From patchwork Tue Dec 5 21:53:25 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joseph Myers X-Patchwork-Id: 24752 Received: (qmail 111541 invoked by alias); 5 Dec 2017 21:53:33 -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 111525 invoked by uid 89); 5 Dec 2017 21:53:32 -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: Tue, 5 Dec 2017 21:53:25 +0000 From: Joseph Myers To: Subject: Use long double not double for _Float64 with old GCC if values the same [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-02.mgc.mentorg.com (139.181.222.2) If double, long double and _Float64 all have the same set of values, TS 18661-3 requires the usual arithmetic conversions on long double and _Float64 to produce _Float64. For this to be the case when building with a compiler without a distinct _Float64 type, _Float64 must be a typedef for long double, not for double. (_Float32x, however, must be double in such a case, not long double, because the usual arithmetic conversions on _Float32x and double must produce double.) This patch adjusts the fallback definition of _Float64 and associated macros accordingly in that case, to fix the build of test-tgmath3 with GCC 6 for such a configuration. Tested in conjunction with _Float64 changes with build-many-glibcs.py for arm-linux-gnueabi, to make sure the issue with test-tgmath3 is fixed. Also tested for x86_64. Committed. 2017-12-05 Joseph Myers * bits/floatn-common.h: Include . [__HAVE_FLOAT64 && (!__GNUC_PREREQ (7, 0) || defined __cplusplus) && __NO_LONG_DOUBLE_MATH] (__f64): Use suffix 'l'. [__HAVE_FLOAT64 && (!__GNUC_PREREQ (7, 0) || defined __cplusplus) && __NO_LONG_DOUBLE_MATH] (__CFLOAT64): Use _Complex long double. [__HAVE_FLOAT64 && (!__GNUC_PREREQ (7, 0) || defined __cplusplus) && __NO_LONG_DOUBLE_MATH] (_Float64): Use long double. [__HAVE_FLOAT64 && !__GNUC_PREREQ (7, 0) && __NO_LONG_DOUBLE_MATH] (__builtin_huge_valf64): Use __builtin_huge_vall. [__HAVE_FLOAT64 && !__GNUC_PREREQ (7, 0) && __NO_LONG_DOUBLE_MATH] (__builtin_inff64): Use __builtin_infl. [__HAVE_FLOAT64 && !__GNUC_PREREQ (7, 0) && __NO_LONG_DOUBLE_MATH] (__builtin_nanf64): Use __builtin_nanl. [__HAVE_FLOAT64 && !__GNUC_PREREQ (7, 0) && __NO_LONG_DOUBLE_MATH] (__builtin_nansf64): Use __builtin_nansl. diff --git a/bits/floatn-common.h b/bits/floatn-common.h index 4d38a91..aa0a757 100644 --- a/bits/floatn-common.h +++ b/bits/floatn-common.h @@ -21,6 +21,7 @@ #define _BITS_FLOATN_COMMON_H #include +#include /* This header should be included at the bottom of each bits/floatn.h. It defines the following macros for each _FloatN and _FloatNx type, @@ -88,7 +89,11 @@ # if __HAVE_FLOAT64 # if !__GNUC_PREREQ (7, 0) || defined __cplusplus -# define __f64(x) x +# ifdef __NO_LONG_DOUBLE_MATH +# define __f64(x) x##l +# else +# define __f64(x) x +# endif # else # define __f64(x) x##f64 # endif @@ -142,7 +147,11 @@ typedef _Complex float __cfloat16 __attribute__ ((__mode__ (__HC__))); # if __HAVE_FLOAT64 # if !__GNUC_PREREQ (7, 0) || defined __cplusplus -# define __CFLOAT64 _Complex double +# ifdef __NO_LONG_DOUBLE_MATH +# define __CFLOAT64 _Complex long double +# else +# define __CFLOAT64 _Complex double +# endif # else # define __CFLOAT64 _Complex _Float64 # endif @@ -209,15 +218,39 @@ typedef float _Float32; # if __HAVE_FLOAT64 -# if !__GNUC_PREREQ (7, 0) || defined __cplusplus +/* If double, long double and _Float64 all have the same set of + values, TS 18661-3 requires the usual arithmetic conversions on + long double and _Float64 to produce _Float64. For this to be the + case when building with a compiler without a distinct _Float64 + type, _Float64 must be a typedef for long double, not for + double. */ + +# ifdef __NO_LONG_DOUBLE_MATH + +# if !__GNUC_PREREQ (7, 0) || defined __cplusplus +typedef long double _Float64; +# endif + +# if !__GNUC_PREREQ (7, 0) +# define __builtin_huge_valf64() (__builtin_huge_vall ()) +# define __builtin_inff64() (__builtin_infl ()) +# define __builtin_nanf64(x) (__builtin_nanl (x)) +# define __builtin_nansf64(x) (__builtin_nansl (x)) +# endif + +# else + +# if !__GNUC_PREREQ (7, 0) || defined __cplusplus typedef double _Float64; -# endif +# endif + +# if !__GNUC_PREREQ (7, 0) +# define __builtin_huge_valf64() (__builtin_huge_val ()) +# define __builtin_inff64() (__builtin_inf ()) +# define __builtin_nanf64(x) (__builtin_nan (x)) +# define __builtin_nansf64(x) (__builtin_nans (x)) +# endif -# if !__GNUC_PREREQ (7, 0) -# define __builtin_huge_valf64() (__builtin_huge_val ()) -# define __builtin_inff64() (__builtin_inf ()) -# define __builtin_nanf64(x) (__builtin_nan (x)) -# define __builtin_nansf64(x) (__builtin_nans (x)) # endif # endif