This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: malloc() and spinlocks
- From: Ulrich Drepper <drepper at redhat dot com>
- To: Wolfram Gloger <Wolfram dot Gloger at dent dot med dot uni-muenchen dot de>
- Cc: aj at suse dot de, libc-alpha at sources dot redhat dot com
- Date: Mon, 13 Jan 2003 02:49:14 -0800
- Subject: Re: malloc() and spinlocks
- Organization: Red Hat, Inc.
- References: <200211251733.14550.l.lunak@suse.cz> <200211261345.OAA22431@max.zk-i.med.uni-muenchen.de> <hoel7h9xtv.fsf@byrd.suse.de> <200301131038.LAA11752@max.zk-i.med.uni-muenchen.de>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Wolfram Gloger wrote:
> I've still got on my TODO to check how NPTL handles fast inline
> spinlocks, to avoid that \approx1.5% penalty for the single-threaded
> case. As I said, the patch could also be applied as-is to get the
> \approx10% improvement for the threaded case on ix86.
The reason I don't like this patch is that spinlocks do not work in RT
processes. Even with multiple arenas you can get into a situation where
nothing works. This is why the nptl version uses futexes and because we
cannot depend on futexes for the LinuxThreads version we should not add it.
- --
- --------------. ,-. 444 Castro Street
Ulrich Drepper \ ,-----------------' \ Mountain View, CA 94041 USA
Red Hat `--' drepper at redhat.com `---------------------------
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQE+Ipmq2ijCOnn/RHQRAthVAKC4BFjsRvP00Ll7lp+a6BSqjIWQ0ACdEgme
ocRsjwW7aD02pLnCgRWI/Pk=
=pVAy
-----END PGP SIGNATURE-----