This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Avoid PLT when calling __sched_getaffinity_new
- From: Joseph Myers <joseph at codesourcery dot com>
- To: Andreas Schwab <schwab at suse dot de>
- Cc: "H.J. Lu" <hjl dot tools at gmail dot com>, Florian Weimer <fweimer at redhat dot com>, GNU C Library <libc-alpha at sourceware dot org>
- Date: Wed, 14 Oct 2015 16:54:10 +0000
- Subject: Re: [PATCH] Avoid PLT when calling __sched_getaffinity_new
- Authentication-results: sourceware.org; auth=none
- References: <20150821170918 dot GA29943 at intel dot com> <561E4D6C dot 9040303 at redhat dot com> <CAMe9rOqT6cGee5iXWbqPfvZqYf=z53O-2LggQ-xGkMwQ+o1XAA at mail dot gmail dot com> <alpine dot DEB dot 2 dot 10 dot 1510141502410 dot 20331 at digraph dot polyomino dot org dot uk> <mvmfv1dtqag dot fsf at hawking dot suse dot de>
On Wed, 14 Oct 2015, Andreas Schwab wrote:
> It's highly x86-specific. Most other architectures don't need a special
> register set up just for doing a PLT call, so it doesn't matter whether
> it's relaxed by the compiler or the linker.
True - the main benefit of an architecture-independent test would be to
make it less likely that people check in changes introducing new
non-hidden calls on x86 because they tested on non-x86, rather than
because it actually affects the generated code on other architectures.
--
Joseph S. Myers
joseph@codesourcery.com