Все переделал в соответствии с "AVR IAR Assembleer"
сделал 2 файла отдельно для макросов и отдельно asm
файл ffft.asm
CODE
MODULE FFFT_ASM
#include "ffftm.s90"
#define FFT_B 8
#define T0L r0
#define T0H r1
#define T2L r2
#define T2H r3
#define T4L r4
#define T4H r5
#define T6L r6
#define T6H r7
#define T8L r8
#define T8H r9
#define T10L r10
#define T10H r11
#define T12L r12
#define T12H r13
#define T14L r14
#define T14H r15
#define AL r16
#define AH r17
#define BL r18
#define BH r19
#define CL r20
#define CH r21
#define DL r22
#define DH r23
#define EL r24
#define EH r25
#define XL r26
#define XH r27
#define YL r28
#define YH r29
#define ZL r30
#define ZH r31
PUBLIC tbl_window
PUBLIC fft_input
PUBLIC fft_execute
PUBLIC fft_output
PUBLIC fmuls_f
EXTERN FFT_N
RSEG CONST
tbl_window ; tbl_window[] = ... (This is a Hamming window)
DW 2621, 2625, 2639, 2662, 2693, 2734, 2784, 2843, 2910, 2987, 3073, 3167, 3270, 3382, 3502, 3631
DW 3768, 3914, 4068, 4230, 4401, 4579, 4765, 4959, 5161, 5370, 5587, 5811, 6042, 6280, 6525, 6777
DW 7036, 7300, 7571, 7849, 8132, 8420, 8715, 9015, 9320, 9630, 9945, 10264, 10588, 10917, 11249, 11586
DW 11926, 12269, 12616, 12966, 13318, 13674, 14031, 14391, 14753, 15117, 15482, 15849, 16216, 16585, 16954, 17324
DW 17694, 18064, 18433, 18802, 19171, 19539, 19905, 20271, 20634, 20996, 21356, 21714, 22069, 22422, 22772, 23118
DW 23462, 23802, 24138, 24471, 24799, 25123, 25443, 25758, 26068, 26373, 26673, 26967, 27256, 27539, 27816, 28087
DW 28352, 28610, 28862, 29107, 29345, 29576, 29800, 30017, 30226, 30428, 30622, 30808, 30987, 31157, 31319, 31473
DW 31619, 31756, 31885, 32006, 32117, 32220, 32315, 32400, 32477, 32545, 32603, 32653, 32694, 32726, 32748, 32762
DW 32766, 32762, 32748, 32726, 32694, 32653, 32603, 32545, 32477, 32400, 32315, 32221, 32117, 32006, 31885, 31757
DW 31619, 31473, 31319, 31157, 30987, 30808, 30622, 30428, 30226, 30017, 29800, 29576, 29345, 29107, 28862, 28610
DW 28352, 28087, 27816, 27539, 27256, 26967, 26673, 26373, 26068, 25758, 25443, 25123, 24799, 24471, 24138, 23802
DW 23462, 23118, 22772, 22422, 22069, 21714, 21356, 20996, 20634, 20271, 19905, 19539, 19171, 18803, 18433, 18064
DW 17694, 17324, 16954, 16585, 16216, 15849, 15482, 15117, 14753, 14391, 14031, 13674, 13318, 12966, 12616, 12269
DW 11926, 11586, 11249, 10917, 10588, 10264, 9945, 9630, 9320, 9015, 8715, 8420, 8132, 7849, 7571, 7300
DW 7036, 6777, 6526, 6280, 6042, 5811, 5587, 5370, 5161, 4959, 4765, 4579, 4401, 4230, 4068, 3914
DW 3768, 3631, 3502, 3382, 3270, 3167, 3073, 2987, 2910, 2843, 2784, 2734, 2693, 2662, 2639, 2625
tbl_cos_sin ; Table of {cos(x),sin(x)}, (0 <= x < pi, in FFT_N/2 steps)
DW 32767, 0, 32757, 804, 32727, 1607, 32678, 2410, 32609, 3211, 32520, 4011, 32412, 4807, 32284, 5601
DW 32137, 6392, 31970, 7179, 31785, 7961, 31580, 8739, 31356, 9511, 31113, 10278, 30851, 11038, 30571, 11792
DW 30272, 12539, 29955, 13278, 29621, 14009, 29268, 14732, 28897, 15446, 28510, 16150, 28105, 16845, 27683, 17530
DW 27244, 18204, 26789, 18867, 26318, 19519, 25831, 20159, 25329, 20787, 24811, 21402, 24278, 22004, 23731, 22594
DW 23169, 23169, 22594, 23731, 22004, 24278, 21402, 24811, 20787, 25329, 20159, 25831, 19519, 26318, 18867, 26789
