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
297
298
299
300
|
//
// Copyright 2011 Ettus Research LLC
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
//
// Buffer pool. Contains 8 buffers, each 2K (512 by 32). Each buffer
// is a dual-ported RAM. Port A on each of them is indirectly connected
// to the wishbone bus by a bridge. Port B may be connected any one of the
// 8 (4 rd, 4 wr) FIFO-like streaming interaces, or disconnected. The wishbone bus
// provides access to all 8 buffers, and also controls the connections
// between the ports and the buffers, allocating them as needed.
// wb_adr is 16 bits --
// bits 13:11 select which buffer
// bits 10:2 select line in buffer
// bits 1:0 are unused (32-bit access only)
// BUF_SIZE is in address lines (i.e. log2 of number of lines).
// For S3 it should be 9 (512 words, 2KB)
// For V5 it should be at least 10 (1024 words, 4KB) or 11 (2048 words, 8KB)
module buffer_pool
#(parameter BUF_SIZE = 9,
parameter SET_ADDR = 64)
(input wb_clk_i,
input wb_rst_i,
input wb_we_i,
input wb_stb_i,
input [15:0] wb_adr_i,
input [31:0] wb_dat_i,
output [31:0] wb_dat_o,
output reg wb_ack_o,
output wb_err_o,
output wb_rty_o,
input stream_clk,
input stream_rst,
input set_stb, input [7:0] set_addr, input [31:0] set_data,
output [31:0] status,
output sys_int_o,
output [31:0] s0, output [31:0] s1, output [31:0] s2, output [31:0] s3,
output [31:0] s4, output [31:0] s5, output [31:0] s6, output [31:0] s7,
// Write Interfaces
input [31:0] wr0_data_i, input [3:0] wr0_flags_i, input wr0_ready_i, output wr0_ready_o,
input [31:0] wr1_data_i, input [3:0] wr1_flags_i, input wr1_ready_i, output wr1_ready_o,
input [31:0] wr2_data_i, input [3:0] wr2_flags_i, input wr2_ready_i, output wr2_ready_o,
input [31:0] wr3_data_i, input [3:0] wr3_flags_i, input wr3_ready_i, output wr3_ready_o,
// Read Interfaces
output [31:0] rd0_data_o, output [3:0] rd0_flags_o, output rd0_ready_o, input rd0_ready_i,
output [31:0] rd1_data_o, output [3:0] rd1_flags_o, output rd1_ready_o, input rd1_ready_i,
output [31:0] rd2_data_o, output [3:0] rd2_flags_o, output rd2_ready_o, input rd2_ready_i,
output [31:0] rd3_data_o, output [3:0] rd3_flags_o, output rd3_ready_o, input rd3_ready_i
);
wire [7:0] sel_a;
wire [BUF_SIZE-1:0] buf_addra = wb_adr_i[BUF_SIZE+1:2]; // ignore address 1:0, 32-bit access only
wire [2:0] which_buf = wb_adr_i[BUF_SIZE+4:BUF_SIZE+2]; // address 15:14 selects the buffer pool
decoder_3_8 dec(.sel(which_buf),.res(sel_a));
genvar i;
wire go;
reg [2:0] port[0:7];
reg [3:0] read_src[0:3];
reg [3:0] write_src[0:3];
wire [7:0] done;
wire [7:0] error;
wire [7:0] idle;
wire [31:0] buf_doa[0:7];
wire [7:0] buf_enb;
wire [7:0] buf_web;
wire [BUF_SIZE-1:0] buf_addrb[0:7];
wire [31:0] buf_dib[0:7];
wire [31:0] buf_dob[0:7];
wire [31:0] wr_data_i[0:7];
wire [3:0] wr_flags_i[0:7];
wire [7:0] wr_ready_i;
wire [7:0] wr_ready_o;
wire [31:0] rd_data_o[0:7];
wire [3:0] rd_flags_o[0:7];
wire [7:0] rd_ready_o;
wire [7:0] rd_ready_i;
assign status = {8'd0,idle[7:0],error[7:0],done[7:0]};
assign s0 = buf_addrb[0];
assign s1 = buf_addrb[1];
assign s2 = buf_addrb[2];
