This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Using crlibm as the default math library in GCC sources
- From: Uros Bizjak <ubizjak at gmail dot com>
- To: Uros Bizjak <ubizjak at gmail dot com>, GCC <gcc at gcc dot gnu dot org>, Florent de Dinechin <Florent dot de dot Dinechin at ens-lyon dot fr>, christoph dot lauter at ens-lyon dot fr, rsandifo at nildram dot co dot uk
- Cc: David Edelsohn <dje at watson dot ibm dot com>
- Date: Sun, 11 Nov 2007 11:05:43 +0100
- Subject: Re: Using crlibm as the default math library in GCC sources
- References: <4730C1CE.2030300@gmail.com> <87hcjujbid.fsf@firetop.home>
Richard Sandiford wrote:
Uros Bizjak <ubizjak@gmail.com> writes:
Forwarded is a short discussion with Mr. de Dinechin (forwarded with
permission), where the possibility to import crlibm as the default gcc
math library is discussed. I would like to ask members of the GCC SC for
the opinion on the idea of adopting crlibm as the default math library.
This way, the library can be compiled using all knowledge that gcc has
about the target processor. As hinted in the attached message -
autovectorization and perhaps recently introduced fixed-point
infrastructure can be used effectively in the implementation of this
library, as well as new AMD's FMA instructions.
That sounds like a great idea. Do you know if the SC is already
considering it? Or is it one of those cases where there needs
to be a bit more technical discussion first?
I don't have any further information from SC side. I have CC'd David in
the hope that perhaps he can update us with some info.
Uros.