Skip to content

X
x393
Commit 5a4a62b0 authored by Andrey Filippov's avatar Andrey Filippov
Browse files
Options

editing DCT-IV

parent 71577936
Hide whitespace changes
Inline Side-by-side
Showing with 39 additions and 63 deletions
dsp/dct_iv8_1d.v
View file @ 5a4a62b0
...@@ -38,6 +38,8 @@ ...@@ -38,6 +38,8 @@
* with at least one of the Free Software programs. * with at least one of the Free Software programs.
*/ */
`timescale 1ns/1ps `timescale 1ns/1ps
// No saturation here, and no rounding as we do not need to match decoder (be bit-precise), skipping rounding adder
// will reduce needed resources
module dct_iv8_1d#( module dct_iv8_1d#(
parameter WIDTH = 24, // input data width parameter WIDTH = 24, // input data width
...@@ -46,8 +48,6 @@ module dct_iv8_1d#( ...@@ -46,8 +48,6 @@ module dct_iv8_1d#(
parameter B_WIDTH = 18, parameter B_WIDTH = 18,
parameter A_WIDTH = 25, parameter A_WIDTH = 25,
parameter P_WIDTH = 48, parameter P_WIDTH = 48,
// parameter M_WIDTH = 43, // actual multiplier width (== (A_WIDTH +B_WIDTH)
// parameter ROUND_OUT = 8, // cut these number of LSBs on the output, round result (in addition to COSINE_SHIFT)
parameter COSINE_SHIFT= 17, parameter COSINE_SHIFT= 17,
parameter COS_01_32 = 130441, // (1<<17) * cos( 1*pi/32) parameter COS_01_32 = 130441, // (1<<17) * cos( 1*pi/32)
parameter COS_03_32 = 125428, // (1<<17) * cos( 3*pi/32) parameter COS_03_32 = 125428, // (1<<17) * cos( 3*pi/32)
...@@ -73,14 +73,10 @@ module dct_iv8_1d#( ...@@ -73,14 +73,10 @@ module dct_iv8_1d#(
// start_out-x-Y0-x-Y7-x-Y1-x-Y6-x-Y3-x-Y4-x-Y2-x-Y5 // start_out-x-Y0-x-Y7-x-Y1-x-Y6-x-Y3-x-Y4-x-Y2-x-Y5
output reg en_out // valid at the same time slot as pre2_start_out (goes active with pre2_start_out) output reg en_out // valid at the same time slot as pre2_start_out (goes active with pre2_start_out)
); );
// localparam TOTAL_RSHIFT= COSINE_SHIFT + ROUND_OUT;
// localparam BEFORE_SAT_WIDTH = P_WIDTH - TOTAL_RSHIFT;
localparam RSHIFT1 = 2; // safe right shift for stage 1 localparam RSHIFT1 = 2; // safe right shift for stage 1
localparam STAGE1_RSHIFT = COSINE_SHIFT + (WIDTH - A_WIDTH) + RSHIFT1; // divide by 4 in stage 1 - never saturates localparam STAGE1_RSHIFT = COSINE_SHIFT + (WIDTH - A_WIDTH) + RSHIFT1; // divide by 4 in stage 1 - never saturates
localparam STAGE2_RSHIFT = COSINE_SHIFT + (OUT_WIDTH - WIDTH) +(OUT_RSHIFT-RSHIFT1); // divide by 4 in stage 1 - never saturates localparam STAGE2_RSHIFT = COSINE_SHIFT + (OUT_WIDTH - WIDTH) +(OUT_RSHIFT-RSHIFT1); // divide by 4 in stage 1 - never saturates
// STAGE2_RSHIFT should be >0 ( >=1 ) for rounding // STAGE2_RSHIFT should be >0 ( >=1 ) for rounding
// TODO: Add rounding to stage 1
// register files on the D-inputs of DSPs // register files on the D-inputs of DSPs
reg signed [A_WIDTH-1:0] dsp_din_1_ram[0:1] ; // just two registers reg signed [A_WIDTH-1:0] dsp_din_1_ram[0:1] ; // just two registers
...@@ -126,23 +122,16 @@ module dct_iv8_1d#( ...@@ -126,23 +122,16 @@ module dct_iv8_1d#(
reg dsp_accum_2; // 0 - use multiplier result, 1 add to accumulator reg dsp_accum_2; // 0 - use multiplier result, 1 add to accumulator
wire signed [P_WIDTH-1:0] dsp_p_2; wire signed [P_WIDTH-1:0] dsp_p_2;
reg [7:0] phase;
reg [3:0] phase_cnt; reg [3:0] phase_cnt;
reg [OUT_WIDTH -1:0] dout_r;
reg [2:0] per_type; // idle/last:0, first cycle - 1, 2-nd - 2, other - 3,... ~en->6 ->7 -> 0 (to generate pre2_start_out)
// Temporarily adding 1 extra latency cycle for rounding/saturation. TODO: Remove when moved to DSP itself
reg pre3_start_out; // 3 clock cycle before F4 output, full dout sequence
// start_out-X-F4-X-F2-X-F6-F5-F0-F3-X-F1-X-F7
reg pre_en_out; // valid at the same time slot as pre2_start_out (goes active with pre2_start_out)
reg run_in; // receiving input data
reg restart; // restarting next block if en was active at phase=14;
