Mercurial > hg > sv-dependency-builds
comparison src/fftw-3.3.3/rdft/simd/common/hc2cfdftv_2.c @ 10:37bf6b4a2645
Add FFTW3
author | Chris Cannam |
---|---|
date | Wed, 20 Mar 2013 15:35:50 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
9:c0fb53affa76 | 10:37bf6b4a2645 |
---|---|
1 /* | |
2 * Copyright (c) 2003, 2007-11 Matteo Frigo | |
3 * Copyright (c) 2003, 2007-11 Massachusetts Institute of Technology | |
4 * | |
5 * This program is free software; you can redistribute it and/or modify | |
6 * it under the terms of the GNU General Public License as published by | |
7 * the Free Software Foundation; either version 2 of the License, or | |
8 * (at your option) any later version. | |
9 * | |
10 * This program is distributed in the hope that it will be useful, | |
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
13 * GNU General Public License for more details. | |
14 * | |
15 * You should have received a copy of the GNU General Public License | |
16 * along with this program; if not, write to the Free Software | |
17 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | |
18 * | |
19 */ | |
20 | |
21 /* This file was automatically generated --- DO NOT EDIT */ | |
22 /* Generated on Sun Nov 25 07:42:29 EST 2012 */ | |
23 | |
24 #include "codelet-rdft.h" | |
25 | |
26 #ifdef HAVE_FMA | |
27 | |
28 /* Generated by: ../../../genfft/gen_hc2cdft_c.native -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -trivial-stores -variables 32 -no-generate-bytw -n 2 -dit -name hc2cfdftv_2 -include hc2cfv.h */ | |
29 | |
30 /* | |
31 * This function contains 5 FP additions, 6 FP multiplications, | |
32 * (or, 3 additions, 4 multiplications, 2 fused multiply/add), | |
33 * 9 stack variables, 1 constants, and 4 memory accesses | |
34 */ | |
35 #include "hc2cfv.h" | |
36 | |
37 static void hc2cfdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms) | |
38 { | |
39 DVK(KP500000000, +0.500000000000000000000000000000000000000000000); | |
40 { | |
41 INT m; | |
42 for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) { | |
43 V T1, T2, T4, T3, T5, T7, T6; | |
44 T1 = LD(&(Rp[0]), ms, &(Rp[0])); | |
45 T2 = LD(&(Rm[0]), -ms, &(Rm[0])); | |
46 T4 = LDW(&(W[0])); | |
47 T3 = VFMACONJ(T2, T1); | |
48 T5 = VZMULIJ(T4, VFNMSCONJ(T2, T1)); | |
49 T7 = VCONJ(VMUL(LDK(KP500000000), VADD(T3, T5))); | |
50 T6 = VMUL(LDK(KP500000000), VSUB(T3, T5)); | |
51 ST(&(Rm[0]), T7, -ms, &(Rm[0])); | |
52 ST(&(Rp[0]), T6, ms, &(Rp[0])); | |
53 } | |
54 } | |
55 VLEAVE(); | |
56 } | |
57 | |
58 static const tw_instr twinstr[] = { | |
59 VTW(1, 1), | |
60 {TW_NEXT, VL, 0} | |
61 }; | |
62 | |
63 static const hc2c_desc desc = { 2, XSIMD_STRING("hc2cfdftv_2"), twinstr, &GENUS, {3, 4, 2, 0} }; | |
64 | |
65 void XSIMD(codelet_hc2cfdftv_2) (planner *p) { | |
66 X(khc2c_register) (p, hc2cfdftv_2, &desc, HC2C_VIA_DFT); | |
67 } | |
68 #else /* HAVE_FMA */ | |
69 | |
70 /* Generated by: ../../../genfft/gen_hc2cdft_c.native -simd -compact -variables 4 -pipeline-latency 8 -trivial-stores -variables 32 -no-generate-bytw -n 2 -dit -name hc2cfdftv_2 -include hc2cfv.h */ | |
71 | |
72 /* | |
73 * This function contains 5 FP additions, 4 FP multiplications, | |
74 * (or, 5 additions, 4 multiplications, 0 fused multiply/add), | |
75 * 10 stack variables, 1 constants, and 4 memory accesses | |
76 */ | |
77 #include "hc2cfv.h" | |
78 | |
79 static void hc2cfdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms) | |
80 { | |
81 DVK(KP500000000, +0.500000000000000000000000000000000000000000000); | |
82 { | |
83 INT m; | |
84 for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) { | |
85 V T4, T6, T1, T3, T2, T5, T7, T8; | |
86 T1 = LD(&(Rp[0]), ms, &(Rp[0])); | |
87 T2 = LD(&(Rm[0]), -ms, &(Rm[0])); | |
88 T3 = VCONJ(T2); | |
89 T4 = VADD(T1, T3); | |
90 T5 = LDW(&(W[0])); | |
91 T6 = VZMULIJ(T5, VSUB(T3, T1)); | |
92 T7 = VCONJ(VMUL(LDK(KP500000000), VSUB(T4, T6))); | |
93 ST(&(Rm[0]), T7, -ms, &(Rm[0])); | |
94 T8 = VMUL(LDK(KP500000000), VADD(T4, T6)); | |
95 ST(&(Rp[0]), T8, ms, &(Rp[0])); | |
96 } | |
97 } | |
98 VLEAVE(); | |
99 } | |
100 | |
101 static const tw_instr twinstr[] = { | |
102 VTW(1, 1), | |
103 {TW_NEXT, VL, 0} | |
104 }; | |
105 | |
106 static const hc2c_desc desc = { 2, XSIMD_STRING("hc2cfdftv_2"), twinstr, &GENUS, {5, 4, 0, 0} }; | |
107 | |
108 void XSIMD(codelet_hc2cfdftv_2) (planner *p) { | |
109 X(khc2c_register) (p, hc2cfdftv_2, &desc, HC2C_VIA_DFT); | |
110 } | |
111 #endif /* HAVE_FMA */ |