From patchwork Wed Sep 13 01:14:25 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joseph Myers X-Patchwork-Id: 813118 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (mailfrom) smtp.mailfrom=sourceware.org (client-ip=209.132.180.131; helo=sourceware.org; envelope-from=libc-alpha-return-84534-incoming=patchwork.ozlabs.org@sourceware.org; receiver=) Authentication-Results: ozlabs.org; dkim=pass (1024-bit key; secure) header.d=sourceware.org header.i=@sourceware.org header.b="wW2TGRhk"; dkim-atps=neutral Received: from sourceware.org (server1.sourceware.org [209.132.180.131]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 3xsNxN436Jz9t4V for ; Wed, 13 Sep 2017 11:14:48 +1000 (AEST) DomainKey-Signature: a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id :list-unsubscribe:list-subscribe:list-archive:list-post :list-help:sender:date:from:to:subject:message-id:mime-version :content-type; q=dns; s=default; b=y3KhoYdY0iA06rtgrarxDHFV7+0/y BnY7MOMOXG+7KDGEcJPB8lxcYH0SMvB50/KMhPi8Ny93wTAHoYqoGO7fauxYRdh5 FFecCEY5GmdN6whg8Y1V/Gk25zy0wwN1mCwsmDcql1OrjGKXcxetIWL0swMnLv/l b83HvmcmQv2JcI= DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id :list-unsubscribe:list-subscribe:list-archive:list-post :list-help:sender:date:from:to:subject:message-id:mime-version :content-type; s=default; bh=qHSdy5fDbDEx5eP1GIeV/zlQi7c=; b=wW2 TGRhkGWpObslcFulqCwWI2KR9TVHrKgGDLCKZVghaPY+/GU5Jr2JIojxGJ6d1ovV OPEPHtz8Swqs/sohouU/wOXeVLx1P97XSwcWyYtjuYv8FHe1gqEuaBUubRbHGETJ oorNwDSMYU+IskTEfODWkCMgULa4bTgNgefmVDR0= Received: (qmail 66826 invoked by alias); 13 Sep 2017 01:14:39 -0000 Mailing-List: contact libc-alpha-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-alpha-owner@sourceware.org Delivered-To: mailing list libc-alpha@sourceware.org Received: (qmail 66817 invoked by uid 89); 13 Sep 2017 01:14:39 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-24.5 required=5.0 tests=AWL, BAYES_00, GIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3, RCVD_IN_DNSWL_NONE, SPF_PASS, URIBL_RED autolearn=ham version=3.3.2 spammy= X-HELO: relay1.mentorg.com Date: Wed, 13 Sep 2017 01:14:25 +0000 From: Joseph Myers To: Subject: Define and use libm_alias_double [committed] Message-ID: User-Agent: Alpine 2.20 (DEB 67 2015-01-07) MIME-Version: 1.0 X-ClientProxiedBy: svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) Continuing the process of setting up common macros for libm function aliases, with a view to using them to define _FloatN / _FloatNx aliases in future, this patch adds a libm_alias_double macro and uses it in the type-generic templates. This macro handles defining aliases for double, and for long double in the NO_LONG_DOUBLE case. It also handles defining compat symbols for long double = double for architectures that changed their long double format. By so doing, it eliminates the need for the M_LIBM_NEED_COMPAT and declare_mgen_libm_compat macros; the single declare_mgen_alias call in each template now suffices to define all required compat symbols. When used for more double functions (not based on type-generic templates), I expect it will eliminate the need for most ldbl-opt wrappers for such functions. A few special cases are needed. __clog10l is a public symbol (for historical reasons) so needs to be given appropriate compat versions for architectures that changed their long double format, but is not defined as an alias using the normal macros since __clog10* are *not* public