This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Mechanism to configure default scheduling model on PA
- From: law at redhat dot com
- To: "John David Anglin" <dave at hiauly1 dot hia dot nrc dot ca>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 31 May 2002 12:58:48 -0600
- Subject: Re: Mechanism to configure default scheduling model on PA
- Reply-to: law at redhat dot com
In message <200205311843.g4VIhRDk006815@hiauly1.hia.nrc.ca>, "John David
Anglin" writes:
> Yes, it seems reasonable to default to the most advanced architecture
> that the target supports. That should give the best overall performance.
> It will also provide a better range of testing for each architecture.
My suggestion was to default to whatever machine you're building the compiler
on. Not the most modern architecture.
> I still like the idea of being able to configure the default in case a
> more portable default is desired.
If we have the model where we default to the architecture you're building
on, then you can change it by explicitly specifying what target you want to
build.
ie, let's pretend we're on an hppa2.0w-hp-hpux11 system and we just
"configure; make bootstrap; make install". That would default to PA2.0.
However, you could get the default to be PA1.0 by doing the following:
"configure hppa1.0-hp-hpux11; make bootstrap; make install".
Similarly if I was on an "hppa1.1-hp-hpux11", then I'm going to get PA1.1
by default. However, I could explicitly ask for something different like
"configure hppa2.0w-hp-hpux11". Of course you couldn't bootstrap as the
compiler would generate PA2.0 instructions which wouldn't work on your
PA1.1 machine :-)
jeff