unofficial mirror of libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella via Libc-alpha <libc-alpha@sourceware.org>
To: Vineet Gupta <Vineet.Gupta1@synopsys.com>, libc-alpha@sourceware.org
Cc: linux-snps-arc@lists.infradead.org
Subject: Re: [PATCH 5/5] powerpc/fpu: use generic fma functions
Date: Mon, 1 Jun 2020 14:10:44 -0300	[thread overview]
Message-ID: <8eb04259-7529-cbb4-81b8-d5b0f56eb6b8@linaro.org> (raw)
In-Reply-To: <20200530020047.5490-6-vgupta@synopsys.com>



On 29/05/2020 23:00, Vineet Gupta wrote:

LGTM, thanks.

This new addition, along with the aarch64 one, makes me wonder if
it would be better to decompose the USE_* defined in multiple files
so the architecture adds only the required ones (instead of copy/paste
the whole file with the generic ones).

Also for powerpc, maybe a future cleanup would to move the logic
to select the builtin to math-use-builtins.h. Something like:

  #ifdef _ARCH_PWR5X
  #  define USE_FLOOR_BUILTIN 1
  #  define USE_FLOORF_BUILTIN 1
  #else
  #  define USE_FLOOR_BUILTIN 0
  #  define USE_FLOORF_BUILTIN 0
  #endif
  #define USE_FLOORL_BUILTIN 0
  #define USE_FLOORF128_BUILTIN 0

Reviewed-by: Adhemerval Zanella  <adhemerval.zanella@linaro.org>

> ---
>  sysdeps/powerpc/fpu/math-use-builtins.h | 70 +++++++++++++++++++++++++
>  sysdeps/powerpc/fpu/s_fma.c             | 27 ----------
>  sysdeps/powerpc/fpu/s_fmaf.c            | 27 ----------
>  3 files changed, 70 insertions(+), 54 deletions(-)
>  create mode 100644 sysdeps/powerpc/fpu/math-use-builtins.h
>  delete mode 100644 sysdeps/powerpc/fpu/s_fma.c
>  delete mode 100644 sysdeps/powerpc/fpu/s_fmaf.c
> 
> diff --git a/sysdeps/powerpc/fpu/math-use-builtins.h b/sysdeps/powerpc/fpu/math-use-builtins.h
> new file mode 100644
> index 000000000000..9bdde66cf1de
> --- /dev/null
> +++ b/sysdeps/powerpc/fpu/math-use-builtins.h
> @@ -0,0 +1,70 @@
> +/* Using math gcc builtins instead of generic implementation.  PowerPC version.
> +   Copyright (C) 2019-2020 Free Software Foundation, Inc.
> +   This file is part of the GNU C Library.
> +
> +   The GNU C Library is free software; you can redistribute it and/or
> +   modify it under the terms of the GNU Lesser General Public
> +   License as published by the Free Software Foundation; either
> +   version 2.1 of the License, or (at your option) any later version.
> +
> +   The GNU C Library is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> +   Lesser General Public License for more details.
> +
> +   You should have received a copy of the GNU Lesser General Public
> +   License along with the GNU C Library; if not, see
> +   <https://www.gnu.org/licenses/>.  */
> +
> +#ifndef MATH_USE_BUILTINS_H
> +#define MATH_USE_BUILTINS_H	1
> +
> +#include <features.h> /* For __GNUC_PREREQ.  */
> +
> +/* Define these macros to 1 to use __builtin_xyz instead of the
> +   generic implementation.  */
> +#define USE_NEARBYINT_BUILTIN 0
> +#define USE_NEARBYINTF_BUILTIN 0
> +#define USE_NEARBYINTL_BUILTIN 0
> +#define USE_NEARBYINTF128_BUILTIN 0
> +
> +#define USE_RINT_BUILTIN 0
> +#define USE_RINTF_BUILTIN 0
> +#define USE_RINTL_BUILTIN 0
> +#define USE_RINTF128_BUILTIN 0
> +
> +#define USE_FLOOR_BUILTIN 0
> +#define USE_FLOORF_BUILTIN 0
> +#define USE_FLOORL_BUILTIN 0
> +#define USE_FLOORF128_BUILTIN 0
> +
> +#define USE_CEIL_BUILTIN 0
> +#define USE_CEILF_BUILTIN 0
> +#define USE_CEILL_BUILTIN 0
> +#define USE_CEILF128_BUILTIN 0
> +
> +#define USE_TRUNC_BUILTIN 0
> +#define USE_TRUNCF_BUILTIN 0
> +#define USE_TRUNCL_BUILTIN 0
> +#define USE_TRUNCF128_BUILTIN 0
> +
> +#define USE_ROUND_BUILTIN 0
> +#define USE_ROUNDF_BUILTIN 0
> +#define USE_ROUNDL_BUILTIN 0
> +#define USE_ROUNDF128_BUILTIN 0
> +
> +#define USE_COPYSIGNL_BUILTIN 1
> +#if __GNUC_PREREQ (7, 0)
> +# define USE_COPYSIGNF128_BUILTIN 1
> +#else
> +# define USE_COPYSIGNF128_BUILTIN 0
> +#endif
> +
> +#define USE_SQRT_BUILTIN 0
> +#define USE_SQRTF_BUILTIN 0
> +
> +#define USE_FMA_BUILTIN 1
> +#define USE_FMAF_BUILTIN 1
> +#define USE_FMAL_BUILTIN 0
> +
> +#endif /* math-use-builtins.h */

Ok.

