clang/win: Start work on getting clang/win working in gn.

clang/win: Start work on getting clang/win working in gn.

This is enough to get base building with clang on Windows.

On Windows, we use clang-cl.exe which is command-line compatible
with cl.exe.

BUG=491209, 404525
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/22c76db4a60e2e23c3f8f22f8e08b9b0cb49e002
Cr-Commit-Position: refs/heads/master@{#332563}

[Nico, 2015-06-02 22:52:58]

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

[Nico, 2015-06-02 22:52:59]

Not sure who's the right person to review this…Scott, giving this to you for now
and cc'ing a few others who might want to comment.

https://codereview.chromium.org/1158763006/diff/1/build/config/compiler/BUILD.gn
File build/config/compiler/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/1/build/config/compiler/BUILD...
build/config/compiler/BUILD.gn:913: "-Wno-inconsistent-missing-override", 
#http://crbug.com/428099
`gn format` insisted that I don't put a space after # here

[scottmg, 2015-06-02 23:02:20]

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

[scottmg, 2015-06-02 23:02:22]

+dpranke

lgtm

https://codereview.chromium.org/1158763006/diff/1/build/config/compiler/BUILD.gn
File build/config/compiler/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/1/build/config/compiler/BUILD...
build/config/compiler/BUILD.gn:913: "-Wno-inconsistent-missing-override", 
#http://crbug.com/428099
On 2015/06/02 22:52:58, Nico wrote:
> `gn format` insisted that I don't put a space after # here

o_O sorry, not sure why
https://code.google.com/p/chromium/issues/detail?id=495862.

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.gn
File build/toolchain/win/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.g...
build/toolchain/win/BUILD.gn:215: if (current_cpu == "x86") {
Dirk should look at this part as he did a lot of wrestling with this. In
particular I think goma was not in here for some complicated reason that I dont'
remember or didn't understand.

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.g...
build/toolchain/win/BUILD.gn:216: msvc_toolchain("x32") {
"x32"? yuck :(

[Nico, 2015-06-02 23:04:30]

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.gn
File build/toolchain/win/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.g...
build/toolchain/win/BUILD.gn:216: msvc_toolchain("x32") {
On 2015/06/02 23:02:21, scottmg wrote:
> "x32"? yuck :(

That should probably be x86. Hopefully some trybot will inform me about this. (I
tried building base with is_clang set to true and false locally, which worked
fine)

[scottmg, 2015-06-02 23:06:20]

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.gn
File build/toolchain/win/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/1/build/toolchain/win/BUILD.g...
build/toolchain/win/BUILD.gn:216: msvc_toolchain("x32") {
On 2015/06/02 23:04:30, Nico wrote:
> On 2015/06/02 23:02:21, scottmg wrote:
> > "x32"? yuck :(
> 
> That should probably be x86. Hopefully some trybot will inform me about this.
(I
> tried building base with is_clang set to true and false locally, which worked
> fine)

Hope so! (x32 seems actively wrong anyway)

[Nico, 2015-06-02 23:06:44]

Done. Apparently gn defaults to 64-bit builds? :-/ (It's the better default, but
having different defaults from gyp confuses me)

[scottmg, 2015-06-02 23:07:41]

On 2015/06/02 23:06:44, Nico wrote:
> Done. Apparently gn defaults to 64-bit builds? :-/ (It's the better default,
but
> having different defaults from gyp confuses me)

Yes and yes (defaults to host cpu I think).

[Nico, 2015-06-02 23:07:53]

Looks like there aren't any gn windows bots. Anyhoo, I checked locally that
things now work if I set target_arch=x86 (and I checked that this didn't work
with ps1)

[scottmg, 2015-06-02 23:09:37]

On 2015/06/02 23:07:53, Nico wrote:
> Looks like there aren't any gn windows bots. Anyhoo, I checked locally that
> things now work if I set target_arch=x86 (and I checked that this didn't work
> with ps1)

There's non-default GN trybots, I added them. (I'm still confused why we used
x32 in other places, but I'll assume you or Dirk know what's going on)

[Nico, 2015-06-02 23:21:42]

Now with even less x32.

[Nico, 2015-06-02 23:22:13]

The CQ bit was checked by thakis@chromium.org

[Nico, 2015-06-02 23:22:13]

The patchset sent to the CQ was uploaded after l-g-t-m from scottmg@chromium.org
Link to the patchset: https://codereview.chromium.org/1158763006/#ps40001
(title: ".")

[Nico, 2015-06-02 23:22:21]

The CQ bit was unchecked by thakis@chromium.org

[Nico, 2015-06-02 23:22:58]

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

[commit-bot: I haz the power, 2015-06-02 23:23:45]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1158763006/40001

[scottmg, 2015-06-02 23:28:46]

lgtm! but Dirk should still look at goma.

[brettw, 2015-06-02 23:49:20]

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

[brettw, 2015-06-02 23:49:20]

LGTM with one change.

https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
File build/toolchain/win/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
build/toolchain/win/BUILD.gn:227: cl = "${goma_prefix}$prefix/clang-cl.exe -m32"
It seems like the -m32/m64 flags should be set in build/config/compiler:compiler
config. This is where the corresponding flags on Posix are set.

