Re: [PATCH] Prefer reg as first operand in commutative operator

["Paulo J. Matos" <paulo at matos-sorge dot com>]

On 14/02/12 14:10, Jakub Jelinek wrote:
> On Tue, Feb 14, 2012 at 02:05:30PM +0000, Paulo J. Matos wrote:
> > > I think the register allocator will generate good code using % if you
> > > make your predicate nonimmediate_operand in operand 1:
> > >
> > > (define_insn "iorqi3"
> > > [(set (match_operand:QI 0 "register_operand" "=c")
> > > (ior:QI (match_operand:QI 1 "nonimmediate_operand" "%0")
> > > (match_operand:QI 2 "general_operand" "cwmi")))
> > > (clobber (reg:CC RCC))]
> > > "register_operand(operands[1], QImode) ||
> > > register_operand(operands[2], QImode)"
> > > "or\\t%0,%f2")
>
> > Yes, I think in general that seems to be the right procedure. In my
> > case, unfortunately it does not work.
> > The reason is that it allows:
> > (set (reg:QI ...)
> >       (ior:QI (mem:QI (reg:QI ...))
> >               (mem:QI (reg:QI ...))))
> >
> > This wouldn't in general be a problem except that my backend only
> > has one register that can be used for a memory dereference, which
> > means that BASE_REG_CLASS is a class with a single register.
>
> It doesn't allow that, because the condition on the insn then fails,
> as neither operand 1 nor operand 2 is register_operand.

It doesn't look like you can actually do that because, according to the 
internals manual:
"For a named pattern, the condition (if present) may not depend on the 
data in the insn being matched, but only the target-machine-type flags. 
The compiler needs to test these conditions during initialization in 
order to learn exactly which named instructions are available in a 
particular run."

And GCC complains with:
insn-opinit.c: In function 'init_all_optabs':
insn-opinit.c:24: error: 'operands' undeclared (first use in this function)
insn-opinit.c:24: error: (Each undeclared identifier is reported only once
insn-opinit.c:24: error: for each function it appears in.)

Did I miss something in the example?


--
PMatos