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

Re: [mpiwg-p2p] Fwd: Re: [MPI Forum] #476: Allow Application Behavior Hints in Info Keys
by Jim Dinan 05 May '15

05 May '15

The following communicator info key tickets have also reached a first draft and could be ready to read at the upcoming meeting: https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/461 https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/463 ~Jim. On Tue, May 5, 2015 at 10:27 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: > Hi All, > > We (Jim and I) have made some progress on the communicator assertions > effort! The new ticket 476 (see below) has a new version of the MPI > Standard with changes designed to > > - oblige the user to comply with any INFO keys applied to any MPI > object or operation - turning all hints into assertions/promises > - prevent the propagation of INFO keys during MPI_COMM_DUP and > MPI_COMM_IDUP operations - to protect legacy libraries from have their > allowed behaviour being restricted by non-library code > > We intend to read this at the face-to-face meeting in June. > > Please review the marked-up changes in the document. Any comments, > suggestions, word-smithing can be sent by email. Jim has offered to run the > teleconference meeting on Monday 11th May at 11am CDT so that the WG can > review and approve this text before the formal reading notice deadline. > > Cheers, > Dan. > > > -------- Forwarded Message -------- Subject: Re: [MPI Forum] #476: Allow > Application Behavior Hints in Info Keys Date: Tue, 05 May 2015 14:17:55 > -0000 From: MPI Forum <mpi-forum(a)lists.mpi-forum.org> > <mpi-forum(a)lists.mpi-forum.org> To: undisclosed-recipients:; > > #476: Allow Application Behavior Hints in Info Keys > -------------------------------------+------------------------------------- > Reporter: jdinan | Owner: jdinan > Type: Text (only) changes | Status: new > Priority: Not ready / author | Milestone: 2015/06/01 > rework | Chicago, USA > Version: MPI 4.0 | Resolution: > Keywords: | Implementation status: Unnecessary > -------------------------------------+------------------------------------- > Changes (by jdinan): > > * cc: dholmes (added) > > > Old description: > > > > New description: > > See attached proposal. > > -- > > -- > Ticket URL: <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/476#comment:1> <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/476#comment:1> > MPI Forum <https://svn.mpi-forum.org/> <https://svn.mpi-forum.org/> > MPI Forum > > > -- > 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 >

1 0

Fwd: Re: [MPI Forum] #476: Allow Application Behavior Hints in Info Keys
by Daniel Holmes 05 May '15

05 May '15

Hi All, We (Jim and I) have made some progress on the communicator assertions effort! The new ticket 476 (see below) has a new version of the MPI Standard with changes designed to * oblige the user to comply with any INFO keys applied to any MPI object or operation - turning all hints into assertions/promises * prevent the propagation of INFO keys during MPI_COMM_DUP and MPI_COMM_IDUP operations - to protect legacy libraries from have their allowed behaviour being restricted by non-library code We intend to read this at the face-to-face meeting in June. Please review the marked-up changes in the document. Any comments, suggestions, word-smithing can be sent by email. Jim has offered to run the teleconference meeting on Monday 11th May at 11am CDT so that the WG can review and approve this text before the formal reading notice deadline. Cheers, Dan. -------- Forwarded Message -------- Subject: Re: [MPI Forum] #476: Allow Application Behavior Hints in Info Keys Date: Tue, 05 May 2015 14:17:55 -0000 From: MPI Forum <mpi-forum(a)lists.mpi-forum.org> To: undisclosed-recipients:; #476: Allow Application Behavior Hints in Info Keys -------------------------------------+------------------------------------- Reporter: jdinan | Owner: jdinan Type: Text (only) changes | Status: new Priority: Not ready / author | Milestone: 2015/06/01 rework | Chicago, USA Version: MPI 4.0 | Resolution: Keywords: | Implementation status: Unnecessary -------------------------------------+------------------------------------- Changes (by jdinan): * cc: dholmes (added) Old description: New description: See attached proposal. -- -- Ticket URL: <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/476#comment:1> MPI Forum <https://svn.mpi-forum.org/> MPI Forum -- 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] Next meeting: 25th May 2015 11am Central US
by Daniel Holmes 27 Apr '15

