Brice bash-3.2$ uname -a AIX pandora1 1 6 00F63F144C00
bash-3.2$ ./lstopo-no-graphics --version lstopo-no-graphics 1.7.2 bash-3.2$ ./lstopo-no-graphics - | grep "PU L#0" PU L#0 (P#0) -erik On Wed, Dec 17, 2014 at 9:34 AM, Brice Goglin <brice.gog...@inria.fr> wrote: > Hello > > I am seeing assert failures on AIX 6.1 because our PU os_index is off by > one. They go from -1 to 62 instead of 0 to 63. > > We have a comment saying > > /* It seems logical processors are numbered from 1 here, while the > * bindprocessor functions numbers them from 0... */ > > This contradicts the doc of rs_getrad() which says index starts from 0 > > index > Specifies the index of the system RAD that should be returned > from among those at the specified sdl. This parameter > must belong to the [0, rs_numrads(rset, sdl, flags)- 1] interval. > > If you have AIX, could you please run > lstopo - | grep "PU L#0" > and report the output as well as you AIX version? > If the line is only "PU L#0", the index is -1 (not displayed), we're > buggy (what I see on AIX 6.1). > If the line is "PU L#0 P#0", the index is OK. > I hope we won't have different behaviors... > > thanks > Brice > > _______________________________________________ > hwloc-users mailing list > hwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2014/12/1135.php -- Erik Schnetter <schnet...@cct.lsu.edu> http://www.perimeterinstitute.ca/personal/eschnetter/