#10456 closed task (fixed)
mesa packages need rebuilt post scheduler
Reported by: | kallisti5 | Owned by: | kallisti5 |
---|---|---|---|
Priority: | normal | Milestone: | R1/beta1 |
Component: | - General | Version: | R1/Development |
Keywords: | Cc: | ||
Blocked By: | #10440 | Blocking: | |
Platform: | All |
Description (last modified by )
This ticket simply tracks packages that need rebuilt post-scheduler.
- mesa-10.0.2 "resolve symbol _get_system_info failed"
Change History (12)
comment:1 by , 11 years ago
Blocked By: | 10440 added |
---|
comment:2 by , 11 years ago
comment:3 by , 11 years ago
Weren't the get_system_info() changes supposed to be binary compatible? If not, they should be.
comment:5 by , 11 years ago
They are on gcc2 builds. I don't think there is much point in keeping deprecated symbols on a builds that are not binary compatible anyway.
comment:6 by , 11 years ago
Makes sense, although we usually do provide limited GCC4 compatibility in order to not have to rebuild all packages with every release. But from time to time, those work-arounds should be removed, anyway.
follow-up: 9 comment:7 by , 11 years ago
Description: | modified (diff) |
---|---|
Summary: | gcc4 packages that need rebuilt post scheduler → mesa packages need rebuilt post scheduler |
yup. python turned out to be due to the latest upgrade to libxml2 not working with python 2.6.x. sorry for the noise.
comment:8 by , 11 years ago
x86 primary packages rebuilt and updated in hrev46732. Still need to do x86 secondary package so leaving this open for now.
comment:9 by , 11 years ago
Replying to kallisti5:
yup. python turned out to be due to the latest upgrade to libxml2 not working with python 2.6.x. sorry for the noise.
See https://bitbucket.org/haikuports/haikuports/issue/88/libxml2_python-291-broken AFAICT the python dependency isn't on a minimum version, hence the problem.
comment:10 by , 11 years ago
Milestone: | R1 → R1/alpha5 |
---|---|
Resolution: | → fixed |
Status: | new → closed |
this was all completed.
comment:12 by , 10 years ago
Milestone: | R1/alpha5 → R1/beta1 |
---|
I had this problem with Python (while trying to build Mesa for gcc4) even before the scheduler merge. It's probably not related to scheduler in any way, and may have been broken long before.
This only happens while using capsules, which I guess is not something widely used.