27 Apr '15

Hi Jim, Yes, that was (and still is!) the intention. I haven't yet created a branch in SVN for the new version of the standard and that's the first step toward getting the text more prepared than we already have it. We are sufficiently close to something we could read that it would be a shame to miss the June deadline. I will attempt to find some time this week to work on this. Cheers, Dan. On 27/04/2015 16:43, Jim Dinan wrote: > Dan, > > At the last meeting, we came up with a proposal to allow info to be > used for assertions. Should we work on pulling that together as a > formal proposal, to read in June? > > ~Jim. > > On Mon, Apr 27, 2015 at 10:29 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk > <mailto:[email protected]>> wrote: > > Hi All, > > I am cancelling the point-to-point teleconference today because I > have not had time to progress the work on communicator assertions > since last time and I am not aware of any other ongoing work for > us to talk about. > > Looking forward to figure out when the next meeting will take > place, I won't be available for the next two teleconferences due > to travelling. This means that the next meeting date would be 25th > May, i.e. one week before the face-to-face meeting in Chicago. > > I will send out a reminder nearer the time and will communicate > any progress in the interim via email. > > Cheers, > Dan. > > On 13/04/2015 10:32, Daniel Holmes wrote: >> Hi All, >> >> The next point-to-point teleconference will be today, Monday 13th >> April 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: >> 0) Discuss/agree the to-do list (see below) that covers the next >> two agenda items. >> 1) "Can INFO keys change the semantic behaviour of MPI?" issue - >> following the discussion at the face-to-face meeting, we seem to >> have a workable way forward on this issue! >> 2) 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 >> 3) Progress on Arecv and Fsend or ReceiveReduce? >> 4) Other stuff? >> >> Cheers, >> Dan. >> >>> Hi All, >>> >>> At the last telco meeting, Jim asked for a to-do list of the >>> tasks needed to move communicator assertions forward. Please >>> find below a first draft of such a task list. Could we agree >>> that this is a good list? If you see something odd/missing, >>> reply to this email group. >>> >>> _To-do list for communicator assertions__ >>> _ >>> 0) Gather evidence to support "no-one will be affected by >>> backwards incompatibility". >>> (e.g. chase Pavan, who volunteered to do this for major MPICH >>> users) >>> 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* >>> propagated to new comm. >>> (include AtoU: if you want to dup INFO use >>> MPI_COMM_DUP_WITH_INFO, with example) >>> 2) Discuss whether attributes (plus, anything else?) should be >>> propagated by MPI_COMM_DUP. >>> (default presumption is yes, they should still be propagated). >>> 3) Find all occurrence of language in the MPI Standard that >>> suggests INFO keys are hints. >>> a) Change or remove any generic statements like "INFO keys >>> cannot change semantics". >>> (e.g. change to "cannot change observable semantics" or remove) >>> b) For each occurrence, determine if it should continue to >>> suggest hint status >>> (e.g. a particular key, most likely for the I/O keys but >>> maybe also some RMA keys) >>> >>> A) Convert ticket 381 into "make comm assertions doable" >>> proposal (by Jun 2015 - Chicago?) >>> B) Get "no wildcard" INFO keys proposal ready for formal reading >>> (by Jun 2015 - Chicago?) >>> C) Get "no ordering" INFO key proposal ready for formal reading >>> (by Jun 2015 - Chicago?) >>> D) Get "no cancel" INFO keys proposal ready for formal reading >>> (by Jun 2015 - Chicago?) >>> E) Discuss other INFO key proposals, e.g. "only wildcard", "no >>> non-blocking", "no underflow" >>> >>> 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 <tel:%2B44%280%29131%20651%203465> >> E:[email protected] <mailto:[email protected]> >> >> *Please consider the environment before printing this email.* > > -- > 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 <tel:%2B44%280%29131%20651%203465> > E:[email protected] <mailto:[email protected]> > > *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 <mailto:[email protected]> > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > -- 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] Next meeting: 25th May 2015 11am Central US
by Jim Dinan 27 Apr '15

