Turns on use_mojo=1 for linux.

Turns on use_mojo=1 for linux.

BUG=353602
TEST=none
R=csharp@chromium.org, mmoss@google.com, thestig@chromium.org, mmoss@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=258062

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=258411

[sky, 2014-03-19 16:01:07]

csharp: *isolate
mmoss/thestiq: the rest

mmoss/thestiq: if you can suggest a way to test this that would be great.

[csharp, 2014-03-19 16:14:12]

*isolate changes LGTM

[mmoss, 2014-03-19 16:34:34]

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/chrom...
File chrome/tools/build/chromeos/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/chrom...
chrome/tools/build/chromeos/FILES.cfg:60: 'filename': 'libmojo_system.so',
lib/libmojo_system.so?

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/linux...
File chrome/tools/build/linux/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/linux...
chrome/tools/build/linux/FILES.cfg:67: 'filename': 'libmojo_system.so',
lib/libmojo_system.so?

[sky, 2014-03-19 16:58:06]

Updated

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/chrom...
File chrome/tools/build/chromeos/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/chrom...
chrome/tools/build/chromeos/FILES.cfg:60: 'filename': 'libmojo_system.so',
On 2014/03/19 16:34:35, mmoss wrote:
> lib/libmojo_system.so?

Done.

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/linux...
File chrome/tools/build/linux/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/30001/chrome/tools/build/linux...
chrome/tools/build/linux/FILES.cfg:67: 'filename': 'libmojo_system.so',
On 2014/03/19 16:34:35, mmoss wrote:
> lib/libmojo_system.so?

Done.

[mmoss, 2014-03-19 17:14:05]

lgtm

[Lei Zhang, 2014-03-19 19:19:26]

lgtm

https://codereview.chromium.org/203963003/diff/10001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/203963003/diff/10001/build/common.gypi#newcod...
build/common.gypi:841: 'use_mojo%': 1,
You probably want to set use_mojo to 0 by default just above the start of the
conditions block, around line 522.

https://codereview.chromium.org/203963003/diff/50001/chrome/test/chromeos/aut...
File
chrome/test/chromeos/autotest/files/client/deps/pyauto_dep/setup_test_links.sh
(right):

https://codereview.chromium.org/203963003/diff/50001/chrome/test/chromeos/aut...
chrome/test/chromeos/autotest/files/client/deps/pyauto_dep/setup_test_links.sh:17:
$(dirname $0)/lib/libmojo_system.so
nit: indentation

[sky, 2014-03-19 19:38:45]

https://codereview.chromium.org/203963003/diff/10001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/203963003/diff/10001/build/common.gypi#newcod...
build/common.gypi:841: 'use_mojo%': 1,
On 2014/03/19 19:19:27, Lei Zhang wrote:
> You probably want to set use_mojo to 0 by default just above the start of the
> conditions block, around line 522.

Actually, it's done at 263.

https://codereview.chromium.org/203963003/diff/50001/chrome/test/chromeos/aut...
File
chrome/test/chromeos/autotest/files/client/deps/pyauto_dep/setup_test_links.sh
(right):

https://codereview.chromium.org/203963003/diff/50001/chrome/test/chromeos/aut...
chrome/test/chromeos/autotest/files/client/deps/pyauto_dep/setup_test_links.sh:17:
$(dirname $0)/lib/libmojo_system.so
On 2014/03/19 19:19:27, Lei Zhang wrote:
> nit: indentation

Done.

[sky, 2014-03-19 19:39:16]

Committed patchset #5 manually as r258062 (presubmit successful).

[Lei Zhang, 2014-03-19 21:13:29]

https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
File chrome/installer/linux/common/installer.include (right):

https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
chrome/installer/linux/common/installer.include:177: install -m 755 -d
"${STAGEDIR}/${INSTALLDIR}/lib/"
Whoops, we forgot do make the lib dir above and the official build's broken now.

[mmoss, 2014-03-19 21:14:12]

On 2014/03/19 21:13:29, Lei Zhang wrote:
>
https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
> File chrome/installer/linux/common/installer.include (right):
> 
>
https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
> chrome/installer/linux/common/installer.include:177: install -m 755 -d
> "${STAGEDIR}/${INSTALLDIR}/lib/"
> Whoops, we forgot do make the lib dir above and the official build's broken
now.

Already fixed.

[mmoss, 2014-03-19 21:14:45]

On 2014/03/19 21:14:12, mmoss wrote:
> On 2014/03/19 21:13:29, Lei Zhang wrote:
> >
>
https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
> > File chrome/installer/linux/common/installer.include (right):
> > 
> >
>
https://codereview.chromium.org/203963003/diff/70001/chrome/installer/linux/c...
> > chrome/installer/linux/common/installer.include:177: install -m 755 -d
> > "${STAGEDIR}/${INSTALLDIR}/lib/"
> > Whoops, we forgot do make the lib dir above and the official build's broken
> now.
> 
> Already fixed.

https://codereview.chromium.org/205253002/

[sky, 2014-03-19 23:25:58]

thestiq/mmoss, could you guys take another look? After the second attempt at
fixing on trunk failed I reverted it all. Diff between patch sets 6 and 7 to see
the changes between what I landed and the fix.

