- mpiwg-p2p - mailman3.cels.anl.gov

Next meeting
by Dan Holmes 19 Jul '15

19 Jul '15

Hi all, I will not be available for the point-to-point WG meeting that should happen tomorrow, Monday 20th July at 11am central US. If others wish to have a meeting anyway, please take notes and add them to the wiki page. Cheers, Dan. -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

Next meeting: 6th July 2015 11am Central US
by Daniel Holmes 03 Jul '15

03 Jul '15

Hi all, The next point-to-point teleconference will be today on Monday 6th July 2015 at 11am Central US via webex. Connection details are on the point-to-point wiki page: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/PtpWikiPage Agenda: 1) Ticket #476: "Can INFO keys change the semantic behaviour of MPI?" issue - the communicator assertions ticket was presented at the June Forum meeting and we got some useful feedback that we should discuss. 2) Tickets #478, #479, #480: "everything you ever wanted to know about MPI send cancel, plus a whole lot more" [new!] 3) New communicator INFO keys - it looks like these will be allowed, so we should define a few and get a proposal ready for the next face-to-face meeting 4) Progress on Arecv and Fsend? 5) Other stuff? Cheers, Dan. -- Dan Holmes Applications Consultant in HPC Research EPCC, The University of Edinburgh James Clerk Maxwell Building The Kings Buildings Peter Guthrie Tait Road Edinburgh EH9 3FD T: +44(0)131 651 3465 E: dholmes(a)epcc.ed.ac.uk *Please consider the environment before printing this email.* The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

Re: [mpiwg-p2p] Minor correction in MPI_WAIT
by William Gropp 29 Jun '15

29 Jun '15

That ticket should never have been left marked as complete (that might be my fault for not checking the status after comment 18). The change never made it into the errata for 30, and commit 1822 was made into mpi-3.0-minutiae/chap-pt2pt/pt2pt.tex , which doesn’t count as getting it into the standard. Bill William Gropp Director, Parallel Computing Institute Thomas M. Siebel Chair in Computer Science Chief Scientist, NCSA University of Illinois Urbana-Champaign On Jun 29, 2015, at 3:20 PM, Thakur, Rajeev <thakur(a)mcs.anl.gov> wrote: > OK, I found the reference. Comment 18 in this ticket seems to indicate that the bug was fixed in the source, but probably not in the right place. > > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/321#comment:18 > > Rajeev > > On Jun 29, 2015, at 7:47 AM, William Gropp <wgropp(a)illinois.edu> > wrote: > >> I added this as a pending errata to the errata-31.tex file - This may help ensure we catch this for the update. >> >> Bill >> >> William Gropp >> Director, Parallel Computing Institute >> Thomas M. Siebel Chair in Computer Science >> Chief Scientist, NCSA >> University of Illinois Urbana-Champaign >> >> >> >> >> >> On Jun 29, 2015, at 2:28 PM, Thakur, Rajeev <thakur(a)mcs.anl.gov> wrote: >> >>> This typo was introduced in MPI 3.0. I seem to remember that someone had reported it earlier, but I can't find the reference. >>> >>> Rajeev >>> >>> On Jun 26, 2015, at 1:54 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: >>> >>>> >>>> FYI. >>>> >>>> >>>> >>>> >>>> >>>> On 6/26/15, 1:46 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: >>>> >>>>> >>>>> MPI-3.1 standard, page 53, lines 24--25: >>>>> >>>>> "Any other type of request is and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>>>> >>>>> >>>>> Should be: >>>>> >>>>> >>>>> "Any other type of request is deallocated and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>>>> >>>>> -- Pavan >>>>> >>>> _______________________________________________ >>>> mpiwg-p2p mailing list >>>> mpiwg-p2p(a)lists.mpi-forum.org >>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p >>> >>> _______________________________________________ >>> mpiwg-p2p mailing list >>> mpiwg-p2p(a)lists.mpi-forum.org >>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p >> >> _______________________________________________ >> mpiwg-p2p mailing list >> mpiwg-p2p(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p

