21
21
/* This file was automatically generated --- DO NOT EDIT */
22
/* Generated on Sun Jul 12 06:45:41 EDT 2009 */
22
/* Generated on Wed Jul 27 06:18:29 EDT 2011 */
24
24
#include "codelet-rdft.h"
28
/* Generated by: ../../../genfft/gen_r2cb -fma -reorder-insns -schedule-for-pipeline -compact -variables 4 -pipeline-latency 4 -sign 1 -n 5 -name r2cb_5 -include r2cb.h */
28
/* Generated by: ../../../genfft/gen_r2cb.native -fma -reorder-insns -schedule-for-pipeline -compact -variables 4 -pipeline-latency 4 -sign 1 -n 5 -name r2cb_5 -include r2cb.h */
31
31
* This function contains 12 FP additions, 10 FP multiplications,
41
41
DK(KP500000000, +0.500000000000000000000000000000000000000000000);
42
42
DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
43
43
DK(KP618033988, +0.618033988749894848204586834365638117720309180);
45
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
46
E T1, T2, T3, Tc, Ta, T8, T9;
52
Tc = FMS(KP618033988, T8, T9);
53
Ta = FMA(KP618033988, T9, T8);
58
R0[0] = FMA(KP2_000000000, T4, T1);
59
T5 = FNMS(KP500000000, T4, T1);
60
T7 = FMA(KP1_118033988, T6, T5);
61
Tb = FNMS(KP1_118033988, T6, T5);
62
R0[WS(rs, 2)] = FMA(KP1_902113032, Ta, T7);
63
R1[0] = FNMS(KP1_902113032, Ta, T7);
64
R1[WS(rs, 1)] = FMA(KP1_902113032, Tc, Tb);
65
R0[WS(rs, 1)] = FNMS(KP1_902113032, Tc, Tb);
46
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
47
E T1, T2, T3, Tc, Ta, T8, T9;
53
Tc = FMS(KP618033988, T8, T9);
54
Ta = FMA(KP618033988, T9, T8);
59
R0[0] = FMA(KP2_000000000, T4, T1);
60
T5 = FNMS(KP500000000, T4, T1);
61
T7 = FMA(KP1_118033988, T6, T5);
62
Tb = FNMS(KP1_118033988, T6, T5);
63
R0[WS(rs, 2)] = FMA(KP1_902113032, Ta, T7);
64
R1[0] = FNMS(KP1_902113032, Ta, T7);
65
R1[WS(rs, 1)] = FMA(KP1_902113032, Tc, Tb);
66
R0[WS(rs, 1)] = FNMS(KP1_902113032, Tc, Tb);
91
93
DK(KP500000000, +0.500000000000000000000000000000000000000000000);
92
94
DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
93
95
DK(KP1_175570504, +1.175570504584946258337411909278145537195304875);
95
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
96
E Ta, Tc, T1, T4, T5, T6, Tb, T7;
101
Ta = FNMS(KP1_902113032, T9, KP1_175570504 * T8);
102
Tc = FMA(KP1_902113032, T8, KP1_175570504 * T9);
107
T5 = FNMS(KP500000000, T4, T1);
108
T6 = KP1_118033988 * (T2 - T3);
98
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
99
E Ta, Tc, T1, T4, T5, T6, Tb, T7;
104
Ta = FNMS(KP1_902113032, T9, KP1_175570504 * T8);
105
Tc = FMA(KP1_902113032, T8, KP1_175570504 * T9);
110
T5 = FNMS(KP500000000, T4, T1);
111
T6 = KP1_118033988 * (T2 - T3);
113
R0[0] = FMA(KP2_000000000, T4, T1);
116
R0[WS(rs, 2)] = Tb + Tc;
118
R0[WS(rs, 1)] = T7 - Ta;
119
R1[WS(rs, 1)] = T7 + Ta;
110
R0[0] = FMA(KP2_000000000, T4, T1);
113
R0[WS(rs, 2)] = Tb + Tc;
115
R0[WS(rs, 1)] = T7 - Ta;
116
R1[WS(rs, 1)] = T7 + Ta;