This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.

Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH] Nuke another 21 .plt slots

From: Jakub Jelinek <jakub at redhat dot com>
To: Ulrich Drepper <drepper at redhat dot com>
Cc: Roland McGrath <roland at frob dot com>, Glibc hackers <libc-hacker at sources dot redhat dot com>
Date: Tue, 6 Aug 2002 08:19:37 +0200
Subject: Re: [PATCH] Nuke another 21 .plt slots
References: <20020805222153.2082A1BA1B@perdition.linnaean.org> <3D4EFF01.1080003@redhat.com>
Reply-to: Jakub Jelinek <jakub at redhat dot com>

On Mon, Aug 05, 2002 at 03:41:05PM -0700, Ulrich Drepper wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Roland McGrath wrote:
> 
> > Then we should make -O3 the default, should we not?
> 
> No.  See my mail to rth from Sunday or Saturday.
> 
> (There is a bug in gcc where it inlines far too much when -O3 is used.)

If -O3 could inline only static or explicit inline routines, it would
be far less useful than it currently is (for the applications which benefit
from automatic inlining, of course, the majority does not).

Aren't we going to have a list of 10 or how many functions which need to
go through .plt and the rest does not? If yes,
then __attribute__ ((noinline)) for those 10 would suffice.

	Jakub

Follow-Ups:
- Re: [PATCH] Nuke another 21 .plt slots
  - From: Ulrich Drepper

References:
- Re: [PATCH] Nuke another 21 .plt slots
  - From: Roland McGrath
- Re: [PATCH] Nuke another 21 .plt slots
  - From: Ulrich Drepper

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]