reg run_out; // running output data
assign dsp_ain_2 = dsp_p_1 [STAGE1_RSHIFT +: A_WIDTH]; assign dsp_ain_2 = dsp_p_1 [STAGE1_RSHIFT +: A_WIDTH];
assign dout = dsp_p_2 [STAGE2_RSHIFT +: OUT_WIDTH]; // dout_r;
assign dout = dout_r;
generate generate
if (A_WIDTH > WIDTH) assign dsp_ain_1 = {{A_WIDTH-WIDTH{d_in[WIDTH-1]}},d_in}; if (A_WIDTH > WIDTH) assign dsp_ain_1 = {{A_WIDTH-WIDTH{d_in[WIDTH-1]}},d_in};
...@@ -161,17 +150,22 @@ module dct_iv8_1d#( ...@@ -161,17 +150,22 @@ module dct_iv8_1d#(
end end
always @ (posedge clk) begin always @ (posedge clk) begin
if (rst) per_type <= 0; if (rst) restart <= 0;
else if (start) per_type <= 3'h1; else restart <= (phase_cnt == 14) && en;
else if (phase[7]) begin
if (!per_type[2] && !en) per_type <= 3'h7; if (rst) run_in <= 0;
else if ((per_type != 0) && (per_type != 3)) per_type <= per_type + 1; else if (start || restart) run_in <= 1;
end else if (phase_cnt==15) run_in <= 0;
phase <= {phase[6:0], start | (phase[7] & (|per_type))};
//TODO: if (rst) run_out <= 0;
else if (phase_cnt == 13) run_out <= run_in;
if (rst || (!run_in && !run_out)) phase_cnt <= 0;
else phase_cnt <= phase_cnt + 1;
pre2_start_out <= run_out && (phase_cnt == 13);
if (rst || start || phase[7]) phase_cnt <= 0; en_out <= run_out && !phase_cnt[0];
else if (|phase[6:0]) phase_cnt <= phase_cnt + 1;
// Cosine table, defined to fit into 17 bits for 18-bit signed DSP B-operand // Cosine table, defined to fit into 17 bits for 18-bit signed DSP B-operand
case (phase_cnt) case (phase_cnt)
...@@ -192,41 +186,25 @@ module dct_iv8_1d#( ...@@ -192,41 +186,25 @@ module dct_iv8_1d#(
4'he: dsp_bin <= COS_12_32; 4'he: dsp_bin <= COS_12_32;
4'hf: dsp_bin <= 'bx; 4'hf: dsp_bin <= 'bx;
endcase endcase
dout_r <= dsp_p_2 [STAGE2_RSHIFT +: OUT_WIDTH] + dsp_p_2 [STAGE2_RSHIFT - 1];
if (rst) pre3_start_out <= 0;
else pre3_start_out <= (per_type == 2) && phase[3];
pre2_start_out <=pre3_start_out;
if (rst || !(en || (|phase))) pre_en_out <= 0;
else if (phase[3]) begin
if (per_type == 2) pre_en_out <= 1;
else if (per_type[2:0]==0) pre_en_out <= 0;
end
en_out <= pre_en_out;
end end
// Control signals for each phase // Control signals for each phase
wire p00 = phase[3:0] == 0; wire p00 = phase_cnt[3:0] == 0;
wire p01 = phase[3:0] == 1; wire p01 = phase_cnt[3:0] == 1;
wire p02 = phase[3:0] == 2; wire p02 = phase_cnt[3:0] == 2;
wire p03 = phase[3:0] == 3; wire p03 = phase_cnt[3:0] == 3;
wire p04 = phase[3:0] == 4; wire p04 = phase_cnt[3:0] == 4;
wire p05 = phase[3:0] == 5; wire p05 = phase_cnt[3:0] == 5;
wire p06 = phase[3:0] == 6; wire p06 = phase_cnt[3:0] == 6;
wire p07 = phase[3:0] == 7; wire p07 = phase_cnt[3:0] == 7;
wire p08 = phase[3:0] == 8; wire p08 = phase_cnt[3:0] == 8;
wire p09 = phase[3:0] == 9; wire p09 = phase_cnt[3:0] == 9;
wire p10 = phase[3:0] == 10; wire p10 = phase_cnt[3:0] == 10;
wire p11 = phase[3:0] == 11; wire p11 = phase_cnt[3:0] == 11;
wire p12 = phase[3:0] == 12; wire p12 = phase_cnt[3:0] == 12;
wire p13 = phase[3:0] == 13; wire p13 = phase_cnt[3:0] == 13;
wire p14 = phase[3:0] == 14; wire p14 = phase_cnt[3:0] == 14;
wire p15 = phase[3:0] == 15; wire p15 = phase_cnt[3:0] == 15;
always @ (posedge clk) begin always @ (posedge clk) begin
// p00 | p01 | p02 | p03 | p04 | p05 | p06 | p07 | p08 | p09 | p10 | p11 | p12 | p13 | p14 | p15 ; // p00 | p01 | p02 | p03 | p04 | p05 | p06 | p07 | p08 | p09 | p10 | p11 | p12 | p13 | p14 | p15 ;
dsp_din_1_we <= p01 | p03 | p08 | p15 | start; dsp_din_1_we <= p01 | p03 | p08 | p15 | start;
...@@ -317,6 +295,4 @@ module dct_iv8_1d#( ...@@ -317,6 +295,4 @@ module dct_iv8_1d#(
.post_add (1'b0), // input .post_add (1'b0), // input
.pout (dsp_p_2) // output[47:0] signed .pout (dsp_p_2) // output[47:0] signed
); );
endmodule endmodule
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment