#6883: [with spkg, needs review] ECL 9.8.4 fails to build in 64-bit mode, OS X
10.5.8
----------------------+-----------------------------------------------------
Reporter: mvngu | Owner: mabshoff
Type: defect | Status: new
Priority: critical | Milestone: sage-4.1.2
Component: packages | Keywords:
Reviewer: | Author:
Merged: |
----------------------+-----------------------------------------------------
Comment(by mvngu):
One of the problems is the following line in the file `spkg-install` of
the ECL spkg:
{{{
./configure --prefix=$SAGE_LOCAL --with-system-gmp --enable-boehm=system
}}}
Deleting the argument `--with-system-gmp` and the build would go OK, but
then it would fail on t2.math. Deleting `--enable-boehm=system` and the
build would go fine. I have created an updated spkg at
[[BR]][[BR]]
http://sage.math.washington.edu/home/mvngu/release/spkg/standard/ecl-9.8.4.p0.spkg
[[BR]][[BR]]
whose file `spkg-install` now only has this configuration line:
{{{
./configure --prefix=$SAGE_LOCAL
}}}
I also did some general clean up such as removing the empty directory
`build` and reformatting comments so that they are no more than 80
characters in length. I'm now building Sage 4.1.2.alpha0 from scratch with
the above updated ECL package. The machines sage.math, mod.math, and
geom.math are all near capacity so I have been experiencing difficulty
getting Sage to compile on those machines since ATLAS complained about
errors while tuning itself. Here are the platforms on which I'm currently
compiling Sage 4.1.2.alpha0 from scratch with my updated
ecl-9.8.4.p0.spkg:
* bsd.math in 32-bit mode
* bsd.math in 64-bit mode
* t2.math
* cicero on SkyNet (x86 Fedora 9 with GCC 4.4.1)
* eno on SkyNet (x86_64 Fedora 9 with GCC 4.4.1)
* lena on SkyNet (x86_64 RHEL Server 5.3 with GeForce GPUs and GCC 4.4.1)
For bsd.math, I have copied fortran-OSX64-20090120.spkg and
ecl-9.8.4.p0.spkg to the directory /Users/mvngu/apps/. Here are the steps
I took to start building Sage 4.1.2.alpha0 in 64-bit mode with
ecl-9.8.4.p0.spkg:
{{{
[mv...@bsd ~]$ pwd
/Users/mvngu
[mv...@bsd ~]$ tar -xf apps/sage-4.1.2.alpha0.tar -C
/scratch/mvngu/sandbox-64/alpha/
[mv...@bsd ~]$ cd /scratch/mvngu/sandbox-64/alpha/sage-4.1.2.alpha0/
[mv...@bsd sage-4.1.2.alpha0]$ rm -rf
spkg/standard/fortran-20071120.p5.spkg
[mv...@bsd sage-4.1.2.alpha0]$ cp -rf /Users/mvngu/apps/fortran-
OSX64-20090120.spkg spkg/standard/
[mv...@bsd sage-4.1.2.alpha0]$ rm -rf spkg/standard/ecl-9.8.4.spkg
[mv...@bsd sage-4.1.2.alpha0]$ cp -rf /Users/mvngu/apps/ecl-9.8.4.p0.spkg
spkg/standard/
[mv...@bsd sage-4.1.2.alpha0]$ export SAGE64=yes
[mv...@bsd sage-4.1.2.alpha0]$ make
}}}
When `make` terminates, it would report success. But actually if you then
start Sage with
{{{
./sage -br main
}}}
you would get many errors. You now need to open the file
{{{
SAGE_ROOT/devel/sage-main/sage/graphs/all.py
}}}
and comment out the line
{{{
from sage.graphs.cliquer import *
}}}
Then start `make` again:
{{{
make
}}}
When it terminates, start Sage again with
{{{
./sage -br main
}}}
This time, Sage should load fine without any errors.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/6883#comment:2>
Sage <http://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
-~----------~----~----~----~------~----~------~--~---
