This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Fix tgmath.h handling of complex integers (bug 21684) [committed]
- From: Joseph Myers <joseph at codesourcery dot com>
- To: <libc-alpha at sourceware dot org>
- Date: Tue, 22 Aug 2017 17:58:00 +0000
- Subject: Fix tgmath.h handling of complex integers (bug 21684) [committed]
- Authentication-results: sourceware.org; auth=none
The tgmath.h macros return a real type not a complex type when an
argument is of complex integer type (a GNU extension) and there are no
arguments of complex floating type. It seems clear that just as real
integers are mapped to double for tgmath.h, so complex integers should
be mapped to _Complex double.
This patch implements such a mapping. The main complication in fixing
this bug is that the tgmath.h macros expand their arguments a large
number of times, resulting in exponential blowup of the size of the
expansion when calls to tgmath.h macros are used in the arguments of
such macros; it would be unfortunate for fixing a bug with a fairly
obscure extension to make the macros expand their arguments even more
times. Thus, this patch optimizes the definitions of the relevant
macros. __tgmath_real_type previously expanded its argument 7 times
and now expands it 3 times. __tgmath_complex_type, used in place of
__tgmath_real_type only for functions that might return either real or
complex types, not for complex functions that always return real types
or always return complex types, expands its argument 5 times. So the
sizes of the macro expansions from nested macro calls are
correspondingly reduced (remembering that each tgmath.h macro expands
__tgmath_real_type, or sometimes now __tgmath_complex_type, several
times).
Sometimes the real return type resulted from calling a complex
function and converting the result to a real type; sometimes it
resulted from calling a real function, because the logic for
determining whether arguments were real or complex, based on sizeof,
was confused by integer promotions applying to e.g. short int but not
_Complex short int. The relevant tests are converted to use a new
macro __expr_is_real, which, by calling __builtin_classify_type rather
than comparing the results of two calls to sizeof, also reduces the
number of times macros expand their arguments.
Although there are reductions in the number of times macros expand
their arguments, I do not consider this to fix bug 21660, since a
proper fix means each macro expanding its arguments only once (via
using new compiler features designed for that purpose).
Tested for x86_64. Committed.
2017-08-22 Joseph Myers <joseph@codesourcery.com>
[BZ #21684]
* math/tgmath.h (__floating_type): Simplify definitions.
(__real_integer_type): New macro.
(__complex_integer_type): Likewise.
(__expr_is_real): Likewise.
(__tgmath_real_type_sub): Update comment to describe handling of
complex types.
(__tgmath_complex_type_sub): New macro.
(__tgmath_complex_type): Likewise.
[__HAVE_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT)]
(__TGMATH_CF128): Use __expr_is_real.
(__TGMATH_UNARY_REAL_IMAG): Use __tgmath_complex_type and
__expr_is_real.
(__TGMATH_BINARY_REAL_IMAG): Likewise.
(__TGMATH_UNARY_REAL_IMAG_RET_REAL): Use __expr_is_real.
* math/gen-tgmath-tests.py (Type.create_type): Create complex
integer types.
diff --git a/math/gen-tgmath-tests.py b/math/gen-tgmath-tests.py
index 9044670..88307a7 100755
--- a/math/gen-tgmath-tests.py
+++ b/math/gen-tgmath-tests.py
@@ -128,9 +128,7 @@ class Type(object):
real_type = Type(name, suffix=suffix, mant_dig=mant_dig,
condition=condition, order=order, integer=integer,
complex=False)
- # Complex integer types currently disabled because of problems
- # in tgmath.h.
- if complex_ok and not integer:
+ if complex_ok:
if complex_name is None:
complex_name = '_Complex %s' % name
complex_type = Type(complex_name, condition=condition,
diff --git a/math/tgmath.h b/math/tgmath.h
index 019f072..68cee64 100644
--- a/math/tgmath.h
+++ b/math/tgmath.h
@@ -48,19 +48,43 @@
/* This is ugly but unless gcc gets appropriate builtins we have to do
something like this. Don't ask how it works. */
-/* 1 if 'type' is a floating type, 0 if 'type' is an integer type.
