This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH i386 AVX512] [76/n] Extend int 2 float conversions.


On Thu, Oct 9, 2014 at 5:01 PM, Kirill Yukhin <kirill.yukhin@gmail.com> wrote:
> Hello,
> This patch extends autogeneration of SI-2-SF
> conversions.
>
> Bootstrapped.
> AVX-512* tests on top of patch-set all pass
> under simulator.
>
> Is it ok for trunk?
>
> gcc/
>         * config/i386/i386.c
>         (ix86_expand_vector_convert_uns_vsivsf): Handle V16SI mode and
>         TARGET_AVX512VL.
> --
> Thanks, K
>
> diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
> index 7c34431..8a7853e 100644
> --- a/gcc/config/i386/i386.c
> +++ b/gcc/config/i386/i386.c
> @@ -18811,6 +18811,19 @@ ix86_expand_vector_convert_uns_vsivsf (rtx target, rtx val)
>    enum machine_mode fltmode = GET_MODE (target);
>    rtx (*cvt) (rtx, rtx);

Please handle this directly in floatuns<sseintvecmodelower><mode>2
expander. The V16SImode is already handled from there.

Uros.

> +  if (intmode == V16SImode)
> +    {
> +      emit_insn (gen_ufloatv16siv16sf2 (target, val));
> +      return;
> +    }
> +  if (TARGET_AVX512VL)
> +    {
> +      if (intmode == V4SImode)
> +        emit_insn (gen_ufloatv4siv4sf2 (target, val));
> +      else
> +        emit_insn (gen_ufloatv8siv8sf2 (target, val));
> +      return;
> +    }
>    if (intmode == V4SImode)
>      cvt = gen_floatv4siv4sf2;
>    else


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]