DW 18204, 27244, 17530, 27683, 16845, 28105, 16150, 28510, 15446, 28897, 14732, 29268, 14009, 29621, 13278, 29955
DW 12539, 30272, 11792, 30571, 11038, 30851, 10278, 31113, 9511, 31356, 8739, 31580, 7961, 31785, 7179, 31970
DW 6392, 32137, 5601, 32284, 4807, 32412, 4011, 32520, 3211, 32609, 2410, 32678, 1607, 32727, 804, 32757
DW 0, 32766, -804, 32757, -1607, 32727, -2410, 32678, -3211, 32609, -4010, 32520, -4807, 32412, -5601, 32284
DW -6392, 32137, -7179, 31970, -7961, 31785, -8739, 31580, -9511, 31356, -10278, 31113, -11038, 30851, -11792, 30571
DW -12539, 30272, -13278, 29955, -14009, 29621, -14732, 29268, -15446, 28897, -16150, 28510, -16845, 28105, -17530, 27683
DW -18204, 27244, -18867, 26789, -19519, 26318, -20159, 25831, -20787, 25329, -21402, 24811, -22004, 24278, -22594, 23731
DW -23169, 23169, -23731, 22594, -24278, 22005, -24811, 21402, -25329, 20787, -25831, 20159, -26318, 19519, -26789, 18867
DW -27244, 18204, -27683, 17530, -28105, 16845, -28510, 16150, -28897, 15446, -29268, 14732, -29620, 14009, -29955, 13278
DW -30272, 12539, -30571, 11792, -30851, 11038, -31113, 10278, -31356, 9511, -31580, 8739, -31784, 7961, -31970, 7179
DW -32137, 6392, -32284, 5601, -32412, 4807, -32520, 4011, -32609, 3211, -32678, 2410, -32727, 1607, -32757, 804
tbl_bitrev ; tbl_bitrev[] = ...
DW 0*4, 128*4, 64*4, 192*4, 32*4, 160*4, 96*4, 224*4, 16*4, 144*4, 80*4, 208*4, 48*4, 176*4, 112*4, 240*4
DW 8*4, 136*4, 72*4, 200*4, 40*4, 168*4, 104*4, 232*4, 24*4, 152*4, 88*4, 216*4, 56*4, 184*4, 120*4, 248*4
DW 4*4, 132*4, 68*4, 196*4, 36*4, 164*4, 100*4, 228*4, 20*4, 148*4, 84*4, 212*4, 52*4, 180*4, 116*4, 244*4
DW 12*4, 140*4, 76*4, 204*4, 44*4, 172*4, 108*4, 236*4, 28*4, 156*4, 92*4, 220*4, 60*4, 188*4, 124*4, 252*4
DW 2*4, 130*4, 66*4, 194*4, 34*4, 162*4, 98*4, 226*4, 18*4, 146*4, 82*4, 210*4, 50*4, 178*4, 114*4, 242*4
DW 10*4, 138*4, 74*4, 202*4, 42*4, 170*4, 106*4, 234*4, 26*4, 154*4, 90*4, 218*4, 58*4, 186*4, 122*4, 250*4
DW 6*4, 134*4, 70*4, 198*4, 38*4, 166*4, 102*4, 230*4, 22*4, 150*4, 86*4, 214*4, 54*4, 182*4, 118*4, 246*4
DW 14*4, 142*4, 78*4, 206*4, 46*4, 174*4, 110*4, 238*4, 30*4, 158*4, 94*4, 222*4, 62*4, 190*4, 126*4, 254*4
ret
RSEG CODE
fft_input
pushw T2H,T2L
pushw AH,AL
pushw YH,YL
movw XL, EL ;X = array_src;
movw YL, DL ;Y = array_bfly;
clr EH ;Zero
ldiw ZH,ZL, tbl_window ;Z = &tbl_window[0];
ldiw AH,AL, FFT_N ;A = FFT_N;
1: lpmw BH,BL, Z+ ;B = *Z++; (window)
ldw CH,CL, X+ ;C = *X++; (I-axis)
FMULS16 DH,DL,T2H,T2L, BH,BL, CH,CL ;D = B * C;
stw Y+, DH,DL ;*Y++ = D;
stw Y+, DH,DL ;*Y++ = D;
subiw AH,AL, 1 ;while(--A)
brne 1b ;/
popw YH,YL
popw AH,AL
popw T2H,T2L
clr r1
ret
fft_execute
pushw T2H,T2L
pushw T4H,T4L
pushw T6H,T6L
pushw T8H,T8L
pushw T10H,T10L
pushw T12H,T12L
pushw T14H,T14L
pushw AH,AL
pushw YH,YL
movw ZL, EL ;Z = array_bfly;
ldiw EH,EL, 1 ;E = 1;
ldiw XH,XL, FFT_N/2 ;X = FFT_N/2;
1: ldi AL, 4 ;T12 = E; (angular speed)
mul EL, AL ;
movw T12L, T0L ;
mul EH, AL ;
add T12H, T0L ;/
movw T14L, EL ;T14 = E;
pushw EH,EL
movw YL, ZL ;Z = &array_bfly[0];
mul XL, AL ;Y = &array_bfly[X];
addw YH,YL, T0H,T0L ;
mul XH, AL ;
add YH, T0L ;/
pushw ZH,ZL
2: clrw T10H,T10L ;T10 = 0 (angle)
clr EH ;Zero reg.