> diff --git a/sysdeps/powerpc/fpu/s_fma.c b/sysdeps/powerpc/fpu/s_fma.c
> deleted file mode 100644
> index 9ddd13253485..000000000000
> --- a/sysdeps/powerpc/fpu/s_fma.c
> +++ /dev/null
> @@ -1,27 +0,0 @@
> -/* Compute x * y + z as ternary operation.  PowerPC version.
> -   Copyright (C) 2019-2020 Free Software Foundation, Inc.
> -   This file is part of the GNU C Library.
> -
> -   The GNU C Library is free software; you can redistribute it and/or
> -   modify it under the terms of the GNU Lesser General Public
> -   License as published by the Free Software Foundation; either
> -   version 2.1 of the License, or (at your option) any later version.
> -
> -   The GNU C Library is distributed in the hope that it will be useful,
> -   but WITHOUT ANY WARRANTY; without even the implied warranty of
> -   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> -   Lesser General Public License for more details.
> -
> -   You should have received a copy of the GNU Lesser General Public
> -   License along with the GNU C Library; if not, see
> -   <https://www.gnu.org/licenses/>.  */
> -
> -#include <sysdep.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__fma (double x, double y, double z)
> -{
> -  return __builtin_fma (x, y, z);
> -}
> -libm_alias_double (__fma, fma)

Ok.

> diff --git a/sysdeps/powerpc/fpu/s_fmaf.c b/sysdeps/powerpc/fpu/s_fmaf.c
> deleted file mode 100644
> index d132c20125bb..000000000000
> --- a/sysdeps/powerpc/fpu/s_fmaf.c
> +++ /dev/null
> @@ -1,27 +0,0 @@
> -/* Compute x * y + z as ternary operation.  PowerPC version.
> -   Copyright (C) 2019-2020 Free Software Foundation, Inc.
> -   This file is part of the GNU C Library.
> -
> -   The GNU C Library is free software; you can redistribute it and/or
> -   modify it under the terms of the GNU Lesser General Public
> -   License as published by the Free Software Foundation; either
> -   version 2.1 of the License, or (at your option) any later version.
> -
> -   The GNU C Library is distributed in the hope that it will be useful,
> -   but WITHOUT ANY WARRANTY; without even the implied warranty of
> -   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> -   Lesser General Public License for more details.
> -
> -   You should have received a copy of the GNU Lesser General Public
> -   License along with the GNU C Library; if not, see
> -   <https://www.gnu.org/licenses/>.  */
> -
> -#include <sysdep.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__fmaf (float x, float y, float z)
> -{
> -  return __builtin_fmaf (x, y, z);
> -}
> -libm_alias_float (__fma, fma)
> 

Ok.

      reply	other threads:[~2020-06-01 17:10 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-30  2:00 [PATCH 0/5] use gcc builtins for sqrt and fma functions Vineet Gupta via Libc-alpha
2020-05-30  2:00 ` [PATCH 1/5] ARC/dl-runtime helper macros Vineet Gupta via Libc-alpha
2020-05-30  2:17   ` Vineet Gupta via Libc-alpha
2020-05-30  2:00 ` [PATCH 2/5] iee754: prvoide gcc builtins based generic sqrt functions Vineet Gupta via Libc-alpha
2020-06-01 14:13   ` Adhemerval Zanella via Libc-alpha
2020-06-01 14:42     ` Adhemerval Zanella via Libc-alpha
2020-06-01 18:12       ` Vineet Gupta via Libc-alpha
2020-06-01 18:32         ` Adhemerval Zanella via Libc-alpha
2020-06-01 14:18   ` Florian Weimer
2020-06-01 18:12     ` Vineet Gupta via Libc-alpha
2020-05-30  2:00 ` [PATCH 3/5] iee754: prvoide gcc builtins based generic fma functions Vineet Gupta via Libc-alpha
2020-06-01 14:20   ` Adhemerval Zanella via Libc-alpha
2020-06-01 16:13   ` Florian Weimer
2020-05-30  2:00 ` [PATCH 4/5] aarch/fpu: use generic sqrt, " Vineet Gupta via Libc-alpha
2020-06-01 16:38   ` Adhemerval Zanella via Libc-alpha
2020-06-01 21:44     ` Vineet Gupta via Libc-alpha
2020-06-01 22:08       ` Ramana Radhakrishnan via Libc-alpha
2020-06-02 18:12       ` Joseph Myers
2020-05-30  2:00 ` [PATCH 5/5] powerpc/fpu: use generic " Vineet Gupta via Libc-alpha
2020-06-01 17:10   ` Adhemerval Zanella via Libc-alpha [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/libc/involved.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8eb04259-7529-cbb4-81b8-d5b0f56eb6b8@linaro.org \
    --to=libc-alpha@sourceware.org \
    --cc=Vineet.Gupta1@synopsys.com \
    --cc=adhemerval.zanella@linaro.org \
    --cc=linux-snps-arc@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).