symbols for _FloatN / _FloatNx types. For scalbn, scalbln and log1p, the changes adding errno setting support for those functions left compat symbols pointing directly to the non-errno-setting implementations. There is no requirement for the compat symbols not to set errno; that just made for the simplest patches at that time. Now, with these common macros, it's natural to redirect the compat symbols to the errno-setting wrappers, which I intend to do in a separate patch. Tested for x86_64, and with build-many-glibcs.py. For ldbl-opt platforms the stripped libm.so binaries are changed (disassembly unchanged) because the details of how the clog10l compat symbol is created mean it ceases to be weak as it was before; for other platforms, stripped libm.so binaries are unchanged. Committed. 2017-09-13 Joseph Myers * sysdeps/generic/libm-alias-double.h: New file. * sysdeps/ieee754/ldbl-opt/libm-alias-double.h: Likewise. * sysdeps/generic/math-type-macros-double.h: Include . [declare_mgen_alias] (declare_mgen_alias): Define to use libm_alias_double. * sysdeps/generic/math-type-macros.h [!M_LIBM_NEED_COMPAT] (M_LIBM_NEED_COMPAT): Remove macro. [!M_LIBM_NEED_COMPAT] (declare_mgen_libm_compat): Likewise. * sysdeps/ieee754/ldbl-opt/math-type-macros-double.h: Remove. * math/cabs_template.c [M_LIBM_NEED_COMPAT]: Remove conditional code. * math/carg_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/cimag_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/conj_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/creal_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_cacos_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_cacosh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_casin_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_casinh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_catan_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_catanh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_ccos_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_ccosh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_cexp_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_clog10_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_clog_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_cpow_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_cproj_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_csin_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_csinh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_csqrt_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_ctan_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_ctanh_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_fdim_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_fmax_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_fmin_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/s_nan_template.c [M_LIBM_NEED_COMPAT]: Likewise. * math/w_ilogb_template.c [M_LIBM_NEED_COMPAT]: Likewise. * sysdeps/ieee754/ldbl-opt/s_clog10.c: New file. * sysdeps/ieee754/ldbl-opt/s_ldexp.c (M_LIBM_NEED_COMPAT): Remove macro. (declare_mgen_alias): New macro. * sysdeps/ieee754/ldbl-opt/w_log1p.c: New file. * sysdeps/ieee754/ldbl-opt/w_scalbln.c: Likewise. * sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim-vis3.c (M_LIBM_NEED_COMPAT): Remove macro. * sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim.c [HAVE_AS_VIS3_SUPPORT]: Include and . [HAVE_AS_VIS3_SUPPORT && LONG_DOUBLE_COMPAT (libm, FIRST_VERSION_libm_fdiml)]: Define fdiml as compat symbol. diff --git a/math/cabs_template.c b/math/cabs_template.c index c9592c1..21b8506 100644 --- a/math/cabs_template.c +++ b/math/cabs_template.c @@ -27,7 +27,3 @@ M_DECL_FUNC (__cabs) (CFLOAT z) } declare_mgen_alias (__cabs, cabs) - -#if M_LIBM_NEED_COMPAT (cabs) -declare_mgen_libm_compat (__cabs, cabs) -#endif diff --git a/math/carg_template.c b/math/carg_template.c index 5d60663..9f7ad14 100644 --- a/math/carg_template.c +++ b/math/carg_template.c @@ -27,7 +27,3 @@ M_DECL_FUNC (__carg) (CFLOAT x) } declare_mgen_alias (__carg, carg) - -#if M_LIBM_NEED_COMPAT (carg) -declare_mgen_libm_compat (__carg, carg) -#endif diff --git a/math/cimag_template.c b/math/cimag_template.c index 6284b57..b24f001 100644 --- a/math/cimag_template.c +++ b/math/cimag_template.c @@ -26,7 +26,3 @@ M_DECL_FUNC (__cimag) (CFLOAT z) } declare_mgen_alias (__cimag, cimag) - -#if M_LIBM_NEED_COMPAT (cimag) -declare_mgen_libm_compat (__cimag, cimag) -#endif diff --git a/math/conj_template.c b/math/conj_template.c index b71f338..c21a1f1 100644 --- a/math/conj_template.c +++ b/math/conj_template.c @@ -26,7 +26,3 @@ M_DECL_FUNC (__conj) (CFLOAT z) } declare_mgen_alias (__conj, conj) - -#if M_LIBM_NEED_COMPAT (conj) -declare_mgen_libm_compat (__conj, conj) -#endif diff --git a/math/creal_template.c b/math/creal_template.c index c658966..8ccbe6f 100644 --- a/math/creal_template.c +++ b/math/creal_template.c @@ -26,7 +26,3 @@ M_DECL_FUNC (__creal) (CFLOAT z) } declare_mgen_alias (__creal, creal) - -#if M_LIBM_NEED_COMPAT (creal) -declare_mgen_libm_compat (__creal, creal) -#endif diff --git a/math/s_cacos_template.c b/math/s_cacos_template.c index 7629e55..29cac62 100644 --- a/math/s_cacos_template.c +++ b/math/s_cacos_template.c @@ -53,7 +53,3 @@ M_DECL_FUNC (__cacos) (CFLOAT x) } declare_mgen_alias (__cacos, cacos); - -#if M_LIBM_NEED_COMPAT (carg) -declare_mgen_libm_compat (__cacos, cacos) -#endif diff --git a/math/s_cacosh_template.c b/math/s_cacosh_template.c index 1fe0f4f..14f0b2a 100644 --- a/math/s_cacosh_template.c +++ b/math/s_cacosh_template.c @@ -90,7 +90,3 @@ M_DECL_FUNC (__cacosh) (CFLOAT x) } declare_mgen_alias (__cacosh, cacosh) - -#if M_LIBM_NEED_COMPAT (cacosh) -declare_mgen_libm_compat (__cacosh, cacosh) -#endif diff --git a/math/s_casin_template.c b/math/s_casin_template.c index faa2f6b..c61c880 100644 --- a/math/s_casin_template.c +++ b/math/s_casin_template.c @@ -61,7 +61,3 @@ M_DECL_FUNC (__casin) (CFLOAT x) } declare_mgen_alias (__casin, casin) - -#if M_LIBM_NEED_COMPAT (casin) -declare_mgen_libm_compat (__casin, casin) -#endif diff --git a/math/s_casinh_template.c b/math/s_casinh_template.c index 8d7cc5c..440e639 100644 --- a/math/s_casinh_template.c +++ b/math/s_casinh_template.c @@ -69,7 +69,3 @@ M_DECL_FUNC (__casinh) (CFLOAT x) } declare_mgen_alias (__casinh, casinh) - -#if M_LIBM_NEED_COMPAT (casinh) -declare_mgen_libm_compat (__casinh, casinh) -#endif diff --git a/math/s_catan_template.c b/math/s_catan_template.c index dce613f..2434870 100644 --- a/math/s_catan_template.c +++ b/math/s_catan_template.c @@ -139,7 +139,3 @@ M_DECL_FUNC (__catan) (CFLOAT x) } declare_mgen_alias (__catan, catan) - -#if M_LIBM_NEED_COMPAT (catan) -declare_mgen_libm_compat (__catan, catan) -#endif diff --git a/math/s_catanh_template.c b/math/s_catanh_template.c index 21e06f5..f4f5605 100644 --- a/math/s_catanh_template.c +++ b/math/s_catanh_template.c @@ -133,7 +133,3 @@ M_DECL_FUNC (__catanh) (CFLOAT x) } declare_mgen_alias (__catanh, catanh) - -#if M_LIBM_NEED_COMPAT (catanh) -declare_mgen_libm_compat (__catanh, catanh) -#endif diff --git a/math/s_ccos_template.c b/math/s_ccos_template.c index 01cbae3..6584450 100644 --- a/math/s_ccos_template.c +++ b/math/s_ccos_template.c @@ -34,7 +34,3 @@ M_DECL_FUNC (__ccos) (CFLOAT x) } declare_mgen_alias (__ccos, ccos); - -#if M_LIBM_NEED_COMPAT (carg) -declare_mgen_libm_compat (__ccos, ccos) -#endif diff --git a/math/s_ccosh_template.c b/math/s_ccosh_template.c index 1ee6014..d3ee8fc 100644 --- a/math/s_ccosh_template.c +++ b/math/s_ccosh_template.c @@ -135,7 +135,3 @@ M_DECL_FUNC (__ccosh) (CFLOAT x) } declare_mgen_alias (__ccosh, ccosh); - -#if M_LIBM_NEED_COMPAT (carg) -declare_mgen_libm_compat (__ccosh, ccosh) -#endif diff --git a/math/s_cexp_template.c b/math/s_cexp_template.c index b2c77c1..6290aca 100644 --- a/math/s_cexp_template.c +++ b/math/s_cexp_template.c @@ -148,7 +148,3 @@ M_DECL_FUNC (__cexp) (CFLOAT x) return retval; } declare_mgen_alias (__cexp, cexp) - -#if M_LIBM_NEED_COMPAT (cexp) -declare_mgen_libm_compat (__cexp, cexp) -#endif diff --git a/math/s_clog10_template.c b/math/s_clog10_template.c index 30d87b5..1b004d1 100644 --- a/math/s_clog10_template.c +++ b/math/s_clog10_template.c @@ -120,9 +120,3 @@ M_DECL_FUNC (__clog10) (CFLOAT x) } declare_mgen_alias (__clog10, clog10) - -#if M_LIBM_NEED_COMPAT (clog10) -/* __clog10 is also a public symbol. */ -declare_mgen_libm_compat (__clog10, __clog10) -declare_mgen_libm_compat (clog10, clog10) -#endif diff --git a/math/s_clog_template.c b/math/s_clog_template.c index feb801f..37746a0 100644 --- a/math/s_clog_template.c +++ b/math/s_clog_template.c @@ -113,7 +113,3 @@ M_DECL_FUNC (__clog) (CFLOAT x) } declare_mgen_alias (__clog, clog) - -#if M_LIBM_NEED_COMPAT (clog) -declare_mgen_libm_compat (__clog, clog) -#endif diff --git a/math/s_cpow_template.c b/math/s_cpow_template.c index 604d095..dd1d08d 100644 --- a/math/s_cpow_template.c +++ b/math/s_cpow_template.c @@ -27,7 +27,3 @@ M_DECL_FUNC (__cpow) (CFLOAT x, CFLOAT c) } declare_mgen_alias (__cpow, cpow) - -#if M_LIBM_NEED_COMPAT (cpow) -declare_mgen_libm_compat (__cpow, cpow) -#endif diff --git a/math/s_cproj_template.c b/math/s_cproj_template.c index 7f9e9f0..0ab2727 100644 --- a/math/s_cproj_template.c +++ b/math/s_cproj_template.c @@ -39,7 +39,3 @@ M_DECL_FUNC (__cproj) (CFLOAT x) } declare_mgen_alias (__cproj, cproj) - -#if M_LIBM_NEED_COMPAT (cproj) -declare_mgen_libm_compat (__cproj, cproj) -#endif diff --git a/math/s_csin_template.c b/math/s_csin_template.c index d52beb9..5a38bf7 100644 --- a/math/s_csin_template.c +++ b/math/s_csin_template.c @@ -159,7 +159,3 @@ M_DECL_FUNC (__csin) (CFLOAT x) } declare_mgen_alias (__csin, csin) - -#if M_LIBM_NEED_COMPAT (csin) -declare_mgen_libm_compat (__csin, csin) -#endif diff --git a/math/s_csinh_template.c b/math/s_csinh_template.c index d5ef5ae..ecea37c 100644 --- a/math/s_csinh_template.c +++ b/math/s_csinh_template.c @@ -154,7 +154,3 @@ M_DECL_FUNC (__csinh) (CFLOAT x) } declare_mgen_alias (__csinh, csinh) - -#if M_LIBM_NEED_COMPAT (csinh) -declare_mgen_libm_compat (__csinh, csinh) -#endif diff --git a/math/s_csqrt_template.c b/math/s_csqrt_template.c index 69cf77c..a94195a 100644 --- a/math/s_csqrt_template.c +++ b/math/s_csqrt_template.c @@ -158,7 +158,3 @@ M_DECL_FUNC (__csqrt) (CFLOAT x) return