assign s3 = buf_addrb[3];
assign s4 = buf_addrb[4];
assign s5 = buf_addrb[5];
assign s6 = buf_addrb[6];
assign s7 = buf_addrb[7];
wire [31:0] fifo_ctrl;
setting_reg #(.my_addr(SET_ADDR))
sreg(.clk(stream_clk),.rst(stream_rst),.strobe(set_stb),.addr(set_addr),.in(set_data),
.out(fifo_ctrl),.changed(go));
integer k;
always @(posedge stream_clk)
if(stream_rst)
for(k=0;k<8;k=k+1)
port[k] <= 4; // disabled
else
for(k=0;k<8;k=k+1)
if(go & (fifo_ctrl[31:28]==k))
port[k] <= fifo_ctrl[27:25];
always @(posedge stream_clk)
if(stream_rst)
for(k=0;k<4;k=k+1)
read_src[k] <= 8; // disabled
else
for(k=0;k<4;k=k+1)
if(go & fifo_ctrl[22] & (fifo_ctrl[27:25]==k))
read_src[k] <= fifo_ctrl[31:28];
always @(posedge stream_clk)
if(stream_rst)
for(k=0;k<4;k=k+1)
write_src[k] <= 8; // disabled
else
for(k=0;k<4;k=k+1)
if(go & fifo_ctrl[23] & (fifo_ctrl[27:25]==k))
write_src[k] <= fifo_ctrl[31:28];
generate
for(i=0;i<8;i=i+1)
begin : gen_buffer
RAMB16_S36_S36 dpram
(.DOA(buf_doa[i]),.ADDRA(buf_addra),.CLKA(wb_clk_i),.DIA(wb_dat_i),.DIPA(4'h0),
.ENA(wb_stb_i & sel_a[i]),.SSRA(0),.WEA(wb_we_i),
.DOB(buf_dob[i]),.ADDRB(buf_addrb[i]),.CLKB(stream_clk),.DIB(buf_dib[i]),.DIPB(4'h0),
.ENB(buf_enb[i]),.SSRB(0),.WEB(buf_web[i]) );
/*
ram_2port #(.DWIDTH(32),.AWIDTH(BUF_SIZE)) buffer
(.clka(wb_clk_i),.ena(wb_stb_i & sel_a[i]),.wea(wb_we_i),
.addra(buf_addra),.dia(wb_dat_i),.doa(buf_doa[i]),
.clkb(stream_clk),.enb(buf_enb[i]),.web(buf_web[i]),
.addrb(buf_addrb[i]),.dib(buf_dib[i]),.dob(buf_dob[i]));
*/
buffer_int #(.BUF_NUM(i),.BUF_SIZE(BUF_SIZE)) buffer_int
(.clk(stream_clk),.rst(stream_rst),
.ctrl_word(fifo_ctrl),.go(go & (fifo_ctrl[31:28]==i)),
.done(done[i]),.error(error[i]),.idle(idle[i]),
.en_o(buf_enb[i]),
.we_o(buf_web[i]),
.addr_o(buf_addrb[i]),
.dat_to_buf(buf_dib[i]),
.dat_from_buf(buf_dob[i]),
.wr_data_i(wr_data_i[i]),
.wr_flags_i(wr_flags_i[i]),
.wr_ready_i(wr_ready_i[i]),
.wr_ready_o(wr_ready_o[i]),
.rd_data_o(rd_data_o[i]),
.rd_flags_o(rd_flags_o[i]),
.rd_ready_o(rd_ready_o[i]),
.rd_ready_i(rd_ready_i[i]) );
mux4 #(.WIDTH(37))
mux4_wr (.en(~port[i][2]),.sel(port[i][1:0]),
.i0({wr0_data_i,wr0_flags_i,wr0_ready_i}),
.i1({wr1_data_i,wr1_flags_i,wr1_ready_i}),
.i2({wr2_data_i,wr2_flags_i,wr2_ready_i}),
.i3({wr3_data_i,wr3_flags_i,wr3_ready_i}),
.o({wr_data_i[i],wr_flags_i[i],wr_ready_i[i]}) );
mux4 #(.WIDTH(1))
mux4_rd (.en(~port[i][2]),.sel(port[i][1:0]),
.i0(rd0_ready_i),.i1(rd1_ready_i),.i2(rd2_ready_i),.i3(rd3_ready_i),
.o(rd_ready_i[i]));
end // block: gen_buffer
endgenerate
//----------------------------------------------------------------------
// Wishbone Outputs
// Use the following lines if ram output and mux can be made fast enough
assign wb_err_o = 1'b0; // Unused for now
assign wb_rty_o = 1'b0; // Unused for now
always @(posedge wb_clk_i)
wb_ack_o <= wb_stb_i & ~wb_ack_o;
assign wb_dat_o = buf_doa[which_buf];
// Use this if we can't make the RAM+MUX fast enough
// reg [31:0] wb_dat_o_reg;
// reg stb_d1;
// always @(posedge wb_clk_i)
// begin
// wb_dat_o_reg <= buf_doa[which_buf];
// stb_d1 <= wb_stb_i;
// wb_ack_o <= (stb_d1 & ~wb_ack_o) | (wb_we_i & wb_stb_i);
// end
