This is the mail archive of the
libc-help@sourceware.org
mailing list for the glibc project.
Re: [RFC][PATCH] Prototype for extended printf hooks
- From: "Ryan S. Arnold" <rsa at us dot ibm dot com>
- To: Jakub Jelinek <jakub at redhat dot com>, Ulrich Drepper <drepper at gmail dot com>
- Cc: libc-help <libc-help at sourceware dot org>, Ulrich Weigand <Ulrich dot Weigand at de dot ibm dot com>, Edmar Wienskoski <edmar at freescale dot com>, "H.J. Lu" <hjl at lucon dot org>, Steven Munroe <munroesj at us dot ibm dot com>, Andreas Krebbel1 <Andreas dot Krebbel at de dot ibm dot com>
- Date: Tue, 21 Oct 2008 13:40:31 -0500
- Subject: Re: [RFC][PATCH] Prototype for extended printf hooks
- Organization: IBM
- References: <1219441510.7917.62.camel@localhost>
- Reply-to: rsa at us dot ibm dot com
On Fri, 2008-08-22 at 16:45 -0500, Ryan S. Arnold wrote:
> Jakub, libc-help, et al.
>
> The linked patch (a GLIBC bugzilla attachment) is a prototype patch
> which adds fully featured printf hooks to GLIBC to support the C-spec
> Draft extension data types: DFP, Altivec/VMX, and VSX.
Ulrich, Jakub, et al.
Here's the third draft of the printf override hooks patch I've been
working on:
http://sourceware.org/bugzilla/attachment.cgi?id=3011&action=view
Associated with bugzilla 6846:
http://sourceware.org/bugzilla/show_bug.cgi?id=6846
This patch does the following:
- Removes the debugging messages.
- Adds __builtin_expect to branch predict around the new hooks.
- Conditionally initializes the arginfo array only if there's a user of
the overrides.
There still remain several issues:
- Adding fields to structs is probably not a good idea without
versioning interfaces.
- The old interfaces either need to be versioned or there need to be
entirely new interfaces for registering overrides.
I figured I'd worked on those once I got some direction on the plumbing
of the patch.
Ryan S. Arnold
IBM Linux Technology Center
Linux Toolchain Development