-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathfft_init_xblock.m
More file actions
297 lines (265 loc) · 14.6 KB
/
fft_init_xblock.m
File metadata and controls
297 lines (265 loc) · 14.6 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
function fft_init_xblock(varargin)
% Set default vararg values.
defaults = { ...
'FFTSize', 5, ...
'n_inputs', 2, ...
'input_bit_width', 18, ...
'coeff_bit_width', 18, ...
'add_latency', 1, ...
'mult_latency', 2, ...
'bram_latency', 2, ...
'conv_latency', 1, ...
'quantization', 'Round (unbiased: +/- Inf)', ...
'overflow', 'Saturate', ...
'arch', 'Virtex5', ...
'opt_target', 'logic', ...
'coeffs_bit_limit', 8, ...
'delays_bit_limit', 8, ...
'specify_mult', 'off', ...
'mult_spec', [2 2 2 2 2], ...
'hardcode_shifts', 'off', ...
'shift_schedule', [1 1 1 1 1], ...
'dsp48_adders', 'off', ...
'unscramble', 'on', ...
'bit_growth_chart', [0 0], ...
};
% Retrieve values from mask fields.
FFTSize = get_var('FFTSize', 'defaults', defaults, varargin{:});
n_inputs = get_var('n_inputs', 'defaults', defaults, varargin{:});
input_bit_width = get_var('input_bit_width', 'defaults', defaults, varargin{:});
coeff_bit_width = get_var('coeff_bit_width', 'defaults', defaults, varargin{:});
add_latency = get_var('add_latency', 'defaults', defaults, varargin{:});
mult_latency = get_var('mult_latency', 'defaults', defaults, varargin{:});
bram_latency = get_var('bram_latency', 'defaults', defaults, varargin{:});
conv_latency = get_var('conv_latency', 'defaults', defaults, varargin{:});
quantization = get_var('quantization', 'defaults', defaults, varargin{:});
overflow = get_var('overflow', 'defaults', defaults, varargin{:});
arch = get_var('arch', 'defaults', defaults, varargin{:});
opt_target = get_var('opt_target', 'defaults', defaults, varargin{:});
coeffs_bit_limit = get_var('coeffs_bit_limit', 'defaults', defaults, varargin{:});
delays_bit_limit = get_var('delays_bit_limit', 'defaults', defaults, varargin{:});
specify_mult = get_var('specify_mult', 'defaults', defaults, varargin{:});
mult_spec = get_var('mult_spec', 'defaults', defaults, varargin{:});
hardcode_shifts = get_var('hardcode_shifts', 'defaults', defaults, varargin{:});
shift_schedule = get_var('shift_schedule', 'defaults', defaults, varargin{:});
dsp48_adders = get_var('dsp48_adders', 'defaults', defaults, varargin{:});
unscramble = get_var('unscramble', 'defaults', defaults, varargin{:});
bit_growth_chart = get_var('bit_growth_chart', 'defaults', defaults, varargin{:});
% for bit growth FFT
bit_growth_chart =[reshape(bit_growth_chart, 1, []) zeros(1,FFTSize)];
bit_growth_chart
bit_growth_biplex = 0;
for i=1:(FFTSize - n_inputs)
bit_growth_biplex = bit_growth_biplex + bit_growth_chart(i);
end
bit_growth_sum = 0;
for i=1:FFTSize,
bit_growth_sum = bit_growth_sum + bit_growth_chart(i);
end
if( strcmp(specify_mult, 'on') && length(mult_spec) ~= FFTSize ),
error('fft_init.m: Multiplier use specification for stages does not match FFT size');
clog('fft_init.m: Multiplier use specification for stages does not match FFT size','error');
return
end
% split up multiplier specification
mults_biplex = 2.*ones(1, FFTSize-n_inputs);
mults_direct = 2.*ones(1, n_inputs);
if strcmp(specify_mult, 'on'),
mults_biplex(1:FFTSize-n_inputs) = mult_spec(1: FFTSize-n_inputs);
mults_direct = mult_spec(FFTSize-n_inputs+1:FFTSize);
end
% split up shift schedule
shifts_biplex = ones(1, FFTSize-n_inputs);
shifts_direct = ones(1, n_inputs);
if strcmp(hardcode_shifts, 'on'),
shifts_biplex(1:FFTSize-n_inputs) = shift_schedule(1: FFTSize-n_inputs);
shifts_direct = shift_schedule(FFTSize-n_inputs+1:FFTSize);
end
%% inports
xlsub2_sync = xInport('sync');
xlsub2_shift = xInport('shift');
if (n_inputs < 1 )
pol0 = xInport('pol0');
pol1 = xInport('pol1');
else
xlsub2_in = cell(1,2^n_inputs);
for i=1:2^n_inputs
xlsub2_in{i} = xInport(['in',num2str(i-1)]);
end
end
%% outports
xlsub2_sync_out = xOutport('sync_out');
xlsub2_of = xOutport('of');
if (n_inputs <1)
out0 = xOutport('out0');
out1 = xOutport('out1');
else
xlsub2_out = cell(1,2^n_inputs);
for i =1:2^n_inputs
xlsub2_out{i} = xOutport(['out',num2str(i-1)]);
end
end
%% diagram
% Add biplex FFTs
if (n_inputs == 0)
% block: fft_xblock_models/fft1/fft_biplex
xlsub2_fft_biplex_out1 = xSignal;
xlsub2_fft_biplex_out2 = xSignal;
xlsub2_fft_biplex_out3 = xSignal;
xlsub2_fft_biplex_out4 = xSignal;
xlsub2_fft_biplex_sub = xBlock(struct('source', str2func('fft_biplex_init_xblock'), 'name', 'fft_biplex','depend',{{'fft_biplex_init_xblock'}}), ...
