[PATCH] Tweak var-tracking to better handle x86-64 lea insns

Jakub Jelinek jakub@redhat.com
Tue Mar 30 15:49:00 GMT 2010


On Tue, Mar 30, 2010 at 08:33:40AM -0700, Richard Henderson wrote:
> On 03/29/2010 11:27 AM, Jakub Jelinek wrote:
> > +/* Helper for adjust_mems.  Return 1 if *loc is unsuitable for
> > +   transformation of wider mode arithmetics to narrower mode,
> > +   -1 if it is suitable and subexpressions shouldn't be
> > +   traversed and 0 if it is suitable and subexpressions should
> > +   be traversed.  */
> > +
> > +static int
> > +use_narrower_mode_p (rtx *loc, void *data)
> 
> Generally ok.  I'd like to find a better name for this function,
> since "_p" suggests that a true return means suitable.  How about
> just "use_narrower_mode_test" and mention in the function comment
> that it is called via for_each_rtx?

Good idea, I wasn't satisfied with that name either very much,
will change that.

	Jakub



More information about the Gcc-patches mailing list