Copy of aRts for Trinity modifications

git-svn-id: svn://anonsvn.kde.org/home/kde/branches/trinity/dependencies/arts@1070145 283d02a7-25f6-0310-bc7c-ecb5cbfe19da

1 files changed, 246 insertions, 0 deletions

diff --git a/flow/gsl/gslieee754.h b/flow/gsl/gslieee754.h
new file mode 100644
index 0000000..48ea0ee
--- /dev/null
+++ b/flow/gsl/gslieee754.h
@@ -0,0 +1,246 @@
+/* GSL - Generic Sound Layer
+ * Copyright (C) 1999, 2001 Tim Janik
+ *
+ * This 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 of the License, or (at your option) any later version.
+ *
+ * This 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 this library; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place, Suite 330,
+ * Boston, MA 02111-1307, USA.
+ */
+#ifndef __GSL_IEEE754_H__
+#define __GSL_IEEE754_H__
+
+#include	<gsl/gsldefs.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif /* __cplusplus */
+
+
+/* IEEE 754 single precision floating point layout:
+ *        31 30           23 22            0
+ * +--------+---------------+---------------+
+ * | s 1bit | e[30:23] 8bit | f[22:0] 23bit |
+ * +--------+---------------+---------------+
+ * B0------------------->B1------->B2-->B3-->
+ *
+ * IEEE 754 double precision floating point layout:
+ *        63 62            52 51            32   31            0
+ * +--------+----------------+----------------+ +---------------+
+ * | s 1bit | e[62:52] 11bit | f[51:32] 20bit | | f[31:0] 32bit |
+ * +--------+----------------+----------------+ +---------------+
+ * B0--------------->B1---------->B2--->B3---->  B4->B5->B6->B7->
+ */
+
+/* floating point type related constants */
+#define GSL_FLOAT_BIAS		 (127)
+#define	GSL_FLOAT_MAX_NORMAL	 (3.40282347e+38)	   /* 7f7fffff */
+#define	GSL_FLOAT_MIN_NORMAL	 (1.17549435e-38)	   /* 00800000 */
+#define	GSL_FLOAT_MAX_SUBNORMAL	 (1.17549421e-38)	   /* 007fffff */
+#define	GSL_FLOAT_MIN_SUBNORMAL	 (1.40129846e-45)	   /* 00000001 */
+#define GSL_DOUBLE_BIAS		 (1023)
+#define	GSL_DOUBLE_MAX_NORMAL	 (1.7976931348623157e+308) /* 7fefffff ffffffff */
+#define	GSL_DOUBLE_MIN_NORMAL	 (2.2250738585072014e-308) /* 00100000 00000000 */
+#define	GSL_DOUBLE_MAX_SUBNORMAL (2.2250738585072009e-308) /* 000fffff ffffffff */
+#define	GSL_DOUBLE_MIN_SUBNORMAL (4.9406564584124654e-324) /* 00000000 00000001 */
+#define	GSL_DOUBLE_INF		 (_gsl_dinf_union.d)
+#define	GSL_DOUBLE_NAN		 (_gsl_dnan_union.d)
+#define	GSL_FLOAT_INF		 (_gsl_finf_union.f)
+#define	GSL_FLOAT_NAN		 (_gsl_fnan_union.f)
+
+/* multiply with base2 exponent to get base10 exponent (for nomal numbers) */
+#define GSL_LOG_2_BASE_10         (0.30102999566398119521)
+
