diff options
Diffstat (limited to 'newlib/libm/machine/spu/headers/coshf4.h')
| -rw-r--r-- | newlib/libm/machine/spu/headers/coshf4.h | 112 |
1 files changed, 112 insertions, 0 deletions
diff --git a/newlib/libm/machine/spu/headers/coshf4.h b/newlib/libm/machine/spu/headers/coshf4.h new file mode 100644 index 000000000..ca43748ca --- /dev/null +++ b/newlib/libm/machine/spu/headers/coshf4.h @@ -0,0 +1,112 @@ +/* -------------------------------------------------------------- */ +/* (C)Copyright 2006,2007, */ +/* International Business Machines 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 _COSHF4_H_ +#define _COSHF4_H_ 1 + +#include <spu_intrinsics.h> + +#include "expf4.h" +#include "recipf4.h" + + +/* + * FUNCTION + * vector float _coshf4(vector float x) + * + * DESCRIPTION + * The _coshf4 function computes the hyperbolic cosines of a vector of + * angles (expressed in radians) to an accuracy of a single precision + * floating point. + * + */ +static __inline vector float _coshf4(vector float x) +{ + // 1.0000 (above this number, use sinh(x) = 0.5 * (e^x - e^-x) + vec_uint4 threshold = (vec_uint4)spu_splats(0x3F800000); + + vec_uint4 sign_mask = (vec_uint4)spu_splats(0x80000000); + + // Coefficents for the Taylor series + vec_float4 f02 = spu_splats(5.0000000000000000E-1f); // 1/2! + vec_float4 f04 = spu_splats(4.1666666666666667E-2f); // 1/4! + vec_float4 f06 = spu_splats(1.3888888888888889E-3f); // 1/6! + vec_float4 f08 = spu_splats(2.4801587301587302E-5f); // 1/8! + vec_float4 f10 = spu_splats(2.7557319223985891E-7f); // 1/10! + vec_float4 f12 = spu_splats(2.0876756987868099E-9f); // 1/12! + + // Perform the calculation as a Taylor series + vec_float4 result; + vector float x2 = spu_mul(x,x); + result = spu_madd(x2,f12,f10); + result = spu_madd(x2,result,f08); + result = spu_madd(x2,result,f06); + result = spu_madd(x2,result,f04); + result = spu_madd(x2,result,f02); + result = spu_madd(x2,result,spu_splats(1.0f)); + + + // Perform calculation as a function of 0.5 * (e^x - e^-x) + vec_float4 ex = _expf4(x); + vec_float4 ex_inv = _recipf4(ex); + + vec_float4 r2= spu_add(ex,ex_inv); + r2 = spu_mul(r2,f02); // we can reused f02 here + + vec_uint4 xabs = spu_andc((vec_uint4)x,sign_mask); + vec_uint4 use_exp = spu_cmpgt(xabs,threshold); + + // Select either the Taylor or exp version + result = spu_sel(result,r2,use_exp); + + return result; +} +#endif /* _COSHF4_H_ */ +#endif /* __SPU__ */ |