This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC] Porting string performance tests into benchtests
- From: "Carlos O'Donell" <carlos at redhat dot com>
- To: Siddhesh Poyarekar <siddhesh dot poyarekar at gmail dot com>
- Cc: Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>, GNU C Library <libc-alpha at sourceware dot org>
- Date: Sat, 06 Apr 2013 09:55:02 -0400
- Subject: Re: [RFC] Porting string performance tests into benchtests
- References: <20130403101130 dot GE20842 at spoyarek dot pnq dot redhat dot com> <20130403 dot 123522 dot 1616212976811705615 dot davem at davemloft dot net> <20130404033719 dot GA14860 at spoyarek dot pnq dot redhat dot com> <20130403 dot 234042 dot 1776194180184022553 dot davem at davemloft dot net> <20130405043801 dot GR14860 at spoyarek dot pnq dot redhat dot com> <515F298D dot 6020908 at linux dot vnet dot ibm dot com> <CAAHN_R3AZKGY-FUK1bjPjAphEMCjUmwJRhvpM-W8-7VgjyLNWQ at mail dot gmail dot com>
On 04/05/2013 09:02 PM, Siddhesh Poyarekar wrote:
> On 6 April 2013 01:14, Adhemerval Zanella <azanella@linux.vnet.ibm.com> wrote:
>> I'm not sure if it is worth, but maybe it would be useful to add a configurable
>> timestamp calculation.
>
> Yes, it would be useful. I don't think I'll spend time on that now.
> I'll first update this to use HP_TIMING and then there's the laborious
> task of adding all functions imaginable to the benchmark list. It
> would be nice to have a baseline for as many functions as possible in
> 2.18.
Agreed. A baseline, even if imprecise, and thus a lower-bound on performance,
is useful for tracking regressions. As the precision of the time measurement
increases the performance should get better.
Cheers,
Carlos.