This is the mail archive of the
mailing list for the GCC project.
Re: [RFC] Combine related fail of gcc.target/powerpc/ti_math1.c
- From: Segher Boessenkool <segher at kernel dot crashing dot org>
- To: gcc at gcc dot gnu dot org
- Date: Fri, 22 May 2015 07:47:29 -0500
- Subject: Re: [RFC] Combine related fail of gcc.target/powerpc/ti_math1.c
- Authentication-results: sourceware.org; auth=none
- References: <20150521103604 dot GK6140 at bubble dot grove dot modra dot org> <20150521123916 dot GA26096 at gate dot crashing dot org> <20150522021841 dot GL6140 at bubble dot grove dot modra dot org>
On Fri, May 22, 2015 at 11:48:42AM +0930, Alan Modra wrote:
> > I think we should add to
> > the canonicalisation rules so that fixed regs sort after other regs.
> > That requires a lot of testing.
> What if you have two hard regs as above? Which of reg 5 and reg 76
> sorts first? If they are sorted by register number, then ca appears
> in the wrong place. Reverse sorting hard regs might work for this
> pattern on powerpc, but that seems an odd choice. And if you say hard
> regs ought to keep their original order in rtl like the above, then it
> is no more difficult to keep all regs in their original order
I said fixed registers, not hard registers. Very different thing :-)
You will not normally see more than one fixed register in a pattern,
and it is always somewhat special.