[mpiwg-tools] Questions for the tools community

Marc-André Hermanns hermanns at itc.rwth-aachen.de
Thu Sep 19 01:08:34 CDT 2019


Dear all,

so we conclude that we will move this discussion to a later date.

As the upcoming weeks may need some special handling (Bill Williams is
not available on usual slot  next week, Oct 3 is a public holiday in
Germany, Oct 10 I have scheduling difficulties) I created a doole poll
with some options.

https://doodle.com/poll/ir35szkfkkausbpf

Please indicate what works for you.

Cheers,
Marc-Andre

On 17.09.19 15:18, William Williams wrote:
> I can probably do something at 1900CET that Thursday but am on
> airplanes for a few hours before that. If that’s not workable for
> others, I’m happy to discuss things over email or a subsequent week as
> folks prefer.
> 
> Bill Williams
> Wiss. Mitarbeiter
> ZIH-IAK, TU-Dresden
> Falkenbrunnen, Room 009
> Chemnitzer Straße 50
> 01187 Dresden
> Phone:  +49 351 463-39976
> e-mail:   william.williams at mailbox.tu-dresden.de
> <mailto:william.williams at mailbox.tu-dresden.de>
> 
> 
> 
>> On 17. Sep 2019, at 14:53, Jeff Squyres (jsquyres)
>> <jsquyres at cisco.com <mailto:jsquyres at cisco.com>> wrote:
>>
>> Pushing to Thursday 26 Sep works for me.
>>
>> Sorry for the schedule complications!
>>
>>
>>> On Sep 17, 2019, at 8:37 AM, Mohror, Kathryn <mohror1 at llnl.gov
>>> <mailto:mohror1 at llnl.gov>> wrote:
>>>
>>> Hi Marc-Andre,
>>>
>>> Pushing to a later Thursday is fine with me. I am not aware of any
>>> other tools people that are planning to join this coming Thursday.
>>>
>>> Kathryn
>>>
>>> -----Original Message-----
>>> From: mpiwg-tools <mpiwg-tools-bounces at lists.mpi-forum.org
>>> <mailto:mpiwg-tools-bounces at lists.mpi-forum.org>> On Behalf Of
>>> Marc-André Hermanns via mpiwg-tools
>>> Sent: Monday, September 16, 2019 11:46 PM
>>> To: Jeff Squyres (jsquyres) <jsquyres at cisco.com
>>> <mailto:jsquyres at cisco.com>>; MPI Tool WG
>>> <mpiwg-tools at lists.mpi-forum.org
>>> <mailto:mpiwg-tools at lists.mpi-forum.org>>; William Williams
>>> <william.williams at mailbox.tu-dresden.de
>>> <mailto:william.williams at mailbox.tu-dresden.de>>
>>> Cc: Marc-André Hermanns <hermanns at itc.rwth-aachen.de
>>> <mailto:hermanns at itc.rwth-aachen.de>>
>>> Subject: Re: [mpiwg-tools] Questions for the tools community
>>>
>>> Hi Jeff,
>>>
>>> sorry for the delay in responding to your email.
>>>
>>> As Martin also has some conflicts for this upcoming Thursday, we
>>> had the idea to push this to another week. How would that work for you?
>>>
>>> @Kathryn: Do you know from anyone at LLNL or other US sites who
>>> planned to come, who we need to inform if we shift to another week?
>>> I put Bill Williams (from TU Dresden) in CC, as he initially
>>> planned to join as well.
>>>
>>> If shifting the topic to a later date doesn't work out for you, you
>>> can certainly sync up with me, but I think first-hand information
>>> is always best in a discussion and it's always a pleasure to have
>>> you on the call ;-)
>>>
>>> Cheers,
>>> Marc-Andre
>>>
>>> On 16.09.19 15:05, Jeff Squyres (jsquyres) wrote:
>>>> Something changed and I'm sorry, I can't be there this Thursday.
>>>>  I'm sorry!
>>>>
>>>> Can I sync up with someone before Thursday to explain what I'm
>>>> asking about?
>>>>
>>>>
>>>>> On Aug 29, 2019, at 11:25 AM, Jeff Squyres (jsquyres)
>>>>> <jsquyres at cisco.com <mailto:jsquyres at cisco.com>> wrote:
>>>>>
>>>>> Sure, I can be there.
>>>>>
>>>>> Thanks!
>>>>>
>>>>>
>>>>>> On Aug 29, 2019, at 11:05 AM, Marc-André Hermanns via
>>>>>> mpiwg-tools <mpiwg-tools at lists.mpi-forum.org
>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>> wrote:
>>>>>>
>>>>>> Hi Jeff,
>>>>>>
>>>>>> sorry about the silence. As Martin already said, your mail led to
>>>>>> some internal discussion and we wanted to reach out beyond the
>>>>>> tools
>>>>>> people on the mailing list because it seems no-one on the list is
>>>>>> currently able to handle Fortran descriptors, but we failed to
>>>>>> keep you informed.
>>>>>>
>>>>>> Would you be available on *Sep 19* for the tools call? That would
>>>>>> give us some time to contact some people outside the working group
>>>>>> and they'd have time to either give feedback or organize their
>>>>>> schedules in a way to attend the call.
>>>>>>
>>>>>> Cheers,
>>>>>> Marc-Andre
>>>>>>
>>>>>> On 29.08.19 16:53, Martin Schulz via mpiwg-tools wrote:
>>>>>>> Hi Jeff,
>>>>>>>
>>>>>>> Due to the summer break and vacation time, especially in
>>>>>>> Europe, we
>>>>>>> haven’t had a chance to address this - Marc-Andre had suggested a
>>>>>>> tools call on this for after EuroMPI and then we’ll know more (we
>>>>>>> need to get tool implementors to the table who are not part of
>>>>>>> the WG).
>>>>>>>
