Re: [O-MPI users] MPI/C++ debugger?

2005-12-15 Thread Konstantin Karganov
> Can anyone share experiences of using a debugger for MPI/C++ programs? > > Information on Ease of use, STL support, Price etc. would be most welcome You might have a look at: 1. TotalView ( http://www.etnus.com/TotalView/index.html ) 2. DDT ( http://www.allinea.com/?page=48 ) 3. Sun Prism ( http

Re: [O-MPI users] Open MPI @SC2005

2005-11-16 Thread Konstantin Karganov
> The Open MPI Team will be at SC this year. > I am co-hosting a BOF entitled "Why MPI Makes You Scream! And how can > we simplify parallel debugging?" By the way (though it may be offtopic here)... Does anybody know some statistics on MPI usage. e.i. : - percentage of C, C++, Fortran MPI program

Re: [O-MPI users] OpenMPI debugging support

2005-10-21 Thread Konstantin Karganov
> My points here were that for at least some debuggers, a > naming scheme is all they need, and we should be able to accommodate > that. Yes, it seems that some advanced "renaming scheme" will fit most of needs. (I mean if it allows to customize not only debugger name & path, but also cmd-line op

Re: [O-MPI users] OpenMPI debugging support

2005-10-21 Thread Konstantin Karganov
> The question was merely how to do it: call "gdb orterun" and catch it > somewhere on breakpoint or attach to orterun later or smth else. Reply to myself: # gdb orterun (gdb) br MPIR_Breakpoint (gdb) run (gdb) (gdb) detach (gdb) exit Am I right? Best regards, Konstantin.

Re: [O-MPI users] OpenMPI debugging support

2005-10-21 Thread Konstantin Karganov
> You and Chris G. raise a good point -- another parallel debugger vendor > has contacted me about the same issue (their debugger does not have an > executable named "totalview"). > <...> > Comments? Actually, the point is deeper than just a debugger naming question. High-quality MPI implementat

Re: [O-MPI users] OpenMPI debugging support

2005-10-20 Thread Konstantin Karganov
> However, we're quite open to other approaches. Because of the nature of > our integration with a variety of different run-time environments, our > startup is not a shell script -- mpirun ("orterun" is its real name; > "mpirun" is a sym link to orterun) is a compiled executable. Surely, I saw th

[O-MPI users] OpenMPI debugging support

2005-10-20 Thread Konstantin Karganov
t together long before MPI_Init. (BTW: what is this - mpirun that is running as a background process becomes "stopped" all the time I try to "bg" it?) What is the "correct" way and how it is supposed to get a debugger attached to all processes of the program? Best regards, Konstantin Karganov.