From patchwork Wed Nov 15 02:09:35 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joseph Myers X-Patchwork-Id: 24247 Received: (qmail 97080 invoked by alias); 15 Nov 2017 02:09:45 -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 97067 invoked by uid 89); 15 Nov 2017 02:09:44 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-24.3 required=5.0 tests=AWL, BAYES_00, GIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3, KB_WAM_FROM_NAME_SINGLEWORD, RCVD_IN_DNSWL_NONE, SPF_PASS, URIBL_RED autolearn=ham version=3.3.2 spammy=3607, looser X-HELO: relay1.mentorg.com Date: Wed, 15 Nov 2017 02:09:35 +0000 From: Joseph Myers To: Subject: Use __builtin_tgmath in tgmath.h with GCC 8 (bug 21660) [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-03.mgc.mentorg.com (139.181.222.3) GCC mainline now supports __builtin_tgmath to allow macro implementations that expand their arguments only once, so avoiding exponential blowup in the size of macro expansions when calls to those macros are nested in arguments to those macros. This patch makes glibc's tgmath.h support using __builtin_tgmath, as a much simpler and more efficient alternative to the existing implementation. (As a side effect, the new feature would make it much more practical to support decimal floating point in with new compilers; currently, libdfp does not provide a implementation, and making decimal arguments cause integer arguments to be considered of type _Decimal64 instead of double would have been very problematic in the old implementation.) Tested for x86_64 (with GCC mainline). Committed. 2017-11-15 Joseph Myers [BZ #21660] * math/tgmath.h (__HAVE_BUILTIN_TGMATH): New macro. [__HAVE_BUILTIN_TGMATH] (__TG_F16_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F32_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F64_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F128_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F32X_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F64X_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TG_F128X_ARG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_FUNCS): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_RCFUNCS): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_1): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_2): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_2STD): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_3): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_1C): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_2C): Likewise. (__tgml): Make conditional on [!__HAVE_BUILTIN_TGMATH]. (__floating_type): Likewise. (__real_integer_type): Likewise. (__complex_integer_type): Likewise. (__expr_is_real): Likewise. (__tgmath_real_type_sub): Likewise. (__tgmath_real_type): Likewise. (__tgmath_complex_type_sub): Likewise. (__tgmath_complex_type): Likewise. (__TGMATH_F128): Likewise. (__TGMATH_CF128): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_ONLY): Define using new macros. [__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_RET_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_FIRST_REAL_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_FIRST_REAL_STD_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_STD_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_RET_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_FIRST_SECOND_REAL_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_REAL_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_TERNARY_FIRST_REAL_RET_ONLY): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_IMAG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_IMAG): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_UNARY_REAL_IMAG_RET_REAL): Likewise. [__HAVE_BUILTIN_TGMATH] (__TGMATH_BINARY_REAL_IMAG): Likewise. (__TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME): New