Skip to content

X
x393
Commit 1b250b27 authored by Andrey Filippov's avatar Andrey Filippov
Browse files
Options

extracted a separate module, 2 alternative implementations

parent 0ec5957a
Hide whitespace changes
Inline Side-by-side
Showing with 246 additions and 0 deletions
dsp/mclt_baeyer_fold_accum.v 0 → 100644
View file @ 1b250b27
/*!
* <b>Module:</b> mclt_baeyer_fold_accum
* @file mclt_baeyer_fold_accum.v
* @date 2017-12-23
* @author eyesis
*
* @brief Alternative implementation of CC and CS folded data accumulators
*
* @copyright Copyright (c) 2017 <set up in Preferences-Verilog/VHDL Editor-Templates> .
*
* <b>License </b>
*
* mclt_baeyer_fold_accum.v 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.
*
* mclt_baeyer_fold_accum.v 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/> .
*
* Additional permission under GNU GPL version 3 section 7:
* If you modify this Program, or any covered work, by linking or combining it
* with independent modules provided by the FPGA vendor only (this permission
* does not extend to any 3-rd party modules, "soft cores" or macros) under
* different license terms solely for the purpose of generating binary "bitstream"
* files and/or simulating the code, the copyright holders of this Program give
* you the right to distribute the covered work without those independent modules
* as long as the source code for them is available from the FPGA vendor free of
* charge, and there is no dependence on any encrypted modules for simulating of
* the combined code. This permission applies to you if the distributed code
* contains all the components and scripts required to completely simulate it
* with at least one of the Free Software programs.
*/
`timescale 1ns/1ps
module mclt_baeyer_fold_accum # (
parameter PIXEL_WIDTH = 16, // input pixel width (unsigned)
parameter WND_WIDTH = 18, // input pixel width (unsigned)
parameter DTT_IN_WIDTH = 25, // bits in DTT input
parameter DSP_B_WIDTH = 18, // signed, output from sin/cos ROM
parameter DSP_A_WIDTH = 25,
parameter DSP_P_WIDTH = 48
)(
input clk,
input rst,
// input [5:0] phases,
input pre_phase,
input signed [PIXEL_WIDTH-1:0] pix_d, //!< pixel data
input [1:0] pix_sgn, //!< bit 0: sign to add to dtt-cc input, bit 1: sign to add to dtt-cs input
input signed [WND_WIDTH-1:0] window,
input var_pre2_first,
output signed [DTT_IN_WIDTH-1:0] dtt_in,
output reg dtt_in_dv
);
reg [PIXEL_WIDTH-1:0] pix_dr; // only for mpy to match dsp
reg var_pre_first;
reg var_first;
reg signed [WND_WIDTH-1:0] window_r;
reg signed [PIXEL_WIDTH-1:0] pix_d_r; // registered pixel data (to be absorbed by MPY)
wire [ 1:0] pix_sgn_d;
reg [ 1:0] pix_sgn_r;
reg var_last;
reg signed [PIXEL_WIDTH + WND_WIDTH - 1:0] pix_wnd_r; // MSB not used: positive[PIXEL_WIDTH]*positive[WND_WIDTH]->positive[PIXEL_WIDTH+WND_WIDTH-1]
reg signed [DTT_IN_WIDTH-1:0] pix_wnd_r2; // pixels (positive) multiplied by window(positive), two MSBs == 2'b0 to prevent overflow
// rounding
// wire signed [DTT_IN_WIDTH-3:0] pix_wnd_r2_w = pix_wnd_r[PIXEL_WIDTH + WND_WIDTH - 2 -: DTT_IN_WIDTH - 2]
wire signed [DTT_IN_WIDTH-2:0] pix_wnd_r2_w = pix_wnd_r[PIXEL_WIDTH + WND_WIDTH - 2 -: DTT_IN_WIDTH - 1]
`ifdef ROUND
// + pix_wnd_r[PIXEL_WIDTH + WND_WIDTH -DTT_IN_WIDTH]
+ pix_wnd_r[PIXEL_WIDTH + WND_WIDTH -DTT_IN_WIDTH -1]
`endif
;
reg [6:0] phases;
reg signed [DTT_IN_WIDTH-1:0] data_cc_r;
reg signed [DTT_IN_WIDTH-1:0] data_sc_r;
reg signed [DTT_IN_WIDTH-1:0] data_sc_r2; // data_sc_r delayed by 1 cycle
reg signed [DTT_IN_WIDTH-1:0] data_dtt_in; // multiplexed DTT input data
reg mode_mux;
assign dtt_in = data_dtt_in;
always @ (posedge clk) begin
phases <= {phases[5:0], pre_phase};
if (rst) dtt_in_dv <= 0;
else dtt_in_dv <= phases[6];