res; } declare_mgen_alias (__csqrt, csqrt) - -#if M_LIBM_NEED_COMPAT (csqrt) -declare_mgen_libm_compat (__csqrt, csqrt) -#endif diff --git a/math/s_ctan_template.c b/math/s_ctan_template.c index baf6030..723156e 100644 --- a/math/s_ctan_template.c +++ b/math/s_ctan_template.c @@ -124,7 +124,3 @@ M_DECL_FUNC (__ctan) (CFLOAT x) } declare_mgen_alias (__ctan, ctan) - -#if M_LIBM_NEED_COMPAT (ctan) -declare_mgen_libm_compat (__ctan, ctan) -#endif diff --git a/math/s_ctanh_template.c b/math/s_ctanh_template.c index 1c00d14..96873dd 100644 --- a/math/s_ctanh_template.c +++ b/math/s_ctanh_template.c @@ -124,7 +124,3 @@ M_DECL_FUNC (__ctanh) (CFLOAT x) } declare_mgen_alias (__ctanh, ctanh) - -#if M_LIBM_NEED_COMPAT (ctanh) -declare_mgen_libm_compat (__ctanh, ctanh) -#endif diff --git a/math/s_fdim_template.c b/math/s_fdim_template.c index 967777a..1fa4b7c 100644 --- a/math/s_fdim_template.c +++ b/math/s_fdim_template.c @@ -34,7 +34,3 @@ M_DECL_FUNC (__fdim) (FLOAT x, FLOAT y) return r; } declare_mgen_alias (__fdim, fdim); - -#if M_LIBM_NEED_COMPAT (fdim) -declare_mgen_libm_compat (__fdim, fdim) -#endif diff --git a/math/s_fmax_template.c b/math/s_fmax_template.c index 0471ab5..fe812c7 100644 --- a/math/s_fmax_template.c +++ b/math/s_fmax_template.c @@ -33,7 +33,3 @@ M_DECL_FUNC (__fmax) (FLOAT x, FLOAT y) } declare_mgen_alias (__fmax, fmax); - -#if M_LIBM_NEED_COMPAT (fmax) -declare_mgen_libm_compat (__fmax, fmax) -#endif diff --git a/math/s_fmin_template.c b/math/s_fmin_template.c index 5f81409..5e8460a 100644 --- a/math/s_fmin_template.c +++ b/math/s_fmin_template.c @@ -33,7 +33,3 @@ M_DECL_FUNC (__fmin) (FLOAT x, FLOAT y) return isnan (y) ? x : y; } declare_mgen_alias (__fmin, fmin); - -#if M_LIBM_NEED_COMPAT (fmin) -declare_mgen_libm_compat (__fmin, fmin) -#endif diff --git a/math/s_nan_template.c b/math/s_nan_template.c index 010dacb..80a9aea 100644 --- a/math/s_nan_template.c +++ b/math/s_nan_template.c @@ -31,7 +31,3 @@ M_DECL_FUNC (__nan) (const char *tagp) } declare_mgen_alias (__nan, nan) - -#if M_LIBM_NEED_COMPAT (nan) -declare_mgen_libm_compat (__nan, nan) -#endif diff --git a/math/w_ilogb_template.c b/math/w_ilogb_template.c index f18dd42..a611f00 100644 --- a/math/w_ilogb_template.c +++ b/math/w_ilogb_template.c @@ -36,7 +36,3 @@ M_DECL_FUNC (__ilogb) (FLOAT x) return r; } declare_mgen_alias (__ilogb, ilogb) - -#if M_LIBM_NEED_COMPAT (ilogb) -declare_mgen_libm_compat (__ilogb, ilogb) -#endif diff --git a/sysdeps/generic/libm-alias-double.h b/sysdeps/generic/libm-alias-double.h new file mode 100644 index 0000000..23b5990 --- /dev/null +++ b/sysdeps/generic/libm-alias-double.h @@ -0,0 +1,42 @@ +/* Define aliases for libm double functions. + Copyright (C) 2017 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 + . */ + +#ifndef _LIBM_ALIAS_DOUBLE_H +#define _LIBM_ALIAS_DOUBLE_H + +/* Define aliases for a double libm function that has internal name + FROM ## R and public names TO ## suffix ## R for each suffix of a + supported floating-point type with the same format as double. This + should only be used for functions where such public names exist for + _FloatN types, not for implementation-namespace exported names + (where there is one name per format, not per type) or for + obsolescent functions not provided for _FloatN types. */ +#ifdef NO_LONG_DOUBLE +# define libm_alias_double_r(from, to, r) \ + weak_alias (from ## r, to ## r) \ + strong_alias (from ## r, from ## l ## r) \ + weak_alias (from ## r, to ## l ## r) +#else +# define libm_alias_double_r(from, to, r) \ + weak_alias (from ## r, to ## r) +#endif + +/* Likewise, but without the R suffix. */ +#define libm_alias_double(from, to) libm_alias_double_r (from, to, ) + +#endif diff --git a/sysdeps/generic/math-type-macros-double.h b/sysdeps/generic/math-type-macros-double.h index f2c1bb8..e978f90 100644 --- a/sysdeps/generic/math-type-macros-double.h +++ b/sysdeps/generic/math-type-macros-double.h @@ -27,14 +27,10 @@ #define CFLOAT _Complex double #define M_STRTO_NAN __strtod_nan -/* Machines without a distinct long double type - alias long double functions to their double - equivalent. */ -#if defined NO_LONG_DOUBLE && !defined declare_mgen_alias -# define declare_mgen_alias(from, to) \ - weak_alias (from, to) \ - strong_alias (from, from ## l) \ - weak_alias (from, to ## l) +#include + +#ifndef declare_mgen_alias +# define declare_mgen_alias(from, to) libm_alias_double (from, to) #endif /* Supply the generic macros. */ diff --git a/sysdeps/generic/math-type-macros.h b/sysdeps/generic/math-type-macros.h index 5656724..6aaa233 100644 --- a/sysdeps/generic/math-type-macros.h +++ b/sysdeps/generic/math-type-macros.h @@ -39,16 +39,7 @@ declare_mgen_alias(from,to) This exposes the appropriate symbol(s) for a - function f of type FLOAT. - - M_LIBM_NEED_COMPAT(func) - This is utilized in macro context to indicate - whether func should declare compat symbols. - - declare_mgen_libm_compat(from,to) - This is used in conjunction with the above macro - outside of macro context to paste whatever is - required to generate a compat symbol. */ + function f of type FLOAT. */ #ifndef M_PFX # error "M_PFX must be defined." @@ -116,10 +107,4 @@ # define declare_mgen_alias(from, to) weak_alias (M_SUF (from), M_SUF (to)) #endif -/* Do not generate anything for compat symbols by default. */ -#ifndef M_LIBM_NEED_COMPAT -# define M_LIBM_NEED_COMPAT(func) 0 -# define declare_mgen_libm_compat(from, to) -#endif - #endif /* _MATH_TYPE_MACROS */ diff --git a/sysdeps/ieee754/ldbl-opt/libm-alias-double.h b/sysdeps/ieee754/ldbl-opt/libm-alias-double.h new file mode 100644 index 0000000..5ae4c8b --- /dev/null +++ b/sysdeps/ieee754/ldbl-opt/libm-alias-double.h @@ -0,0 +1,44 @@ +/* Define aliases for libm double functions. ldbl-opt version. + Copyright (C) 2017 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 + . */ + +#ifndef _LIBM_ALIAS_DOUBLE_H +#define _LIBM_ALIAS_DOUBLE_H + +#include +#include +#include + +/* Define aliases for a double libm function that has internal name + FROM ## R and public names TO ## suffix ## R for each suffix of a + supported floating-point type with the same format as double. This + should only be used for functions where such public names exist for + _FloatN types, not for implementation-namespace exported names + (where there is one name per format, not per type) or for + obsolescent functions not provided for _FloatN types. */ +#define libm_alias_double_r(from, to, r) \ + weak_alias (from ## r, to ## r) \ + LONG_DOUBLE_COMPAT_CHOOSE_libm_ ## to ## l ## r \ + (compat_symbol (libm, \ + from ## r, \ + to ## l ## r, \ + FIRST_VERSION_libm_ ## to ## l ## r), ); + +/* Likewise, but without the R suffix. */ +#define libm_alias_double(from, to) libm_alias_double_r (from, to, ) + +#endif diff --git a/sysdeps/ieee754/ldbl-opt/math-type-macros-double.h b/sysdeps/ieee754/ldbl-opt/math-type-macros-double.h