1 0

Re: [mpiwg-p2p] Minor correction in MPI_WAIT
by Thakur, Rajeev 29 Jun '15

29 Jun '15

OK, I found the reference. Comment 18 in this ticket seems to indicate that the bug was fixed in the source, but probably not in the right place. https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/321#comment:18 Rajeev On Jun 29, 2015, at 7:47 AM, William Gropp <wgropp(a)illinois.edu> wrote: > I added this as a pending errata to the errata-31.tex file - This may help ensure we catch this for the update. > > Bill > > William Gropp > Director, Parallel Computing Institute > Thomas M. Siebel Chair in Computer Science > Chief Scientist, NCSA > University of Illinois Urbana-Champaign > > > > > > On Jun 29, 2015, at 2:28 PM, Thakur, Rajeev <thakur(a)mcs.anl.gov> wrote: > >> This typo was introduced in MPI 3.0. I seem to remember that someone had reported it earlier, but I can't find the reference. >> >> Rajeev >> >> On Jun 26, 2015, at 1:54 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: >> >>> >>> FYI. >>> >>> >>> >>> >>> >>> On 6/26/15, 1:46 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: >>> >>>> >>>> MPI-3.1 standard, page 53, lines 24--25: >>>> >>>> "Any other type of request is and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>>> >>>> >>>> Should be: >>>> >>>> >>>> "Any other type of request is deallocated and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>>> >>>> -- Pavan >>>> >>> _______________________________________________ >>> mpiwg-p2p mailing list >>> mpiwg-p2p(a)lists.mpi-forum.org >>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p >> >> _______________________________________________ >> mpiwg-p2p mailing list >> mpiwg-p2p(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p

1 0

Re: [mpiwg-p2p] Minor correction in MPI_WAIT
by William Gropp 29 Jun '15

29 Jun '15

I added this as a pending errata to the errata-31.tex file - This may help ensure we catch this for the update. Bill William Gropp Director, Parallel Computing Institute Thomas M. Siebel Chair in Computer Science Chief Scientist, NCSA University of Illinois Urbana-Champaign On Jun 29, 2015, at 2:28 PM, Thakur, Rajeev <thakur(a)mcs.anl.gov> wrote: > This typo was introduced in MPI 3.0. I seem to remember that someone had reported it earlier, but I can't find the reference. > > Rajeev > > On Jun 26, 2015, at 1:54 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: > >> >> FYI. >> >> >> >> >> >> On 6/26/15, 1:46 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: >> >>> >>> MPI-3.1 standard, page 53, lines 24--25: >>> >>> "Any other type of request is and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>> >>> >>> Should be: >>> >>> >>> "Any other type of request is deallocated and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >>> >>> -- Pavan >>> >> _______________________________________________ >> mpiwg-p2p mailing list >> mpiwg-p2p(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p

1 0

Re: [mpiwg-p2p] Minor correction in MPI_WAIT
by Thakur, Rajeev 29 Jun '15

29 Jun '15

This typo was introduced in MPI 3.0. I seem to remember that someone had reported it earlier, but I can't find the reference. Rajeev On Jun 26, 2015, at 1:54 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: > > FYI. > > > > > > On 6/26/15, 1:46 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: > >> >> MPI-3.1 standard, page 53, lines 24--25: >> >> "Any other type of request is and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >> >> >> Should be: >> >> >> "Any other type of request is deallocated and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." >> >> -- Pavan >> > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p

1 0

FW: Minor correction in MPI_WAIT
by Balaji, Pavan 26 Jun '15

26 Jun '15

FYI. On 6/26/15, 1:46 PM, "Balaji, Pavan" <balaji(a)anl.gov> wrote: > >MPI-3.1 standard, page 53, lines 24--25: > >"Any other type of request is and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." > > >Should be: > > >"Any other type of request is deallocated and the request handle is set to MPI_REQUEST_NULL. MPI_WAIT is a non-local operation." > > -- Pavan >

1 0

Re: [mpiwg-p2p] Next meeting: 22nd June 2015 11am Central US
by Balaji, Pavan 22 Jun '15

