On Mon, Mar 28, 2022 at 6:55 AM liuhongt <hongtao....@intel.com> wrote:
>
> pinsrw is available for both reg and mem operand under sse2.
> pextrw requires sse4.1 for mem operands.
>
> The patch change attr "isa" for pinsrw mem alternative from sse4_noavx
> to noavx, will enable below optimization.
>
> -        movzwl  (%rdi), %eax
>          pxor    %xmm1, %xmm1
> -        pinsrw  $0, %eax, %xmm1
> +        pinsrw  $0, (%rdi), %xmm1
>          movdqa  %xmm1, %xmm0
>
> Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}.
> Ok for trunk?
>
> gcc/ChangeLog:
>
>         PR target/105066
>         * config/i386/sse.md (vec_set<mode>_0): Change attr "isa" of
>         alternative 4 from sse4_noavx to noavx.
>
> gcc/testsuite/ChangeLog:
>
>         * gcc.target/i386/pr105066.c: New test.

OK.

Thanks,
Uros.

> ---
>  gcc/config/i386/sse.md                   |  4 ++--
>  gcc/testsuite/gcc.target/i386/pr105066.c | 10 ++++++++++
>  2 files changed, 12 insertions(+), 2 deletions(-)
>  create mode 100644 gcc/testsuite/gcc.target/i386/pr105066.c
>
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index a9e18d38323..27e9629f4b0 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -10617,9 +10617,9 @@ (define_insn "vec_set<mode>_0"
>    [(set (attr "isa")
>         (cond [(eq_attr "alternative" "0,1,2")
>                  (const_string "avx512fp16")
> -              (eq_attr "alternative" "3")
> +              (eq_attr "alternative" "3,4")
>                  (const_string "noavx")
> -              (eq_attr "alternative" "4,5,6")
> +              (eq_attr "alternative" "5,6")
>                  (const_string "sse4_noavx")
>                (eq_attr "alternative" "7,8,9")
>                  (const_string "avx")
> diff --git a/gcc/testsuite/gcc.target/i386/pr105066.c 
> b/gcc/testsuite/gcc.target/i386/pr105066.c
> new file mode 100644
> index 00000000000..c5c5b9e12de
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/pr105066.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -msse2 -mno-sse4.1" } */
> +/* { dg-final { scan-assembler-not "movzwl" } } */
> +/* { dg-final { scan-assembler {(?n)pinsrw[ \t]+\$0.*\(%} } } */
> +
> +#include <immintrin.h>
> +
> +__m128i load16(void *p){
> +    return _mm_loadu_si16(p);
> +}
> --
> 2.18.1
>

Reply via email to