Thanks for the tip!
I tried setting the stack size to 64MB (ulimit -a reads 64MB at any
rate) and I still get a seg fault.
What's the next step?
Cheers
Jon
Jon Erik Ween, MD, MS
Scientist, Kunin-Lunenfeld Applied Research Unit
Director, Stroke Clinic, Brain Health Clinic
Baycrest Centre for Geriatric Care
Assistant Professor, Dept. of Medicine, Div. of Neurology
University of Toronto Faculty of Medicine
Posluns Building, 6th Floor, Room 644
Baycrest Centre for Geriatric Care
3560 Bathurst Street
Toronto, Ontario M6A 2E1
Canada
Phone: 416-785-2500 x3636
Fax: 416-785-2484
Email: jween@xxxxxxxxxxxxxxxxxxxx
On 5-Apr-07, at 7:25 AM, Bud Davis wrote:
--- Jon Erik Ween <jween@xxxxxxxxxxxxxxxxxxxx> wrote:
HI!
I'm trying to compile packages for the R stats
system (on a vanilla
FC6 with the full development environment installed)
but keep getting
seg faults for various compilations that seem to
have to do with
Fortran. Not being very well versed with Fortran and
not finding
anything on the R lists I was wondering if the gcc
crowd had any
ideas? Here's the output, not much, but all there
is:
[root@luther Desktop]# R CMD INSTALL ./mvtnorm
* Installing *source* package 'mvtnorm' ...
** libs
gfortran -fpic -O2 -g -c mvt.f -o mvt.o
mvt.f: In function 'mvkrsv':
mvt.f:1480: internal compiler error: Segmentation
fault
This also happens if I try to compile "mvtnorm"
outside the R
environment.
Thanks!
Jon
let's check the easiest thing first. The program you
are compiling may need a lot of stack space, and by
default shell's do not allocate very much.
if you are using tcsh the command is:
"limit stacksize unlimited"
bash and the other shells have similar commands.
give this a try, and if the problem still exists then
we may have a problem in gfortran.
HTH,
bud davis