pix_dr <= pix_d;
if (phases[1]) begin
pix_d_r <= pix_dr;
window_r <= window;
end
if (phases[2]) pix_wnd_r <= pix_d_r * window_r; // 1 MSB is extra
if (phases[2]) begin
var_pre_first <= var_pre2_first;
end
if (phases[3]) begin
pix_wnd_r2 <= {pix_wnd_r2_w[DTT_IN_WIDTH-2],pix_wnd_r2_w};
pix_sgn_r <= pix_sgn_d;
var_first <= var_pre_first;
end
var_last <= var_first & phases[4];
if (phases[4]) begin
data_cc_r <= (var_first ? {DTT_IN_WIDTH{1'b0}} : data_cc_r) + (pix_sgn_r[0]?(-pix_wnd_r2):pix_wnd_r2) ;
data_sc_r <= (var_first ? {DTT_IN_WIDTH{1'b0}} : data_sc_r) + (pix_sgn_r[1]?(-pix_wnd_r2):pix_wnd_r2) ;
data_sc_r2 <= data_sc_r;
end
if (phases[5]) data_sc_r2 <= data_sc_r;
if (var_last) mode_mux <= 0;
else if (phases[6]) mode_mux <= mode_mux + 1;
if (phases[6]) case (mode_mux)
1'b0: data_dtt_in <= data_cc_r;
1'b1: data_dtt_in <= data_sc_r2;
endcase
end
dly_var #(
.WIDTH(2),
.DLY_WIDTH(4)
) dly_pix_sgn_i (
.clk (clk), // input
.rst (rst), // input
.dly (4'h1), // input[3:0]
.din (pix_sgn), // input[0:0]
.dout (pix_sgn_d) // output[0:0]
);
// wire sub_a1=0,sub_a2=0;
wire neg_m1, neg_m2;
// wire accum1= !var_pre_first;
// wire accum2= !var_first;
wire accum1= !var_pre2_first;
wire accum2= !var_pre_first;
wire [DSP_P_WIDTH-1:0] pout1;
wire [DSP_P_WIDTH-1:0] pout2;
wire signed [DTT_IN_WIDTH-1:0] dtt_in_dsp_w = (var_last ?
pout1 [PIXEL_WIDTH + WND_WIDTH - 1 -: DTT_IN_WIDTH] :
pout2 [PIXEL_WIDTH + WND_WIDTH - 1 -: DTT_IN_WIDTH])
`ifdef ROUND
+ (var_last ?
pout1 [PIXEL_WIDTH + WND_WIDTH -DTT_IN_WIDTH -1] :
pout2 [PIXEL_WIDTH + WND_WIDTH -DTT_IN_WIDTH -1)
`endif
;
reg signed [DTT_IN_WIDTH-1:0] dtt_in_dsp;
// wire signed [DTT_IN_WIDTH-2:0] pix_wnd_r2_w = pix_wnd_r[PIXEL_WIDTH + WND_WIDTH - 2 -: DTT_IN_WIDTH - 1]
always @ (posedge clk) begin
if (phases[5]) dtt_in_dsp <= dtt_in_dsp_w;
end
dsp_ma_preadd #(
.B_WIDTH(DSP_B_WIDTH),
.A_WIDTH(DSP_A_WIDTH),
.P_WIDTH(DSP_P_WIDTH),
.AREG(1),
.BREG(1)
) dsp_fold_cc_i (
.clk (clk), // input
.rst (rst), // input
.bin (window), // input[17:0] signed
.ceb1 (1'b0), // input
.ceb2 (phases[1]), // input
.selb (1'b1), // input
.ain ({{(DSP_A_WIDTH-PIXEL_WIDTH){pix_d[PIXEL_WIDTH-1]}},pix_d}), // input[24:0] signed
.cea1 (1'b0), // input
.cea2 (phases[0]), // input
.din (25'b0), // input[24:0] signed
.ced (1'b0), // input
.cead (phases[1]), // input
.sela (1'b1), // input
.en_a (1'b1), // input
.en_d (1'b0), // input
.sub_a (1'b0), // input
.neg_m (neg_m1), // input
.accum (accum1), // input
.pout (pout1) // output[47:0] signed
);
dsp_ma_preadd #(
.B_WIDTH(DSP_B_WIDTH),
.A_WIDTH(DSP_A_WIDTH),
.P_WIDTH(DSP_P_WIDTH),
.AREG(2), // delayed by 1
.BREG(2)
) dsp_fold_cs_i (
.clk (clk), // input
.rst (rst), // input
.bin (window), // input[17:0] signed
.ceb1 (phases[1]), // input
.ceb2 (phases[2]), // input
.selb (1'b1), // input
.ain ({{(DSP_A_WIDTH-PIXEL_WIDTH){pix_d[PIXEL_WIDTH-1]}},pix_d}), // input[24:0] signed
.cea1 (phases[0]), // input
.cea2 (phases[1]), // input
.din (25'b0), // input[24:0] signed
.ced (1'b0), // input
.cead (phases[2]), // input
.sela (1'b1), // input
.en_a (1'b1), // input
.en_d (1'b0), // input
.sub_a (1'b0), // input
.neg_m (neg_m2), // input
.accum (accum2), // input
.pout (pout2) // output[47:0] signed
);
dly_var #(
.WIDTH(1),
.DLY_WIDTH(4)
) dly_neg_m1_i (
.clk (clk), // input
.rst (rst), // input
// .dly (4'h1), // input[3:0]
.dly (4'h0), // input[3:0]
.din (pix_sgn[0]), // input[0:0]
.dout (neg_m1) // output[0:0]
);
dly_var #(
.WIDTH(1),
.DLY_WIDTH(4)
) dly_neg_m2_i (
.clk (clk), // input
.rst (rst), // input
// .dly (4'h2), // input[3:0]
.dly (4'h1), // input[3:0]
.din (pix_sgn[1]), // input[0:0]
.dout (neg_m2) // output[0:0]
);
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