Make testing i18n the default.

Make testing i18n the default.

If i18n support is on when building with make, the intl test suite is automatically run with "check".

On the bots, we'll turn it off explicitly, where required.

R=jochen@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=16785

[jochen (gone - plz use gerrit), 2013-09-17 10:53:23]

https://codereview.chromium.org/23702044/diff/7001/Makefile
File Makefile (right):

https://codereview.chromium.org/23702044/diff/7001/Makefile#newcode129
Makefile:129: GYPFLAGS += -Dv8_enable_i18n_support=1
as soon as i18n is on by default, the flag will test for off

Shouldn't run-tests.py then also default to on, and have an option to turn the
intl tests off?

[Michael Achenbach, 2013-09-17 11:12:56]

On 2013/09/17 10:53:23, jochen wrote:
> https://codereview.chromium.org/23702044/diff/7001/Makefile
> File Makefile (right):
> 
> https://codereview.chromium.org/23702044/diff/7001/Makefile#newcode129
> Makefile:129: GYPFLAGS += -Dv8_enable_i18n_support=1
> as soon as i18n is on by default, the flag will test for off
> 
> Shouldn't run-tests.py then also default to on, and have an option to turn the
> intl tests off?

Done. Like that, NaCl will go red on the bots after landing this and before
changing the bot, but that's ok. I'll rebase with
https://codereview.chromium.org/23003028 as soon as it lands.

PTAL

[jochen (gone - plz use gerrit), 2013-09-17 12:29:43]

On 2013/09/17 11:12:56, machenbach wrote:
> On 2013/09/17 10:53:23, jochen wrote:
> > https://codereview.chromium.org/23702044/diff/7001/Makefile
> > File Makefile (right):
> > 
> > https://codereview.chromium.org/23702044/diff/7001/Makefile#newcode129
> > Makefile:129: GYPFLAGS += -Dv8_enable_i18n_support=1
> > as soon as i18n is on by default, the flag will test for off
> > 
> > Shouldn't run-tests.py then also default to on, and have an option to turn
the
> > intl tests off?
> 
> Done. Like that, NaCl will go red on the bots after landing this and before
> changing the bot, but that's ok. I'll rebase with
> https://codereview.chromium.org/23003028 as soon as it lands.
> 
> PTAL

Why not just add --noi18n to Makefile.nacl?

[Michael Achenbach, 2013-09-17 13:26:31]

On 2013/09/17 12:29:43, jochen wrote:
> On 2013/09/17 11:12:56, machenbach wrote:
> > On 2013/09/17 10:53:23, jochen wrote:
> > > https://codereview.chromium.org/23702044/diff/7001/Makefile
> > > File Makefile (right):
> > > 
> > > https://codereview.chromium.org/23702044/diff/7001/Makefile#newcode129
> > > Makefile:129: GYPFLAGS += -Dv8_enable_i18n_support=1
> > > as soon as i18n is on by default, the flag will test for off
> > > 
> > > Shouldn't run-tests.py then also default to on, and have an option to turn
> the
> > > intl tests off?
> > 
> > Done. Like that, NaCl will go red on the bots after landing this and before
> > changing the bot, but that's ok. I'll rebase with
> > https://codereview.chromium.org/23003028 as soon as it lands.
> > 
> > PTAL
> 
> Why not just add --noi18n to Makefile.nacl?

The bots call run-tests.py directly without going through a Makefile.

[jochen (gone - plz use gerrit), 2013-09-17 13:35:11]

On 2013/09/17 13:26:31, machenbach wrote:
> On 2013/09/17 12:29:43, jochen wrote:
> > On 2013/09/17 11:12:56, machenbach wrote:
> > > On 2013/09/17 10:53:23, jochen wrote:
> > > > https://codereview.chromium.org/23702044/diff/7001/Makefile
> > > > File Makefile (right):
> > > > 
> > > > https://codereview.chromium.org/23702044/diff/7001/Makefile#newcode129
> > > > Makefile:129: GYPFLAGS += -Dv8_enable_i18n_support=1
> > > > as soon as i18n is on by default, the flag will test for off
> > > > 
> > > > Shouldn't run-tests.py then also default to on, and have an option to
turn
> > the
> > > > intl tests off?
> > > 
> > > Done. Like that, NaCl will go red on the bots after landing this and
before
> > > changing the bot, but that's ok. I'll rebase with
> > > https://codereview.chromium.org/23003028 as soon as it lands.
> > > 
> > > PTAL
> > 
> > Why not just add --noi18n to Makefile.nacl?
> 
> The bots call run-tests.py directly without going through a Makefile.

What about developers doing make nacl_32.check

[jochen (gone - plz use gerrit), 2013-09-17 13:46:12]

lgtm

[Michael Achenbach, 2013-09-18 08:26:00]

Rebased

[Michael Achenbach, 2013-09-18 08:58:09]

Committed patchset #5 manually as r16785.