mirror of https://github.com/marqs85/ossc.git
SL Multiplication:
- Hybrid value based on Y (Y calculated according to YCoCg appr.) - Use 8bit input as p-factor
This commit is contained in:
borti4938
parent
3154a83847
commit
b10d7f3762
|
|
@ -44,7 +44,7 @@ module lpm_mult_4_sl (
|
|||
|
||||
input clock;
|
||||
input [7:0] dataa;
|
||||
input [3:0] datab;
|
||||
input [7:0] datab;
|
||||
output [7:0] result;
|
||||
|
||||
wire [7:0] sub_wire0;
|
||||
|
|
@ -65,7 +65,7 @@ module lpm_mult_4_sl (
|
|||
lpm_mult_component.lpm_representation = "UNSIGNED",
|
||||
lpm_mult_component.lpm_type = "LPM_MULT",
|
||||
lpm_mult_component.lpm_widtha = 8,
|
||||
lpm_mult_component.lpm_widthb = 4,
|
||||
lpm_mult_component.lpm_widthb = 8,
|
||||
lpm_mult_component.lpm_widthp = 8;
|
||||
|
||||
|
||||
|
|
@ -85,7 +85,7 @@ endmodule
|
|||
// Retrieval info: PRIVATE: USE_MULT NUMERIC "1"
|
||||
// Retrieval info: PRIVATE: ValidConstant NUMERIC "0"
|
||||
// Retrieval info: PRIVATE: WidthA NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: WidthB NUMERIC "4"
|
||||
// Retrieval info: PRIVATE: WidthB NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: WidthP NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: aclr NUMERIC "0"
|
||||
// Retrieval info: PRIVATE: clken NUMERIC "0"
|
||||
|
|
@ -97,15 +97,15 @@ endmodule
|
|||
// Retrieval info: CONSTANT: LPM_REPRESENTATION STRING "UNSIGNED"
|
||||
// Retrieval info: CONSTANT: LPM_TYPE STRING "LPM_MULT"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHA NUMERIC "8"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHB NUMERIC "4"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHB NUMERIC "8"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHP NUMERIC "8"
|
||||
// Retrieval info: USED_PORT: clock 0 0 0 0 INPUT NODEFVAL "clock"
|
||||
// Retrieval info: USED_PORT: dataa 0 0 8 0 INPUT NODEFVAL "dataa[7..0]"
|
||||
// Retrieval info: USED_PORT: datab 0 0 4 0 INPUT NODEFVAL "datab[3..0]"
|
||||
// Retrieval info: USED_PORT: datab 0 0 8 0 INPUT NODEFVAL "datab[7..0]"
|
||||
// Retrieval info: USED_PORT: result 0 0 8 0 OUTPUT NODEFVAL "result[7..0]"
|
||||
// Retrieval info: CONNECT: @clock 0 0 0 0 clock 0 0 0 0
|
||||
// Retrieval info: CONNECT: @dataa 0 0 8 0 dataa 0 0 8 0
|
||||
// Retrieval info: CONNECT: @datab 0 0 4 0 datab 0 0 4 0
|
||||
// Retrieval info: CONNECT: @datab 0 0 8 0 datab 0 0 8 0
|
||||
// Retrieval info: CONNECT: result 0 0 8 0 @result 0 0 8 0
|
||||
// Retrieval info: GEN_FILE: TYPE_NORMAL lpm_mult_4_sl.v TRUE
|
||||
// Retrieval info: GEN_FILE: TYPE_NORMAL lpm_mult_4_sl.inc FALSE
|
||||
|
|
|
|||
|
|
@ -39,7 +39,7 @@ module lpm_mult_4_sl (
|
|||
|
||||
input clock;
|
||||
input [7:0] dataa;
|
||||
input [3:0] datab;
|
||||
input [7:0] datab;
|
||||
output [7:0] result;
|
||||
|
||||
endmodule
|
||||
|
|
@ -58,7 +58,7 @@ endmodule
|
|||
// Retrieval info: PRIVATE: USE_MULT NUMERIC "1"
|
||||
// Retrieval info: PRIVATE: ValidConstant NUMERIC "0"
|
||||
// Retrieval info: PRIVATE: WidthA NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: WidthB NUMERIC "4"
|
||||
// Retrieval info: PRIVATE: WidthB NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: WidthP NUMERIC "8"
|
||||
// Retrieval info: PRIVATE: aclr NUMERIC "0"
|
||||
// Retrieval info: PRIVATE: clken NUMERIC "0"