>>>>>>> Personally, though, I fear you are right with your assumption, but
>>>>>>> I am happy to be proven wrong.
>>>>>>>
>>>>>>> Martin
>>>>>>>
>>>>>>>
>>>>>>>>>>>>>> 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 <mailto:schulzm at in.tum.de>
>>>>>>> <mailto:schulzm at in.tum.de>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>> On 29. Aug 2019, at 16:46, Jeff Squyres (jsquyres) via
>>>>>>>> mpiwg-tools
>>>>>>>> <mpiwg-tools at lists.mpi-forum.org
>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>
>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>> wrote:
>>>>>>>>
>>>>>>>> Should I infer from the lack of reply that no tools properly
>>>>>>>> handle either Fortran descriptors or the descriptor-based symbols?
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>> On Aug 25, 2019, at 11:31 AM, Jeff Squyres (jsquyres) via
>>>>>>>>> mpiwg-tools <mpiwg-tools at lists.mpi-forum.org
>>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>
>>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>> wrote:
>>>>>>>>>
>>>>>>>>> In Open MPI, we are finally working towards supporting Fortran
>>>>>>>>> descriptors for choice buffers (i.e., "TYPE(*), DIMENSION(..)").
>>>>>>>>>
>>>>>>>>> Per MPI-3.1 p613 table 17.1, this means ***changing all the
>>>>>>>>> back-end symbols*** that are used for PMPI-style interception.
>>>>>>>>>
>>>>>>>>> A concrete example:
>>>>>>>>>
>>>>>>>>> - in mpif.h / the mpi module, MPI_SEND will become MPI_SEND_FTS
>>>>>>>>> - in the mpi_f08 module, MPI_SEND will become MPI_SEND_F08TS
>>>>>>>>>
>>>>>>>>> I.e., anywhere the application calls MPI_SEND, it will be
>>>>>>>>> transmogrified to either MPI_SEND_FTS or MPI_SEND_F08TS.  And
>>>>>>>>> the
>>>>>>>>> choice buffer passed will be a descriptor (which may
>>>>>>>>> therefore be
>>>>>>>>> a subarray, not a contiguous buffer), not a naked pointer.
>>>>>>>>>
>>>>>>>>> Here's my questions:
>>>>>>>>>
>>>>>>>>> 1. Do any tools support Fortran descriptors as choice buffers?
>>>>>>>>> 2. Do any tools intercept the symbols as described in MPI-3.1
>>>>>>>>> table 17.1?
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Jeff Squyres
>>>>>>>>> jsquyres at cisco.com <mailto:jsquyres at cisco.com>
>>>>>>>>> <mailto:jsquyres at cisco.com>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> mpiwg-tools mailing list
>>>>>>>>> mpiwg-tools at lists.mpi-forum.org
>>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>
>>>>>>>>> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-tools
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Jeff Squyres
>>>>>>>> jsquyres at cisco.com <mailto:jsquyres at cisco.com>
>>>>>>>> <mailto:jsquyres at cisco.com>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> mpiwg-tools mailing list
>>>>>>>> mpiwg-tools at lists.mpi-forum.org
>>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>
>>>>>>>> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-tools
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> mpiwg-tools mailing list
>>>>>>> mpiwg-tools at lists.mpi-forum.org
>>>>>>> <mailto:mpiwg-tools at lists.mpi-forum.org>
>>>>>>> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-tools
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> Dr. rer. nat. Marc-André Hermanns
>>>>>>
>>>>>> IT Center
>>>>>> Group: High Performance Computing
>>>>>> Division: Computational Science and Engineering RWTH Aachen
>>>>>> University Seffenter Weg 23
>>>>>> 52074 Aachen
>>>>>> Phone: +49 241 80-24381
>>>>>> hermanns at itc.rwth-aachen.de <mailto:hermanns at itc.rwth-aachen.de>
>>>>>> www.itc.rwth-aachen.de
>>>>>>
>>>>>> _______________________________________________
>>>>>> mpiwg-tools mailing list
>>>>>> mpiwg-tools at lists.mpi-forum.org
>>>>>> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-tools
>>>>>
>>>>>
>>>>> --
>>>>> Jeff Squyres
>>>>> jsquyres at cisco.com <mailto:jsquyres at cisco.com>
>>>>>
>>>>
>>>>
>>>
>>> --
>>> Dr. rer. nat. Marc-André Hermanns
>>>
>>> IT Center
>>> Group: High Performance Computing
>>> Division: Computational Science and Engineering RWTH Aachen
>>> University Seffenter Weg 23
>>> 52074 Aachen
>>> Phone: +49 241 80-24381
>>> hermanns at itc.rwth-aachen.de <mailto:hermanns at itc.rwth-aachen.de>
>>> www.itc.rwth-aachen.de
>>>
>>
>>
>> -- 
>> Jeff Squyres
>> jsquyres at cisco.com <mailto:jsquyres at cisco.com>
>>
> 

-- 
Dr. rer. nat. Marc-André Hermanns

IT Center
Group: High Performance Computing
Division: Computational Science and Engineering
RWTH Aachen University
Seffenter Weg 23
52074 Aachen
Phone: +49 241 80-24381
hermanns at itc.rwth-aachen.de
www.itc.rwth-aachen.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5336 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-tools/attachments/20190919/c26d54aa/attachment.p7s>


More information about the mpiwg-tools mailing list