glibc/benchtests
Siddhesh Poyarekar 43fe811b73 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.
2013-05-13 13:44:32 +05:30
..
Makefile Use HP_TIMING for benchmarks if available 2013-05-13 13:44:32 +05:30
atan-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
bench-modf.c Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
bench-skeleton.c Use HP_TIMING for benchmarks if available 2013-05-13 13:44:32 +05:30
bench-timing.h Use HP_TIMING for benchmarks if available 2013-05-13 13:44:32 +05:30
cos-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
exp-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
pow-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
rint-inputs PowerPC: remove branch prediction from rint implementation 2013-04-01 06:36:51 -05:00
sin-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30
tan-inputs Allow multiple input domains to be run in the same benchmark program 2013-04-30 14:17:57 +05:30