22 Jun '15

I'll be at the SC PC meeting. Can someone cover send cancel? -- Pavan Sent from my iPhone > On Jun 22, 2015, at 10:17 AM, Jeff Hammond <jeff.science(a)gmail.com> wrote: > > There will be no update on ARECV+FSEND. No progress has been made since April. > > Jeff > >> On Mon, Jun 22, 2015 at 3:15 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: >> Hi all, >> >> The next point-to-point teleconference will be today on Monday 22th June >> 2015 at 11am Central US via webex. >> >> Connection details are on the point-to-point wiki page: >> https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/PtpWikiPage >> >> Agenda: >> 1) Ticket #476: "Can INFO keys change the semantic behaviour of MPI?" issue >> - the communicator assertions ticket was presented at the June Forum meeting >> and we got some useful feedback that we should discuss. >> >> 2) Tickets #478, #479, #480: "everything you ever wanted to know about MPI >> send cancel, plus a whole lot more" [new!] >> >> 3) New communicator INFO keys - it looks like these will be allowed, so we >> should define a few and get a proposal ready for the next face-to-face >> meeting >> >> 4) Progress on Arecv and Fsend? >> >> 5) Other stuff? >> >> Cheers, >> Dan. >> >> -- >> Dan Holmes >> Applications Consultant in HPC Research >> EPCC, The University of Edinburgh >> James Clerk Maxwell Building >> The Kings Buildings >> Peter Guthrie Tait Road >> Edinburgh >> EH9 3FD >> T: +44(0)131 651 3465 >> E: dholmes(a)epcc.ed.ac.uk >> >> *Please consider the environment before printing this email.* >> >> >> The University of Edinburgh is a charitable body, registered in >> Scotland, with registration number SC005336. >> >> _______________________________________________ >> mpiwg-p2p mailing list >> mpiwg-p2p(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p

1 0

Re: [mpiwg-p2p] Next meeting: 22nd June 2015 11am Central US
by Jeff Hammond 22 Jun '15

22 Jun '15

There will be no update on ARECV+FSEND. No progress has been made since April. Jeff On Mon, Jun 22, 2015 at 3:15 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: > Hi all, > > The next point-to-point teleconference will be today on Monday 22th June > 2015 at 11am Central US via webex. > > Connection details are on the point-to-point wiki page: > https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/PtpWikiPage > > Agenda: > 1) Ticket #476: "Can INFO keys change the semantic behaviour of MPI?" issue > - the communicator assertions ticket was presented at the June Forum meeting > and we got some useful feedback that we should discuss. > > 2) Tickets #478, #479, #480: "everything you ever wanted to know about MPI > send cancel, plus a whole lot more" [new!] > > 3) New communicator INFO keys - it looks like these will be allowed, so we > should define a few and get a proposal ready for the next face-to-face > meeting > > 4) Progress on Arecv and Fsend? > > 5) Other stuff? > > Cheers, > Dan. > > -- > Dan Holmes > Applications Consultant in HPC Research > EPCC, The University of Edinburgh > James Clerk Maxwell Building > The Kings Buildings > Peter Guthrie Tait Road > Edinburgh > EH9 3FD > T: +44(0)131 651 3465 > E: dholmes(a)epcc.ed.ac.uk > > *Please consider the environment before printing this email.* > > > The University of Edinburgh is a charitable body, registered in > Scotland, with registration number SC005336. > > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

Next meeting: 22nd June 2015 11am Central US
by Daniel Holmes 22 Jun '15

22 Jun '15

Hi all, The next point-to-point teleconference will be today on Monday 22th June 2015 at 11am Central US via webex. Connection details are on the point-to-point wiki page: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/PtpWikiPage Agenda: 1) Ticket #476: "Can INFO keys change the semantic behaviour of MPI?" issue - the communicator assertions ticket was presented at the June Forum meeting and we got some useful feedback that we should discuss. 2) Tickets #478, #479, #480: "everything you ever wanted to know about MPI send cancel, plus a whole lot more" [new!] 3) New communicator INFO keys - it looks like these will be allowed, so we should define a few and get a proposal ready for the next face-to-face meeting 4) Progress on Arecv and Fsend? 5) Other stuff? Cheers, Dan. -- Dan Holmes Applications Consultant in HPC Research EPCC, The University of Edinburgh James Clerk Maxwell Building The Kings Buildings Peter Guthrie Tait Road Edinburgh EH9 3FD T: +44(0)131 651 3465 E: dholmes(a)epcc.ed.ac.uk *Please consider the environment before printing this email.* The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0