27 Apr '15

Dan, At the last meeting, we came up with a proposal to allow info to be used for assertions. Should we work on pulling that together as a formal proposal, to read in June? ~Jim. On Mon, Apr 27, 2015 at 10:29 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: > Hi All, > > I am cancelling the point-to-point teleconference today because I have not > had time to progress the work on communicator assertions since last time > and I am not aware of any other ongoing work for us to talk about. > > Looking forward to figure out when the next meeting will take place, I > won't be available for the next two teleconferences due to travelling. This > means that the next meeting date would be 25th May, i.e. one week before > the face-to-face meeting in Chicago. > > I will send out a reminder nearer the time and will communicate any > progress in the interim via email. > > Cheers, > Dan. > > On 13/04/2015 10:32, Daniel Holmes wrote: > > Hi All, > > The next point-to-point teleconference will be today, Monday 13th April > 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: > 0) Discuss/agree the to-do list (see below) that covers the next two > agenda items. > 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following > the discussion at the face-to-face meeting, we seem to have a workable way > forward on this issue! > 2) 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 > 3) Progress on Arecv and Fsend or ReceiveReduce? > 4) Other stuff? > > Cheers, > Dan. > > Hi All, > > At the last telco meeting, Jim asked for a to-do list of the tasks needed > to move communicator assertions forward. Please find below a first draft of > such a task list. Could we agree that this is a good list? If you see > something odd/missing, reply to this email group. > > *To-do list for communicator assertions* > > 0) Gather evidence to support "no-one will be affected by backwards > incompatibility". > (e.g. chase Pavan, who volunteered to do this for major MPICH users) > 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* propagated > to new comm. > (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, with > example) > 2) Discuss whether attributes (plus, anything else?) should be propagated > by MPI_COMM_DUP. > (default presumption is yes, they should still be propagated). > 3) Find all occurrence of language in the MPI Standard that suggests INFO > keys are hints. > a) Change or remove any generic statements like "INFO keys cannot change > semantics". > (e.g. change to "cannot change observable semantics" or remove) > b) For each occurrence, determine if it should continue to suggest hint > status > (e.g. a particular key, most likely for the I/O keys but maybe also > some RMA keys) > > A) Convert ticket 381 into "make comm assertions doable" proposal (by Jun > 2015 - Chicago?) > B) Get "no wildcard" INFO keys proposal ready for formal reading (by Jun > 2015 - Chicago?) > C) Get "no ordering" INFO key proposal ready for formal reading (by Jun > 2015 - Chicago?) > D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun > 2015 - Chicago?) > E) Discuss other INFO key proposals, e.g. "only wildcard", "no > non-blocking", "no underflow" > > 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.* > > > -- > 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 >

1 0

Re: [mpiwg-p2p] Next meeting: 25th May 2015 11am Central US
by Daniel Holmes 27 Apr '15

27 Apr '15

Hi All, I am cancelling the point-to-point teleconference today because I have not had time to progress the work on communicator assertions since last time and I am not aware of any other ongoing work for us to talk about. Looking forward to figure out when the next meeting will take place, I won't be available for the next two teleconferences due to travelling. This means that the next meeting date would be 25th May, i.e. one week before the face-to-face meeting in Chicago. I will send out a reminder nearer the time and will communicate any progress in the interim via email. Cheers, Dan. On 13/04/2015 10:32, Daniel Holmes wrote: > Hi All, > > The next point-to-point teleconference will be today, Monday 13th > April 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: > 0) Discuss/agree the to-do list (see below) that covers the next two > agenda items. > 1) "Can INFO keys change the semantic behaviour of MPI?" issue - > following the discussion at the face-to-face meeting, we seem to have > a workable way forward on this issue! > 2) 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 > 3) Progress on Arecv and Fsend or ReceiveReduce? > 4) Other stuff? > > Cheers, > Dan. > >> Hi All, >> >> At the last telco meeting, Jim asked for a to-do list of the tasks >> needed to move communicator assertions forward. Please find below a >> first draft of such a task list. Could we agree that this is a good >> list? If you see something odd/missing, reply to this email group. >> >> _To-do list for communicator assertions__ >> _ >> 0) Gather evidence to support "no-one will be affected by backwards >> incompatibility". >> (e.g. chase Pavan, who volunteered to do this for major MPICH users) >> 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* >> propagated to new comm. >> (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, >> with example) >> 2) Discuss whether attributes (plus, anything else?) should be >> propagated by MPI_COMM_DUP. >> (default presumption is yes, they should still be propagated). >> 3) Find all occurrence of language in the MPI Standard that suggests >> INFO keys are hints. >> a) Change or remove any generic statements like "INFO keys cannot >> change semantics". >> (e.g. change to "cannot change observable semantics" or remove) >> b) For each occurrence, determine if it should continue to suggest >> hint status >> (e.g. a particular key, most likely for the I/O keys but maybe >> also some RMA keys) >> >> A) Convert ticket 381 into "make comm assertions doable" proposal (by >> Jun 2015 - Chicago?) >> B) Get "no wildcard" INFO keys proposal ready for formal reading (by >> Jun 2015 - Chicago?) >> C) Get "no ordering" INFO key proposal ready for formal reading (by >> Jun 2015 - Chicago?) >> D) Get "no cancel" INFO keys proposal ready for formal reading (by >> Jun 2015 - Chicago?) >> E) Discuss other INFO key proposals, e.g. "only wildcard", "no >> non-blocking", "no underflow" >> >> 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:[email protected] > > *Please consider the environment before printing this email.* -- 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] Next meeting: 13th April 2015 11am Central US
by Jeff Hammond 13 Apr '15

