diff options
author | Siddhesh Poyarekar <siddhesh@sourceware.org> | 2017-10-05 21:49:37 +0530 |
---|---|---|
committer | Siddhesh Poyarekar <siddhesh@sourceware.org> | 2017-10-05 22:20:23 +0530 |
commit | 9ec87fd2b11ffce5c27c8eacb9deaa20e2373ff5 (patch) | |
tree | 5b1ea6ee495759a1935f054af2f62b14d9dc0ecb /benchtests/sqrt-inputs | |
parent | 1e9522c61c7a544d59db32cb7fbbd42e6793d848 (diff) | |
download | glibc-9ec87fd2b11ffce5c27c8eacb9deaa20e2373ff5.tar glibc-9ec87fd2b11ffce5c27c8eacb9deaa20e2373ff5.tar.gz glibc-9ec87fd2b11ffce5c27c8eacb9deaa20e2373ff5.tar.bz2 glibc-9ec87fd2b11ffce5c27c8eacb9deaa20e2373ff5.zip |
benchtests: Memory walking benchmark for memcpy
This benchmark is an attempt to eliminate cache effects from string
benchmarks. The benchmark walks both ways through a large memory area
and copies different sizes of memory and alignments one at a time
instead of looping around in the same memory area. This is a good
metric to have alongside the other memcpy benchmarks, especially for
larger sizes where the likelihood of the call being done only once is
pretty high.
* benchtests/bench-memcpy-walk.c: New file.
* benchtests/Makefile (string-benchset): Add it.
Diffstat (limited to 'benchtests/sqrt-inputs')
0 files changed, 0 insertions, 0 deletions