This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

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: C++ compile time (again)

From: Diego Novillo <dnovillo at redhat dot com>
To: Wolfgang Bangerth <bangerth at ices dot utexas dot edu>
Cc: Karel Gardas <kgardas at objectsecurity dot com>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>
Date: 17 Jun 2003 09:55:49 -0400
Subject: Re: C++ compile time (again)
Organization: Red Hat Canada
References: <Pine.LNX.4.43.0306171529510.26402-100000@thinkpad.c0202001.roe.itnq.net> <1055857168.8479.97.camel@frodo.toronto.redhat.com> <200306170853.05157.bangerth@ices.utexas.edu>

On Tue, 2003-06-17 at 09:53, Wolfgang Bangerth wrote:
> > > gcc2.95.4: 6m4s
> > > gcc3.4:    9m24s
> > > gcc3.5:   11m17s
> >
> > You haven't used optimization?  If not, then timings are not including
> > the SSA transformations. 
> 
> I thought the same, but then what's going on? No SSA used and still roughly 
> 20% slower?
> 
Gimplification tends to create lots more RTL for the back end to go
through.  We need to address that.


Diego.

Follow-Ups:
- Re: C++ compile time (again)
  - From: Wolfgang Bangerth

References:
- Re: C++ compile time (again)
  - From: Karel Gardas
- Re: C++ compile time (again)
  - From: Diego Novillo
- Re: C++ compile time (again)
  - From: Wolfgang Bangerth

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