haz the power, 2017-03-03 22:16:05]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Nico, 2017-03-03 22:17:47]

thakis@chromium.org changed reviewers:
+ scottmg@chromium.org

[Nico, 2017-03-03 22:17:48]

I could imagine that other things might go wrong when using clang for libyuv
without the global is_clang set, but at least locally libyuv builds fine with
just this change. We'll see how annoying maintaining this will be over time.

https://codereview.chromium.org/2728233002/diff/1/build/toolchain/win/setup_t...
File build/toolchain/win/setup_toolchain.py (right):

https://codereview.chromium.org/2728233002/diff/1/build/toolchain/win/setup_t...
build/toolchain/win/setup_toolchain.py:204: print 'include_flags_I = ' +
gn_helpers.ToGNString(include_I)
this is currently not read anywhere, but if the MS branch where to use this flag
instead of the env subprocess, this is how we'd do it. seemed a bit nicer to
output this too, but I can drop include_I if you think YAGNI is better. (I
usually do think that.)

[scottmg, 2017-03-03 22:33:40]

lgtm

[Nico, 2017-03-03 22:37:17]

The CQ bit was unchecked by thakis@chromium.org

[Nico, 2017-03-03 22:37:22]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2017-03-03 22:37:30]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-03 23:14:40]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1488580642362980, "parent_rev":
"0471e6079c754e7478a61ac81765cec789f5249a", "commit_rev":
"cbdbc2fd19dcde3cf06296bb9bad8f41978214b5"}

[commit-bot: I haz the power, 2017-03-03 23:15:54]

Description was changed from

==========
win: Decide how to include system dirs at toolchain time instead of globally.

Before this patch, we used the global is_clang setting to decide if
we include system libraries via /I or -imsvc in the clang toolchains.
Now, always use -imsvc in the clang toolchains.

The motivation is that we (maybe) want to build libyuv with a clang
toolchain even when is_clang is globally set to false.

BUG=692600
==========

to

==========
win: Decide how to include system dirs at toolchain time instead of globally.

Before this patch, we used the global is_clang setting to decide if
we include system libraries via /I or -imsvc in the clang toolchains.
Now, always use -imsvc in the clang toolchains.

The motivation is that we (maybe) want to build libyuv with a clang
toolchain even when is_clang is globally set to false.

BUG=692600

Review-Url: https://codereview.chromium.org/2728233002
Cr-Commit-Position: refs/heads/master@{#454704}
Committed:
https://chromium.googlesource.com/chromium/src/+/cbdbc2fd19dcde3cf06296bb9bad...
==========

[commit-bot: I haz the power, 2017-03-03 23:15:54]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/cbdbc2fd19dcde3cf06296bb9bad...