macro. (carg): Use __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME. (cimag): Likewise. (creal): Likewise. diff --git a/math/tgmath.h b/math/tgmath.h index 108c1a1..1216177 100644 --- a/math/tgmath.h +++ b/math/tgmath.h @@ -31,22 +31,76 @@ #include -/* Since `complex' is currently not really implemented in most C compilers - and if it is implemented, the implementations differ. This makes it - quite difficult to write a generic implementation of this header. We - do not try this for now and instead concentrate only on GNU CC. Once - we have more information support for other compilers might follow. */ +/* There are two variant implementations of type-generic macros in + this file: one for GCC 8 and later, using __builtin_tgmath and + where each macro expands each of its arguments only once, and one + for older GCC, using other compiler extensions but with macros + expanding their arguments many times (so resulting in exponential + blowup of the size of expansions when calls to such macros are + nested inside arguments to such macros). */ + +#define __HAVE_BUILTIN_TGMATH __GNUC_PREREQ (8, 0) #if __GNUC_PREREQ (2, 7) -# ifdef __NO_LONG_DOUBLE_MATH -# define __tgml(fct) fct -# else -# define __tgml(fct) fct ## l -# endif +# if __HAVE_BUILTIN_TGMATH -/* This is ugly but unless gcc gets appropriate builtins we have to do - something like this. Don't ask how it works. */ +# if __HAVE_FLOAT16 && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F16_ARG(X) X ## f16, +# else +# define __TG_F16_ARG(X) +# endif +# if __HAVE_FLOAT32 && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F32_ARG(X) X ## f32, +# else +# define __TG_F32_ARG(X) +# endif +# if __HAVE_FLOAT64 && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F64_ARG(X) X ## f64, +# else +# define __TG_F64_ARG(X) +# endif +# if __HAVE_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F128_ARG(X) X ## f128, +# else +# define __TG_F128_ARG(X) +# endif +# if __HAVE_FLOAT32X && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F32X_ARG(X) X ## f32x, +# else +# define __TG_F32X_ARG(X) +# endif +# if __HAVE_FLOAT64X && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F64X_ARG(X) X ## f64x, +# else +# define __TG_F64X_ARG(X) +# endif +# if __HAVE_FLOAT128X && __GLIBC_USE (IEC_60559_TYPES_EXT) +# define __TG_F128X_ARG(X) X ## f128x, +# else +# define __TG_F128X_ARG(X) +# endif + +# define __TGMATH_FUNCS(X) X ## f, X, X ## l, \ + __TG_F16_ARG (X) __TG_F32_ARG (X) __TG_F64_ARG (X) __TG_F128_ARG (X) \ + __TG_F32X_ARG (X) __TG_F64X_ARG (X) __TG_F128X_ARG (X) +# define __TGMATH_RCFUNCS(F, C) __TGMATH_FUNCS (F) __TGMATH_FUNCS (C) +# define __TGMATH_1(F, X) __builtin_tgmath (__TGMATH_FUNCS (F) (X)) +# define __TGMATH_2(F, X, Y) __builtin_tgmath (__TGMATH_FUNCS (F) (X), (Y)) +# define __TGMATH_2STD(F, X, Y) __builtin_tgmath (F ## f, F, F ## l, (X), (Y)) +# define __TGMATH_3(F, X, Y, Z) __builtin_tgmath (__TGMATH_FUNCS (F) \ + (X), (Y), (Z)) +# define __TGMATH_1C(F, C, X) __builtin_tgmath (__TGMATH_RCFUNCS (F, C) (X)) +# define __TGMATH_2C(F, C, X, Y) __builtin_tgmath (__TGMATH_RCFUNCS (F, C) \ + (X), (Y)) + +# else /* !