[v2,3/4] aarch/fpu: use generic builtins based math functions

  introduce sysdep header math-use-builtins.h to replace aarch64
impementations with corresponding generic ones

 - newly inroduced generic sqrt{,f}, fma{,f}
 - existing floor{,f}, nearbyint{,f}, rint{,f}, round{,f}, trunc{,f}
 - Note that generic copysign was already enabled (via generic
   math-use-builtins.h) now thru sysdep header

This is supposed to be a non functional change
Passes build-many for aarch64-linux-gnu
---
 sysdeps/aarch64/fpu/e_sqrt.c            | 27 ----------
 sysdeps/aarch64/fpu/e_sqrtf.c           | 27 ----------
 sysdeps/aarch64/fpu/math-use-builtins.h | 71 +++++++++++++++++++++++++
 sysdeps/aarch64/fpu/s_floor.c           | 29 ----------
 sysdeps/aarch64/fpu/s_floorf.c          | 29 ----------
 sysdeps/aarch64/fpu/s_fma.c             | 28 ----------
 sysdeps/aarch64/fpu/s_fmaf.c            | 28 ----------
 sysdeps/aarch64/fpu/s_nearbyint.c       | 28 ----------
 sysdeps/aarch64/fpu/s_nearbyintf.c      | 28 ----------
 sysdeps/aarch64/fpu/s_rint.c            | 29 ----------
 sysdeps/aarch64/fpu/s_rintf.c           | 29 ----------
 sysdeps/aarch64/fpu/s_round.c           | 29 ----------
 sysdeps/aarch64/fpu/s_roundf.c          | 29 ----------
 sysdeps/aarch64/fpu/s_trunc.c           | 29 ----------
 sysdeps/aarch64/fpu/s_truncf.c          | 29 ----------
 15 files changed, 71 insertions(+), 398 deletions(-)
 delete mode 100644 sysdeps/aarch64/fpu/e_sqrt.c
 delete mode 100644 sysdeps/aarch64/fpu/e_sqrtf.c
 create mode 100644 sysdeps/aarch64/fpu/math-use-builtins.h
 delete mode 100644 sysdeps/aarch64/fpu/s_floor.c
 delete mode 100644 sysdeps/aarch64/fpu/s_floorf.c
 delete mode 100644 sysdeps/aarch64/fpu/s_fma.c
 delete mode 100644 sysdeps/aarch64/fpu/s_fmaf.c
 delete mode 100644 sysdeps/aarch64/fpu/s_nearbyint.c
 delete mode 100644 sysdeps/aarch64/fpu/s_nearbyintf.c
 delete mode 100644 sysdeps/aarch64/fpu/s_rint.c
 delete mode 100644 sysdeps/aarch64/fpu/s_rintf.c
 delete mode 100644 sysdeps/aarch64/fpu/s_round.c
 delete mode 100644 sysdeps/aarch64/fpu/s_roundf.c
 delete mode 100644 sysdeps/aarch64/fpu/s_trunc.c
 delete mode 100644 sysdeps/aarch64/fpu/s_truncf.c
  

On 01/06/2020 21:35, Vineet Gupta wrote:
> introduce sysdep header math-use-builtins.h to replace aarch64
> impementations with corresponding generic ones

s/impementations/implementations and missing ':'.

> 
>  - newly inroduced generic sqrt{,f}, fma{,f}
>  - existing floor{,f}, nearbyint{,f}, rint{,f}, round{,f}, trunc{,f}
>  - Note that generic copysign was already enabled (via generic
>    math-use-builtins.h) now thru sysdep header
> 
> This is supposed to be a non functional change
> Passes build-many for aarch64-linux-gnu

LGTM (just double check if the objects generated by the removed 
implementations did not changed).

