Use HP_TIMING for benchmarks if available

HP_TIMING uses native timestamping instructions if available, thus
greatly reducing the overhead of recording start and end times for
function calls.  For architectures that don't have HP_TIMING
available, we fall back to the clock_gettime bits.  One may also
override this by invoking the benchmark as follows:

  make USE_CLOCK_GETTIME=1 bench

and get the benchmark results using clock_gettime.  One has to do
`make bench-clean` to ensure that the benchmark programs are rebuilt.

1 files changed, 7 insertions, 0 deletions

diff --git a/ChangeLog b/ChangeLog
index 8a12bf9898..ea678c8295 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,12 @@
 2013-05-13  Siddhesh Poyarekar  <siddhesh@redhat.com>
 
+	* benchtests/Makefile (CPPFLAGS-nonlib): Add
+	-DUSE_CLOCK_GETTIME if USE_CLOCK_GETTIME is defined.
+	(bench-deps): Add bench-timing.h.
+	* benchtests-bench-skeleton.c: Include bench-timing.h.
+	(main): Use TIMING_* macros instead of clock_gettime.
+	* benchtests/bench-timing.h: New file.
+
 	[BZ #14582]
 	* sysdeps/ieee754/s_lib_version.c (_LIB_VERSION_INTERNAL):
 	Renamed from _LIB_VERSION.