[mpiwg-tools] The path to decode MPI_T event buffers?

[Marc-Andre Hermanns]

Hi all,

regarding decoding copied MPI_T event buffers, I was wondering whether
we could kill two birds with one stone and combine this with the
design for the 'handle introspection' interface we initially thought
for debuggers.

After all, characteristics are similar in the sense that I need an
interface to query structure and potentially some crude form of
semantics (arrays, groups, ranks, ...) from a 'blob' or a memory
address, respectively.

One difference is though that event buffers may exist before and after
MPI_Init and MPI_Finalize, respectively, but the normal handles (comm,
file, win, ...) a debugger may want to decode do not.

What do you think? Should we explore/design in this direction?

Personally, I think if we come up with a common solution, we'd reduce
the number of function calls needed to enable both approaches as well
as the number of 'concepts' a user needs to understand.

Cheers,
Marc-Andre

PS: I would like to postpone the actual work on such an interface
until we have more progress our active proposals. I don't want to get
tangled up in juggling too many large proposals at the same time.
-- 
Marc-Andre Hermanns
Jülich Aachen Research Alliance,
High Performance Computing (JARA-HPC)
Jülich Supercomputing Centre (JSC)

Wilhelm-Johnen-Str.
52425 Jülich
Germany

Phone: +49 2461 61 2509 | +49 241 80 24381
Fax: +49 2461 80 6 99753
www.jara.org/jara-hpc
email: hermanns at jara.rwth-aachen.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4899 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-tools/attachments/20171006/6518bf58/attachment.p7s>