This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: porting EGCS to the Cray T3E
- To: law at cygnus dot com
- Subject: Re: porting EGCS to the Cray T3E
- From: "Julian C. Cummings" <julianc at acl dot lanl dot gov>
- Date: Tue, 14 Jul 1998 11:46:10 -0600
- Cc: egcs at cygnus dot com
- References: <16363.900436834@hurl.cygnus.com>
On Jul 14, 11:20am, Jeffrey A Law wrote:
> Subject: Re: porting EGCS to the Cray T3E
>
> In message <9807141122.ZM4885@vapor.acl.lanl.gov>you write:
> > I guess I don't understand your comment. There are other Cray platforms
listed
> > as possibilities in the config.sub file, such as the Cray X-MP, Y-MP,
Cray 2,
> > and Cray [C,J,T]90. Why would these options be listed if Cray is
> > not supported?
> They may be supported as host or build machines, but that does not
> mean they are supported as a target.
>
> ie, gcc does not know how to generate code for any cray that I'm aware of.
OK. As far as I can tell, the egcs build instructions do not make any sort of
distinction like this between "host" and "target". The instructions imply that
these are the same thing. Nevertheless, I tried the suggestion I received of
setting the host to "alpha-cray-unicosmk" to see if that works. It does not.
I get the exact same behavior as before. It works for a while, then says
Configuration alpha-cray-unicosmk not supported
Configure in /usr/tmp/julianc/EGCS/egcs-objdir-alpha/gcc failed, exiting.
It sounds like you're saying that egcs cannot generate code for a Cray. But
these are DEC Alpha processors, so I don't see why not.
Is there some sort of verbosity switch I can throw that might tell me why this
is failing? There is nothing telltale in the config.log file, as you saw.
Julian C.
--
Julian C. Cummings
Advanced Computing Laboratory
Los Alamos National Laboratory
(505) 667-6064
julianc@acl.lanl.gov