Re: HEADS UP: GCC 3.4.6 update in progress

From: Steve Kargl <sgk_at_troutmask.apl.washington.edu>
Date: Wed, 30 Aug 2006 12:17:22 -0700
On Tue, Aug 29, 2006 at 09:59:36PM -0500, Conrad J. Sabatier wrote:
> On Tue, 29 Aug 2006 19:33:36 -0700, Steve Kargl
> <sgk_at_troutmask.apl.washington.edu> wrote:
> > On Tue, Aug 29, 2006 at 08:29:16PM -0500, Conrad J. Sabatier wrote:
> > > On Tue, 29 Aug 2006 16:04:47 -0700, Steve Kargl
> > > <sgk_at_troutmask.apl.washington.edu> wrote:
> > > 
> > > > On Tue, Aug 29, 2006 at 05:50:22PM -0500, Conrad J. Sabatier
> > > > wrote:
> > > > > 
> > > > > Incidentally, I've been wondering for the longest time now:
> > > > > is/are there any particular reason(s) why we exclude gcj from
> > > > > the gcc tree/build? Sure would be a cool thing to have.
> > > > > 
> > > > 
> > > > 1) It's not needed to build anything in the base system.
> > > > 2) It takes a long time to build.
> > > > 3) You really want to use the gcc41 or gcc42 port.
> > > > 
> > > > Of course, IMHO.
> > > 
> > > I've tried both ports, but gcj fails to run, complaining about a
> > > missing "spec" file (I forget the exact filename, maybe
> > > "gcjlib.spec" or something similar).  :-(
> > > 
> > > Is this maybe an arch-specific (amd64) problem?  Or?
> > > 
> > 
> > Contact Gerald.  He is fairly quick at addressing port
> > issues.  I don't build the ports because I build GCC
> > svn sources for 4.1.x and trunk several times a week.
> > I'll note that I don't normally build gjc.  I can try
> > a bootstrap tomorrow.
> 
> OK, will do.  Thanks.

For those interesting in bootstrap time.

% time gmake bootstrap
     4456.51 real      4098.55 user       349.61 sys

This includes C, C++, Fortran, and Java.  These numbers are from
a dual-processor, dual-core opteron system running at 2.4 GHz
and with 16 GB of memory.  The system was otherwise idle.
-- 
Steve
Received on Wed Aug 30 2006 - 17:17:31 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:59 UTC