Don't build base with nacl glibc and newlib toolchains

Don't build base with nacl glibc and newlib toolchains

It's unnecessary because nothing appears to depend on it,
and it blocks use of C++11 in base.

The IRT build is still enabled because the IRT is the primary user of base_nacl.
The pnacl_newlib build is still enabled because the remoting client uses it.

BUG=360096
R=thakis@chromium.org,mseaborn@chromium.org

Committed: https://crrev.com/47e06f8940c019e513f9b9f40b96f38ac46a18f6
Cr-Commit-Position: refs/heads/master@{#293442}

[Derek Schuff, 2014-09-04 16:21:58]

dschuff@chromium.org changed reviewers:
+ mseaborn@chromium.org, thakis@chromium.org

[Nico, 2014-09-04 16:27:03]

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp
File base/base_nacl.gyp (right):

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp#newcode50
base/base_nacl.gyp:50: 'build_newlib': 1,
What about this one?

[Derek Schuff, 2014-09-04 16:31:51]

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp
File base/base_nacl.gyp (right):

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp#newcode50
base/base_nacl.gyp:50: 'build_newlib': 1,
On 2014/09/04 16:27:02, Nico (hiding) wrote:
> What about this one?

Done.

[Nico, 2014-09-04 16:34:52]

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp
File base/base_nacl.gyp (right):

https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp#newcode50
base/base_nacl.gyp:50: 'build_newlib': 1,
On 2014/09/04 16:31:51, Derek Schuff wrote:
> On 2014/09/04 16:27:02, Nico (hiding) wrote:
> > What about this one?
> 
> Done.

What about these:
https://code.google.com/p/chromium/codesearch#search/&q=build_(glibc%7Cnewlib...
?

(Why did we build them all these years if the aren't needed?)

[Derek Schuff, 2014-09-04 16:38:26]

On 2014/09/04 16:34:52, Nico (hiding) wrote:
> https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp
> File base/base_nacl.gyp (right):
> 
> https://codereview.chromium.org/540093002/diff/1/base/base_nacl.gyp#newcode50
> base/base_nacl.gyp:50: 'build_newlib': 1,
> On 2014/09/04 16:31:51, Derek Schuff wrote:
> > On 2014/09/04 16:27:02, Nico (hiding) wrote:
> > > What about this one?
> > 
> > Done.
> 
> What about these:
>
https://code.google.com/p/chromium/codesearch#search/&q=build_(glibc%7Cnewlib...
> ?
> 
> (Why did we build them all these years if the aren't needed?)

I haven't looked at all of them yet but most of those go into the SDK and
browser tests, which we do need to build with all the different compilers.

[Nico, 2014-09-04 16:47:32]

>
> I haven't looked at all of them yet but most of those go into the SDK and
> browser tests, which we do need to build with all the different compilers.
>

For tests files, that don't depend on anything, I suppose this is fine (is
there an eventual upgrade path for newlib / glibc?), but
testing/gtest_nacl.gyp, third_party/liblouis/liblouis_nacl.gyp,
third_party/libjingle/libjingle_nacl.gyp,
third_party/modp_b64/modp_b64_nacl.gyp,
and maybe ppapi/ppapi_nacl.gyp look like "regular" code and shouldn't have
this, right? (Especially gtest, which apparently wants to depend on c++11
stuff badly).

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Derek Schuff, 2014-09-04 17:05:30]

On 2014/09/04 16:47:32, Nico (hiding) wrote:
> >
> > I haven't looked at all of them yet but most of those go into the SDK and
> > browser tests, which we do need to build with all the different compilers.
> >
> 
> For tests files, that don't depend on anything, I suppose this is fine (is
> there an eventual upgrade path for newlib / glibc?), but
> testing/gtest_nacl.gyp, third_party/liblouis/liblouis_nacl.gyp,
> third_party/libjingle/libjingle_nacl.gyp,
> third_party/modp_b64/modp_b64_nacl.gyp,
> and maybe ppapi/ppapi_nacl.gyp look like "regular" code and shouldn't have
> this, right? (Especially gtest, which apparently wants to depend on c++11
> stuff badly).
> 
> To unsubscribe from this group and stop receiving emails from it, send an
email
> to mailto:chromium-reviews+unsubscribe@chromium.org.

ppapi_nacl goes into the SDK and will certainly have to keep building with
whatever compilers we ship in the SDK (and yes, there is an eventual upgrade
path). Presumably gtest also? The others, I don't know about. But now is
actually probably a good time to bring the question up again on nacl-eng.

[Nico, 2014-09-04 17:07:05]

On 2014/09/04 17:05:30, Derek Schuff wrote:
> On 2014/09/04 16:47:32, Nico (hiding) wrote:
> > >
> > > I haven't looked at all of them yet but most of those go into the SDK and
> > > browser tests, which we do need to build with all the different compilers.
> > >
> > 
> > For tests files, that don't depend on anything, I suppose this is fine (is
> > there an eventual upgrade path for newlib / glibc?), but
> > testing/gtest_nacl.gyp, third_party/liblouis/liblouis_nacl.gyp,
> > third_party/libjingle/libjingle_nacl.gyp,
> > third_party/modp_b64/modp_b64_nacl.gyp,
> > and maybe ppapi/ppapi_nacl.gyp look like "regular" code and shouldn't have
> > this, right? (Especially gtest, which apparently wants to depend on c++11
> > stuff badly).
> > 
> > To unsubscribe from this group and stop receiving emails from it, send an
> email
> > to mailto:chromium-reviews+unsubscribe@chromium.org.
> 
> ppapi_nacl goes into the SDK and will certainly have to keep building with
> whatever compilers we ship in the SDK (and yes, there is an eventual upgrade
> path). Presumably gtest also? The others, I don't know about. But now is
> actually probably a good time to bring the question up again on nacl-eng.