+/* the following macros work only on variables
+ * and evaluate arguments multiple times
+ */
+
+/* single precision value checks */
+#define	GSL_FLOAT_IS_ZERO(f)		((f) == 0.0)	/* compiler knows this one */
+#define	GSL_FLOAT_IS_NORMAL(f)		(GSL_FLOAT_PARTS (f).mpn.biased_exponent > 0 && \
+				         GSL_FLOAT_PARTS (f).mpn.biased_exponent < 255)
+#define	GSL_FLOAT_IS_SUBNORMAL(f)	(GSL_FLOAT_PARTS (f).mpn.biased_exponent == 0 && \
+					 GSL_FLOAT_PARTS (f).mpn.mantissa != 0)
+#define	GSL_FLOAT_IS_NANINF(f)		(GSL_FLOAT_PARTS (f).mpn.biased_exponent == 255)
+#define	GSL_FLOAT_IS_NAN(f)		(GSL_FLOAT_IS_NANINF (f) && GSL_FLOAT_PARTS (f).mpn.mantissa != 0)
+#define	GSL_FLOAT_IS_INF(f)		(GSL_FLOAT_IS_NANINF (f) && GSL_FLOAT_PARTS (f).mpn.mantissa == 0)
+#define	GSL_FLOAT_IS_INF_POSITIVE(f)	(GSL_FLOAT_IS_INF (f) && GSL_FLOAT_PARTS (f).mpn.sign == 0)
+#define	GSL_FLOAT_IS_INF_NEGATIVE(f)	(GSL_FLOAT_IS_INF (f) && GSL_FLOAT_PARTS (f).mpn.sign == 1)
+#define	GSL_FLOAT_SIGN(f)		(GSL_FLOAT_PARTS (f).mpn.sign)
+
+/* double precision value checks */
+#define	GSL_DOUBLE_IS_ZERO(d)		((d) == 0.0)	/* compiler knows this one */
+#define	GSL_DOUBLE_IS_NORMAL(d)		(GSL_DOUBLE_PARTS (d).mpn.biased_exponent > 0 && \
+					 GSL_DOUBLE_PARTS (d).mpn.biased_exponent < 2047)
+#define	GSL_DOUBLE_IS_SUBNORMAL(d)	(GSL_DOUBLE_PARTS (d).mpn.biased_exponent == 0 && \
+                                         (GSL_DOUBLE_PARTS (d).mpn.mantissa_low != 0 || \
+					  GSL_DOUBLE_PARTS (d).mpn.mantissa_high != 0))
+#define	GSL_DOUBLE_IS_NANINF(d)		(GSL_DOUBLE_PARTS (d).mpn.biased_exponent == 2047)
+#define	GSL_DOUBLE_IS_NAN(d)		(GSL_DOUBLE_IS_NANINF (d) && \
+					 (GSL_DOUBLE_PARTS (d).mpn.mantissa_low != 0 || \
+					  GSL_DOUBLE_PARTS (d).mpn.mantissa_high != 0))
+#define	GSL_DOUBLE_IS_INF(d)		(GSL_DOUBLE_IS_NANINF (d) && \
+					 GSL_DOUBLE_PARTS (d).mpn.mantissa_low == 0 && \
+					 GSL_DOUBLE_PARTS (d).mpn.mantissa_high == 0)
+#define	GSL_DOUBLE_IS_INF_POSITIVE(d)	(GSL_DOUBLE_IS_INF (d) && GSL_DOUBLE_PARTS (d).mpn.sign == 0)
+#define	GSL_DOUBLE_IS_INF_NEGATIVE(d)	(GSL_DOUBLE_IS_INF (d) && GSL_DOUBLE_PARTS (d).mpn.sign == 1)
+#define	GSL_DOUBLE_SIGN(d)		(GSL_DOUBLE_PARTS (d).mpn.sign)
+
+/* get structured parts of floating point numbers */
+#define	GSL_FLOAT_PARTS(f)		(*((GslFloatIEEE754*) &(f)))
+#define	GSL_DOUBLE_PARTS(d)		(*((GslDoubleIEEE754*) &(d)))
+
+/* --- rounding --- */
+typedef	unsigned short int	GslFpuState;
+#if defined (__i386__) && defined (__GNUC__)
+/* setting/restoring rounding mode shouldn't actually
+ * be necessary as round-to-nearest is the hardware
+ * default (can be checked with gsl_fpu_okround()).
