[O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Ferris McCormick
Sorry if this is old news, configuration problem, or whatever. I have been tied up with other things, and have not been able to follow ompi very closely. I just built openmpi-1.0a1r7305 for testing, and notice that ompi_info (and all other ompi tests) give mca: base: components_open: component li

Re: [O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Brian Barrett
Thanks for the heads up. We are not seeing this on other platforms, so it might be a Sparc-specific issue. Any chance you could compile with debugging symbols and generate a backtrace? Also, could you send the contents of /proc/cpuinfo (long story...)? Thanks! Brian On Sep 12, 2005, at

Re: [O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Ferris McCormick
On Mon, 2005-09-12 at 11:14 -0500, Brian Barrett wrote: > Thanks for the heads up. We are not seeing this on other platforms, > so it might be a Sparc-specific issue. Any chance you could compile > with debugging symbols and generate a backtrace? Also, could you > send the contents of /pro

Re: [O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Jeff Squyres
On Sep 12, 2005, at 2:05 PM, Ferris McCormick wrote: HOWEVER: If I configure with --enable-debug, two things happen: 1. I have to build ompi/mca/rcache/rb by hand because of incorrect CFLAGS; FWIW, the rcache guys are currently off working in a /tmp branch, and they have fixed this problem

Re: [O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Brian Barrett
Ok, I see what's happening, although I'm not sure the two problems are actually related. The first is that the component to provide high resolution timer support on Linux is disabling itself because: 1) it doesn't know how to figure out the clock rate of the CPU 2) there's no assembly fo

Re: [O-MPI devel] ompi_info Seg Fault, missing component -- linux

2005-09-12 Thread Ferris McCormick
On Mon, 2005-09-12 at 13:34 -0500, Brian Barrett wrote: > Ok, I see what's happening, although I'm not sure the two problems > are actually related. The first is that the component to provide > high resolution timer support on Linux is disabling itself because: > >1) it doesn't know how t

[O-MPI devel] svn merge: lessons learned

2005-09-12 Thread Jeff Squyres
Lesson learned the hard way... If you're going to make a branch into /tmp, it is STRONGLY ADVISED to cp an ***UNMODIFIED /trunk*** (i.e., do not have any local edits on the /trunk that you're copying). Then make/apply all your changes in a new checkout of your /tmp tree and go from there. T

[O-MPI devel] 64bit shared library problems

2005-09-12 Thread Nathan DeBardeleben
I've been having this problem for a week or so and I've been asking other people to weigh in if they know what I'm doing wrong. I've gotten no where on this so I figure I'll finally drop it out on the list. First, here's the important info: The machine: [sparkplug]~ > cat /etc/issue Welcom

Re: [O-MPI devel] 64bit shared library problems

2005-09-12 Thread Jeff Squyres
Maybe I'm dense -- I thought you couldn't use --shared when linking to a static library...? If you want to build OMPI as a shared library, then ditch the --enable-static --disable-shared from your configure line (building OMPI as shared is the default, which is how I build 95% of the time).