This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
[PATCH 0/1] benchtest: Run benchmark only once for each input
- From: Ashwin Sekhar T K <ashwin dot sekhar at caviumnetworks dot com>
- To: libc-alpha at sourceware dot org
- Cc: Ashwin Sekhar T K <ashwin dot sekhar at caviumnetworks dot com>
- Date: Thu, 6 Jul 2017 01:51:51 -0700
- Subject: [PATCH 0/1] benchtest: Run benchmark only once for each input
- Authentication-results: sourceware.org; auth=none
- Authentication-results: sourceware.org; dkim=none (message not signed) header.d=none;sourceware.org; dmarc=none action=none header.from=caviumnetworks.com;
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
This patch addresses the change proposed to the benchtest framework in the
following discussion.
https://sourceware.org/ml/libc-alpha/2017-06/msg00540.html
Basically this patch avoids running the benchmark for each input in a loop.
Instead it runs the benchmark only once for each input, so that the effects
of branches, caches etc. are not nullified.
Ashwin Sekhar T K (1):
benchtest: Run benchmark only once for each input
benchtests/bench-skeleton.c | 47 +++++++++++++++++++++------------------------
benchtests/scripts/bench.py | 12 ++++++++++--
2 files changed, 32 insertions(+), 27 deletions(-)
--
2.12.2