Hi Iain,
on 2023/7/22 23:58, Iain Sandoe wrote:
> Hi Kewen,
>
> This patch breaks bootstrap on powerpc-darwin (which has Altivec, but not
> VSX) while building libgfortran.
>
>> On 3 Jul 2023, at 04:19, Kewen.Lin via Gcc-patches
>> wrote:
>
> Please see
Hi Kewen,
This patch breaks bootstrap on powerpc-darwin (which has Altivec, but not VSX)
while building libgfortran.
> On 3 Jul 2023, at 04:19, Kewen.Lin via Gcc-patches
> wrote:
Please see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110776
thanks
Iain
Hi Richard,
on 2023/7/2 16:58, Richard Sandiford wrote:
> Kewen Lin writes:
>> @@ -9743,11 +9739,23 @@ vectorizable_load (vec_info *vinfo,
>>unsigned int n_groups = 0;
>>for (j = 0; j < ncopies; j++)
>> {
>> - if (nloads > 1)
>> + if (nloads > 1 && !costing_p)
>>
Kewen Lin writes:
> @@ -9743,11 +9739,23 @@ vectorizable_load (vec_info *vinfo,
>unsigned int n_groups = 0;
>for (j = 0; j < ncopies; j++)
> {
> - if (nloads > 1)
> + if (nloads > 1 && !costing_p)
> vec_alloc (v, nloads);
> gimple *new_stmt =
This patch adjusts the cost handling on VMAT_ELEMENTWISE
and VMAT_STRIDED_SLP in function vectorizable_load. We
don't call function vect_model_load_cost for them any more.
As PR82255 shows, we don't always need a vector construction
there, moving costing next to the transform can make us only