[Dirk Pranke, 2015-06-03 00:39:55]

lgtm w/ brett's change, but make sure you send this through all of the GN
trybots as well as the normal CQ set.

[Nico, 2015-06-03 02:38:44]

Thanks!

https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
File build/toolchain/win/BUILD.gn (right):

https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
build/toolchain/win/BUILD.gn:227: cl = "${goma_prefix}$prefix/clang-cl.exe -m32"
On 2015/06/02 23:49:20, brettw wrote:
> It seems like the -m32/m64 flags should be set in
build/config/compiler:compiler
> config. This is where the corresponding flags on Posix are set.

I think here fits better with the non-clang windows build: foo/cl.exe is msvc's
32-bit compiler, foo/amd64/cl.exe is msvc's 64-bit compiler. Likewise,
"bin/clang-cl.exe -m32" is the 32-bit clang compiler, "bin/clang-cl.exe -m64" is
the 64-bit clang compiler (they happen to be the same binary, but that's
details).

But done, have a look at patch set 4. If you like that better, hit cq.


Dirk: I think patch set 3 has try jobs for the right bots already. If you think
there are bots missing, please add them. (win8_chromium_gn_rel and a few linux
gn bots; I tried adding win_chromium_gn_x64_rel  since that's in the popup, but
apparently that doesn't exist; at least the job ended up grey)

[Dirk Pranke, 2015-06-03 02:46:47]

On 2015/06/03 02:38:44, Nico wrote:
> Dirk: I think patch set 3 has try jobs for the right bots already. If you
think
> there are bots missing, please add them. (win8_chromium_gn_rel and a few linux
> gn bots; I tried adding win_chromium_gn_x64_rel  since that's in the popup,
but
> apparently that doesn't exist; at least the job ended up grey)

Please add the bots to CQ_EXTRA_TRYBOTS just to be explicit, and add the
win8_chromium_gn_dbg
bot and the mac_chromium_gn_{dbg,rel} bots (mac because you're tweaking the
clang defs, which
are used there as well but not in the CQ by default yet).

The win_chromium_gn_x64_{rel,dbg} bots as well are theoretically available:

http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_gn_x...
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_gn_x...

have done builds today; that pool might just be overloaded or something, dunno.
We might
need to bug a trooper about them; feel free to ignore them for purposes of
landing this if
they don't seem to be responding.

[brettw, 2015-06-03 04:28:07]

On 2015/06/03 02:38:44, Nico wrote:
> Thanks!
> 
>
https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
> File build/toolchain/win/BUILD.gn (right):
> 
>
https://codereview.chromium.org/1158763006/diff/40001/build/toolchain/win/BUI...
> build/toolchain/win/BUILD.gn:227: cl = "${goma_prefix}$prefix/clang-cl.exe
-m32"
> On 2015/06/02 23:49:20, brettw wrote:
> > It seems like the -m32/m64 flags should be set in
> build/config/compiler:compiler
> > config. This is where the corresponding flags on Posix are set.
> 
> I think here fits better with the non-clang windows build: foo/cl.exe is
msvc's
> 32-bit compiler, foo/amd64/cl.exe is msvc's 64-bit compiler. Likewise,
> "bin/clang-cl.exe -m32" is the 32-bit clang compiler, "bin/clang-cl.exe -m64"
is
> the 64-bit clang compiler (they happen to be the same binary, but that's
> details).
> 
> But done, have a look at patch set 4. If you like that better, hit cq.

I like it the new way for consistency. I kind of get what you're saying but I
think the same thing could apply to gcc and we should be consitent.

[brettw, 2015-06-03 04:29:35]

The CQ bit was checked by brettw@chromium.org

[brettw, 2015-06-03 04:29:36]

lgtm

I added the bots I use for triple-checking GN changes.

[brettw, 2015-06-03 04:29:36]

The patchset sent to the CQ was uploaded after l-g-t-m from
dpranke@chromium.org, scottmg@chromium.org
Link to the patchset: https://codereview.chromium.org/1158763006/#ps60001
(title: "comment")

[commit-bot: I haz the power, 2015-06-03 04:31:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1158763006/60001

[Nico, 2015-06-03 05:12:19]

On 2015/06/03 04:29:36, brettw wrote:
> lgtm
> 
> I added the bots I use for triple-checking GN changes.

Thanks!

[commit-bot: I haz the power, 2015-06-03 06:01:26]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-06-03 06:01:27]

Try jobs failed on following builders:
  win8_chromium_rel on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_rel...)

[Nico, 2015-06-03 06:14:36]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2015-06-03 06:14:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1158763006/60001

[commit-bot: I haz the power, 2015-06-03 08:11:53]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-06-03 08:12:48]

Patchset 4 (id:??) landed as
https://crrev.com/22c76db4a60e2e23c3f8f22f8e08b9b0cb49e002
Cr-Commit-Position: refs/heads/master@{#332563}