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

> ---
>  sysdeps/aarch64/fpu/e_sqrt.c            | 27 ----------
>  sysdeps/aarch64/fpu/e_sqrtf.c           | 27 ----------
>  sysdeps/aarch64/fpu/math-use-builtins.h | 71 +++++++++++++++++++++++++
>  sysdeps/aarch64/fpu/s_floor.c           | 29 ----------
>  sysdeps/aarch64/fpu/s_floorf.c          | 29 ----------
>  sysdeps/aarch64/fpu/s_fma.c             | 28 ----------
>  sysdeps/aarch64/fpu/s_fmaf.c            | 28 ----------
>  sysdeps/aarch64/fpu/s_nearbyint.c       | 28 ----------
>  sysdeps/aarch64/fpu/s_nearbyintf.c      | 28 ----------
>  sysdeps/aarch64/fpu/s_rint.c            | 29 ----------
>  sysdeps/aarch64/fpu/s_rintf.c           | 29 ----------
>  sysdeps/aarch64/fpu/s_round.c           | 29 ----------
>  sysdeps/aarch64/fpu/s_roundf.c          | 29 ----------
>  sysdeps/aarch64/fpu/s_trunc.c           | 29 ----------
>  sysdeps/aarch64/fpu/s_truncf.c          | 29 ----------
>  15 files changed, 71 insertions(+), 398 deletions(-)
>  delete mode 100644 sysdeps/aarch64/fpu/e_sqrt.c
>  delete mode 100644 sysdeps/aarch64/fpu/e_sqrtf.c
>  create mode 100644 sysdeps/aarch64/fpu/math-use-builtins.h
>  delete mode 100644 sysdeps/aarch64/fpu/s_floor.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_floorf.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_fma.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_fmaf.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_nearbyint.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_nearbyintf.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_rint.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_rintf.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_round.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_roundf.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_trunc.c
>  delete mode 100644 sysdeps/aarch64/fpu/s_truncf.c
> 
> diff --git a/sysdeps/aarch64/fpu/e_sqrt.c b/sysdeps/aarch64/fpu/e_sqrt.c
> deleted file mode 100644
> index abb67ef7b061..000000000000
> --- a/sysdeps/aarch64/fpu/e_sqrt.c
> +++ /dev/null
> @@ -1,27 +0,0 @@
> -/* Square root of floating point number.
> -   Copyright (C) 2015-2020 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 <math_private.h>
> -#include <libm-alias-finite.h>
> -
> -double
> -__ieee754_sqrt (double d)
> -{
> -  return __builtin_sqrt (d);
> -}
> -libm_alias_finite (__ieee754_sqrt, __sqrt)

Ok.

> diff --git a/sysdeps/aarch64/fpu/e_sqrtf.c b/sysdeps/aarch64/fpu/e_sqrtf.c
> deleted file mode 100644
> index 13008a4f45d6..000000000000
> --- a/sysdeps/aarch64/fpu/e_sqrtf.c
> +++ /dev/null
> @@ -1,27 +0,0 @@
> -/* Single-precision floating point square root.
> -   Copyright (C) 2015-2020 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 <math_private.h>
> -#include <libm-alias-finite.h>
> -
> -float
> -__ieee754_sqrtf (float s)
> -{
> -  return __builtin_sqrtf (s);
> -}
> -libm_alias_finite (__ieee754_sqrtf, __sqrtf)

Ok.

> diff --git a/sysdeps/aarch64/fpu/math-use-builtins.h b/sysdeps/aarch64/fpu/math-use-builtins.h
> new file mode 100644
> index 000000000000..50a1ba0db5ea
> --- /dev/null
> +++ b/sysdeps/aarch64/fpu/math-use-builtins.h
> @@ -0,0 +1,71 @@
> +/* Using math gcc builtins instead of generic implementation.  aarch64 version.
> +   Copyright (C) 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 1
> +#define USE_NEARBYINTF_BUILTIN 1
> +#define USE_NEARBYINTL_BUILTIN 0
> +#define USE_NEARBYINTF128_BUILTIN 0
> +
> +#define USE_RINT_BUILTIN 1
> +#define USE_RINTF_BUILTIN 1
> +#define USE_RINTL_BUILTIN 0
> +#define USE_RINTF128_BUILTIN 0
> +
> +#define USE_FLOOR_BUILTIN 1
> +#define USE_FLOORF_BUILTIN 1
> +#define USE_FLOORL_BUILTIN 0
> +#define USE_FLOORF128_BUILTIN 0
> +
> +#define USE_CEIL_BUILTIN 1
> +#define USE_CEILF_BUILTIN 1
> +#define USE_CEILL_BUILTIN 0
> +#define USE_CEILF128_BUILTIN 0
> +
> +#define USE_TRUNC_BUILTIN 1
> +#define USE_TRUNCF_BUILTIN 1
> +#define USE_TRUNCL_BUILTIN 0
> +#define USE_TRUNCF128_BUILTIN 0
> +
> +#define USE_ROUND_BUILTIN 1
> +#define USE_ROUNDF_BUILTIN 1
> +#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 1
> +#define USE_SQRTF_BUILTIN 1
> +
> +#define USE_FMA_BUILTIN 1
> +#define USE_FMAF_BUILTIN 1
> +#define USE_FMAL_BUILTIN 0
> +#define USE_FMAF128_BUILTIN 0
> +
> +#endif

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_floor.c b/sysdeps/aarch64/fpu/s_floor.c
> deleted file mode 100644
> index d7d3e9336a6e..000000000000
> --- a/sysdeps/aarch64/fpu/s_floor.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__floor (double x)
> -{
> -  return __builtin_floor (x);
> -}
> -
> -libm_alias_double (__floor, floor)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_floorf.c b/sysdeps/aarch64/fpu/s_floorf.c
> deleted file mode 100644
> index 625ff4375702..000000000000
> --- a/sysdeps/aarch64/fpu/s_floorf.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__floorf (float x)
> -{
> -  return __builtin_floorf (x);
> -}
> -
> -libm_alias_float (__floor, floor)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_fma.c b/sysdeps/aarch64/fpu/s_fma.c
> deleted file mode 100644
> index 48bb40482dc9..000000000000
> --- a/sysdeps/aarch64/fpu/s_fma.c
> +++ /dev/null
> @@ -1,28 +0,0 @@
> -/* Copyright (C) 1996-2020 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 <math.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/aarch64/fpu/s_fmaf.c b/sysdeps/aarch64/fpu/s_fmaf.c
> deleted file mode 100644
> index 544f32e27aec..000000000000
> --- a/sysdeps/aarch64/fpu/s_fmaf.c
> +++ /dev/null
> @@ -1,28 +0,0 @@
> -/* Copyright (C) 2011-2020 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 <math.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.

