Opened 11 years ago
Closed 11 years ago
#9599 closed enhancement (fixed)
re-merge #1396 and insure that Sage starts + passes doc tests on t2.math
Reported by: | ddrake | Owned by: | drkirkby |
---|---|---|---|
Priority: | major | Milestone: | sage-4.6.2 |
Component: | porting: Solaris | Keywords: | |
Cc: | wjp, SimonKing, malb, jhpalmieri | Merged in: | sage-4.6.2.alpha3 |
Authors: | Simon King, John Perry, Martin Albrecht | Reviewers: | Martin Albrecht |
Report Upstream: | N/A | Work issues: | |
Branch: | Commit: | ||
Dependencies: | Stopgaps: |
Description (last modified by )
A patch at #1396 was merged in 4.5.2.alpha0, but that caused Sage to segfault when starting on t2.math. That ticket was backed out in 4.5.2.alpha1, and should be re-merged. See #9583 for discussion.
In order to be of help for anyone looking at this ticket in some months or years time, the following description of t2.math may be useful.
Hardware and software configuration of t2.math.washington.edu
- Sun SPARC Enterprise T5240 Server
- 2 x 1167 MHz UltraSPARC T2 PLUS processors. (16 cores and 128 hardware threads in total).
- 32 GB RAM
- No swap devices configured.
- Solaris 10 update 7 (5/09)
- gcc 4.4.1 configured to use the Sun linker and Sun assembler.
Attachments (1)
Change History (15)
comment:1 Changed 11 years ago by
- Cc wjp added
comment:2 Changed 11 years ago by
comment:3 Changed 11 years ago by
- Cc SimonKing malb added
comment:4 Changed 11 years ago by
- Cc jhpalmieri added
- Description modified (diff)
- Summary changed from re-merge #1396 and insure that Sage starts on t2.math to re-merge #1396 and insure that Sage starts + passes doc tests on t2.math
comment:5 Changed 11 years ago by
See
http://groups.google.com/group/libsingular-devel/browse_thread/thread/e49fe19e034ec774
for upstream's reaction.
comment:6 Changed 11 years ago by
- Status changed from new to needs_review
comment:7 Changed 11 years ago by
The rebased patch passes doctests for me. It still needs testing on t2.
comment:8 Changed 11 years ago by
comment:9 Changed 11 years ago by
- Status changed from needs_review to positive_review
The patch applies fine to 4.6.1.rc1 and doctests pass on t2, except:
sage -t -long -force_lib devel/sage/sage/tests/cmdline.py # 19 doctests failed, timeouts sage -t -long -force_lib devel/sage/sage/parallel/decorate.py # 1 doctests failed, OSError: [Errno 12] Not enough space sage -t -long -force_lib devel/sage/sage/misc/trace.py # 2 doctests failed,
These don't seem related to this patch.
The patch did have a positive review before and the only issue was a SIGSEGV on startup on t2. I'll thus set this patch to a positive review.
comment:10 Changed 11 years ago by
- Reviewers set to Martin Albrecht
Is there a reason the milestone is sage-5.0?
comment:12 Changed 11 years ago by
- Work issues set to commit message
The commit message will have to be changed to reflect number, also you should split it over multiple lines if it's that long (but make sure the first line makes sense by itself).
Changed 11 years ago by
comment:14 Changed 11 years ago by
- Merged in set to sage-4.6.2.alpha3
- Resolution set to fixed
- Status changed from positive_review to closed
To summarize the lengthy discussion on #9583:
#1396 exposed a symbol clash between Singular and Pari with some linkers.
To fix this, we rename Singular's
mu
toKstd1_mu
. This could be done either with the SPKG at #9583 which changes only this, or with the SPKG at #8059 which upgrades Singular to a new version that includes the renamedmu
.