This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: Darwin specific 1-liners (1)
- From: Richard Henderson <rth at redhat dot com>
- To: David Edelsohn <dje at watson dot ibm dot com>
- Cc: Daniel Berlin <dan at cgsoftware dot com>, Stan Shebs <shebs at apple dot com>, dalej at apple dot com, gcc-patches at gcc dot gnu dot org, +gcc/macos at makai dot watson dot ibm dot com
- Date: Thu, 6 Dec 2001 10:26:03 -0800
- Subject: Re: PATCH: Darwin specific 1-liners (1)
- References: <rth@redhat.com> <200112061813.NAA25764@makai.watson.ibm.com>
On Thu, Dec 06, 2001 at 01:13:36PM -0500, David Edelsohn wrote:
> I think that the first "c,l" so that GCC prefers CTR and does not use LR
> during regrename should be sufficient.
This isn't the way preference works anymore. The problem is with ports
that have different mnemonics -- particularly for move insns -- and so
really have to use different alternatives. So now we look over the set
of alternatives and see if there's a superclass that contains them, and
if so, use that as a preference.
I recommend "c,*l", since that will cause CTR to be preferred in the
register allocator, but reload will happily use LR if the value
happens to already be there.
r~