[Openmp-commits] [openmp] r320152 - Use hyperbarrier by default on all architectures

[Jonas Hahnfeld via Openmp-commits]

Author: hahnfeld
Date: Fri Dec  8 07:07:07 2017
New Revision: 320152

URL: http://llvm.org/viewvc/llvm-project?rev=320152&view=rev
Log:
Use hyperbarrier by default on all architectures

All architectures except x86_64 used the linear barrier implementation
by default which doesn't give good performance for a larger number
of threads.

Improvements for PARALLEL overhead (EPCC) with this patch on a Power8
system (2 sockets x 10 cores x 8 threads, OMP_PLACES=cores)

 20 threads:  4.55us -> 3.49us
 40 threads:  8.84us -> 4.06us
 80 threads: 19.18us -> 4.74us
160 threads: 54.22us -> 6.73us

Differential Revision: https://reviews.llvm.org/D40358

Modified:
    openmp/trunk/runtime/src/kmp_global.cpp

Modified: openmp/trunk/runtime/src/kmp_global.cpp
URL: http://llvm.org/viewvc/llvm-project/openmp/trunk/runtime/src/kmp_global.cpp?rev=320152&r1=320151&r2=320152&view=diff
==============================================================================
--- openmp/trunk/runtime/src/kmp_global.cpp (original)
+++ openmp/trunk/runtime/src/kmp_global.cpp Fri Dec  8 07:07:07 2017
@@ -76,25 +76,16 @@ size_t __kmp_malloc_pool_incr = KMP_DEFA
 
 // Barrier method defaults, settings, and strings.
 // branch factor = 2^branch_bits (only relevant for tree & hyper barrier types)
-#if KMP_ARCH_X86_64
 kmp_uint32 __kmp_barrier_gather_bb_dflt = 2;
 /* branch_factor = 4 */ /* hyper2: C78980 */
 kmp_uint32 __kmp_barrier_release_bb_dflt = 2;
 /* branch_factor = 4 */ /* hyper2: C78980 */
-#else
-kmp_uint32 __kmp_barrier_gather_bb_dflt = 2;
-/* branch_factor = 4 */ /* communication in core for MIC */
-kmp_uint32 __kmp_barrier_release_bb_dflt = 2;
-/* branch_factor = 4 */ /* communication in core for MIC */
-#endif // KMP_ARCH_X86_64
-#if KMP_ARCH_X86_64
-kmp_bar_pat_e __kmp_barrier_gather_pat_dflt = bp_hyper_bar; /* hyper2: C78980 */
-kmp_bar_pat_e __kmp_barrier_release_pat_dflt =
-    bp_hyper_bar; /* hyper2: C78980 */
-#else
-kmp_bar_pat_e __kmp_barrier_gather_pat_dflt = bp_linear_bar;
-kmp_bar_pat_e __kmp_barrier_release_pat_dflt = bp_linear_bar;
-#endif
+
+kmp_bar_pat_e __kmp_barrier_gather_pat_dflt = bp_hyper_bar;
+/* hyper2: C78980 */
+kmp_bar_pat_e __kmp_barrier_release_pat_dflt = bp_hyper_bar;
+/* hyper2: C78980 */
+
 kmp_uint32 __kmp_barrier_gather_branch_bits[bs_last_barrier] = {0};
 kmp_uint32 __kmp_barrier_release_branch_bits[bs_last_barrier] = {0};
 kmp_bar_pat_e __kmp_barrier_gather_pattern[bs_last_barrier] = {bp_linear_bar};