+ */
+static inline void	gsl_fpu_setround	(GslFpuState		*cw);
+static inline int	gsl_fpu_okround		(void);
+static inline void	gsl_fpu_restore		(GslFpuState		 cv);
+static inline int	gsl_ftoi /* nearest */	(register float		 f)  G_GNUC_CONST;
+static inline int	gsl_dtoi /* nearest */	(register double	 f)  G_GNUC_CONST;
+/* fallbacks for the !386 case are below */
+#endif
+
+
+/* --- implementation bits --- */
+#if G_BYTE_ORDER == G_LITTLE_ENDIAN
+typedef union
+{
+  float		v_float;
+  struct {
+    unsigned int mantissa : 23;
+    unsigned int biased_exponent : 8;
+    unsigned int sign : 1;
+  } mpn;
+} GslFloatIEEE754;
+typedef union
+{
+  double	v_double;
+  struct {
+    unsigned int mantissa_low : 32;
+    unsigned int mantissa_high : 20;
+    unsigned int biased_exponent : 11;
+    unsigned int sign : 1;
+  } mpn;
+} GslDoubleIEEE754;
+#define	_GSL_DOUBLE_INF_BYTES	{ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xf0, 0x7f }
+#define	_GSL_DOUBLE_NAN_BYTES	{ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xf8, 0x7f }
+#define	_GSL_FLOAT_INF_BYTES	{ 0x00, 0x00, 0x80, 0x7f }
+#define	_GSL_FLOAT_NAN_BYTES	{ 0x00, 0x00, 0xc0, 0x7f }
+#elif G_BYTE_ORDER == G_BIG_ENDIAN
+typedef union
+{
+  float         v_float;
+  struct {
+    unsigned int sign : 1;
+    unsigned int biased_exponent : 8;
+    unsigned int mantissa : 23;
+  } mpn;
+} GslFloatIEEE754;
+typedef union
+{
+  double        v_double;
+  struct {
+    unsigned int sign : 1;
+    unsigned int biased_exponent : 11;
+    unsigned int mantissa_high : 20;
+    unsigned int mantissa_low : 32;
+  } mpn;
+} GslDoubleIEEE754;
+#define	_GSL_DOUBLE_INF_BYTES	{ 0x7f, 0xf0, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }
+#define	_GSL_DOUBLE_NAN_BYTES	{ 0x7f, 0xf8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }
+#define	_GSL_FLOAT_INF_BYTES	{ 0x7f, 0x80, 0x00, 0x00 }
+#define	_GSL_FLOAT_NAN_BYTES	{ 0x7f, 0xc0, 0x00, 0x00 }
+#else /* !G_LITTLE_ENDIAN && !G_BIG_ENDIAN */
+#error unknown ENDIAN type
+#endif /* !G_LITTLE_ENDIAN && !G_BIG_ENDIAN */
+
+static const union { unsigned char c[8]; double d; } _gsl_dnan_union = { _GSL_DOUBLE_NAN_BYTES };
+static const union { unsigned char c[8]; double d; } _gsl_dinf_union = { _GSL_DOUBLE_INF_BYTES };
+static const union { unsigned char c[4]; float f; }  _gsl_fnan_union = { _GSL_FLOAT_NAN_BYTES };
+static const union { unsigned char c[4]; float f; }  _gsl_finf_union = { _GSL_FLOAT_INF_BYTES };
+
+#if defined (__i386__) && defined (__GNUC__)
+static inline void
+gsl_fpu_setround (GslFpuState *cw)
+{
+  GslFpuState cv;
+  
+  __asm__ ("fnstcw %0"
+	   : "=m" (*&cv));
+  *cw = cv;
+  cv &= ~0x0c00;
+  __asm__ ("fldcw %0"
+	   :
+	   : "m" (*&cv));
+}
+static inline int
+gsl_fpu_okround (void)
+{
+  GslFpuState cv;
+  
+  __asm__ ("fnstcw %0"
+	   : "=m" (*&cv));
+  return !(cv & 0x0c00);
+}
+static inline void
+gsl_fpu_restore (GslFpuState cv)
+{
+  __asm__ ("fldcw %0"
+	   :
+	   : "m" (*&cv));
+}
+static inline int G_GNUC_CONST
+gsl_ftoi (register float f)
+{
+  int r;
+  
+  __asm__ ("fistl %0"
+	   : "=m" (r)
+	   : "t" (f));
+  return r;
+}
+static inline int G_GNUC_CONST
+gsl_dtoi (register double f)
+{
+  int r;
+  
+  __asm__ ("fistl %0"
+	   : "=m" (r)
+	   : "t" (f));
+  return r;
+}
+#else   /* !386 */
+#  define gsl_fpu_setround(p)   ((void) (p));
+#  define gsl_fpu_okround()     (1)
+#  define gsl_fpu_restore(x)    /* nop */
+static inline int G_GNUC_CONST
+gsl_ftoi (register float v)
+{
+  return v < -0.0 ? v - 0.5 : v + 0.5;
+}
+static inline int G_GNUC_CONST
+gsl_dtoi (register double v)
+{
+  return v < -0.0 ? v - 0.5 : v + 0.5;
+}
+#endif
+
+#ifdef __cplusplus
+}
+#endif /* __cplusplus */
+
+#endif /* __GSL_IEEE754_H__ */		/* vim: set ts=8 sw=2 sts=2: */