On Mon, Oct 12, 2020 at 1:31 PM Martin Liška wrote:
>
> On 10/12/20 1:27 PM, Richard Biener wrote:
> > Btw, POLY_INT_CST can likely be handled the same as INTEGER_CST - I suppose
> > you tried that? (it might need further adjustments downstream).
>
> Yes, it can. But it seemed to me like an incorr
On 10/12/20 1:27 PM, Richard Biener wrote:
Btw, POLY_INT_CST can likely be handled the same as INTEGER_CST - I suppose
you tried that? (it might need further adjustments downstream).
Yes, it can. But it seemed to me like an incorrect match:
/home/marxin/Programming/testcases/pr97079-2.c:10:12:
On Mon, Oct 12, 2020 at 1:26 PM Richard Biener
wrote:
>
> On Mon, Oct 12, 2020 at 12:59 PM Martin Liška wrote:
> >
> > The patch fixes the following 2 issues:
> >
> >.MASK_STORE_LANES (&a, 4B, max_mask_34, vect_array.12);
> >
> > here we miss to return the last argument as stored value.
> >
>
On Mon, Oct 12, 2020 at 12:59 PM Martin Liška wrote:
>
> The patch fixes the following 2 issues:
>
>.MASK_STORE_LANES (&a, 4B, max_mask_34, vect_array.12);
>
> here we miss to return the last argument as stored value.
>
> ivtmp_32 = ivtmp_31 + POLY_INT_CST [4, 4];
>
> here we miss a bail out i
