*[PATCH] math: Remove bogus math implementations@ 2024-01-30 16:54 Adhemerval Zanella2024-01-31 12:23 ` Joseph Myers 0 siblings, 1 reply; 2+ messages in thread From: Adhemerval Zanella @ 2024-01-30 16:54 UTC (permalink / raw) To: libc-alpha;+Cc:Joseph Myers The exp10, exp10l, fma, fmaf, and fmal default implementation do not implement the appropriate semantics nor with an reasonable accuracy. They are also not used by any supported port. --- math/e_exp10.c | 33 --------------------------------- math/e_exp10l.c | 33 --------------------------------- math/s_fma.c | 36 ------------------------------------ math/s_fmaf.c | 30 ------------------------------ math/s_fmal.c | 32 -------------------------------- 5 files changed, 164 deletions(-) delete mode 100644 math/e_exp10.c delete mode 100644 math/e_exp10l.c delete mode 100644 math/s_fma.c delete mode 100644 math/s_fmaf.c delete mode 100644 math/s_fmal.c diff --git a/math/e_exp10.c b/math/e_exp10.c deleted file mode 100644 index 82fbde520d..0000000000 --- a/math/e_exp10.c +++ /dev/null @@ -1,33 +0,0 @@ -/* Copyright (C) 1998-2024 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 <math_private.h> -#include <float.h> -#include <libm-alias-finite.h> - -double -__ieee754_exp10 (double arg) -{ - if (isfinite (arg) && arg < DBL_MIN_10_EXP - DBL_DIG - 10) - return DBL_MIN * DBL_MIN; - else - /* This is a very stupid and inprecise implementation. It'll get - replaced sometime (soon?). */ - return __ieee754_exp (M_LN10 * arg); -} -libm_alias_finite (__ieee754_exp10, __exp10) diff --git a/math/e_exp10l.c b/math/e_exp10l.c deleted file mode 100644 index 2786ca34b4..0000000000 --- a/math/e_exp10l.c +++ /dev/null @@ -1,33 +0,0 @@ -/* Copyright (C) 1998-2024 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 <math_private.h> -#include <float.h> -#include <libm-alias-finite.h> - -long double -__ieee754_exp10l (long double arg) -{ - if (isfinite (arg) && arg < LDBL_MIN_10_EXP - LDBL_DIG - 10) - return LDBL_MIN * LDBL_MIN; - else - /* This is a very stupid and inprecise implementation. It'll get - replaced sometime (soon?). */ - return __ieee754_expl (M_LN10l * arg); -} -libm_alias_finite (__ieee754_exp10, __exp10) diff --git a/math/s_fma.c b/math/s_fma.c deleted file mode 100644 index e036a1c809..0000000000 --- a/math/s_fma.c +++ /dev/null @@ -1,36 +0,0 @@ -/* Compute x * y + z as ternary operation. - Copyright (C) 1997-2024 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 -#define dfmal __hide_dfmal -#define f32xfmaf64 __hide_f32xfmaf64 -#include <math.h> -#undef dfmal -#undef f32xfmaf64 -#include <libm-alias-double.h> -#include <math-narrow-alias.h> - -double -__fma (double x, double y, double z) -{ - return (x * y) + z; -} -#ifndef __fma -libm_alias_double (__fma, fma) -libm_alias_double_narrow (__fma, fma) -#endif diff --git a/math/s_fmaf.c b/math/s_fmaf.c deleted file mode 100644 index 407531a40c..0000000000 --- a/math/s_fmaf.c +++ /dev/null @@ -1,30 +0,0 @@ -/* Compute x * y + z as ternary operation. - Copyright (C) 1997-2024 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 -__fmaf (float x, float y, float z) -{ - return (x * y) + z; -} -#ifndef __fmaf -libm_alias_float (__fma, fma) -#endif diff --git a/math/s_fmal.c b/math/s_fmal.c deleted file mode 100644 index 05fd756b32..0000000000 --- a/math/s_fmal.c +++ /dev/null @@ -1,32 +0,0 @@ -/* Compute x * y + z as ternary operation. - Copyright (C) 1997-2024 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 -#define f64xfmaf128 __hide_f64xfmaf128 -#include <math.h> -#undef f64xfmaf128 -#include <libm-alias-ldouble.h> -#include <math-narrow-alias.h> - -long double -__fmal (long double x, long double y, long double z) -{ - return (x * y) + z; -} -libm_alias_ldouble (__fma, fma) -libm_alias_ldouble_narrow (__fma, fma) -- 2.34.1 ^ permalink raw reply [flat|nested] 2+ messages in thread

*Re: [PATCH] math: Remove bogus math implementations2024-01-30 16:54 [PATCH] math: Remove bogus math implementations Adhemerval Zanella@ 2024-01-31 12:23 ` Joseph Myers0 siblings, 0 replies; 2+ messages in thread From: Joseph Myers @ 2024-01-31 12:23 UTC (permalink / raw) To: Adhemerval Zanella;+Cc:libc-alpha On Tue, 30 Jan 2024, Adhemerval Zanella wrote: > The exp10, exp10l, fma, fmaf, and fmal default implementation do not > implement the appropriate semantics nor with an reasonable accuracy. > They are also not used by any supported port. OK. -- Joseph S. Myers josmyers@redhat.com ^ permalink raw reply [flat|nested] 2+ messages in thread

end of thread, other threads:[~2024-01-31 12:25 UTC | newest]Thread overview:2+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2024-01-30 16:54 [PATCH] math: Remove bogus math implementations Adhemerval Zanella 2024-01-31 12:23 ` Joseph Myers

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).