deleted file mode 100644 index 67b5268..0000000 --- a/sysdeps/ieee754/ldbl-opt/math-type-macros-double.h +++ /dev/null @@ -1,38 +0,0 @@ -/* Overrides for ldbl-opt versioning for double types. - Copyright (C) 2016-2017 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 - . */ - -#ifndef _MATH_TYPE_MACROS_DOUBLE - -#include -#include - -/* Define compat symbols for long double on platforms - where it was not always a distinct type. */ -#if !defined M_LIBM_NEED_COMPAT -# define M_LIBM_NEED_COMPAT(f) \ - LONG_DOUBLE_COMPAT (libm, FIRST_VERSION_libm_ ## f ## l) -#endif - -#if !defined declare_mgen_libm_compat -# define declare_mgen_libm_compat(from, to) \ - compat_symbol (libm, from, to ## l, \ - FIRST_VERSION_libm_ ## to ## l); -#endif - -#include_next -#endif diff --git a/sysdeps/ieee754/ldbl-opt/s_clog10.c b/sysdeps/ieee754/ldbl-opt/s_clog10.c new file mode 100644 index 0000000..ec52218 --- /dev/null +++ b/sysdeps/ieee754/ldbl-opt/s_clog10.c @@ -0,0 +1,28 @@ +/* Define __clog10l compat symbol for clog10 for ldbl-opt. + Copyright (C) 2017 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 + . */ + +#include +#include +#include + +#include + +#if LONG_DOUBLE_COMPAT (libm, FIRST_VERSION_libm___clog10l) +strong_alias (__clog10, __clog10l_alias) +compat_symbol (libm, __clog10l_alias, __clog10l, FIRST_VERSION_libm___clog10l); +#endif diff --git a/sysdeps/ieee754/ldbl-opt/s_ldexp.c b/sysdeps/ieee754/ldbl-opt/s_ldexp.c index 809080a..5a875ae 100644 --- a/sysdeps/ieee754/ldbl-opt/s_ldexp.c +++ b/sysdeps/ieee754/ldbl-opt/s_ldexp.c @@ -17,7 +17,7 @@ License along with the GNU C Library; if not, see . */ -#define M_LIBM_NEED_COMPAT(f) 0 +#define declare_mgen_alias(from, to) weak_alias (from, to) #include #include diff --git a/sysdeps/ieee754/ldbl-opt/w_log1p.c b/sysdeps/ieee754/ldbl-opt/w_log1p.c new file mode 100644 index 0000000..ad05247 --- /dev/null +++ b/sysdeps/ieee754/ldbl-opt/w_log1p.c @@ -0,0 +1,3 @@ +#define declare_mgen_alias(from, to) weak_alias (from, to) +#include +#include diff --git a/sysdeps/ieee754/ldbl-opt/w_scalbln.c b/sysdeps/ieee754/ldbl-opt/w_scalbln.c new file mode 100644 index 0000000..495169f --- /dev/null +++ b/sysdeps/ieee754/ldbl-opt/w_scalbln.c @@ -0,0 +1,3 @@ +#define declare_mgen_alias(from, to) weak_alias (from, to) +#include +#include diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim-vis3.c b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim-vis3.c index 8c3666d..daa477b 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim-vis3.c +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim-vis3.c @@ -20,6 +20,5 @@ #define __fdim __fdim_vis3 #define declare_mgen_alias(t, f) -#define M_LIBM_NEED_COMPAT(f) 0 #include diff --git a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim.c b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim.c index 417b869..ff3acd4 100644 --- a/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim.c +++ b/sysdeps/sparc/sparc32/sparcv9/fpu/multiarch/s_fdim.c @@ -17,6 +17,8 @@ . */ #ifdef HAVE_AS_VIS3_SUPPORT +# include +# include # include # include @@ -25,6 +27,9 @@ extern double __fdim_generic (double, double); sparc_libm_ifunc(__fdim, hwcap & HWCAP_SPARC_VIS3 ? __fdim_vis3 : __fdim_generic); weak_alias (__fdim, fdim) +# if LONG_DOUBLE_COMPAT (libm, FIRST_VERSION_libm_fdiml) +compat_symbol (libm, __fdim, fdiml, FIRST_VERSION_libm_fdiml); +# endif # define __fdim __fdim_generic # define declare_mgen_alias(t, f)