struct( 'FFTSize', FFTSize-n_inputs,...
'input_bit_width', input_bit_width,...
'coeff_bit_width', coeff_bit_width,...
'quantization', quantization,...
'overflow', overflow,...
'add_latency', add_latency,...
'mult_latency', mult_latency,...
'bram_latency', bram_latency,...
'conv_latency', conv_latency, ...
'arch', arch, ...
'opt_target', opt_target, ...
'coeffs_bit_limit', coeffs_bit_limit, ...
'delays_bit_limit', delays_bit_limit, ...
'hardcore_shifts',hardcode_shifts,...
'shift_schedule', shift_schedule, ...
'specify_mult', specify_mult, ...
'mult_spec', mult_spec, ...
'dsp48_adders', dsp48_adders, ...
'bit_growth_chart', bit_growth_chart), ...
{xlsub2_sync, xlsub2_shift, pol0, pol1}, ...
{xlsub2_fft_biplex_out1, xlsub2_fft_biplex_out2, xlsub2_fft_biplex_out3, xlsub2_fft_biplex_out4});
elseif (n_inputs ~= FFTSize)
% block: fft_xblock_models/fft1/of_or
xlsub2_fft_direct_xblock_out4 = xSignal;
xlsub2_fft_biplex_sub = cell(1,2^(n_inputs-1));
xlsub2_fft_biplex_out1 = cell(1,2^(n_inputs-1));
xlsub2_fft_biplex_out2 = cell(1,2^(n_inputs-1));
xlsub2_fft_biplex_out3 = cell(1,2^(n_inputs-1));
xlsub2_fft_biplex_out4 = cell(1,2^(n_inputs-1));
for i=1:2^(n_inputs-1),
xlsub2_fft_biplex_out1{i} = xSignal;
xlsub2_fft_biplex_out2{i} = xSignal;
xlsub2_fft_biplex_out3{i} = xSignal;
xlsub2_fft_biplex_out4{i} = xSignal;
xlsub2_fft_biplex_sub{i} = xBlock(struct('source', str2func('fft_biplex_init_xblock'), 'name', ['fft_biplex',num2str(i)],'depend',{{'fft_biplex_init_xblock'}}), ...
struct( 'FFTSize', FFTSize-n_inputs,...
'input_bit_width', input_bit_width,...
'coeff_bit_width', coeff_bit_width,...
'quantization', quantization,...
'overflow', overflow,...
'add_latency', add_latency,...
'mult_latency', mult_latency,...
'bram_latency', bram_latency,...
'conv_latency', conv_latency, ...
'arch', arch, ...
'opt_target', opt_target, ...
'coeffs_bit_limit', coeffs_bit_limit, ...
'delays_bit_limit', delays_bit_limit, ...
'hardcore_shifts',hardcode_shifts,...
'shift_schedule', shift_schedule, ...
'specify_mult', specify_mult, ...
'mult_spec', mult_spec, ...
'dsp48_adders', dsp48_adders, ...
'bit_growth_chart', bit_growth_chart), ...
{xlsub2_sync, xlsub2_shift, xlsub2_in{2*i-1}, xlsub2_in{2*i}}, ...
{xlsub2_fft_biplex_out1{i}, xlsub2_fft_biplex_out2{i}, xlsub2_fft_biplex_out3{i}, xlsub2_fft_biplex_out4{i}});
end
xlsub2_of_or = xBlock(struct('source', 'Logical', 'name', 'of_or'), ...
struct('logical_function', 'OR', ...
'inputs',2^(n_inputs-1)+1, ...
'latency', 1), ...
[{xlsub2_fft_direct_xblock_out4}, xlsub2_fft_biplex_out4], ...
