Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] Trunk fails to build with --disable-dlopen
From: Nathan Hjelm (hjelmn_at_[hidden])
Date: 2014-07-28 18:00:47


Ok, got --disable-dlopen working again. I removed the code in question
and changed how coll/sm shares the segment data.

-Nathan

On Mon, Jul 28, 2014 at 02:41:37PM -0600, Nathan Hjelm wrote:
>
> Or pull it into coll/sm. Though I think we can do better here since
> point-to-point messaging can be used in coll/sm. We can use the
> netpatterns code to share the segment information.
>
> -Nathan
>
> On Mon, Jul 28, 2014 at 08:37:15PM +0000, Jeff Squyres (jsquyres) wrote:
> > Perhaps that RML code can go back up in ompi/common/sm...? (since only ompi/coll/sm uses it)
> >
> >
> > On Jul 28, 2014, at 4:34 PM, Nathan Hjelm <hjelmn_at_[hidden]> wrote:
> >
> > >
> > > Damn, spoke too soon. coll/sm uses it:
> > >
> > > ./ompi/mca/coll/sm/coll_sm_module.c: mca_common_sm_init_group(comm->c_local_group, size, fullpath,
> > > ./ompi/mca/coll/sm/coll_sm_module.c: "coll:sm:enable:bootstrap comm (%d/%s): mca_common_sm_init_group failed",
> > >
> > >
> > > Let me see if I can resolve that one.
> > >
> > > -Nathan
> > >
> > > On Mon, Jul 28, 2014 at 02:14:36PM -0600, Nathan Hjelm wrote:
> > >>
> > >> Looks like you are correct. The function that calls the rml code is
> > >> mca_common_sm_init which is no longer called by anything (other than
> > >> mca_common_sm_init_group.. which isn't called either). Let me see if I
> > >> can fix this. I need this build working again with --disable-dlopen.
> > >>
> > >> mu-fey:/usr/projects/hpctools/hjelmn/ompi-trunk-git hjelmn$ find opal -name \*.[ch] | xargs grep mca_common_sm_init
> > >> opal/mca/common/sm/common_sm.c:mca_common_sm_init(opal_proc_t **procs,
> > >> opal/mca/common/sm/common_sm.c:mca_common_sm_init_group(ompi_group_t *group,
> > >> opal/mca/common/sm/common_sm.c: /* let mca_common_sm_init take care of the rest ... */
> > >> opal/mca/common/sm/common_sm.c: ret = mca_common_sm_init(procs, group_size, size, file_name,
> > >> opal/mca/common/sm/common_sm.h:mca_common_sm_init(opal_proc_t **procs,
> > >> opal/mca/common/sm/common_sm.h:mca_common_sm_init_group(ompi_group_t *group,
> > >>
> > >>
> > >> On Mon, Jul 28, 2014 at 01:05:50PM -0700, Ralph Castain wrote:
> > >>> I'm not sure the sm actually relies on the RML any more - I thought we had removed that dependency, though the file may not have been deleted.
> > >>>
> > >>> On Jul 28, 2014, at 1:02 PM, Nathan Hjelm <hjelmn_at_[hidden]> wrote:
> > >>>
> > >>>>
> > >>>> The trunk is totally broken and it might not be easy to fix. I am seeing
> > >>>> this error when building with --disable-dlopen (the LANL default):
> > >>>>
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/mca/common/sm/common_sm_rml.c: In function 'mca_common_sm_rml_info_bcast':
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/mca/common/sm/common_sm_rml.c:110: warning: passing argument 1 of 'orte_rml.send_buffer_nb' from incompatible pointer type
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/mca/common/sm/common_sm_rml.c:110: note: expected 'struct orte_process_name_t *' but argument is of type 'opal_process_name_t *'
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/mca/common/sm/common_sm_rml.c:127: warning: passing argument 1 of 'orte_rml.recv_buffer_nb' from incompatible pointer type
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/mca/common/sm/common_sm_rml.c:127: note: expected 'struct orte_process_name_t *' but argument is of type 'opal_process_name_t *'
> > >>>> /usr/projects/hpctools/hjelmn/ompi-trunk-git/opal/runtime/opal_init.c:69:2: warning: #ident is a GCC extension
> > >>>> ../../../opal/.libs/libopen-pal.so: undefined reference to `orte_rml'
> > >>>> ../../../opal/.libs/libopen-pal.so: undefined reference to `orte_proc_is_bound'
> > >>>> ../../../opal/.libs/libopen-pal.so: undefined reference to `orte_rml_send_callback'
> > >>>>
> > >>>>
> > >>>> common/sm relies on the RML for some communication. This will have to be
> > >>>> rewritten to get it to work.
> > >>>>
> > >>>> -Nathan
> > >>>> _______________________________________________
> > >>>> devel mailing list
> > >>>> devel_at_[hidden]
> > >>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> > >>>> Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15292.php
> > >>>
> > >>> _______________________________________________
> > >>> devel mailing list
> > >>> devel_at_[hidden]
> > >>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> > >>> Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15294.php
> > >
> > >
> > >
> > >> _______________________________________________
> > >> devel mailing list
> > >> devel_at_[hidden]
> > >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> > >> Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15295.php
> > >
> > > _______________________________________________
> > > devel mailing list
> > > devel_at_[hidden]
> > > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> > > Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15296.php
> >
> >
> > --
> > Jeff Squyres
> > jsquyres_at_[hidden]
> > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
> >
> > _______________________________________________
> > devel mailing list
> > devel_at_[hidden]
> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> > Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15297.php

> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> Link to this post: http://www.open-mpi.org/community/lists/devel/2014/07/15298.php



  • application/pgp-signature attachment: stored