13 Apr '15

i am working on SC15 papers and will not attend. On Mon, Apr 13, 2015 at 2:32 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: > Hi All, > > The next point-to-point teleconference will be today, Monday 13th April 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: > 0) Discuss/agree the to-do list (see below) that covers the next two agenda > items. > 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following > the discussion at the face-to-face meeting, we seem to have a workable way > forward on this issue! > 2) 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 > 3) Progress on Arecv and Fsend or ReceiveReduce? > 4) Other stuff? > > Cheers, > Dan. > > Hi All, > > At the last telco meeting, Jim asked for a to-do list of the tasks needed to > move communicator assertions forward. Please find below a first draft of > such a task list. Could we agree that this is a good list? If you see > something odd/missing, reply to this email group. > > To-do list for communicator assertions > > 0) Gather evidence to support "no-one will be affected by backwards > incompatibility". > (e.g. chase Pavan, who volunteered to do this for major MPICH users) > 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* propagated > to new comm. > (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, with > example) > 2) Discuss whether attributes (plus, anything else?) should be propagated by > MPI_COMM_DUP. > (default presumption is yes, they should still be propagated). > 3) Find all occurrence of language in the MPI Standard that suggests INFO > keys are hints. > a) Change or remove any generic statements like "INFO keys cannot change > semantics". > (e.g. change to "cannot change observable semantics" or remove) > b) For each occurrence, determine if it should continue to suggest hint > status > (e.g. a particular key, most likely for the I/O keys but maybe also some > RMA keys) > > A) Convert ticket 381 into "make comm assertions doable" proposal (by Jun > 2015 - Chicago?) > B) Get "no wildcard" INFO keys proposal ready for formal reading (by Jun > 2015 - Chicago?) > C) Get "no ordering" INFO key proposal ready for formal reading (by Jun 2015 > - Chicago?) > D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun 2015 > - Chicago?) > E) Discuss other INFO key proposals, e.g. "only wildcard", "no > non-blocking", "no underflow" > > 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

Re: [mpiwg-p2p] Next meeting: 13th April 2015 11am Central US
by Daniel Holmes 13 Apr '15

13 Apr '15

Hi All, The next point-to-point teleconference will be today, Monday 13th April 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: 0) Discuss/agree the to-do list (see below) that covers the next two agenda items. 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! 2) 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 3) Progress on Arecv and Fsend or ReceiveReduce? 4) Other stuff? Cheers, Dan. > Hi All, > > At the last telco meeting, Jim asked for a to-do list of the tasks > needed to move communicator assertions forward. Please find below a > first draft of such a task list. Could we agree that this is a good > list? If you see something odd/missing, reply to this email group. > > _To-do list for communicator assertions__ > _ > 0) Gather evidence to support "no-one will be affected by backwards > incompatibility". > (e.g. chase Pavan, who volunteered to do this for major MPICH users) > 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* > propagated to new comm. > (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, > with example) > 2) Discuss whether attributes (plus, anything else?) should be > propagated by MPI_COMM_DUP. > (default presumption is yes, they should still be propagated). > 3) Find all occurrence of language in the MPI Standard that suggests > INFO keys are hints. > a) Change or remove any generic statements like "INFO keys cannot > change semantics". > (e.g. change to "cannot change observable semantics" or remove) > b) For each occurrence, determine if it should continue to suggest > hint status > (e.g. a particular key, most likely for the I/O keys but maybe also > some RMA keys) > > A) Convert ticket 381 into "make comm assertions doable" proposal (by > Jun 2015 - Chicago?) > B) Get "no wildcard" INFO keys proposal ready for formal reading (by > Jun 2015 - Chicago?) > C) Get "no ordering" INFO key proposal ready for formal reading (by > Jun 2015 - Chicago?) > D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun > 2015 - Chicago?) > E) Discuss other INFO key proposals, e.g. "only wildcard", "no > non-blocking", "no underflow" > > 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] Next meeting: 30th March 2015 11am Central US
by Jeff Hammond 30 Mar '15