- Allows for _Bool. Expands to an integer constant expression. */
+/* __floating_type expands to 1 if TYPE is a floating type (including
+ complex floating types), 0 if TYPE is an integer type (including
+ complex integer types). __real_integer_type expands to 1 if TYPE
+ is a real integer type. __complex_integer_type expands to 1 if
+ TYPE is a complex integer type. All these macros expand to integer
+ 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)
-# define __floating_type(type) \
- (__builtin_classify_type ((type) 0) == 8 \
- || (__builtin_classify_type ((type) 0) == 9 \
- && __builtin_classify_type (__real__ ((type) 0)) == 8))
+/* __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) \
+ (__builtin_classify_type (__real__ ((type) 0)) == 8)
+# define __real_integer_type(type) \
+ (__builtin_classify_type ((type) 0) == 1)
+# define __complex_integer_type(type) \
+ (__builtin_classify_type ((type) 0) == 9 \
+ && __builtin_classify_type (__real__ ((type) 0)) == 1)
# else
-# define __floating_type(type) (((type) 0.25) && ((type) 0.25 - 1))
+/* 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) \
+ (((type) (1.25 + _Complex_I)) == (1 + _Complex_I))
# endif
-/* The tgmath real type for T, where E is 0 if T is an integer type and
- 1 for a floating type. */
+/* Whether an expression (of arithmetic type) has a real type. */
+# 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) \
__typeof__ (*(0 ? (__typeof__ (0 ? (double *) 0 : (void *) (E))) 0 \
: (__typeof__ (0 ? (T *) 0 : (void *) (!(E)))) 0))
@@ -70,6 +94,27 @@
__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) \
+ __typeof__ (*(0 \
+ ? (__typeof__ (0 ? (T *) 0 : (void *) (!(E1)))) 0 \
+ : (__typeof__ (0 \
+ ? (__typeof__ (0 \
+ ? (double *) 0 \
+ : (void *) (!(E2)))) 0 \
+ : (__typeof__ (0 \
+ ? (_Complex double *) 0 \
+ : (void *) (!(E3)))) 0)) 0))
+
+/* The tgmath complex type of 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))))
+
/* 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. */
@@ -79,7 +124,7 @@
? fct ## f128 arg_call :
# define __TGMATH_CF128(arg_comb, fct, cfct, arg_call) \
__builtin_types_compatible_p (__typeof (+__real__ (arg_comb)), _Float128) \
- ? (sizeof (+__real__ (arg_comb)) == sizeof (+(arg_comb)) \
+ ? (__expr_is_real (arg_comb) \
? fct ## f128 arg_call \
: cfct ## f128 arg_call) :
# else
@@ -244,19 +289,20 @@
# define __TGMATH_UNARY_REAL_IMAG(Val, Fct, Cfct) \
(__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \
|| __builtin_classify_type (__real__ (Val)) != 8) \
- ? ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
- ? (__tgmath_real_type (Val)) Fct (Val) \
- : (__tgmath_real_type (Val)) Cfct (Val)) \
+ ? (__expr_is_real (Val) \
+ ? (__tgmath_complex_type (Val)) Fct (Val) \
+ : (__tgmath_complex_type (Val)) Cfct (Val)) \
: (sizeof (+__real__ (Val)) == sizeof (float)) \
- ? ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
- ? (__tgmath_real_type (Val)) Fct##f (Val) \
- : (__tgmath_real_type (Val)) Cfct##f (Val)) \
- : __TGMATH_CF128 ((Val), (__tgmath_real_type (Val)) Fct, \
- (__tgmath_real_type (Val)) Cfct, \
+ ? (__expr_is_real (Val) \
+ ? (__tgmath_complex_type (Val)) Fct##f (Val) \
+ : (__tgmath_complex_type (Val)) Cfct##f (Val)) \
+ : __TGMATH_CF128 ((Val), \
+ (__tgmath_complex_type (Val)) Fct, \
