This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Add TUNABLES to control IFUNC selection and cache sizes
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Siddhesh Poyarekar <siddhesh at gotplt dot org>
- Cc: GNU C Library <libc-alpha at sourceware dot org>, Erich Elsen <eriche at google dot com>
- Date: Fri, 9 Jun 2017 04:16:42 -0700
- Subject: Re: [PATCH] Add TUNABLES to control IFUNC selection and cache sizes
- Authentication-results: sourceware.org; auth=none
- References: <20170608200335.GA30996@gmail.com> <464b2275-e431-4888-1047-9523ceec3790@gotplt.org>
On Thu, Jun 8, 2017 at 10:45 PM, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:
> On Friday 09 June 2017 01:33 AM, H.J. Lu wrote:
>> The current IFUNC selection is based on microbenchmarks in glibc. It
>> should give the best performance for most workloads. But other choices
>> may have better performance for a particular workload or on the hardware
>> which wasn't available at the selection was made. The environment
>> variable, GLIBC_IFUNC=-xxx,yyy,-zzz...., can be used to enable CPU/ARCH
>> feature yyy, disable CPU/ARCH feature yyy and zzz, where the feature
>> name is case-sensitive and has to match the ones in cpu-features.h. It
>> can be used by glibc developers to override the IFUNC selection to tune
>> for a new processor or improve performance for a particular workload.
>> It isn't intended for normal end users.
>
> We have agreed in the past that env_alias is only a compatibility hack
> to allow existing environment variables to be ported into tunables so
> please don't add new ones. glibc.tune.ifunc should be the only way to
> select ifuncs.
>
> Siddhesh
Does that mean tunables must be enabled to use it? Will tunables
be enabled by default?
--
H.J.