Re: [patches] Re: get_attr_* speedups

[Jan Hubicka <jh at suse dot cz>]

> On Mon, Aug 28, 2000 at 01:37:42AM +0200, Jan Hubicka wrote:
> > + /* Wrappers around recog, split, peephole2 functions from insn-recog.c that
> > +    does dirty work.  These functions just clear the recog_data cache.  */
> > + int
> > + recog (x0, insn, pnum_clobbers)
> > + 	rtx x0, insn;
> > + 	int *pnum_clobbers;
> > + {
> > +   recog_data.insn = NULL;
> > +   which_alternative = -1;
> > +   return recog_1 (x0, insn, pnum_clobbers);
> > + }
> 
> I think you should just emit this code into the regular recog, etc.
I just tought that it is handy to have wrappers - several times I had
tendency to put something there, but was lazy to dig into genrecog.
It seems to be little bit tricky to force recog to put these only to the
root functions, not to the subfunctions.
> I don't like splitting this into constrain_operands.  Is there any
> reason not to have the -1 test in extract_constrain_insn_cached?
Not. The extract_constrain_insn_cached was done later and I forgot about
that issue.

I will send updated patch.

Honza
> 
> 
> r~