An option and a tentative fix for 32-bit Clang build.

An option and a tentative fix for 32-bit Clang build.


BUG=162816
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=178249

[Dai Mikurube (NOT FULLTIME), 2013-01-18 07:23:18]

Hi Nico,

What do you think about just adding a 32-bit Clang build option?

I agree that we don't have 32-bit Clang on the tree (we discussed it at
https://codereview.chromium.org/11348245/), but it is hard and complex to change
the gyp for every build (especially in automated testing environments) in
contrast to adding 32-bit Clang into the local tree.

[Dai Mikurube (NOT FULLTIME), 2013-01-18 07:40:41]

https://codereview.chromium.org/11280181/diff/12001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/11280181/diff/12001/build/common.gypi#newcode848
build/common.gypi:848: 'make_clang_dir%':
'third_party/llvm-build/Release+Asserts',
Finally when the type profiler is upstreamed (I'm trying), I wonder if we could
move this dir to third_party/llvm-build/Linux_x64/Release+Asserts to compare
32-bit and 64-bit easily in the same *local* tree.  But, it may have some side
effects.  What do you think?

[Nico, 2013-01-18 19:02:33]

Is it hard to keep a local "32bitclang" git branch and just rebase that on top
of trunk every time you need to use it?

[Dai Mikurube (NOT FULLTIME), 2013-01-21 01:30:46]

On 2013/01/18 19:02:33, Nico wrote:
> Is it hard to keep a local "32bitclang" git branch and just rebase that on top
> of trunk every time you need to use it?

In my experience, common.gypi is sometimes hard to rebase.  It's a little hard
in an automated environment...

[Nico, 2013-01-22 16:15:04]

Ok, lgtm

[Dai Mikurube (NOT FULLTIME), 2013-01-23 02:59:06]

On 2013/01/22 16:15:04, Nico wrote:
> Ok, lgtm

Thanks.

[commit-bot: I haz the power, 2013-01-23 02:59:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dmikurube@chromium.org/11280181/18001

[commit-bot: I haz the power, 2013-01-23 04:47:34]

Change committed as 178249