Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(675)

Issue 2434153002: gyp: Add support for all big endian platforms (Closed)

Created:
4 years, 2 months ago by JaideepBajwa
Modified:
4 years, 1 month ago
Reviewers:
jungshik at Google
CC:
chromium-reviews, miran.karic_imgtec.com, JoranSiu, john.yan, michael_dawson, MTBrandyberry
Target Ref:
refs/heads/master
Project:
icu
Visibility:
Public.

Description

gyp: Add support for all big endian platform Update config to use icudtb.dat if the byteorder is big endian, this will work for other big endian platforms such as PPC and s390. Original CL for BE support: https://codereview.chromium.org/2162393003 R=jshin@chromium.org BUG=v8:5567 Committed: https://chromium.googlesource.com/chromium/deps/icu/+/c1a237113f525a1561d4b322d7653e1083f79aaa

Patch Set 1 #

Total comments: 1

Patch Set 2 : added TODO in BUILD.gn #

Unified diffs Side-by-side diffs Delta from patch set Stats (+19 lines, -8 lines) Patch
M BUILD.gn View 1 1 chunk +1 line, -0 lines 0 comments Download
M icu.gyp View 5 chunks +18 lines, -8 lines 0 comments Download

Messages

Total messages: 14 (2 generated)
JaideepBajwa
ptal
4 years, 2 months ago (2016-10-20 01:34:11 UTC) #1
jungshik at Google
Thank you for a CL. Please, file a new bug against v8 instead of referring ...
4 years, 2 months ago (2016-10-21 07:54:41 UTC) #2
jungshik at Google
https://codereview.chromium.org/2434153002/diff/1/icu.gyp File icu.gyp (right): https://codereview.chromium.org/2434153002/diff/1/icu.gyp#newcode113 icu.gyp:113: [ 'v8_host_byteorder=="big"', { Given that gyp is not used ...
4 years, 2 months ago (2016-10-21 08:01:04 UTC) #3
JaideepBajwa
ptal. I've updated the bug and added a todo in BUILD.gn (it seems byteorder is ...
4 years, 1 month ago (2016-10-25 21:25:00 UTC) #5
JaideepBajwa
ptal, this would enable us to build/test v8 with i18n on PPC/s390 big-endian platform. Thanks.
4 years, 1 month ago (2016-11-03 14:33:45 UTC) #6
jungshik at Google
On 2016/11/03 14:33:45, JaideepBajwa wrote: > ptal, this would enable us to build/test v8 with ...
4 years, 1 month ago (2016-11-08 19:28:43 UTC) #7
JaideepBajwa
On 2016/11/08 19:28:43, jungshik at google wrote: > On 2016/11/03 14:33:45, JaideepBajwa wrote: > > ...
4 years, 1 month ago (2016-11-08 19:44:51 UTC) #8
JaideepBajwa
On 2016/11/08 19:44:51, JaideepBajwa wrote: > On 2016/11/08 19:28:43, jungshik at google wrote: > > ...
4 years, 1 month ago (2016-11-08 19:47:01 UTC) #9
jungshik at Google
On 2016/11/08 19:47:01, JaideepBajwa wrote: > On 2016/11/08 19:44:51, JaideepBajwa wrote: > > On 2016/11/08 ...
4 years, 1 month ago (2016-11-11 21:36:31 UTC) #10
jungshik at Google
Committed patchset #2 (id:20001) manually as c1a237113f525a1561d4b322d7653e1083f79aaa (presubmit successful).
4 years, 1 month ago (2016-11-11 21:39:09 UTC) #12
JaideepBajwa
On 2016/11/11 21:36:31, jungshik at google wrote: > On 2016/11/08 19:47:01, JaideepBajwa wrote: > > ...
4 years, 1 month ago (2016-11-11 21:56:21 UTC) #13
miran.karic
4 years, 1 month ago (2016-11-18 12:10:21 UTC) #14
Message was sent while issue was closed.
On 2016/11/11 21:56:21, JaideepBajwa wrote:
> On 2016/11/11 21:36:31, jungshik at google wrote:
> > On 2016/11/08 19:47:01, JaideepBajwa wrote:
> > > On 2016/11/08 19:44:51, JaideepBajwa wrote:
> > > > On 2016/11/08 19:28:43, jungshik at google wrote:
> > > > > On 2016/11/03 14:33:45, JaideepBajwa wrote:
> > > > > > ptal, this would enable us to build/test v8 with i18n on PPC/s390
> > > big-endian
> > > > > > platform. Thanks.
> > > > > 
> > > > > Ok. Thank you for the update. LGTM
> > > > 
> > > > Thank you.
> > > 
> > > I haven't committed to icu project before. How should I commit it?
> > > "Project "icu" does not have a commit queue."
> > 
> > Sorry about that. I'll land it on your behalf.
> 
> Thank you.

This CL has broken things for MIPS. Variable v8_host_byteorder seems to be
problematic, it is initialized in toolchain.gypi with
'v8_host_byteorder%': '<!(python -c "import sys; print sys.byteorder")',
so this doesn't work when cross compiling, it reads "little" and copies the
wrong ICU data file.
I think I can make the fix CL by checking for mips/mips64 first but I am not
sure that is the best solution.
How do you suggest to proceed?

Powered by Google App Engine
This is Rietveld 408576698