[mpiwg-tools] PMPI and sessions init
schulzm at in.tum.de
Thu Nov 5 17:39:26 CST 2020
Hi Joachim, all,
Yes, I think you are right - we probably didn't think this all the way through on the TelCon: as long as all items are loaded, the interceptions with PMPI should work, no matter when and where MPI calls are made. The only issue would be the load order of a tool vs. a library using MPI_Sessions, which could lead to situations where a tool wouldn't be able to catch the init calls.
PS: what do you mean with the environment variable in this scenario?
Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems
Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching
Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ)
Email: schulzm at in.tum.de
On 05.11.20, 19:00, "mpiwg-tools on behalf of Joachim Protze via mpiwg-tools" <mpiwg-tools-bounces at lists.mpi-forum.org on behalf of mpiwg-tools at lists.mpi-forum.org> wrote:
if I understood the discussion about MPI_Session_init and PMPI right, I
don't think there exists an issue.
By ld-preloading the PMPI tool, the tool can also intercept a call to
MPI_Session_init, which might come from a library constructor before
main. Even a function called from the library constructor of a static
library can be intercepted by a ld-preloaded tool.
For libraries, which are explicitly loaded by the MPI runtime during
initialization, this would also work. For others, it depends on the
ordering of calling the library constructor (might work by chance :).
This was one of my main motivation of having such an environmental
Dipl.-Inf. Joachim Protze
Group: High Performance Computing
Division: Computational Science and Engineering
RWTH Aachen University
Seffenter Weg 23
D 52074 Aachen (Germany)
Tel: +49 241 80- 24765
Fax: +49 241 80-624765
protze at itc.rwth-aachen.de
More information about the mpiwg-tools