> diff --git a/sysdeps/aarch64/fpu/s_nearbyint.c b/sysdeps/aarch64/fpu/s_nearbyint.c
> deleted file mode 100644
> index 59a57dd39ba1..000000000000
> --- a/sysdeps/aarch64/fpu/s_nearbyint.c
> +++ /dev/null
> @@ -1,28 +0,0 @@
> -/* Copyright (C) 2011-2020 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 <math.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__nearbyint (double x)
> -{
> -  return __builtin_nearbyint (x);
> -}
> -
> -libm_alias_double (__nearbyint, nearbyint)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_nearbyintf.c b/sysdeps/aarch64/fpu/s_nearbyintf.c
> deleted file mode 100644
> index f52968ae2bbd..000000000000
> --- a/sysdeps/aarch64/fpu/s_nearbyintf.c
> +++ /dev/null
> @@ -1,28 +0,0 @@
> -/* Copyright (C) 2011-2020 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 <math.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__nearbyintf (float x)
> -{
> -  return __builtin_nearbyintf (x);
> -}
> -
> -libm_alias_float (__nearbyint, nearbyint)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_rint.c b/sysdeps/aarch64/fpu/s_rint.c
> deleted file mode 100644
> index bdd757167e0c..000000000000
> --- a/sysdeps/aarch64/fpu/s_rint.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__rint (double x)
> -{
> -  return __builtin_rint (x);
> -}
> -
> -libm_alias_double (__rint, rint)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_rintf.c b/sysdeps/aarch64/fpu/s_rintf.c
> deleted file mode 100644
> index 8cf7c63d6dda..000000000000
> --- a/sysdeps/aarch64/fpu/s_rintf.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__rintf (float x)
> -{
> -  return __builtin_rintf (x);
> -}
> -
> -libm_alias_float (__rint, rint)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_round.c b/sysdeps/aarch64/fpu/s_round.c
> deleted file mode 100644
> index d2fd5f34117a..000000000000
> --- a/sysdeps/aarch64/fpu/s_round.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__round (double x)
> -{
> -  return __builtin_round (x);
> -}
> -
> -libm_alias_double (__round, round)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_roundf.c b/sysdeps/aarch64/fpu/s_roundf.c
> deleted file mode 100644
> index e51214e86cd2..000000000000
> --- a/sysdeps/aarch64/fpu/s_roundf.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__roundf (float x)
> -{
> -  return __builtin_roundf (x);
> -}
> -
> -libm_alias_float (__round, round)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_trunc.c b/sysdeps/aarch64/fpu/s_trunc.c
> deleted file mode 100644
> index bac23a5e269d..000000000000
> --- a/sysdeps/aarch64/fpu/s_trunc.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-double.h>
> -
> -double
> -__trunc (double x)
> -{
> -  return __builtin_trunc (x);
> -}
> -
> -libm_alias_double (__trunc, trunc)

Ok.

> diff --git a/sysdeps/aarch64/fpu/s_truncf.c b/sysdeps/aarch64/fpu/s_truncf.c
> deleted file mode 100644
> index 6a025c084a5a..000000000000
> --- a/sysdeps/aarch64/fpu/s_truncf.c
> +++ /dev/null
> @@ -1,29 +0,0 @@
> -/* Copyright (C) 2011-2020 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/>.  */
> -
> -#define NO_MATH_REDIRECT
> -#include <math.h>
> -#include <libm-alias-float.h>
> -
> -float
> -__truncf (float x)
> -{
> -  return __builtin_truncf (x);
> -}
> -
> -libm_alias_float (__trunc, trunc)
> 


Ok.
  

On 6/2/20 10:31 AM, Adhemerval Zanella via Libc-alpha wrote:
> 
> 
> On 01/06/2020 21:35, Vineet Gupta wrote:
>> introduce sysdep header math-use-builtins.h to replace aarch64
>> impementations with corresponding generic ones
> 
> s/impementations/implementations and missing ':'.

Fixed.

>>  - newly inroduced generic sqrt{,f}, fma{,f}
>>  - existing floor{,f}, nearbyint{,f}, rint{,f}, round{,f}, trunc{,f}
>>  - Note that generic copysign was already enabled (via generic
>>    math-use-builtins.h) now thru sysdep header
>>
>> This is supposed to be a non functional change
>> Passes build-many for aarch64-linux-gnu
> 
> LGTM (just double check if the objects generated by the removed 
> implementations did not changed).

I will add the diff output to v3.