#7426: fix mpir spkg to correctly detect OS X 10.6
-------------------------------+--------------------------------------------
Reporter: was | Owner: tbd
Type: defect | Status: positive_review
Priority: blocker | Milestone: sage-4.2.1
Component: build | Keywords:
Work_issues: | Author: William Stein
Reviewer: Minh Van Nguyen | Merged:
-------------------------------+--------------------------------------------
Changes (by newvalueoldvalue):
* status: needs_review => positive_review
* reviewer: => Minh Van Nguyen
* author: => William Stein
Comment:
On bsd.math which now runs OS X 10.6.2, Sage 4.2.1.alpha0 fails to compile
with the message (full
[http://sage.math.washington.edu/home/mvngu/doc/sage/install/sage-4.2.1/4.2.1.alpha0/install-4.2.1.alpha0
-intel-osx-10.6.2.log.bz2 install log] is up on sage.math):
{{{
mv mach_desc.h ../include/NTL/mach_desc.h
sh MakeGetTime "gcc -I../include -I. -O2 -g -fno-common " "-lm"
does anybody really know what time it is?
gcc -I../include -I. -O2 -g -fno-common -o TestGetTime TestGetTime.c
GetTime1.c -lm
running
using GetTime1.c
gcc -I../include -I. -O2 -g -fno-common
-I/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/include -o
gen_lip_gmp_aux gen_lip_gmp_aux.c
-L/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/lib -lgmp -lm
ld: warning: in
/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/lib/libgmp.dylib, file is
not of required architecture
./gen_lip_gmp_aux > lip_gmp_aux_impl.h
NTL_GMP_HACK flag not set.
gcc -I../include -I. -O2 -g -fno-common
-I/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/include -o gen_gmp_aux
gen_gmp_aux.c -L/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/lib -lgmp
-lm
ld: warning: in
/scratch/mvngu/sandbox/sage-4.2.1.alpha0/local/lib/libgmp.dylib, file is
not of required architecture
Undefined symbols:
"___gmp_bits_per_limb", referenced from:
_main in ccgka6B4.o
ld: symbol(s) not found
collect2: ld returned 1 exit status
make[2]: *** [setup3] Error 1
Failed building setup3 of NTL
real 0m0.775s
user 0m0.457s
sys 0m0.206s
sage: An error occurred while installing ntl-5.4.2.p9
Please email sage-devel http://groups.google.com/group/sage-devel
explaining the problem and send the relevant part of
of /scratch/mvngu/sandbox/sage-4.2.1.alpha0/install.log. Describe your
computer, operating system, etc.
If you want to try to fix the problem yourself, *don't* just cd to
/scratch/mvngu/sandbox/sage-4.2.1.alpha0/spkg/build/ntl-5.4.2.p9 and type
'make'.
Instead type "/scratch/mvngu/sandbox/sage-4.2.1.alpha0/sage -sh"
in order to set all environment variables correctly, then cd to
/scratch/mvngu/sandbox/sage-4.2.1.alpha0/spkg/build/ntl-5.4.2.p9
(When you are done debugging, you can type "exit" to leave the
subshell.)
make[1]: *** [installed/ntl-5.4.2.p9] Error 1
real 13m10.576s
user 7m49.302s
sys 5m4.192s
Error building Sage.
}}}
I took Sage 4.2.1.alpha0 and replaced its MPIR package mpir-1.2.p7.spkg
with the updated package mpir-1.2.p8.spkg. I then compiled this modified
source tarball and doctested the whole Sage library. Here are the results
of my tests:
* bsd.math, MacIntel OS X 10.6.2 --- compiled OK; many more doctest
failures than on OS X 10.6.1. Full
[http://sage.math.washington.edu/home/mvngu/doc/sage/doctest/sage-4.2.1/sage-4.2.1.alpha0/doctest-4.2.1.alpha0-7426
-intel-osx-10.6.2.log doctest log] is up on sage.math.
