Rich, It appears the the udapl btl is not even building in your tmp branch.
-DON ... "btl_udapl.c", line 305: syntax error before or at: ; "btl_udapl.c", line 315: syntax error before or at: ) "btl_udapl.c", line 374: cannot recover from previous errors cc: acomp failed for btl_udapl.c make[2]: *** [btl_udapl.lo] Error 1 make[2]: *** Waiting for unfinished jobs.... make[2]: Leaving directory `/opt/mtt/64/non-threaded/free-list-branch-testing/installs/3m5g/src/free_list/ompi/mca/btl/udapl' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/opt/mtt/64/non-threaded/free-list-branch-testing/installs/3m5g/src/free_list/ompi' make: *** [all-recursive] Error 1 Richard Graham wrote:
We are looking at making some changes to the ompi free list in ompi/class/ ompi_free_list.[c,h] , and are trying to decide if to go ahead with an interface change that will allow separate control over alignment of the frag and payload data structures. We are aware of several implementations of btl's and mtl's that are not part of the Open MPI repository, and want to minimize changes to these. We have an option to make these changes by duplicating code, and creating a new set of functions to handle the alignment (and other cleanup) issues, not perturbing existing code. The questions are, does anyone have code out there that uses the ompi free lists, and if so would would an interface change to the init functions cause you trouble ? Thanks, Rich _______________________________________________ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel
