Build crash_service_win64 with clang-cl when is_clang=true

Build crash_service_win64 with clang-cl when is_clang=true

Otherwise this gn assertion fails:
  ERROR at //build/config/sanitizers/sanitizers.gni:85:1: Assertion failed.
  assert(!using_sanitizer || is_clang,
  ^-----
  Sanitizers (is_*san) require setting is_clang = true in 'gn args'

R=dpranke@chromium.org,scottmg@chromium.org
BUG=598761
Committed: https://crrev.com/9065ac8a2b45154c8e5dbef97bddd01829680f88
Cr-Commit-Position: refs/heads/master@{#383808}

[Dirk Pranke, 2016-03-29 18:56:08]

you'll still need owners approval from Scott ...

https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
File chrome/tools/crash_service/BUILD.gn (right):

https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
chrome/tools/crash_service/BUILD.gn:33:
":crash_service(//build/toolchain/win:${maybe_clang}x64)"
lgtm. 

There's a related issue in that is_clang is a toolchain-scoped variable (i.e.,
the value may change per toolchain) and so if this target got pulled in via
multiple toolchains we might end up declaring it multiple times. 

However, oddly, it looks like this target is really only referenced via the
chromium_builder_perf and telemetry_chrome_test targets, which can't be right,
so I'll try to clean this stuff up in a different follow-on CL.

[scottmg, 2016-03-29 19:03:23]

On 2016/03/29 18:56:08, Dirk Pranke wrote:
> you'll still need owners approval from Scott ...
> 
>
https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
> File chrome/tools/crash_service/BUILD.gn (right):
> 
>
https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
> chrome/tools/crash_service/BUILD.gn:33:
> ":crash_service(//build/toolchain/win:${maybe_clang}x64)"
> lgtm. 
> 
> There's a related issue in that is_clang is a toolchain-scoped variable (i.e.,
> the value may change per toolchain) and so if this target got pulled in via
> multiple toolchains we might end up declaring it multiple times. 
> 
> However, oddly, it looks like this target is really only referenced via the
> chromium_builder_perf and telemetry_chrome_test targets, which can't be right,
> so I'll try to clean this stuff up in a different follow-on CL.

I think that might be right. I don't really know what crash_service is, so it's
probably something to do with Breakpad, which probably means it's only used in
content_shell or some non-Chrome case.

[scottmg, 2016-03-29 19:04:37]

On 2016/03/29 19:03:23, scottmg wrote:
> On 2016/03/29 18:56:08, Dirk Pranke wrote:
> > you'll still need owners approval from Scott ...
> > 
> >
>
https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
> > File chrome/tools/crash_service/BUILD.gn (right):
> > 
> >
>
https://codereview.chromium.org/1838953003/diff/1/chrome/tools/crash_service/...
> > chrome/tools/crash_service/BUILD.gn:33:
> > ":crash_service(//build/toolchain/win:${maybe_clang}x64)"
> > lgtm. 
> > 
> > There's a related issue in that is_clang is a toolchain-scoped variable
(i.e.,
> > the value may change per toolchain) and so if this target got pulled in via
> > multiple toolchains we might end up declaring it multiple times. 
> > 
> > However, oddly, it looks like this target is really only referenced via the
> > chromium_builder_perf and telemetry_chrome_test targets, which can't be
right,
> > so I'll try to clean this stuff up in a different follow-on CL.
> 
> I think that might be right. I don't really know what crash_service is, so
it's
> probably something to do with Breakpad, which probably means it's only used in
> content_shell or some non-Chrome case.

Oh! It's the other-other-thing. We should just delete it, so ... "lgtm" in that
it doesn't matter if it helps you get by a build error and Dirk says it's ok
gn-wise.

[Dirk Pranke, 2016-03-29 19:10:26]

On 2016/03/29 19:04:37, scottmg wrote:
> Oh! It's the other-other-thing. We should just delete it, so 

?? It's not actually used?

We actually go out of our way to build this and start it on the win bots. Is
that pointless?

Deleting this would certainly simplify a couple things ...

[Reid Kleckner, 2016-03-29 19:57:04]

The CQ bit was checked by rnk@chromium.org

[commit-bot: I haz the power, 2016-03-29 19:57:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1838953003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1838953003/1

[Reid Kleckner, 2016-03-29 19:58:55]

On 2016/03/29 19:04:37, scottmg wrote:
> Oh! It's the other-other-thing. We should just delete it, so ... "lgtm" in
that
> it doesn't matter if it helps you get by a build error and Dirk says it's ok
> gn-wise.

OK, thanks. :)

[scottmg, 2016-03-29 20:19:48]

On 2016/03/29 19:10:26, Dirk Pranke wrote:
> On 2016/03/29 19:04:37, scottmg wrote:
> > Oh! It's the other-other-thing. We should just delete it, so 
> 
> ?? It's not actually used?
> 
> We actually go out of our way to build this and start it on the win bots. Is
> that pointless?
> 
> Deleting this would certainly simplify a couple things ...

I'm not 100% sure. I think there's two things called crash_service, one that's
part of Breakpad/content_shell and another one (this one) that was intended to
be part of a previous grand vision to use Crashpad but be something more. that
I'm not too familiar with either, but will try to look at it at some point soon
to hopefully delete one/both of them.

[commit-bot: I haz the power, 2016-03-29 20:43:02]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-03-29 20:44:34]

Description was changed from

==========
Build crash_service_win64 with clang-cl when is_clang=true

Otherwise this gn assertion fails:
  ERROR at //build/config/sanitizers/sanitizers.gni:85:1: Assertion failed.
  assert(!using_sanitizer || is_clang,
  ^-----
  Sanitizers (is_*san) require setting is_clang = true in 'gn args'

R=dpranke@chromium.org,scottmg@chromium.org
BUG=598761
==========

to

==========
Build crash_service_win64 with clang-cl when is_clang=true

Otherwise this gn assertion fails:
  ERROR at //build/config/sanitizers/sanitizers.gni:85:1: Assertion failed.
  assert(!using_sanitizer || is_clang,
  ^-----
  Sanitizers (is_*san) require setting is_clang = true in 'gn args'

R=dpranke@chromium.org,scottmg@chromium.org
BUG=598761

Committed: https://crrev.com/9065ac8a2b45154c8e5dbef97bddd01829680f88
Cr-Commit-Position: refs/heads/master@{#383808}
==========

[commit-bot: I haz the power, 2016-03-29 20:44:35]

Patchset 1 (id:??) landed as
https://crrev.com/9065ac8a2b45154c8e5dbef97bddd01829680f88
Cr-Commit-Position: refs/heads/master@{#383808}