Dan Strother is sharing code with you
Bitbucket is a code hosting site. Unlimited public and private repositories. Free for small teams.
Don't show this again1 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 | //
// Copyright (c) 2011, Daniel Strother < http://danstrother.com/ >
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
// - Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
// - Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
// - The name of the author may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED
// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
// TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
// LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Module Description:
// Sort of a catch-all for reconciling pipeline data:
// - finds best overall SAD/disparity (between all MULT_D pipes and previous-
// best in buffer; 'sad'/'disp').. this is the only mandatory feature
// - finds 2nd-best (excluding ones adjacent to the best; 'thresh').. used
// for uniqueness ratio checking in dlsc_stereobm_postprocess_uniqueness
// - finds SAD values for the disparities above ('sad_hi') and below ('sad_lo')
// the best one.. used for sub-pixel interpolation in
// dlsc_stereobm_postprocess_subpixel
// - performs texture-filtering ('filtered')
//
// This block is the top-level for this functionality. It has control/sequencing
// logic and the RAM used for the disparity row buffer (which is used for saving
// state across multiple passes).
module dlsc_stereobm_disparity #(
parameter IMG_WIDTH = 320,
parameter DISP_BITS = 6,
parameter DISPARITIES = (2**DISP_BITS),
parameter TEXTURE = 0,
parameter SUB_BITS = 4,
parameter UNIQUE_MUL = 1,
parameter MULT_D = 4,
parameter MULT_R = 3,
parameter SAD = 9,
parameter SAD_BITS = 16,
parameter PIPELINE_RD = 0, // enable pipeline register on BRAM read path
parameter PIPELINE_WR = 0, // enable pipeline register on BRAM write path
parameter PIPELINE_LUT4 = 0,
// derived parameters; don't touch
parameter DISP_BITS_R = (DISP_BITS *MULT_R),
parameter SAD_BITS_R = ( SAD_BITS *MULT_R),
parameter SAD_BITS_D = ( SAD_BITS *MULT_D),
parameter SAD_BITS_RD = ( SAD_BITS_R*MULT_D)
) (
input wire clk,
input wire rst,
// inputs from stereo pipeline
input wire in_valid,
input wire [ SAD_BITS_RD-1:0] in_sad,
// outputs to post-processing
output wire out_valid,
output wire [DISP_BITS_R -1:0] out_disp,
output wire [ SAD_BITS_R -1:0] out_sad,
// outputs in sub-pixel mode
output wire [ SAD_BITS_R -1:0] out_lo,
output wire [ SAD_BITS_R -1:0] out_hi,
// outputs in uniqueness mode
output wire [ SAD_BITS_R -1:0] out_thresh,
// outputs for texture filtering
output wire [ MULT_R -1:0] out_filtered
);
`include "dlsc_clog2.vh"
// number of valid pixels that make it to the end
localparam END_WIDTH = IMG_WIDTH - (DISPARITIES-1) - (SAD-1);
localparam END_WIDTH_BITS = `dlsc_clog2(END_WIDTH);
localparam LOHI_EN = (SUB_BITS>0||UNIQUE_MUL>0);
localparam BUF_BITS = DISP_BITS + (1 + (LOHI_EN>0?3:0) + (UNIQUE_MUL>0?1:0)) * SAD_BITS;
localparam BUF_BITS_R = BUF_BITS*MULT_R;
// ** compute delays to output **
localparam MULT_D_BITS = `dlsc_clog2(MULT_D);
localparam MULT_D1_BITS = `dlsc_clog2(MULT_D+1);
localparam MULT_D3_BITS = `dlsc_clog2(MULT_D+3);
localparam CYCLE_C4 = (PIPELINE_RD>0?4:3);
localparam CYCLE_C5 = CYCLE_C4 + 1;
// cycle at output of main sorter
localparam CYCLE_CM0 = CYCLE_C5 + 2 + (MULT_D1_BITS-1) * (PIPELINE_LUT4>0?2:1);
localparam CYCLE_CM1 = CYCLE_CM0 + 1;
localparam CYCLE_CM2 = CYCLE_CM1 + (PIPELINE_LUT4>0?1:0);
// cycle at output of threshold sorter
localparam CYCLE_CT0 = CYCLE_CM2 + 2 + (MULT_D3_BITS-1) * (PIPELINE_LUT4>0?2:1);
// cycle at output of this block
localparam CYCLE_CW = UNIQUE_MUL>0 ? CYCLE_CT0 : CYCLE_CM1;
// ** control **