//assign wb_dat_o = wb_dat_o_reg;
mux8 #(.WIDTH(1))
mux8_wr0(.en(~write_src[0][3]),.sel(write_src[0][2:0]),
.i0(wr_ready_o[0]), .i1(wr_ready_o[1]), .i2(wr_ready_o[2]), .i3(wr_ready_o[3]),
.i4(wr_ready_o[4]), .i5(wr_ready_o[5]), .i6(wr_ready_o[6]), .i7(wr_ready_o[7]),
.o(wr0_ready_o));
mux8 #(.WIDTH(1))
mux8_wr1(.en(~write_src[1][3]),.sel(write_src[1][2:0]),
.i0(wr_ready_o[0]), .i1(wr_ready_o[1]), .i2(wr_ready_o[2]), .i3(wr_ready_o[3]),
.i4(wr_ready_o[4]), .i5(wr_ready_o[5]), .i6(wr_ready_o[6]), .i7(wr_ready_o[7]),
.o(wr1_ready_o));
mux8 #(.WIDTH(1))
mux8_wr2(.en(~write_src[2][3]),.sel(write_src[2][2:0]),
.i0(wr_ready_o[0]), .i1(wr_ready_o[1]), .i2(wr_ready_o[2]), .i3(wr_ready_o[3]),
.i4(wr_ready_o[4]), .i5(wr_ready_o[5]), .i6(wr_ready_o[6]), .i7(wr_ready_o[7]),
.o(wr2_ready_o));
mux8 #(.WIDTH(1))
mux8_wr3(.en(~write_src[3][3]),.sel(write_src[3][2:0]),
.i0(wr_ready_o[0]), .i1(wr_ready_o[1]), .i2(wr_ready_o[2]), .i3(wr_ready_o[3]),
.i4(wr_ready_o[4]), .i5(wr_ready_o[5]), .i6(wr_ready_o[6]), .i7(wr_ready_o[7]),
.o(wr3_ready_o));
mux8 #(.WIDTH(37))
mux8_rd0(.en(~read_src[0][3]),.sel(read_src[0][2:0]),
.i0({rd_data_o[0],rd_flags_o[0],rd_ready_o[0]}),
.i1({rd_data_o[1],rd_flags_o[1],rd_ready_o[1]}),
.i2({rd_data_o[2],rd_flags_o[2],rd_ready_o[2]}),
.i3({rd_data_o[3],rd_flags_o[3],rd_ready_o[3]}),
.i4({rd_data_o[4],rd_flags_o[4],rd_ready_o[4]}),
.i5({rd_data_o[5],rd_flags_o[5],rd_ready_o[5]}),
.i6({rd_data_o[6],rd_flags_o[6],rd_ready_o[6]}),
.i7({rd_data_o[7],rd_flags_o[7],rd_ready_o[7]}),
.o({rd0_data_o,rd0_flags_o,rd0_ready_o}));
mux8 #(.WIDTH(37))
mux8_rd1(.en(~read_src[1][3]),.sel(read_src[1][2:0]),
.i0({rd_data_o[0],rd_flags_o[0],rd_ready_o[0]}),
.i1({rd_data_o[1],rd_flags_o[1],rd_ready_o[1]}),
.i2({rd_data_o[2],rd_flags_o[2],rd_ready_o[2]}),
.i3({rd_data_o[3],rd_flags_o[3],rd_ready_o[3]}),
.i4({rd_data_o[4],rd_flags_o[4],rd_ready_o[4]}),
.i5({rd_data_o[5],rd_flags_o[5],rd_ready_o[5]}),
.i6({rd_data_o[6],rd_flags_o[6],rd_ready_o[6]}),
.i7({rd_data_o[7],rd_flags_o[7],rd_ready_o[7]}),
.o({rd1_data_o,rd1_flags_o,rd1_ready_o}));
mux8 #(.WIDTH(37))
mux8_rd2(.en(~read_src[2][3]),.sel(read_src[2][2:0]),
.i0({rd_data_o[0],rd_flags_o[0],rd_ready_o[0]}),
.i1({rd_data_o[1],rd_flags_o[1],rd_ready_o[1]}),
.i2({rd_data_o[2],rd_flags_o[2],rd_ready_o[2]}),
.i3({rd_data_o[3],rd_flags_o[3],rd_ready_o[3]}),
.i4({rd_data_o[4],rd_flags_o[4],rd_ready_o[4]}),
.i5({rd_data_o[5],rd_flags_o[5],rd_ready_o[5]}),
.i6({rd_data_o[6],rd_flags_o[6],rd_ready_o[6]}),
.i7({rd_data_o[7],rd_flags_o[7],rd_ready_o[7]}),
.o({rd2_data_o,rd2_flags_o,rd2_ready_o}));
mux8 #(.WIDTH(37))
mux8_rd3(.en(~read_src[3][3]),.sel(read_src[3][2:0]),
.i0({rd_data_o[0],rd_flags_o[0],rd_ready_o[0]}),
.i1({rd_data_o[1],rd_flags_o[1],rd_ready_o[1]}),
.i2({rd_data_o[2],rd_flags_o[2],rd_ready_o[2]}),
.i3({rd_data_o[3],rd_flags_o[3],rd_ready_o[3]}),
.i4({rd_data_o[4],rd_flags_o[4],rd_ready_o[4]}),
.i5({rd_data_o[5],rd_flags_o[5],rd_ready_o[5]}),
.i6({rd_data_o[6],rd_flags_o[6],rd_ready_o[6]}),
.i7({rd_data_o[7],rd_flags_o[7],rd_ready_o[7]}),
.o({rd3_data_o,rd3_flags_o,rd3_ready_o}));
assign sys_int_o = (|error) | (|done);
endmodule // buffer_pool
|