> your description above hints at that the actual modes involved in the
> vec_pack_sbool_trunc are the same so the TYPE_MODE (narrow_vectype)
> and TYPE_MODE (vectype) are not the actual modes participating. I think
> it would be way better to fix that.
>
> I suppose that since we know TYPE_VECTOR
On Tue, Jan 18, 2022 at 7:26 AM liuhongt wrote:
>
> For testcase in PR, the patch supports QI:4 -> HI:16 pack with
> multi steps(first pack QI:4 -> QI:8 through vec_pack_sbool_trunc_qi,
> then pack QI:8 -> HI:16 through vec_pack_trunc_hi).
> Similar for QI:2 -> HI:16 which is test4 in mask-pack-pr
For testcase in PR, the patch supports QI:4 -> HI:16 pack with
multi steps(first pack QI:4 -> QI:8 through vec_pack_sbool_trunc_qi,
then pack QI:8 -> HI:16 through vec_pack_trunc_hi).
Similar for QI:2 -> HI:16 which is test4 in mask-pack-prefer-128.c.
Bootstrapped both with and w/o '--with-arch=na