__HAVE_BUILTIN_TGMATH. */ + +# ifdef __NO_LONG_DOUBLE_MATH +# define __tgml(fct) fct +# else +# define __tgml(fct) fct ## l +# endif /* __floating_type expands to 1 if TYPE is a floating type (including complex floating types), 0 if TYPE is an integer type (including @@ -56,48 +110,48 @@ constant expressions. All these macros can assume their argument has an arithmetic type (not vector, decimal floating-point or fixed-point), valid to pass to tgmath.h macros. */ -# if __GNUC_PREREQ (3, 1) +# if __GNUC_PREREQ (3, 1) /* __builtin_classify_type expands to an integer constant expression in GCC 3.1 and later. Default conversions applied to the argument of __builtin_classify_type mean it always returns 1 for real integer types rather than ever returning different values for character, boolean or enumerated types. */ -# define __floating_type(type) \ +# define __floating_type(type) \ (__builtin_classify_type (__real__ ((type) 0)) == 8) -# define __real_integer_type(type) \ +# define __real_integer_type(type) \ (__builtin_classify_type ((type) 0) == 1) -# define __complex_integer_type(type) \ +# define __complex_integer_type(type) \ (__builtin_classify_type ((type) 0) == 9 \ && __builtin_classify_type (__real__ ((type) 0)) == 1) -# else +# else /* GCC versions predating __builtin_classify_type are also looser on what counts as an integer constant expression. */ -# define __floating_type(type) (((type) 1.25) != 1) -# define __real_integer_type(type) (((type) (1.25 + _Complex_I)) == 1) -# define __complex_integer_type(type) \ +# define __floating_type(type) (((type) 1.25) != 1) +# define __real_integer_type(type) (((type) (1.25 + _Complex_I)) == 1) +# define __complex_integer_type(type) \ (((type) (1.25 + _Complex_I)) == (1 + _Complex_I)) -# endif +# endif /* Whether an expression (of arithmetic type) has a real type. */ -# define __expr_is_real(E) (__builtin_classify_type (E) != 9) +# define __expr_is_real(E) (__builtin_classify_type (E) != 9) /* The tgmath real type for T, where E is 0 if T is an integer type and 1 for a floating type. If T has a complex type, it is unspecified whether the return type is real or complex (but it has the correct corresponding real type). */ -# define __tgmath_real_type_sub(T, E) \ +# define __tgmath_real_type_sub(T, E) \ __typeof__ (*(0 ? (__typeof__ (0 ? (double *) 0 : (void *) (E))) 0 \ : (__typeof__ (0 ? (T *) 0 : (void *) (!(E)))) 0)) /* The tgmath real type of EXPR. */ -# define __tgmath_real_type(expr) \ +# define __tgmath_real_type(expr) \ __tgmath_real_type_sub (__typeof__ ((__typeof__ (+(expr))) 0), \ __floating_type (__typeof__ (+(expr)))) /* The tgmath complex type for T, where E1 is 1 if T has a floating type and 0 otherwise, E2 is 1 if T has a real integer type and 0 otherwise, and E3 is 1 if T has a complex type and 0 otherwise. */ -# define __tgmath_complex_type_sub(T, E1, E2, E3) \ +# define __tgmath_complex_type_sub(T, E1, E2, E3) \ __typeof__ (*(0 \ ? (__typeof__ (0 ? (T *) 0 : (void *) (!(E1)))) 0 \ : (__typeof__ (0 \ @@ -109,61 +163,94 @@ : (void *) (!