3: lddw AH,AL, Z+0 ;A = *Z - *Y; *Z++ += *Y;
asrw AH,AL ;
lddw DH,DL, Y+0 ;
asrw DH,DL ;
movw CL, AL ;
subw AH,AL, DH,DL ;
addw CH,CL, DH,DL ;
stw Z+, CH,CL ;/
lddw BH,BL, Z+0 ;B = *Z - *Y; *Z++ += *Y;
asrw BH,BL ;
lddw DH,DL, Y+2 ;
asrw DH,DL ;
movw CL, BL ;
subw BH,BL, DH,DL ;
addw CH,CL, DH,DL ;
stw Z+, CH,CL ;/
movw T0L, ZL
ldiw ZH,ZL, tbl_cos_sin ;C = cos(T10); D = sin(T10);
addw ZH,ZL, T10H,T10L ;
lpmw CH,CL, Z+ ;
lpmw DH,DL, Z+ ;/
movw ZL, T0L
FMULS16 T4H,T4L,T2H,T2L, AH,AL, CH,CL ;*Y++ = A * C + B * D;
FMULS16 T8H,T8L,T6H,T6L, BH,BL, DH,DL ;
addd T4H,T4L,T2H,T2L, T8H,T8L,T6H,T6L;
stw Y+, T4H,T4L ;/
FMULS16 T4H,T4L,T2H,T2L, BH,BL, CH,CL ;*Y++ = B * C - A * D;
FMULS16 T8H,T8L,T6H,T6L, AH,AL, DH,DL ;
subd T4H,T4L,T2H,T2L, T8H,T8L,T6H,T6L;
stw Y+, T4H,T4L ;/
addw T10H,T10L, T12H,T12L ;T10 += T12; (next angle)
sbrs T10H, FFT_B - 7 ;while(T10 < pi)
rjmp 3b ;/
ldi AL, 4 ;Y += X; Z += X; (skip split segment)
mul XL, AL
addw YH,YL, T0H,T0L ;
addw ZH,ZL, T0H,T0L ;
mul XH, AL ;
add YH, T0L ;
add ZH, T0L ;/
ldi EL, 1 ;while(--T14)
subw T14H,T14L, EH,EL ;
rjne 2b ;/
popw ZH,ZL
popw EH,EL
lslw EH,EL ;E *= 2;
lsrw XH,XL ;while(X /= 2)
adiw XL, 0 ;
rjne 1b ;/
popw YH,YL
popw AH,AL
popw T14H,T14L
popw T12H,T12L
popw T10H,T10L
popw T8H,T8L
popw T6H,T6L
popw T4H,T4L
popw T2H,T2L
ret
fft_output
pushw T2H,T2L
pushw T4H,T4L
pushw T6H,T6L
pushw T8H,T8L
pushw T10H,T10L
pushw AH,AL
pushw YH,YL
movw T10L, EL ;T10 = array_bfly;
movw YL, DL ;Y = array_output;
ldiw ZH,ZL, tbl_bitrev ;Z = tbl_bitrev;
clr EH ;Zero
ldiw AH,AL, FFT_N / 2 ;A = FFT_N / 2; (plus only)
1: lpmw XH,XL, Z+ ;X = *Z++;
addw XH,XL, T10H,T10L ;X += array_bfly;
ldw BH,BL, X+ ;B = *X++;
ldw CH,CL, X+ ;C = *X++;
FMULS16 T4H,T4L,T2H,T2L, BH,BL, BH,BL ;T4:T2 = B * B;
FMULS16 T8H,T8L,T6H,T6L, CH,CL, CH,CL ;T8:T6 = C * C;
addd T4H,T4L,T2H,T2L, T8H,T8L,T6H,T6L;T4:T2 += T8:T6;
SQRT32 ;B = sqrt(T4:T2);
stw Y+, BH,BL ;*Y++ = B;
subiw AH,AL, 1 ;while(--A)
rjne 1b ;/
popw YH,YL
popw AH,AL
popw T10H,T10L
popw T8H,T8L
popw T6H,T6L
popw T4H,T4L
popw T2H,T2L
clr r1
ret
fmuls_f
movw CL, EL ;C = E;
clr EH ;Zero
FMULS16 ZH,ZL,XH,XL, CH,CL, DH,DL ;Z:X = C * D;
movw EL, ZL
clr r1
ret
END
Далее файл макросов ffftm.s90
CODE
ldiw macro dh,dl, abs
ldi \dl, lo8(\abs)
ldi \dh, hi8(\abs)
endm