30 Mar '15

I only tried "call me" 2x. Life is too short... Sent from my iPhone > On Mar 30, 2015, at 9:16 AM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > > Sadness. :-( > > I had a little trouble connecting today, too -- had to click the "call me" 3-4 times before it succeeded. Or try just the direct phone dialin that Jim succeeded. > > >> On Mar 30, 2015, at 12:11 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: >> >> Webex trash refused to let me connect by 3+ means so I cannot participate. >> >> Jeff >> >> Sent from my iPhone >> >>> On Mar 30, 2015, at 5:37 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: >>> >>> Hi All, >>> >>> The next point-to-point teleconference will be today, Monday 30th March 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: >>> 0) Discuss/agree the to-do list (see below) that covers the next two agenda items. >>> 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >>> 2) 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 >>> 3) Progress on Arecv and Fsend or ReceiveReduce? >>> 4) Other stuff? >>> >>> Cheers, >>> Dan. >>> >>> >>>> On 23/03/2015 11:59, Dan Holmes wrote: >>>> Hi All, >>>> >>>> At the last telco meeting, Jim asked for a to-do list of the tasks needed to move communicator assertions forward. Please find below a first draft of such a task list. Could we agree that this is a good list? If you see something odd/missing, reply to this email group. >>>> >>>> To-do list for communicator assertions >>>> >>>> 0) Gather evidence to support "no-one will be affected by backwards incompatibility". >>>> (e.g. chase Pavan, who volunteered to do this for major MPICH users) >>>> 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* propagated to new comm. >>>> (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, with example) >>>> 2) Discuss whether attributes (plus, anything else?) should be propagated by MPI_COMM_DUP. >>>> (default presumption is yes, they should still be propagated). >>>> 3) Find all occurrence of language in the MPI Standard that suggests INFO keys are hints. >>>> a) Change or remove any generic statements like "INFO keys cannot change semantics". >>>> (e.g. change to "cannot change observable semantics" or remove) >>>> b) For each occurrence, determine if it should continue to suggest hint status >>>> (e.g. a particular key, most likely for the I/O keys but maybe also some RMA keys) >>>> >>>> A) Convert ticket 381 into "make comm assertions doable" proposal (by Jun 2015 - Chicago?) >>>> B) Get "no wildcard" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> C) Get "no ordering" INFO key proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> E) Discuss other INFO key proposals, e.g. "only wildcard", "no non-blocking", "no underflow" >>>> >>>> Cheers, >>>> Dan. >>>> >>>>> On 13/03/2015 19:22, Daniel Holmes wrote: >>>>> Hi All, >>>>> >>>>> The next point-to-point teleconference will be on Monday 16th March 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) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >>>>> 2) 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 >>>>> 3) Progress on Arecv and Fsend or ReceiveReduce? >>>>> 4) 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 >> _______________________________________________ >> mpiwg-p2p mailing list >> mpiwg-p2p(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ > > _______________________________________________ > 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: 30th March 2015 11am Central US
by Jeff Hammond 30 Mar '15

30 Mar '15

Like I said, I tried 3+ ways... Sent from my iPhone > On Mar 30, 2015, at 9:12 AM, Jim Dinan <james.dinan(a)gmail.com> wrote: > > You can dial in at: +1-866-432-9903 Call-in toll-free number (US/Canada) > Meeting code: 201 343 896 > >> On Mon, Mar 30, 2015 at 12:11 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: >> Webex trash refused to let me connect by 3+ means so I cannot participate. >> >> Jeff >> >> Sent from my iPhone >> >>> On Mar 30, 2015, at 5:37 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: >>> >>> Hi All, >>> >>> The next point-to-point teleconference will be today, Monday 30th March 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: >>> 0) Discuss/agree the to-do list (see below) that covers the next two agenda items. >>> 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >>> 2) 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 >>> 3) Progress on Arecv and Fsend or ReceiveReduce? >>> 4) Other stuff? >>> >>> Cheers, >>> Dan. >>> >>> >>>> On 23/03/2015 11:59, Dan Holmes wrote: >>>> Hi All, >>>> >>>> At the last telco meeting, Jim asked for a to-do list of the tasks needed to move communicator assertions forward. Please find below a first draft of such a task list. Could we agree that this is a good list? If you see something odd/missing, reply to this email group. >>>> >>>> To-do list for communicator assertions >>>> >>>> 0) Gather evidence to support "no-one will be affected by backwards incompatibility". >>>> (e.g. chase Pavan, who volunteered to do this for major MPICH users) >>>> 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* propagated to new comm. >>>> (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, with example) >>>> 2) Discuss whether attributes (plus, anything else?) should be propagated by MPI_COMM_DUP. >>>> (default presumption is yes, they should still be propagated). >>>> 3) Find all occurrence of language in the MPI Standard that suggests INFO keys are hints. >>>> a) Change or remove any generic statements like "INFO keys cannot change semantics". >>>> (e.g. change to "cannot change observable semantics" or remove) >>>> b) For each occurrence, determine if it should continue to suggest hint status >>>> (e.g. a particular key, most likely for the I/O keys but maybe also some RMA keys) >>>> >>>> A) Convert ticket 381 into "make comm assertions doable" proposal (by Jun 2015 - Chicago?) >>>> B) Get "no wildcard" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> C) Get "no ordering" INFO key proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>>> E) Discuss other INFO key proposals, e.g. "only wildcard", "no non-blocking", "no underflow" >>>> >>>> Cheers, >>>> Dan. >>>> >>>>> On 13/03/2015 19:22, Daniel Holmes wrote: >>>>> Hi All, >>>>> >>>>> The next point-to-point teleconference will be on Monday 16th March 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) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >>>>> 2) 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 >>>>> 3) Progress on Arecv and Fsend or ReceiveReduce? >>>>> 4) 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 >> >> _______________________________________________ >> 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] Next meeting: 30th March 2015 11am Central US
by Jeff Squyres (jsquyres) 30 Mar '15

