Objet : Developers list for StarPU
Archives de la liste
- From: Samuel Thibault <samuel.thibault@inria.fr>
- To: Florent Pruvost <florent.pruvost@inria.fr>
- Cc: starpu-devel@lists.gforge.inria.fr
- Subject: Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool
- Date: Fri, 26 Jun 2015 17:00:56 +0200
- List-archive: <http://lists.gforge.inria.fr/pipermail/starpu-devel/>
- List-id: "Developers list. For discussion of new features, code changes, etc." <starpu-devel.lists.gforge.inria.fr>
Does it work on plafrim1?
How can I connect to plafrim2?
Florent Pruvost, le Fri 26 Jun 2015 16:43:10 +0200, a écrit :
> (gdb) bt
> #0 0x00007ffff63c4a94 in vfprintf () from /lib64/libc.so.6
> #1 0x00007ffff63cdb77 in fprintf () from /lib64/libc.so.6
> #2 0x00007ffff7b7ece6 in _starpu_fxt_dag_set_task_done (job_id=<optimized
> out>, label=<optimized out>, color=<optimized out>) at
> ../../src/debug/traces/starpu_fxt_dag.c:97
I guess the problematic pointer is the "name" pointer? What is the
content of the ev struct, which is the source of this pointer? The trace
files themselves dont't show anything odd: when param[3] is 1, param[4]
always contain a name, so it shouldn't segfault.
Samuel
- [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Florent Pruvost, 26/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Samuel Thibault, 26/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Florent Pruvost, 26/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Samuel Thibault, 29/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Samuel Thibault, 29/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Florent Pruvost, 29/06/2015
- Re: [Starpu-devel] segfault MPI execution and starpu_fxt_tool, Samuel Thibault, 26/06/2015
Archives gérées par MHonArc 2.6.19+.