Rainer Keller 4e6a6fc146 - Check, whether the compiler supports __builtin_clz (count leading ... zeroes); if so, use it for bit-operations like opal_cube_dim and opal_hibit. Implement two versions of power-of-two. In case of opal_next_poweroftwo, this reduces the average execution time from 83 cycles to 4 cycles (Intel Nehalem, icc, -O2, inlining, measured rdtsc, with loop over 2^27 values). Numbers for other functions are similar (but of course heavily depend on the usage, e.g. opal_hibit() with a start of 4 does not save much). The bsr instruction on AMD Opteron is also not as fast. - Replace various places where the next power-of-two is computed. Tested on Intel Nehalem Cluster with openib, compilers GNU-4.6.1 and Intel-12.0.4 using mpi_testsuite -t "Collective" with 128 processes. This commit was SVN r25270.		2011-10-11 22:49:01 +00:00
..
coll_tuned_allgather.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_allgatherv.c	- Split the datatype engine into two parts: an MPI specific part in	2009-07-13 04:56:31 +00:00
coll_tuned_allreduce.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_alltoall.c	Don't forget to initialize "line" in all cases.	2010-05-19 21:19:45 +00:00
coll_tuned_alltoallv.c	- Split the datatype engine into two parts: an MPI specific part in	2009-07-13 04:56:31 +00:00
coll_tuned_barrier.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_bcast.c	- Split the datatype engine into two parts: an MPI specific part in	2009-07-13 04:56:31 +00:00
coll_tuned_component.c	Rework the selection logic for the tuned collectives. All supported collectives	2009-08-14 21:06:23 +00:00
coll_tuned_decision_dynamic.c	Rework the selection logic for the tuned collectives. All supported collectives	2009-08-14 21:06:23 +00:00
coll_tuned_decision_fixed.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_dynamic_file.c	... Delayed due to notifier commits earlier this day ...	2009-04-29 01:32:14 +00:00
coll_tuned_dynamic_file.h	- Replace combinations of	2009-08-20 11:42:18 +00:00
coll_tuned_dynamic_rules.c	Rework the selection logic for the tuned collectives. All supported collectives	2009-08-14 21:06:23 +00:00
coll_tuned_dynamic_rules.h	- Replace combinations of	2009-08-20 11:42:18 +00:00
coll_tuned_gather.c	Only get the receive datatype extent on the root process, as every	2010-02-17 16:01:50 +00:00
coll_tuned_module.c	Ensure that the com_rules[] array entries are initialized to NULL in	2010-07-07 14:04:18 +00:00
coll_tuned_reduce_scatter.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_reduce.c	Rework the selection logic for the tuned collectives. All supported collectives	2009-08-14 21:06:23 +00:00
coll_tuned_scatter.c	- Split the datatype engine into two parts: an MPI specific part in	2009-07-13 04:56:31 +00:00
coll_tuned_topo.c	- Check, whether the compiler supports __builtin_clz (count leading	2011-10-11 22:49:01 +00:00
coll_tuned_topo.h	- Replace combinations of	2009-08-20 11:42:18 +00:00
coll_tuned_util.c	Cleanups.	2011-02-25 00:28:32 +00:00
coll_tuned_util.h	- Replace combinations of	2009-08-20 11:42:18 +00:00
coll_tuned.h	Rework the selection logic for the tuned collectives. All supported collectives	2009-08-14 21:06:23 +00:00
Makefile.am	WARNING: Work on the temp branch being merged here encountered problems with bugs in subversion. Considerable effort has gone into validating the branch. However, not all conditions can be checked, so users are cautioned that it may be advisable to not update from the trunk for a few days to allow MTT to identify platform-specific issues.	2010-09-17 23:04:06 +00:00