30 Mar '15

Sadness. :-( I had a little trouble connecting today, too -- had to click the "call me" 3-4 times before it succeeded. Or try just the direct phone dialin that Jim succeeded. > On Mar 30, 2015, at 12:11 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: > > Webex trash refused to let me connect by 3+ means so I cannot participate. > > Jeff > > Sent from my iPhone > > On Mar 30, 2015, at 5:37 AM, Daniel Holmes <dholmes(a)epcc.ed.ac.uk> wrote: > >> Hi All, >> >> The next point-to-point teleconference will be today, Monday 30th March 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: >> 0) Discuss/agree the to-do list (see below) that covers the next two agenda items. >> 1) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >> 2) 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 >> 3) Progress on Arecv and Fsend or ReceiveReduce? >> 4) Other stuff? >> >> Cheers, >> Dan. >> >> >> On 23/03/2015 11:59, Dan Holmes wrote: >>> Hi All, >>> >>> At the last telco meeting, Jim asked for a to-do list of the tasks needed to move communicator assertions forward. Please find below a first draft of such a task list. Could we agree that this is a good list? If you see something odd/missing, reply to this email group. >>> >>> To-do list for communicator assertions >>> >>> 0) Gather evidence to support "no-one will be affected by backwards incompatibility". >>> (e.g. chase Pavan, who volunteered to do this for major MPICH users) >>> 1) Change behaviour of MPI_COMM_DUP so that INFO keys are *not* propagated to new comm. >>> (include AtoU: if you want to dup INFO use MPI_COMM_DUP_WITH_INFO, with example) >>> 2) Discuss whether attributes (plus, anything else?) should be propagated by MPI_COMM_DUP. >>> (default presumption is yes, they should still be propagated). >>> 3) Find all occurrence of language in the MPI Standard that suggests INFO keys are hints. >>> a) Change or remove any generic statements like "INFO keys cannot change semantics". >>> (e.g. change to "cannot change observable semantics" or remove) >>> b) For each occurrence, determine if it should continue to suggest hint status >>> (e.g. a particular key, most likely for the I/O keys but maybe also some RMA keys) >>> >>> A) Convert ticket 381 into "make comm assertions doable" proposal (by Jun 2015 - Chicago?) >>> B) Get "no wildcard" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>> C) Get "no ordering" INFO key proposal ready for formal reading (by Jun 2015 - Chicago?) >>> D) Get "no cancel" INFO keys proposal ready for formal reading (by Jun 2015 - Chicago?) >>> E) Discuss other INFO key proposals, e.g. "only wildcard", "no non-blocking", "no underflow" >>> >>> Cheers, >>> Dan. >>> >>> On 13/03/2015 19:22, Daniel Holmes wrote: >>>> Hi All, >>>> >>>> The next point-to-point teleconference will be on Monday 16th March 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) "Can INFO keys change the semantic behaviour of MPI?" issue - following the discussion at the face-to-face meeting, we seem to have a workable way forward on this issue! >>>> 2) 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 >>>> 3) Progress on Arecv and Fsend or ReceiveReduce? >>>> 4) 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 > _______________________________________________ > mpiwg-p2p mailing list > mpiwg-p2p(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-p2p -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0