diff options
| author | Aliaksey Kandratsenka <alk@tut.by> | 2015-05-03 12:55:47 -0700 |
|---|---|---|
| committer | Aliaksey Kandratsenka <alk@tut.by> | 2015-08-02 16:53:19 -0700 |
| commit | 64e0133901a20f83c41adb36748fd19d21228515 (patch) | |
| tree | 86d91f813f7c800fc00d4027419d1bc9170b0ead /Makefile.am | |
| parent | e1d1311cfb6312cd44e086c879f3e95cbfa0eb9d (diff) | |
| download | gperftools-64e0133901a20f83c41adb36748fd19d21228515.tar.gz | |
added trivial malloc fast-path benchmark
While this is not good representation of real-world production malloc
behavior, it is representative of length (instruction-wise and well as
cycle-wise) of fast-path. So this is better than nothing.
Diffstat (limited to 'Makefile.am')
| -rwxr-xr-x | Makefile.am | 12 |
1 files changed, 12 insertions, 0 deletions
diff --git a/Makefile.am b/Makefile.am index a5a9af8..bae605a 100755 --- a/Makefile.am +++ b/Makefile.am @@ -847,6 +847,18 @@ endif WITH_STACK_TRACE endif WITH_DEBUGALLOC +noinst_PROGRAMS += malloc_bench malloc_bench_shared + +malloc_bench_SOURCES = benchmark/malloc_bench.cc +malloc_bench_CXXFLAGS = $(PTHREAD_CFLAGS) $(AM_CXXFLAGS) $(NO_BUILTIN_CXXFLAGS) +malloc_bench_LDFLAGS = $(PTHREAD_CFLAGS) $(TCMALLOC_FLAGS) -static +malloc_bench_LDADD = libtcmalloc_minimal.la $(PTHREAD_LIBS) + +malloc_bench_shared_SOURCES = benchmark/malloc_bench.cc +malloc_bench_shared_CXXFLAGS = $(PTHREAD_CFLAGS) $(AM_CXXFLAGS) $(NO_BUILTIN_CXXFLAGS) +malloc_bench_shared_LDFLAGS = $(PTHREAD_CFLAGS) $(TCMALLOC_FLAGS) +malloc_bench_shared_LDADD = libtcmalloc_minimal.la $(PTHREAD_LIBS) + ### ------- tcmalloc (thread-caching malloc + heap profiler + heap checker) |