* 32-bit openSUSE 11.0, AMD Opteron(tm) Processor 148 --- compiled OK;
all doctests passed.
* 32-bit Ubuntu 9.10, Intel(R) Pentium(R) 4 CPU 3.00GHz --- compiled OK;
all doctests passed.
* 64-bit Ubuntu 8.04.3 LTS, Intel(R) Xeon(R) CPU X7460 @ 2.66GHz ---
compiled OK; all doctests passed.
* 32-bit Fedora 9, Intel(R) Pentium(R) 4 CPU 2.66GHz --- compiled OK; the
following tests failed (full
[http://sage.math.washington.edu/home/mvngu/doc/sage/doctest/sage-4.2.1/sage-4.2.1.alpha0/doctest-4.2.1.alpha0-7426-cicero.log
doctest log] up on sage.math):
{{{
sage -t -long "devel/sage/sage/misc/randstate.pyx"
sage -t -long "devel/sage/sage/interfaces/expect.py"
sage -t -long "devel/sage/sage/interfaces/sage0.py"
}}}
* 64-bit Red Hat Enterprise Linux Server 5.4, Intel(R) Xeon(R) CPU X7460
@ 2.66GHz --- compiled OK; all doctests passed.
* 64-bit Fedora 9, Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz ---
compiled OK; all doctests passed.
* 64-bit Red Hat Enterprise Linux Server 5.3, AMD Phenom(tm) II X4 940
Processor --- compiled OK; all doctests passed.
* 64-bit openSUSE 11.1, Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz,
menas --- compiled OK; one doctest failure:
{{{
[mv...@menas sage-4.2.1.alpha0-7426]$ ./sage -t -long -verbose devel/sage-
main/sage/interfaces/ecm.py
sage -t -long -verbose "devel/sage-main/sage/interfaces/ecm.py"
Traceback (most recent call last):
File "/home/mvngu/.sage//tmp/.doctest_ecm.py", line 2, in <module>
from sage.all_cmdline import *;
File
"/home/mvngu/usr/menas/sandbox/sage-4.2.1.alpha0-7426/local/lib/python
/site-packages/sage/all_cmdline.py", line 14, in <module>
from sage.all import *
File
"/home/mvngu/usr/menas/sandbox/sage-4.2.1.alpha0-7426/local/lib/python
/site-packages/sage/all.py", line 68, in <module>
from sage.libs.all import *
File
"/home/mvngu/usr/menas/sandbox/sage-4.2.1.alpha0-7426/local/lib/python
/site-packages/sage/libs/all.py", line 3, in <module>
import sage.libs.ntl.all as ntl
File
"/home/mvngu/usr/menas/sandbox/sage-4.2.1.alpha0-7426/local/lib/python
/site-packages/sage/libs/ntl/all.py", line 26, in <module>
from sage.libs.ntl.ntl_ZZ import (
ImportError: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.11' not found
(required by
/home/mvngu/usr/menas/sandbox/sage-4.2.1.alpha0-7426/local/lib/libgmpxx.so.3)
[0.9 s]
exit code: 1024
}}}
* MacIntel OS X 10.4.11 --- compiled OK; the following doctest failed
([http://sage.math.washington.edu/home/mvngu/doc/sage/doctest/sage-4.2.1/sage-4.2.1.alpha0/doctest-4.2.1.alpha0-7426
-intel-osx-10.4.11.log doctest log] up on sage.math):
{{{
sage -t -long -verbose "devel/sage-main/sage/interfaces/maxima.py"
Trying:
f**g###line 2625:_sage_ >>> f^g
Expecting:
1/sin(*** *** Error: TIMED OUT! PROCESS KILLED! *** ***
*** *** Error: TIMED OUT! *** ***
[1800.5 s]
exit code: 768
}}}
Since the updated MPIR package now allows Sage 4.2.1.alpha0 to build on OS
X 10.6.2, I give it a positive review.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/7426#comment:3>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en
-~----------~----~----~----~------~----~------~--~---