Ok, will do. This CL LGTM if tests look happy (but I don't actually know what
I'm doing here, so if you want a review from someone qualified, maybe wait for
Mark to chime in too.)

[Nico, 2014-09-04 17:36:58]

> > ppapi_nacl goes into the SDK and will certainly have to keep building with
> > whatever compilers we ship in the SDK (and yes, there is an eventual upgrade
> > path). Presumably gtest also? The others, I don't know about. But now is
> > actually probably a good time to bring the question up again on nacl-eng.
> 
> Ok, will do.

(I'll wait for this change to go in and other things to look good first.)

[Mark Seaborn, 2014-09-04 17:56:09]

https://codereview.chromium.org/540093002/diff/20001/base/base_nacl.gyp
File base/base_nacl.gyp (right):

https://codereview.chromium.org/540093002/diff/20001/base/base_nacl.gyp#newco...
base/base_nacl.gyp:24: 'build_newlib': 0,
Does "build_newlib" mean "x86+newlib+gcc" or "newlib+gcc for user code (but not
IRT code)"?

That is, what happens for ARM?  Can you run the ARM bots?

[Derek Schuff, 2014-09-04 18:00:09]

On 2014/09/04 17:56:09, Mark Seaborn wrote:
> https://codereview.chromium.org/540093002/diff/20001/base/base_nacl.gyp
> File base/base_nacl.gyp (right):
> 
>
https://codereview.chromium.org/540093002/diff/20001/base/base_nacl.gyp#newco...
> base/base_nacl.gyp:24: 'build_newlib': 0,
> Does "build_newlib" mean "x86+newlib+gcc" or "newlib+gcc for user code (but
not
> IRT code)"?
> 
> That is, what happens for ARM?  Can you run the ARM bots?

build_newlib means "gcc-newlib for user code (but not IRT code)" so this should
also disable arm-nacl-gcc builds that don't go into the IRT. I added the arm and
SDK bots.

[Mark Seaborn, 2014-09-04 18:20:25]

LGTM if clobber builds pass on the bots.  (Normal trybot runs don't test the
removal of build targets.  Ninja doesn't remove targets it doesn't know about,
though it would be nice if it did.)

[Derek Schuff, 2014-09-04 20:57:09]

On 2014/09/04 18:20:25, Mark Seaborn wrote:
> LGTM if clobber builds pass on the bots.  (Normal trybot runs don't test the
> removal of build targets.  Ninja doesn't remove targets it doesn't know about,
> though it would be nice if it did.)

clobber builds are clean, modulo several things that are currently broken on tip
of tree, so I'm going to land this.

[Derek Schuff, 2014-09-04 20:57:13]

The CQ bit was checked by dschuff@chromium.org

[Nico, 2014-09-04 20:58:03]

The failure on
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_arm_cross...
is unrelated, yes? (Just to check)


On Thu, Sep 4, 2014 at 1:57 PM, <dschuff@chromium.org> wrote:

> On 2014/09/04 18:20:25, Mark Seaborn wrote:
>
>> LGTM if clobber builds pass on the bots.  (Normal trybot runs don't test
>> the
>> removal of build targets.  Ninja doesn't remove targets it doesn't know
>> about,
>> though it would be nice if it did.)
>>
>
> clobber builds are clean, modulo several things that are currently broken
> on tip
> of tree, so I'm going to land this.
>
> https://codereview.chromium.org/540093002/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[commit-bot: I haz the power, 2014-09-04 20:58:23]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dschuff@chromium.org/540093002/20001

[Derek Schuff, 2014-09-04 21:00:27]

On 2014/09/04 20:58:03, Nico (hiding) wrote:
> The failure on
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_arm_cross...
> is unrelated, yes? (Just to check)

Yeah, I wasn't 100% sure so I ran a whitespace CL just to be sure.

[Nico, 2014-09-04 23:15:49]

The CQ bit was unchecked by thakis@chromium.org

[Nico, 2014-09-04 23:16:29]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2014-09-04 23:19:07]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dschuff@chromium.org/540093002/20001

[commit-bot: I haz the power, 2014-09-05 01:08:21]

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

[commit-bot: I haz the power, 2014-09-05 01:08:22]

Try jobs failed on following builders:
  win_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/win_gpu_triggered...)

[Nico, 2014-09-05 01:09:37]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2014-09-05 01:11:45]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dschuff@chromium.org/540093002/20001

[commit-bot: I haz the power, 2014-09-05 04:17:46]

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

[commit-bot: I haz the power, 2014-09-05 04:17:47]

Try jobs failed on following builders:
  mac_chromium_rel_swarming on tryserver.chromium.mac
(http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[Nico, 2014-09-05 04:19:29]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2014-09-05 04:20:37]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dschuff@chromium.org/540093002/20001

[commit-bot: I haz the power, 2014-09-05 04:56:15]

Committed patchset #2 (id:20001) as 6eaee8fccc67e360f533f831dc7e568a78755767

[commit-bot: I haz the power, 2014-09-10 03:36:53]

Patchset 2 (id:??) landed as
https://crrev.com/47e06f8940c019e513f9b9f40b96f38ac46a18f6
Cr-Commit-Position: refs/heads/master@{#293442}