This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: PATCH: Fix the MIPS ISA ELF setting (Re: RedHat 7.1/mips update)
- To: cgd at broadcom dot com
- Subject: Re: PATCH: Fix the MIPS ISA ELF setting (Re: RedHat 7.1/mips update)
- From: "H . J . Lu" <hjl at lucon dot org>
- Date: Thu, 1 Nov 2001 09:27:21 -0800
- Cc: jim at jtan dot com, binutils at sources dot redhat dot com
- References: <20011026001025.A7648@lucon.org> <20011026093248.A16729@lucon.org> <1004664608.18027.7.camel@ghostwheel.cygnus.com> <20011031213759.A9712@lucon.org> <20011101044439.A7502@neurosis.mit.edu> <mailpost.1004607896.9172@postal.sibyte.com> <yov5668ucss0.fsf@broadcom.com>
On Thu, Nov 01, 2001 at 09:20:15AM -0800, cgd@broadcom.com wrote:
> jim@jtan.com ("Jim Paris") writes:
> > > I was told he was happy with my changes.
> >
> > Yep, that's correct. It doesn't really matter to me whether the
> > compiler is doing any tricky platform-specific stuff; I just needed
> > access to particular instructions from inline assembly. H.J's patch and
> >
> > -mips2 -Wa,-m4100
> >
> > work perfectly.
>
> I'd suggest that, at least to my mind, supporting ".set <random
> architecture>" is the right way to accomplish this.
>
> Right now you can do e.g.:
>
> .set push
> .set mips4
> instructions
> .set pop
>
> but you can't do that for any architecture string. I don't think it'd
> be hard to do, it's just a matter of doing it.
I believe it is the wrong approach. If you do this, you may not get the
architecture bit in the ELF header, which is very much misleading. I
do want that bit if the binary contains instructions specific to that
architecture..
H.J.