|
||||
|
|
@ -70,15 +70,15 @@ endmodule
|
|||
// Retrieval info: CONSTANT: LPM_REPRESENTATION STRING "UNSIGNED"
|
||||
// Retrieval info: CONSTANT: LPM_TYPE STRING "LPM_MULT"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHA NUMERIC "8"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHB NUMERIC "4"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHB NUMERIC "8"
|
||||
// Retrieval info: CONSTANT: LPM_WIDTHP NUMERIC "8"
|
||||
// Retrieval info: USED_PORT: clock 0 0 0 0 INPUT NODEFVAL "clock"
|
||||
// Retrieval info: USED_PORT: dataa 0 0 8 0 INPUT NODEFVAL "dataa[7..0]"
|
||||
// Retrieval info: USED_PORT: datab 0 0 4 0 INPUT NODEFVAL "datab[3..0]"
|
||||
// Retrieval info: USED_PORT: datab 0 0 8 0 INPUT NODEFVAL "datab[7..0]"
|
||||
// Retrieval info: USED_PORT: result 0 0 8 0 OUTPUT NODEFVAL "result[7..0]"
|
||||
// Retrieval info: CONNECT: @clock 0 0 0 0 clock 0 0 0 0
|
||||
// Retrieval info: CONNECT: @dataa 0 0 8 0 dataa 0 0 8 0
|
||||
// Retrieval info: CONNECT: @datab 0 0 4 0 datab 0 0 4 0
|
||||
// Retrieval info: CONNECT: @datab 0 0 8 0 datab 0 0 8 0
|
||||
// Retrieval info: CONNECT: result 0 0 8 0 @result 0 0 8 0
|
||||
// Retrieval info: GEN_FILE: TYPE_NORMAL lpm_mult_4_sl.v TRUE
|
||||
// Retrieval info: GEN_FILE: TYPE_NORMAL lpm_mult_4_sl.inc FALSE
|
||||
|
|
|
|||
|
|
@ -185,34 +185,35 @@ assign pclk_lock = {pclk_2x_lock, pclk_3x_lock};
|
|||
|
||||
|
||||
//Scanline generation
|
||||
reg [7:0] R_sl_hybrid_ref, G_sl_hybrid_ref, B_sl_hybrid_ref;
|
||||
reg [7:0] R_sl_str, G_sl_str, B_sl_str;
|
||||
reg [8:0] Y_rb_tmp;
|
||||
reg [9:0] Y;
|
||||
reg [7:0] Y_sl_hybrid_ref, R_sl_hybrid_ref, G_sl_hybrid_ref, B_sl_hybrid_ref;
|
||||
reg [7:0] Y_sl_str, R_sl_str, G_sl_str, B_sl_str;
|
||||
|
||||
reg [7:0] R_sl_sub, G_sl_sub, B_sl_sub;
|
||||
wire [7:0] R_sl_mult, G_sl_mult, B_sl_mult;
|
||||
|
||||
lpm_mult_4_sl R_sl_mult_u
|
||||
(
|
||||
.clock(pclk_act),
|
||||
.dataa(R_pp7),
|
||||
.datab(~R_sl_str[7:4]),
|
||||
.datab(~Y_sl_str),
|
||||
.result(R_sl_mult)
|
||||
);
|
||||
lpm_mult_4_sl G_sl_mult_u
|
||||
(
|
||||
.clock(pclk_act),
|
||||
.dataa(G_pp7),
|
||||
.datab(~G_sl_str[7:4]),
|
||||
.datab(~Y_sl_str),
|
||||
.result(G_sl_mult)
|
||||
);
|
||||
lpm_mult_4_sl B_sl_mult_u
|
||||
(
|
||||
.clock(pclk_act),
|
||||
.dataa(B_pp7),
|
||||
.datab(~B_sl_str[7:4]),
|
||||
.datab(~Y_sl_str),
|
||||
.result(B_sl_mult)
|
||||
);
|
||||
|
||||
reg [7:0] R_sl_lin, G_sl_lin, B_sl_lin;
|
||||
reg draw_sl;
|
||||
|
||||
//LT box / border generation
|
||||
|
|
@ -506,6 +507,8 @@ begin
|
|||
R_prev_pp4 <= R_prev_pp3;
|
||||
G_prev_pp4 <= G_prev_pp3;
|
||||
B_prev_pp4 <= B_prev_pp3;
|
||||
// calculate Y step 1/2
|
||||
Y_rb_tmp <= {1'b0,R_pp3} + {1'b0,B_pp3};
|
||||
// Reverse LPF step2
|
||||
R_diff_pp4 <= (R_diff_pp3 * X_REV_LPF_STR);
|
||||
G_diff_pp4 <= (G_diff_pp3 * X_REV_LPF_STR);
|
||||
|
|
@ -514,6 +517,8 @@ begin
|
|||
R_pp5 <= apply_reverse_lpf(X_REV_LPF_ENABLE, R_pp4, R_prev_pp4, R_diff_pp4);
|
||||
G_pp5 <= apply_reverse_lpf(X_REV_LPF_ENABLE, G_pp4, G_prev_pp4, G_diff_pp4);
|
||||
B_pp5 <= apply_reverse_lpf(X_REV_LPF_ENABLE, B_pp4, B_prev_pp4, B_diff_pp4);
|
||||
// calculate Y step 2/2
|
||||
Y <= {1'b0,Y_rb_tmp} + {1'b0,G_pp4,1'b0};
|
||||
HSYNC_pp5 <= HSYNC_pp4;
|
||||
VSYNC_pp5 <= VSYNC_pp4;
