diff options
Diffstat (limited to 'newlib/libm/machine/spu/headers/atanf4.h')
| -rw-r--r-- | newlib/libm/machine/spu/headers/atanf4.h | 136 |
1 files changed, 136 insertions, 0 deletions
diff --git a/newlib/libm/machine/spu/headers/atanf4.h b/newlib/libm/machine/spu/headers/atanf4.h new file mode 100644 index 000000000..a62b88c28 --- /dev/null +++ b/newlib/libm/machine/spu/headers/atanf4.h @@ -0,0 +1,136 @@ +/* -------------------------------------------------------------- */ +/* (C)Copyright 2006,2007, */ +/* International Business Machines Corporation, */ +/* Sony Computer Entertainment, Incorporated, */ +/* Toshiba Corporation, */ +/* */ +/* All Rights Reserved. */ +/* */ +/* Redistribution and use in source and binary forms, with or */ +/* without modification, are permitted provided that the */ +/* following conditions are met: */ +/* */ +/* - Redistributions of source code must retain the above copyright*/ +/* notice, this list of conditions and the following disclaimer. */ +/* */ +/* - Redistributions in binary form must reproduce the above */ +/* copyright notice, this list of conditions and the following */ +/* disclaimer in the documentation and/or other materials */ +/* provided with the distribution. */ +/* */ +/* - Neither the name of IBM Corporation nor the names of its */ +/* contributors may be used to endorse or promote products */ +/* derived from this software without specific prior written */ +/* permission. */ +/* Redistributions of source code must retain the above copyright */ +/* notice, this list of conditions and the following disclaimer. */ +/* */ +/* Redistributions in binary form must reproduce the above */ +/* copyright notice, this list of conditions and the following */ +/* disclaimer in the documentation and/or other materials */ +/* provided with the distribution. */ +/* */ +/* Neither the name of IBM Corporation nor the names of its */ +/* contributors may be used to endorse or promote products */ +/* derived from this software without specific prior written */ +/* permission. */ +/* */ +/* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND */ +/* CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, */ +/* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF */ +/* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE */ +/* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR */ +/* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, */ +/* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT */ +/* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; */ +/* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) */ +/* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN */ +/* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR */ +/* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, */ +/* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ +/* -------------------------------------------------------------- */ +/* PROLOG END TAG zYx */ +#ifdef __SPU__ + +#ifndef _ATANF4_H_ +#define _ATANF4_H_ 1 + +#include <spu_intrinsics.h> + +#include "simdmath.h" +#include "recipf4.h" + +/* + * FUNCTION + * vector float _atanf4(vector float x) + * + * DESCRIPTION + * The _atanf4 function computes the arc tangent of a vector of values x; + * that is the values whose tangent is x. + * + * The _atanf4 function returns the arc tangents in radians and the value + * is mathematically defined to be in the range -PI/2 to PI/2. + * + * The arc tangent function is computed using a polynomial approximation + * (B. Carlson, M. Goldstein, Los Alamos Scientific Laboratory, 1955). + * __8__ + * \ + * \ + * atan(x) = / Ci*x^(2*i+1) + * /____ + * i=0 + * + * for x in the range -1 to 1. The remaining regions are defined to be: + * + * [1, infinity] : PI/2 + atanf(-1/x) + * [-infinity, -1] : -PI/2 + atanf(-1/x) + * + */ +static __inline vector float _atanf4(vector float x) +{ + vector float bias; + vector float x2, x3, x4, x8, x9; + vector float hi, lo; + vector float result; + vector float inv_x; + vector unsigned int sign; + vector unsigned int select; + + sign = spu_sl(spu_rlmask((vector unsigned int)x, -31), 31); + inv_x = _recipf4(x); + inv_x = (vector float)spu_xor((vector unsigned int)inv_x, spu_splats(0x80000000)); + + select = (vector unsigned int)spu_cmpabsgt(x, spu_splats(1.0f)); + bias = (vector float)spu_or(sign, (vector unsigned int)(spu_splats((float)SM_PI_2))); + bias = (vector float)spu_and((vector unsigned int)bias, select); + + x = spu_sel(x, inv_x, select); + + /* Instruction counts can be reduced if the polynomial was + * computed entirely from nested (dependent) fma's. However, + * to reduce the number of pipeline stalls, the polygon is evaluated + * in two halves(hi and lo). + */ + bias = spu_add(bias, x); + x2 = spu_mul(x, x); + x3 = spu_mul(x2, x); + x4 = spu_mul(x2, x2); + x8 = spu_mul(x4, x4); + x9 = spu_mul(x8, x); + hi = spu_madd(spu_splats(0.0028662257f), x2, spu_splats(-0.0161657367f)); + hi = spu_madd(hi, x2, spu_splats(0.0429096138f)); + hi = spu_madd(hi, x2, spu_splats(-0.0752896400f)); + hi = spu_madd(hi, x2, spu_splats(0.1065626393f)); + lo = spu_madd(spu_splats(-0.1420889944f), x2, spu_splats(0.1999355085f)); + lo = spu_madd(lo, x2, spu_splats(-0.3333314528f)); + lo = spu_madd(lo, x3, bias); + + result = spu_madd(hi, x9, lo); + + return (result); +} + +#endif /* _ATANF4_H_ */ +#endif /* __SPU__ */ + + |