tfv: inline the signed multiply

gr-sim/tfv_flying_6502.c

@@ -333,11 +333,18 @@ static void init_table(void) {
 
 static unsigned int product[4];
 
-static int fixed_mul_unsigned(
+static void fixed_mul(unsigned char x_i, unsigned char x_f,
-		unsigned char x_i, unsigned char x_f,
 		unsigned char y_i, unsigned char y_f,
 		unsigned char *z_i, unsigned char *z_f,
-			int debug, int reuse) {
+		int reuse) {
+
+	int debug=0;
+
+//static int fixed_mul_unsigned(
+//		unsigned char x_i, unsigned char x_f,
+//	 	unsigned char y_i, unsigned char y_f,
+//		unsigned char *z_i, unsigned char *z_f,
+//			int debug, int reuse) {
 
 //	<T1 * <T2 = AAaa
 //	<T1 * >T2 = BBbb
@@ -551,24 +558,24 @@ static int fixed_mul_unsigned(
 //		printf("DDdd      %02x:%02x\n",_DD,_dd);
 //	}
 
-					cycles.multiply+=6;
+//					cycles.multiply+=6;
 
-	return (product[3]<<24)|(product[2]<<16)|(product[1]<<8)|product[0];
+//	return (product[3]<<24)|(product[2]<<16)|(product[1]<<8)|product[0];
 						// rts			; 6
-}
+//}
 
 /* signed */
-static void fixed_mul(unsigned char x_i, unsigned char x_f,
+//static void fixed_mul(unsigned char x_i, unsigned char x_f,
-		unsigned char y_i, unsigned char y_f,
+//		unsigned char y_i, unsigned char y_f,
-		unsigned char *z_i, unsigned char *z_f, int reuse) {
+//		unsigned char *z_i, unsigned char *z_f, int reuse) {
 
-	int a,c;
 
-	fixed_mul_unsigned(x_i,x_f,y_i,y_f,z_i,z_f,0,reuse);
+
-					// jsr multiply_16bit_unsigned	; 6
+//	fixed_mul_unsigned(x_i,x_f,y_i,y_f,z_i,z_f,0,reuse);
+					// jsr multiply_16bit_unsigned	;
 
 	a=(x_i&0xff);			// lda T1+1			; 3
-					cycles.multiply+=12;
+					cycles.multiply+=6;
 	if ((a&0x80)==0) goto x_positive;	// bpl :+		;^3/2nt
 
 	c=1;				// sec				; 2

gr-sim/tfv_multiply.c

@@ -719,10 +719,15 @@ static void init_table(void) {
 
 static unsigned int product[4];
 
-int fixed_mul_unsigned(int x_i,int x_f,
+int fixed_mul(int x_i,int x_f,
 		int y_i, int y_f,
+		int *z_i, int *z_f,
 			int debug) {
 
+//int fixed_mul_unsigned(int x_i,int x_f,
+//		int y_i, int y_f,
+//			int debug) {
+
 //	<T1 * <T2 = AAaa
 //	<T1 * >T2 = BBbb
 //	>T1 * <T2 = CCcc
@@ -935,20 +940,20 @@ int fixed_mul_unsigned(int x_i,int x_f,
 //		printf("DDdd      %02x:%02x\n",_DD,_dd);
 //	}
 
-	return 0;
+//	return 0;
 //(product[3]<<24)|(product[2]<<16)|(product[1]<<8)|product[0];
 						// rts			; 6
-}
+//}
 
 /* signed */
-int fixed_mul(int x_i,int x_f,
+//int fixed_mul(int x_i,int x_f,
-		int y_i, int y_f,
+//		int y_i, int y_f,
-		int *z_i, int *z_f,
+//		int *z_i, int *z_f,
-			int debug) {
+//			int debug) {
 
-	int a,c;
+//	int a,c;
 
-	fixed_mul_unsigned(x_i,x_f,y_i,y_f,debug);
+//	fixed_mul_unsigned(x_i,x_f,y_i,y_f,debug);
 					// jsr multiply_16bit_unsigned	; 6
 
 	a=(x_i&0xff);			// lda T1+1			; 3

tfv/OPTIMIZATION

@@ -170,5 +170,21 @@ Remove unneeded precision in the 8.8 x 8.8 fixed point multiply
 	Total =                 187,189
 	Frame Rate = 5.34 fps
 
+In-line unsigned multiply inside of signed multiply (save 12 cycles)
+	Multiplying 1.0 * 2.0 = 2.0, took 198 cycles
+	Multiplying ff.ff * ff.ff = 0.0, took 218 cycles
+
+	Cycles: flying=             187
+	Cycles: getkey=              46
+	Cycles: page_flip=           26
+	Cycles: multiply=        40,888
+	Cycles: mode7=          118,034
+	Cycles: lookup_map=      22,747
+	Cycles: put_sprite=       2,561
+	================================
+	Total =                 184,489
+	Frame Rate = 5.42 fps
+
+
 Each cycle removed from inner X loop saves
 	32*40=1280 cycles

tfv/tfv_multiply.s

@@ -114,7 +114,9 @@ second_table:
 ;	 ----------
 ;	   RESULT
 
-fixed_16x16_mul_unsigned:
+;fixed_16x16_mul_unsigned:
+
+multiply:
 
 	bcc	num1_same_as_last_time					; 2nt/3
 
@@ -290,16 +292,16 @@ dd_no_carry2:
 ;	*z_i=product[1];
 ;	*z_f=product[0];
 
-	rts								; 6
+;	rts								; 6
 
 
 	;=================
 	; Signed multiply
 	;=================
 
-multiply:
+;multiply:
 
-	jsr fixed_16x16_mul_unsigned					; 6
+;	jsr fixed_16x16_mul_unsigned					; 6
 
 	lda	NUM1H		; x_i					; 3
 								;===========