(E3)))) 0)) 0)) /* The tgmath complex type of EXPR. */ -# define __tgmath_complex_type(expr) \ +# define __tgmath_complex_type(expr) \ __tgmath_complex_type_sub (__typeof__ ((__typeof__ (+(expr))) 0), \ __floating_type (__typeof__ (+(expr))), \ __real_integer_type (__typeof__ (+(expr))), \ __complex_integer_type (__typeof__ (+(expr)))) -# if (__HAVE_DISTINCT_FLOAT16 \ +# if (__HAVE_DISTINCT_FLOAT16 \ || __HAVE_DISTINCT_FLOAT32 \ || __HAVE_DISTINCT_FLOAT64 \ || __HAVE_DISTINCT_FLOAT32X \ || __HAVE_DISTINCT_FLOAT64X \ || __HAVE_DISTINCT_FLOAT128X) -# error "Unsupported _FloatN or _FloatNx types for ." -# endif +# error "Unsupported _FloatN or _FloatNx types for ." +# endif /* Expand to text that checks if ARG_COMB has type _Float128, and if so calls the appropriately suffixed FCT (which may include a cast), or FCT and CFCT for complex functions, with arguments ARG_CALL. */ -# if __HAVE_DISTINCT_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT) -# if (!__HAVE_FLOAT64X \ +# if __HAVE_DISTINCT_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT) +# if (!__HAVE_FLOAT64X \ || __HAVE_FLOAT64X_LONG_DOUBLE \ || !__HAVE_FLOATN_NOT_TYPEDEF) -# define __TGMATH_F128(arg_comb, fct, arg_call) \ +# define __TGMATH_F128(arg_comb, fct, arg_call) \ __builtin_types_compatible_p (__typeof (+(arg_comb)), _Float128) \ ? fct ## f128 arg_call : -# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) \ +# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) \ __builtin_types_compatible_p (__typeof (+__real__ (arg_comb)), _Float128) \ ? (__expr_is_real (arg_comb) \ ? fct ## f128 arg_call \ : cfct ## f128 arg_call) : -# else +# else /* _Float64x is a distinct type at the C language level, which must be handled like _Float128. */ -# define __TGMATH_F128(arg_comb, fct, arg_call) \ +# define __TGMATH_F128(arg_comb, fct, arg_call) \ (__builtin_types_compatible_p (__typeof (+(arg_comb)), _Float128) \ || __builtin_types_compatible_p (__typeof (+(arg_comb)), _Float64x)) \ ? fct ## f128 arg_call : -# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) \ +# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) \ (__builtin_types_compatible_p (__typeof (+__real__ (arg_comb)), _Float128) \ || __builtin_types_compatible_p (__typeof (+__real__ (arg_comb)), \ _Float64x)) \ ? (__expr_is_real (arg_comb) \ ? fct ## f128 arg_call \ : cfct ## f128 arg_call) : +# endif +# else +# define __TGMATH_F128(arg_comb, fct, arg_call) /* Nothing. */ +# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) /* Nothing. */ # endif -# else -# define __TGMATH_F128(arg_comb, fct, arg_call) /* Nothing. */ -# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) /* Nothing. */ -# endif +# endif /* !__HAVE_BUILTIN_TGMATH. */ /* We have two kinds of generic macros: to support functions which are only defined on real valued parameters and those which are defined for complex functions as well. */ -# define __TGMATH_UNARY_REAL_ONLY(Val, Fct) \ +# if __HAVE_BUILTIN_TGMATH + +# define __TGMATH_UNARY_REAL_ONLY(Val, Fct) __TGMATH_1 (Fct, (Val)) +# define __TGMATH_UNARY_REAL_RET_ONLY(Val, Fct) __TGMATH_1 (Fct, (Val)) +# define __TGMATH_BINARY_FIRST_REAL_ONLY(Val1, Val2, Fct) \ + __TGMATH_2 (Fct, (Val1), (Val2)) +# define __TGMATH_BINARY_FIRST_REAL_STD_ONLY(Val1, Val2, Fct) \ + __TGMATH_2STD (Fct, (Val1), (Val2)) +# define __TGMATH_BINARY_REAL_ONLY(Val1, Val2, Fct) \ + __TGMATH_2 (Fct, (Val1), (Val2)) +# define __TGMATH_BINARY_REAL_STD_ONLY(Val1, Val2, Fct) \ + __TGMATH_2STD (Fct, (Val1), (Val2)) +# define __TGMATH_BINARY_REAL_RET_ONLY(Val1, Val2, Fct) \ + __TGMATH_2 (Fct, (Val1), (Val2)) +# define __TGMATH_TERNARY_FIRST_SECOND_REAL_ONLY(Val1, Val2, Val3, Fct) \ + __TGMATH_3 (Fct, (Val1), (Val2), (Val3)) +# define __TGMATH_TERNARY_REAL_ONLY(Val1, Val2, Val3, Fct) \ + __TGMATH_3 (Fct, (Val1), (Val2), (Val3)) +# define __TGMATH_TERNARY_FIRST_REAL_RET_ONLY(Val1, Val2, Val3, Fct) \ + __TGMATH_3 (Fct, (Val1), (Val2), (Val3)) +# define __TGMATH_UNARY_REAL_IMAG(Val, Fct, Cfct) \ + __TGMATH_1C (Fct, Cfct, (Val)) +# define __TGMATH_UNARY_IMAG(Val, Cfct) __TGMATH_1 (Cfct, (Val)) +# define __TGMATH_UNARY_REAL_IMAG_RET_REAL(Val, Fct, Cfct) \ + __TGMATH_1C (Fct, Cfct, (Val)) +# define __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME(Val, Cfct) \ + __TGMATH_1 (Cfct, (Val)) +# define __TGMATH_BINARY_REAL_IMAG(Val1, Val2, Fct, Cfct) \ + __TGMATH_2C (Fct, Cfct, (Val1), (Val2)) + +# else /* !__HAVE_BUILTIN_TGMATH. */ + +# define __TGMATH_UNARY_REAL_ONLY(Val, Fct) \ (__extension__ ((sizeof (+(Val)) == sizeof (double) \ || __builtin_classify_type (Val) != 8) \ ? (__tgmath_real_type (Val)) Fct (Val) \ @@ -173,7 +260,7 @@ (Val)) \ (__tgmath_real_type (Val)) __tgml(Fct) (Val))) -# define __TGMATH_UNARY_REAL_RET_ONLY(Val, Fct) \ +# define __TGMATH_UNARY_REAL_RET_ONLY(Val, Fct) \ (__extension__ ((sizeof (+(Val)) == sizeof (double) \ || __builtin_classify_type (Val) != 8) \ ? Fct (Val) \ @@ -182,7 +269,7 @@ : __TGMATH_F128 ((Val), Fct, (Val)) \ __tgml(Fct) (Val))) -# define __TGMATH_BINARY_FIRST_REAL_ONLY(Val1, Val2, Fct) \ +# define __TGMATH_BINARY_FIRST_REAL_ONLY(Val1, Val2, Fct) \ (__extension__ ((sizeof (+(Val1)) == sizeof (double) \ || __builtin_classify_type (Val1) != 8) \ ? (__tgmath_real_type (Val1)) Fct (Val1, Val2) \ @@ -192,7 +279,7 @@ (Val1, Val2)) \ (__tgmath_real_type (Val1)) __tgml(Fct) (Val1, Val2))) -# define __TGMATH_BINARY_FIRST_REAL_STD_ONLY(Val1, Val2, Fct) \ +# define __TGMATH_BINARY_FIRST_REAL_STD_ONLY(Val1, Val2, Fct) \ (__extension__ ((sizeof (+(Val1)) == sizeof (double) \ || __builtin_classify_type (Val1) != 8) \ ? (__tgmath_real_type (Val1)) Fct (Val1, Val2) \ @@ -200,7 +287,7 @@ ? (__tgmath_real_type (Val1)) Fct##f (Val1, Val2) \ : (__tgmath_real_type (Val1)) __tgml(Fct) (Val1, Val2))) -# define __TGMATH_BINARY_REAL_ONLY(Val1, Val2, Fct) \ +# define __TGMATH_BINARY_REAL_ONLY(Val1, Val2, Fct) \ (__extension__ ((sizeof ((Val1) + (Val2)) > sizeof (double) \ && __builtin_classify_type ((Val1) + (Val2)) == 8) \ ? __TGMATH_F128 ((Val1) + (Val2), \ @@ -222,7 +309,7 @@ + (__tgmath_real_type (Val2)) 0)) \ Fct##f (Val1, Val2))) -# define __TGMATH_BINARY_REAL_STD_ONLY(Val1, Val2, Fct) \ +# define __TGMATH_BINARY_REAL_STD_ONLY(Val1, Val2, Fct) \ (__extension__ ((sizeof ((Val1) + (Val2)) > sizeof (double) \ && __builtin_classify_type ((Val1) + (Val2)) == 8) \ ? (__typeof ((__tgmath_real_type (Val1)) 0 \ @@ -239,7 +326,7 @@ + (__tgmath_real_type (Val2)) 0)) \ Fct##f (Val1, Val2))) -# define __TGMATH_BINARY_REAL_RET_ONLY(Val1, Val2, Fct) \ +# define __TGMATH_BINARY_REAL_RET_ONLY(Val1, Val2, Fct) \ (__extension__ ((sizeof ((Val1) + (Val2)) > sizeof (double) \ && __builtin_classify_type ((Val1) + (Val2)) == 8) \ ? __TGMATH_F128 ((Val1) + (Val2), Fct, (Val1, Val2)) \ @@ -251,7 +338,7 @@ ? Fct (Val1, Val2) \ : Fct##f (Val1, Val2))) -# define __TGMATH_TERNARY_FIRST_SECOND_REAL_ONLY(Val1, Val2, Val3, Fct) \ +# define __TGMATH_TERNARY_FIRST_SECOND_REAL_ONLY(Val1, Val2, Val3, Fct) \ (__extension__ ((sizeof ((Val1) + (Val2)) > sizeof (double) \ && __builtin_classify_type ((Val1) + (Val2)) == 8) \ ? __TGMATH_F128 ((Val1) + (Val2), \ @@ -273,7 +360,7 @@ + (__tgmath_real_type (Val2)) 0)) \ Fct##f (Val1, Val2, Val3))) -# define __TGMATH_TERNARY_REAL_ONLY(Val1, Val2, Val3, Fct) \ +# define __TGMATH_TERNARY_REAL_ONLY(Val1, Val2, Val3, Fct) \ (__extension__ ((sizeof ((Val1) + (Val2) + (Val3)) > sizeof (double) \ && __builtin_classify_type ((Val1) + (Val2) + (Val3)) \ == 8) \ @@ -302,7 +389,7 @@ + (__tgmath_real_type (Val3)) 0)) \ Fct##f (Val1, Val2, Val3))) -# define __TGMATH_TERNARY_FIRST_REAL_RET_ONLY(Val1, Val2, Val3, Fct) \ +# define __TGMATH_TERNARY_FIRST_REAL_RET_ONLY(Val1, Val2, Val3, Fct) \ (__extension__ ((sizeof (+(Val1)) == sizeof (double) \ || __builtin_classify_type (Val1) != 8) \ ? Fct (Val1, Val2, Val3) \ @@ -313,7 +400,7 @@ /* XXX This definition has to be changed as soon as the compiler understands the imaginary keyword. */ -# define __TGMATH_UNARY_REAL_IMAG(Val, Fct, Cfct) \ +# define __TGMATH_UNARY_REAL_IMAG(Val, Fct, Cfct) \ (__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \ || __builtin_classify_type (__real__ (Val)) != 8) \ ? (__expr_is_real (Val) \ @@ -331,7 +418,7 @@ ? (__tgmath_complex_type (Val)) __tgml(Fct) (Val) \ : (__tgmath_complex_type (Val)) __tgml(Cfct) (Val)))) -# define __TGMATH_UNARY_IMAG(Val, Cfct) \ +# define __TGMATH_UNARY_IMAG(Val, Cfct) \ (__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \ || __builtin_classify_type (__real__ (Val)) != 8) \ ? (__typeof__ ((__tgmath_real_type (Val)) 0 \ @@ -348,7 +435,7 @@ /* XXX This definition has to be changed as soon as the compiler understands the imaginary keyword. */ -# define __TGMATH_UNARY_REAL_IMAG_RET_REAL(Val, Fct, Cfct) \ +# define __TGMATH_UNARY_REAL_IMAG_RET_REAL(Val, Fct, Cfct) \ (__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \ || __builtin_classify_type (__real__ (Val)) != 8) \ ? (__expr_is_real (Val) \ @@ -375,10 +462,12 @@ __tgml(Fct) (Val) \ : (__typeof__ (__real__ (__tgmath_real_type (Val)) 0)) \ __tgml(Cfct) (Val)))) +# define __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME(Val, Cfct) \ + __TGMATH_UNARY_REAL_IMAG_RET_REAL ((Val), Cfct, Cfct) /* XXX This definition has to be changed as soon as the compiler understands the imaginary keyword. */ -# define __TGMATH_BINARY_REAL_IMAG(Val1, Val2, Fct, Cfct) \ +# define __TGMATH_BINARY_REAL_IMAG(Val1, Val2, Fct, Cfct) \ (__extension__ ((sizeof (__real__ (Val1) \ + __real__ (Val2)) > sizeof (double) \ && __builtin_classify_type (__real__ (Val1) \ @@ -418,6 +507,7 @@ : (__typeof ((__tgmath_complex_type (Val1)) 0 \ + (__tgmath_complex_type (Val2)) 0)) \ Cfct##f (Val1, Val2)))) +# endif /* !__HAVE_BUILTIN_TGMATH. */ #else # error "Unsupported compiler; you cannot use " #endif @@ -654,7 +744,7 @@ /* Absolute value, conjugates, and projection. */ /* Argument value of Z. */ -#define carg(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL (Val, carg, carg) +#define carg(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME (Val, carg) /* Complex conjugate of Z. */ #define conj(Val) __TGMATH_UNARY_IMAG (Val, conj) @@ -666,9 +756,9 @@ /* Decomposing complex values. */ /* Imaginary part of Z. */ -#define cimag(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL (Val, cimag, cimag) +#define cimag(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME (Val, cimag) /* Real part of Z. */ -#define creal(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL (Val, creal, creal) +#define creal(Val) __TGMATH_UNARY_REAL_IMAG_RET_REAL_SAME (Val, creal) #endif /* tgmath.h */