subiw macro dh,dl, abs
subi \dl, lo8(\abs)
sbci \dh, hi8(\abs)
endm
addw macro dh,dl, sh,sl
add \dl, \sl
adc \dh, \sh
endm
addd macro d3,d2,d1,d0, s3,s2,s1,s0
add \d0, \s0
adc \d1, \s1
adc \d2, \s2
adc \d3, \s3
endm
subw macro dh,dl, sh,sl
sub \dl, \sl
sbc \dh, \sh
endm
subd macro d3,d2,d1,d0, s3,s2,s1,s0
sub \d0, \s0
sbc \d1, \s1
sbc \d2, \s2
sbc \d3, \s3
endm
lddw macro dh,dl, src
ldd \dl, \src
ldd \dh, \src+1
endm
ldw macro dh,dl, src
ld \dl, \src
ld \dh, \src
endm
stw macro dst, sh,sl
st \dst, \sl
st \dst, \sh
endm
clrw macro dh, dl
clr \dh
clr \dl
endm
lsrw macro dh, dl
lsr \dh
ror \dl
endm
asrw macro dh, dl
asr \dh
ror \dl
endm
lslw macro dh, dl
lsl \dl
rol \dh
endm
pushw macro dh, dl
push \dh
push \dl
endm
popw macro dh, dl
pop \dl
pop \dh
endm
lpmw macro dh,dl, src
lpm \dl, \src
lpm \dh, \src
endm
rjne macro lbl
breq 99f
rjmp \lbl
99:
endm
FMULS16 macro d3,d2,d1,d0 ,s1h,s1l, s2h,s2l ; Fractional Multiply (19clk)
fmuls \s1h, \s2h
movw \d2, T0L
fmul \s1l, \s2l
movw \d0, T0L
adc \d2, EH ;EH: zero reg.
fmulsu \s1h, \s2l
sbc \d3, EH
add \d1, T0L
adc \d2, T0H
adc \d3, EH
fmulsu \s2h, \s1l
sbc \d3, EH
add \d1, T0L
adc \d2, T0H
adc \d3, EH
endm
SQRT32 macro ; 32bit square root (526..542clk)
clr T6L
clr T6H
clr T8L
clr T8H
ldi BL, 1
ldi BH, 0
clr CL
clr CH
ldi DH, 16
90: lsl T2L
rol T2H
rol T4L
rol T4H
rol T6L
rol T6H
rol T8L
rol T8H
lsl T2L
rol T2H
rol T4L
rol T4H
rol T6L
rol T6H
rol T8L
rol T8H
brpl 91f
add T6L, BL
adc T6H, BH
adc T8L, CL
adc T8H, CH
rjmp 92f
91: sub T6L, BL
sbc T6H, BH
sbc T8L, CL
sbc T8H, CH
92: lsl BL
rol BH
rol CL
andi BL, 0xF8
ori BL, 0x05
sbrc T8H, 7
subi BL, 2
dec DH
brne 90b
lsr CL
ror BH
ror BL
lsr CL
ror BH
ror BL
endm
В ffft.h указал вот так
CODE
#define int16_t signed int
#define uint16_t unsigned int
#define prog_int16_t unsigned int
typedef struct _tag_complex_t {
int16_t r;
int16_t i;
} complex_t;
extern void fft_input (const int16_t *, complex_t *);
extern void fft_execute (complex_t *);
extern void fft_output (const complex_t *, uint16_t *);
extern int16_t fmuls_f (int16_t, int16_t);
extern const prog_int16_t tbl_window[];
однако при компиляции выдается ошибка Error[4]: Unexpected end of file encountered F:\WORK\PROCESSOR_PROGRAMM\TESTFFT_LCD\ffft.asm 293
Где закралась ошибка никак понять не могу ?