reg ctrl_first; // first pass; must ignore memory contents
reg ctrl_last; // last pass; must output results
reg [DISP_BITS-1:0] ctrl_disp; // base disparity level for this pass
reg [DISP_BITS-1:0] ctrl_disp_prev; // base disparity level for previous pass
reg [END_WIDTH_BITS-1:0] ctrl_addr;
reg ctrl_addr_last; // ctrl_addr == (END_WIDTH-1)
/* verilator lint_off WIDTH */
always @(posedge clk) begin
if(rst) begin
ctrl_first <= 1'b1;
ctrl_last <= 1'b0;
ctrl_disp <= DISPARITIES-MULT_D; // start at maximum disparity
ctrl_disp_prev <= 0;
ctrl_addr <= 0;
ctrl_addr_last <= 1'b0;
end else if(in_valid) begin
ctrl_addr_last <= (ctrl_addr == (END_WIDTH-2));
if(!ctrl_addr_last) begin
ctrl_addr <= ctrl_addr + 1;
end else begin
ctrl_addr <= 0;
ctrl_disp_prev <= ctrl_disp;
if(!ctrl_last) begin
// not on last yet; but at least done with first
ctrl_first <= 1'b0;
if(TEXTURE == 0) begin
ctrl_last <= (ctrl_disp == MULT_D); // on 2nd-to-last; will be on last next cycle
end else begin
// need extra pass when performing texture filtering
ctrl_last <= (ctrl_disp == 0 ); // on 2nd-to-last; will be on last next cycle
end
ctrl_disp <= ctrl_disp - MULT_D;
end else begin
// on last; will be on first next cycle
ctrl_first <= 1'b1;
ctrl_last <= 1'b0;
ctrl_disp <= DISPARITIES-MULT_D; // start at maximum disparity
end
end
end
end
/* verilator lint_on WIDTH */
// ** pipeline delays **
wire [DISP_BITS-1:0] c3_ctrl_disp_prev;
wire c3_ctrl_first;
wire c3_ctrl_last;
dlsc_pipedelay #(
.DATA ( DISP_BITS + 2 ),
.DELAY ( (PIPELINE_RD>0) ? 3 : 2 )
) dlsc_pipedelay_inst_c3_ctrl (
.clk ( clk ),
.in_data ( { ctrl_disp_prev , ctrl_first , ctrl_last } ),
.out_data ( { c3_ctrl_disp_prev , c3_ctrl_first , c3_ctrl_last } )
);
wire [DISP_BITS-1:0] cm_ctrl_disp;
dlsc_pipedelay #(
.DATA ( DISP_BITS ),
.DELAY ( CYCLE_CM0 - 1 )
) dlsc_pipedelay_inst_cm_ctrl_disp (
.clk ( clk ),
.in_data ( ctrl_disp ),
.out_data ( cm_ctrl_disp )
);
// ** slices **
wire [BUF_BITS_R-1:0] c3_buf_data;
wire [BUF_BITS_R-1:0] cw_buf_data;
generate
genvar j;
for(j=0;j<MULT_R;j=j+1) begin:GEN_SLICES
dlsc_stereobm_disparity_slice #(
.DISP_BITS ( DISP_BITS ),
.TEXTURE ( TEXTURE ),
.SUB_BITS ( SUB_BITS ),
.UNIQUE_MUL ( UNIQUE_MUL ),
.MULT_D ( MULT_D ),
.SAD_BITS ( SAD_BITS ),
.PIPELINE_RD ( PIPELINE_RD ),
.PIPELINE_LUT4 ( PIPELINE_LUT4 )
) dlsc_stereobm_disparity_slice_inst (
.clk ( clk ),
.rst ( rst ),
.cm_ctrl_disp ( cm_ctrl_disp ),
.c3_ctrl_disp_prev ( c3_ctrl_disp_prev ),
.c3_ctrl_first ( c3_ctrl_first ),
.c3_ctrl_last ( c3_ctrl_last ),
.c0_pipe_sad ( in_sad [ (j* SAD_BITS_D) +: SAD_BITS_D ] ),
.c3_buf_data ( c3_buf_data [ (j* BUF_BITS ) +: BUF_BITS ] ),
.cw_buf_data ( cw_buf_data [ (j* BUF_BITS ) +: BUF_BITS ] ),
.cw_out_disp ( out_disp [ (j*DISP_BITS ) +: DISP_BITS ] ),
.cw_out_sad ( out_sad [ (j* SAD_BITS ) +: SAD_BITS ] ),
.cw_out_lo ( out_lo [ (j* SAD_BITS ) +: SAD_BITS ] ),
.cw_out_hi ( out_hi [ (j* SAD_BITS ) +: SAD_BITS ] ),
.cw_out_thresh ( out_thresh [ (j* SAD_BITS ) +: SAD_BITS ] ),
.cw_out_filtered ( out_filtered[ j ] )
);
end
endgenerate
// ** buffer **
dlsc_ram_dp #(
.DATA ( BUF_BITS_R ),
.ADDR ( END_WIDTH_BITS ),
.DEPTH ( END_WIDTH ),
.PIPELINE_WR ( CYCLE_CW + (PIPELINE_WR>0?1:0) ),
.PIPELINE_WR_DATA ( PIPELINE_WR>0?1:0 ),
.PIPELINE_RD ( 3 )
) dlsc_ram_dp_inst (
.write_clk ( clk ),
.write_en ( in_valid && !ctrl_last ), // no point in writing on last pass
.write_addr ( ctrl_addr ),
.write_data ( cw_buf_data ),
.read_clk ( clk ),
.read_en ( in_valid && !ctrl_first ), // no point in reading on first pass
.read_addr ( ctrl_addr ),
.read_data ( c3_buf_data )
);
// ** output **
dlsc_pipedelay_rst #(
.DATA ( 1 ),
.DELAY ( CYCLE_CW ),
.RESET ( 1'b0 )
) dlsc_pipedelay_valid_inst_out_valid (
.clk ( clk ),
.rst ( rst ),
.in_data ( in_valid && ctrl_last ),
.out_data ( out_valid )
);
endmodule
|