On Fri, Oct 5, 2018 at 4:29 AM Jan Hubicka wrote:
>
> >
> > The following fixes bogus differences in scalar iteration cost
> > computed by the vectorizer when comparing 128 and 256 bit vectorizations.
> > This was caused by the hook looking at the vector types mode even
> > for kind == scalar_stmt
>
> The following fixes bogus differences in scalar iteration cost
> computed by the vectorizer when comparing 128 and 256 bit vectorizations.
> This was caused by the hook looking at the vector types mode even
> for kind == scalar_stmt and thus returning vector operation costs
> for things like a
On Thu, 4 Oct 2018, Richard Biener wrote:
>
> The following fixes bogus differences in scalar iteration cost
> computed by the vectorizer when comparing 128 and 256 bit vectorizations.
> This was caused by the hook looking at the vector types mode even
> for kind == scalar_stmt and thus returning
The following fixes bogus differences in scalar iteration cost
computed by the vectorizer when comparing 128 and 256 bit vectorizations.
This was caused by the hook looking at the vector types mode even
for kind == scalar_stmt and thus returning vector operation costs
for things like add or negat
