On Thu, 30 Nov 2017, Jakub Jelinek wrote:
> On Thu, Nov 30, 2017 at 10:26:33AM +0100, Richard Biener wrote:
> > On Wed, 29 Nov 2017, Jakub Jelinek wrote:
> >
> > > Hi!
> > >
> > > Even if target has an umulv4_optab pattern like i?86 mul[lq]; jo,
> > > if one argument is constant power of two, us
On Thu, Nov 30, 2017 at 10:26:33AM +0100, Richard Biener wrote:
> On Wed, 29 Nov 2017, Jakub Jelinek wrote:
>
> > Hi!
> >
> > Even if target has an umulv4_optab pattern like i?86 mul[lq]; jo,
> > if one argument is constant power of two, using two shifts, or
> > for multiplication by 2 just shl w
On Wed, 29 Nov 2017, Jakub Jelinek wrote:
> Hi!
>
> Even if target has an umulv4_optab pattern like i?86 mul[lq]; jo,
> if one argument is constant power of two, using two shifts, or
> for multiplication by 2 just shl with js on previous value is faster
> (though, for -Os mul[lq]; jo wins).
>
>
Hi!
Even if target has an umulv4_optab pattern like i?86 mul[lq]; jo,
if one argument is constant power of two, using two shifts, or
for multiplication by 2 just shl with js on previous value is faster
(though, for -Os mul[lq]; jo wins).
Bootstrapped/regtested on x86_64-linux and i686-linux, ok f