{xlsub2_of});
end
% Add direct FFTs
if (n_inputs == 0)
xlsub2_sync_out.bind(xlsub2_fft_biplex_out1);
out0.bind(xlsub2_fft_biplex_out2);
out1.bind(xlsub2_fft_biplex_out3);
xlsub2_of.bind(xlsub2_fft_biplex_out4);
elseif (n_inputs == FFTSize)
xlsub2_fft_direct_xblock_out1 = xSignal;
xlsub2_fft_direct_xblock_out2 = xSignal;
xlsub2_fft_direct_xblock_out3 = xSignal;
xlsub2_fft_direct_xblock_sub = xBlock(struct('source', str2func('fft_direct_init_xblock'), 'name', 'fft_direct_xblock','depend',{{'fft_direct_init_xblock'}}), ...
struct('FFTSize', n_inputs, ...
'input_bit_width', input_bit_width + bit_growth_biplex, ...
'coeff_bit_width', coeff_bit_width + bit_growth_biplex, ...
'map_tail', 'off', ...
'add_latency', add_latency, ...
'mult_latency', mult_latency, ...
'bram_latency', bram_latency, ...
'conv_latency', conv_latency, ...
'quantization', quantization, ...
'overflow', overflow, ...
'arch', arch, ...
'opt_target', opt_target, ...
'coeffs_bit_limit', coeffs_bit_limit, ...
'specify_mult',specify_mult, ...
'mult_spec', mults_direct, ...
'hardcode_shifts', hardcode_shifts, ...
'shift_schedule', shifts_direct, ...
'dsp48_adders', dsp48_adders, ...
'bit_growth_chart',bit_growth_chart((FFTSize - n_inputs + 1):end)), ...
[{xlsub2_sync},{xlsub2_shift}, xlsub2_in], ...
[{xlsub2_sync_out},xlsub2_out,{xlsub2_of}]);
else
% block: fft_xblock_models/fft1/slice
xlsub2_slice_out1 = xSignal;
xlsub2_slice = xBlock(struct('source', 'Slice', 'name', 'slice'), ...
struct('mode', 'Lower Bit Location + Width', ...
'bit0', FFTSize-n_inputs, ...
'nbits', n_inputs), ...
{xlsub2_shift}, ...
{xlsub2_slice_out1});
fft_direct_inputs = cell(1,2^n_inputs);
fft_direct_outputs = cell(1,2^n_inputs);
for i = 1:2^n_inputs,
fft_direct_inputs{i} = xSignal;
fft_direct_outputs{i} = xSignal;
end
fft_direct_output1 = xSignal;
xlsub2_fft_direct_xblock_sub = xBlock(struct('source', str2func('fft_direct_init_xblock'), 'name', 'fft_direct_xblock','depend',{{'fft_direct_init_xblock'}}), ...
struct('FFTSize', n_inputs, ...
'input_bit_width', input_bit_width + bit_growth_biplex, ...
'coeff_bit_width', coeff_bit_width + bit_growth_biplex, ...
'map_tail', 'on', ...
'LargerFFTSize', FFTSize, ...
'StartStage',FFTSize-n_inputs+1, ...
'add_latency', add_latency, ...
'mult_latency', mult_latency, ...
'bram_latency', bram_latency, ...
'conv_latency', conv_latency, ...
'quantization', quantization, ...
'overflow', overflow, ...
'arch', arch, ...
'opt_target', opt_target, ...
'coeffs_bit_limit', coeffs_bit_limit, ...
'specify_mult',specify_mult, ...
'mult_spec', mults_direct, ...
'hardcode_shifts', hardcode_shifts, ...
'shift_schedule', shifts_direct, ...
'dsp48_adders', dsp48_adders, ...
'bit_growth_chart',bit_growth_chart((FFTSize - n_inputs + 1):end)), ...
[{xlsub2_fft_biplex_out1{1}}, {xlsub2_slice_out1}, fft_direct_inputs], ...
[{fft_direct_output1},fft_direct_outputs,{xlsub2_fft_direct_xblock_out4}]);
for i=1:2^(n_inputs-1)
xlsub2_fft_biplex_out2{i}.bind(fft_direct_inputs{2*i-1});
xlsub2_fft_biplex_out3{i}.bind(fft_direct_inputs{2*i});
end
% Add Unscrambler
if strcmp(unscramble, 'on')
% block: fft_xblock_models/fft1/fft_unscrambler_xblock
xlsub2_fft_unscrambler_xblock_sub = xBlock(struct('source', str2func('fft_unscrambler_init_xblock'), 'name', 'fft_unscrambler_xblock','depend',{{'fft_unscrambler_init_xblock'}}), ...
{FFTSize, n_inputs, bram_latency}, ...
[{fft_direct_output1}, fft_direct_outputs], ...
[{xlsub2_sync_out}, xlsub2_out]);
else
xlsub2_sync_out.bind(fft_direct_output1);
for i=1:2^n_inputs,
xlsub2_out{i}.bind(fft_direct_outputs{i});
end
end
end
end