+ (__tgmath_complex_type (Val)) Cfct, \
(Val)) \
- ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
- ? (__tgmath_real_type (Val)) __tgml(Fct) (Val) \
- : (__tgmath_real_type (Val)) __tgml(Cfct) (Val))))
+ (__expr_is_real (Val) \
+ ? (__tgmath_complex_type (Val)) __tgml(Fct) (Val) \
+ : (__tgmath_complex_type (Val)) __tgml(Cfct) (Val))))
# define __TGMATH_UNARY_IMAG(Val, Cfct) \
(__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \
@@ -278,13 +324,13 @@
# define __TGMATH_UNARY_REAL_IMAG_RET_REAL(Val, Fct, Cfct) \
(__extension__ ((sizeof (+__real__ (Val)) == sizeof (double) \
|| __builtin_classify_type (__real__ (Val)) != 8) \
- ? ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
+ ? (__expr_is_real (Val) \
? (__typeof__ (__real__ (__tgmath_real_type (Val)) 0))\
Fct (Val) \
: (__typeof__ (__real__ (__tgmath_real_type (Val)) 0))\
Cfct (Val)) \
: (sizeof (+__real__ (Val)) == sizeof (float)) \
- ? ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
+ ? (__expr_is_real (Val) \
? (__typeof__ (__real__ (__tgmath_real_type (Val)) 0))\
Fct##f (Val) \
: (__typeof__ (__real__ (__tgmath_real_type (Val)) 0))\
@@ -297,7 +343,7 @@
(__real__ \
(__tgmath_real_type (Val)) 0)) Cfct, \
(Val)) \
- ((sizeof (+__real__ (Val)) == sizeof (+(Val))) \
+ (__expr_is_real (Val) \
? (__typeof__ (__real__ (__tgmath_real_type (Val)) 0)) \
__tgml(Fct) (Val) \
: (__typeof__ (__real__ (__tgmath_real_type (Val)) 0)) \
@@ -312,41 +358,38 @@
+ __real__ (Val2)) == 8) \
? __TGMATH_CF128 ((Val1) + (Val2), \
(__typeof \
- ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Fct, \
(__typeof \
- ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Cfct, \
(Val1, Val2)) \
- ((sizeof (+__real__ (Val1)) == sizeof (+(Val1)) \
- && sizeof (+__real__ (Val2)) == sizeof (+(Val2))) \
- ? (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ (__expr_is_real ((Val1) + (Val2)) \
+ ? (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
__tgml(Fct) (Val1, Val2) \
- : (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ : (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
__tgml(Cfct) (Val1, Val2)) \
: (sizeof (+__real__ (Val1)) == sizeof (double) \
|| sizeof (+__real__ (Val2)) == sizeof (double) \
|| __builtin_classify_type (__real__ (Val1)) != 8 \
|| __builtin_classify_type (__real__ (Val2)) != 8) \
- ? ((sizeof (+__real__ (Val1)) == sizeof (+(Val1)) \
- && sizeof (+__real__ (Val2)) == sizeof (+(Val2))) \
- ? (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ ? (__expr_is_real ((Val1) + (Val2)) \
+ ? (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Fct (Val1, Val2) \
- : (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ : (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Cfct (Val1, Val2)) \
- : ((sizeof (+__real__ (Val1)) == sizeof (+(Val1)) \
- && sizeof (+__real__ (Val2)) == sizeof (+(Val2))) \
- ? (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ : (__expr_is_real ((Val1) + (Val2)) \
+ ? (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Fct##f (Val1, Val2) \
- : (__typeof ((__tgmath_real_type (Val1)) 0 \
- + (__tgmath_real_type (Val2)) 0)) \
+ : (__typeof ((__tgmath_complex_type (Val1)) 0 \
+ + (__tgmath_complex_type (Val2)) 0)) \
Cfct##f (Val1, Val2))))
#else
# error "Unsupported compiler; you cannot use <tgmath.h>"
--
Joseph S. Myers
joseph@codesourcery.com