This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: 64-bit PowerPC code quality regression
- To: David Edelsohn <dje at watson dot ibm dot com>
- Subject: Re: 64-bit PowerPC code quality regression
- From: Richard Henderson <rth at redhat dot com>
- Date: Sat, 8 Sep 2001 11:43:13 -0700
- Cc: gcc at gcc dot gnu dot org
- References: <rth@redhat.com> <200109080149.VAA24862@makai.watson.ibm.com>
On Fri, Sep 07, 2001 at 09:49:52PM -0400, David Edelsohn wrote:
> Yes, I know the splitter can match. I mentioned that in my
> original note. What I don't understand is why GCC is applying the
> splitter when it already has a match.
You mean an insn match? Because they're completely separate
operations. Because in addition to splitting "invalid" insns,
define_split is used for lowering code by splitting compound
patterns into their constituant parts.
r~