|
||||
DE_pp5 <= DE_pp4;
|
||||
|
|
@ -525,7 +530,11 @@ begin
|
|||
R_pp6 <= R_pp5;
|
||||
G_pp6 <= G_pp5;
|
||||
B_pp6 <= B_pp5;
|
||||
// modify scanline strength for linear method step 1
|
||||
// modify scanline strength for linear method step 1/2
|
||||
Y_sl_hybrid_ref <= (X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_HIGH) ? (Y[9:2] - (Y[9:2] >> 3)) :
|
||||
(X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_MED) ? ((Y[9:2] >> 1) + (Y[9:2] >> 3)) :
|
||||
(X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_LOW) ? (Y[9:2] >> 1) :
|
||||
8'h0;
|
||||
R_sl_hybrid_ref <= (X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_HIGH) ? (R_pp5 - (R_pp5 >> 3)) :
|
||||
(X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_MED) ? ((R_pp5 >> 1) + (R_pp5 >> 3)) :
|
||||
(X_SCANLINESTR_HYBR_CONTR == `SCANLINES_HYBR_CONTR_LOW) ? (R_pp5 >> 1) :
|
||||
|
|
@ -549,7 +558,8 @@ begin
|
|||
R_pp7 <= R_pp6;
|
||||
G_pp7 <= G_pp6;
|
||||
B_pp7 <= B_pp6;
|
||||
// modify scanline strength for linear method step 2
|
||||
// modify scanline strength for linear method step 2/2
|
||||
Y_sl_str <= (X_SCANLINESTR > Y_sl_hybrid_ref) ? X_SCANLINESTR - Y_sl_hybrid_ref : 8'h0;
|
||||
R_sl_str <= (X_SCANLINESTR > R_sl_hybrid_ref) ? X_SCANLINESTR - R_sl_hybrid_ref : 8'h0;
|
||||
G_sl_str <= (X_SCANLINESTR > G_sl_hybrid_ref) ? X_SCANLINESTR - G_sl_hybrid_ref : 8'h0;
|
||||
B_sl_str <= (X_SCANLINESTR > B_sl_hybrid_ref) ? X_SCANLINESTR - B_sl_hybrid_ref : 8'h0;
|
||||
|
|
@ -564,10 +574,11 @@ begin
|
|||
R_pp8 <= R_pp7;
|
||||
G_pp8 <= G_pp7;
|
||||
B_pp8 <= B_pp7;
|
||||
// R_sl_mult, G_sl_mult and B_sl_mult are registered output of IP blocks (line 191-213)
|
||||
R_sl_lin <= (R_pp7 > R_sl_str) ? (R_pp7-R_sl_str) : 8'h00;
|
||||
G_sl_lin <= (G_pp7 > G_sl_str) ? (G_pp7-G_sl_str) : 8'h00;
|
||||
B_sl_lin <= (B_pp7 > B_sl_str) ? (B_pp7-B_sl_str) : 8'h00;
|
||||
// R_sl_mult, G_sl_mult and B_sl_mult are registered output of IP blocks (line 194-215)
|
||||
// perform subtraction
|
||||
R_sl_sub <= (R_pp7 > R_sl_str) ? (R_pp7-R_sl_str) : 8'h00;
|
||||
G_sl_sub <= (G_pp7 > G_sl_str) ? (G_pp7-G_sl_str) : 8'h00;
|
||||
B_sl_sub <= (B_pp7 > B_sl_str) ? (B_pp7-B_sl_str) : 8'h00;
|
||||
draw_sl <= |{(V_SCANLINEMODE == `SCANLINES_H) && (V_SCANLINEID & (5'h1<<line_id_pp7)),
|
||||
(V_SCANLINEMODE == `SCANLINES_V) && (5'h0 == col_id_pp7),
|
||||
(V_SCANLINEMODE == `SCANLINES_ALT) && (V_SCANLINEID & (5'h1<<(line_id_pp7^FID_1x)))};
|
||||
|
|
@ -577,9 +588,9 @@ begin
|
|||
border_enable_pp8 <= border_enable_pp7;
|
||||
lt_box_enable_pp8 <= lt_box_enable_pp7;
|
||||
|
||||
R_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? R_sl_lin : R_sl_mult) : R_pp8;
|
||||
G_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? G_sl_lin : G_sl_mult) : G_pp8;
|
||||
B_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? B_sl_lin : B_sl_mult) : B_pp8;
|
||||
R_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? R_sl_sub : R_sl_mult) : R_pp8;
|
||||
G_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? G_sl_sub : G_sl_mult) : G_pp8;
|
||||
B_pp9 <= draw_sl ? (X_SCANLINESTR_METHOD ? B_sl_sub : B_sl_mult) : B_pp8;
|
||||
HSYNC_pp9 <= HSYNC_pp8;
|
||||
VSYNC_pp9 <= VSYNC_pp8;
|
||||
DE_pp9 <= DE_pp8;
|
||||
|
|
|
|||
Loading…
Reference in New Issue