Also, it appears to get chromeos working is going to require changes in another
repo. I'll figure that out before landing this.

[Lei Zhang, 2014-03-19 23:29:00]

On 2014/03/19 23:25:58, sky wrote:
> thestiq/mmoss, could you guys take another look? After the second attempt at
> fixing on trunk failed I reverted it all. Diff between patch sets 6 and 7 to
see
> the changes between what I landed and the fix.
> 
> Also, it appears to get chromeos working is going to require changes in
another
> repo. I'll figure that out before landing this.

lgtm

Maybe turn off use_mojo for OS==Linux && chromeos==1 ?

[sky, 2014-03-20 00:07:16]

On 2014/03/19 23:29:00, Lei Zhang wrote:
> On 2014/03/19 23:25:58, sky wrote:
> > thestiq/mmoss, could you guys take another look? After the second attempt at
> > fixing on trunk failed I reverted it all. Diff between patch sets 6 and 7 to
> see
> > the changes between what I landed and the fix.
> > 
> > Also, it appears to get chromeos working is going to require changes in
> another
> > repo. I'll figure that out before landing this.
> 
> lgtm
> 
> Maybe turn off use_mojo for OS==Linux && chromeos==1 ?

Ya, that is likely safest for the time being. This resulted in lots of changes
to the patch though. Take another look?

[Lei Zhang, 2014-03-20 00:18:30]

https://codereview.chromium.org/203963003/diff/140001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/203963003/diff/140001/build/common.gypi#newco...
build/common.gypi:850: ['OS=="linux" and chromeos==1', {
Did you mean to put chromeos==0 here? As in, this CL is a no-op on ChromeOS,
since you have to deal with things for CrOS in another repo.

[sky, 2014-03-20 01:49:16]

GAH! Updated.

[sky, 2014-03-20 01:58:06]

Also, is there a try bot that does an official build so I don't have
to break the tree again?

On Wed, Mar 19, 2014 at 6:49 PM,  <sky@chromium.org> wrote:
> GAH! Updated.
>
> https://codereview.chromium.org/203963003/

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

[mmoss, 2014-03-20 02:17:00]

I don't think there is :(


On Wed, Mar 19, 2014 at 6:58 PM, Scott Violet <sky@chromium.org> wrote:

> Also, is there a try bot that does an official build so I don't have
> to break the tree again?
>
> On Wed, Mar 19, 2014 at 6:49 PM,  <sky@chromium.org> wrote:
> > GAH! Updated.
> >
> > https://codereview.chromium.org/203963003/
>

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

[mmoss, 2014-03-20 02:24:09]

https://codereview.chromium.org/203963003/diff/160001/chrome/installer/linux/...
File chrome/installer/linux/common/installer.include (right):

https://codereview.chromium.org/203963003/diff/160001/chrome/installer/linux/...
chrome/installer/linux/common/installer.include:155: if [ -f
"${BUILDDIR}/lib/libmojo_system.so" ]; then
Does this need to be conditional? The common.gypi change means it will always be
built for "real" Linux, right (which is the only time these packaging scripts
are run)?

https://codereview.chromium.org/203963003/diff/160001/chrome/tools/build/linu...
File chrome/tools/build/linux/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/160001/chrome/tools/build/linu...
chrome/tools/build/linux/FILES.cfg:68: 'buildtype': ['dev', 'official'],
If this is in fact optional for Linux, then you should add:
'optional': ['dev', 'official'],

[sky, 2014-03-20 14:51:39]

https://codereview.chromium.org/203963003/diff/160001/chrome/installer/linux/...
File chrome/installer/linux/common/installer.include (right):

https://codereview.chromium.org/203963003/diff/160001/chrome/installer/linux/...
chrome/installer/linux/common/installer.include:155: if [ -f
"${BUILDDIR}/lib/libmojo_system.so" ]; then
On 2014/03/20 02:24:09, mmoss wrote:
> Does this need to be conditional? The common.gypi change means it will always
be
> built for "real" Linux, right (which is the only time these packaging scripts
> are run)?

I was hoping to make it so less changes are needed if we flip the bit such that
use_mojo is 0.

https://codereview.chromium.org/203963003/diff/160001/chrome/tools/build/linu...
File chrome/tools/build/linux/FILES.cfg (right):

https://codereview.chromium.org/203963003/diff/160001/chrome/tools/build/linu...
chrome/tools/build/linux/FILES.cfg:68: 'buildtype': ['dev', 'official'],
On 2014/03/20 02:24:09, mmoss wrote:
> If this is in fact optional for Linux, then you should add:
> 'optional': ['dev', 'official'],

Added along with a TODO.

[mmoss, 2014-03-20 16:48:00]

lgtm

[sky, 2014-03-20 21:17:00]

Committed patchset #14 manually as r258411 (presubmit successful).