6 //13 VIDEO PIXEL OPERATIONS
\r
9 //Dreg = ALIGN8 ( Dreg, Dreg ) ; /* overlay 1 byte (b) */
\r
10 R0 = ALIGN8(R0, R0);
\r
11 R0 = ALIGN8(R0, R1);
\r
12 R0 = ALIGN8(R1, R0);
\r
13 R0 = ALIGN8(R1, R1);
\r
14 R0 = ALIGN8(R1, R2);
\r
15 R3 = ALIGN8(R4, R5);
\r
16 R6 = ALIGN8(R7, R0);
\r
17 R1 = ALIGN8(R2, R3);
\r
18 R4 = ALIGN8(R5, R6);
\r
19 R7 = ALIGN8(R0, R1);
\r
20 R2 = ALIGN8(R3, R4);
\r
21 R5 = ALIGN8(R6, R7);
\r
23 //Dreg = ALIGN16 ( Dreg, Dreg ) ; /* overlay 2 bytes (b) */
\r
24 R0 = ALIGN16(R0, R0);
\r
25 R0 = ALIGN16(R0, R1);
\r
26 R0 = ALIGN16(R1, R0);
\r
27 R0 = ALIGN16(R1, R1);
\r
28 R0 = ALIGN16(R1, R2);
\r
29 R3 = ALIGN16(R4, R5);
\r
30 R6 = ALIGN16(R7, R0);
\r
31 R1 = ALIGN16(R2, R3);
\r
32 R4 = ALIGN16(R5, R6);
\r
33 R7 = ALIGN16(R0, R1);
\r
34 R2 = ALIGN16(R3, R4);
\r
35 R5 = ALIGN16(R6, R7);
\r
37 //Dreg = ALIGN24 ( Dreg, Dreg ) ; /* overlay 3 bytes (b) */
\r
38 R0 = ALIGN24(R0, R0);
\r
39 R0 = ALIGN24(R0, R1);
\r
40 R0 = ALIGN24(R1, R0);
\r
41 R0 = ALIGN24(R1, R1);
\r
42 R0 = ALIGN24(R1, R2);
\r
43 R3 = ALIGN24(R4, R5);
\r
44 R6 = ALIGN24(R7, R0);
\r
45 R1 = ALIGN24(R2, R3);
\r
46 R4 = ALIGN24(R5, R6);
\r
47 R7 = ALIGN24(R0, R1);
\r
48 R2 = ALIGN24(R3, R4);
\r
49 R5 = ALIGN24(R6, R7);
\r
51 DISALGNEXCPT ; /* (b) */
\r
53 /* forward byte order operands */
\r
54 //Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (LO) ; /* sum into low bytes (b) */
\r
55 //Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (HI) ; /* sum into high bytes (b) */
\r
56 /* reverse byte order operands */
\r
57 //Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (LO, R) ; /* sum into low bytes (b) */
\r
58 //Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (HI, R) ; /* sum into high bytes (b) */
\r
60 r0 = byteop3p (r1:0, r3:2) (lo) ;
\r
61 r1 = byteop3p (r1:0, r3:2) (hi) ;
\r
62 r2 = byteop3p (r1:0, r3:2) (lo, r) ;
\r
63 r3 = byteop3p (r1:0, r3:2) (hi, r) ;
\r
64 r4 = byteop3p (r3:2, r1:0) (lo) ;
\r
65 r5 = byteop3p (r3:2, r1:0) (hi) ;
\r
66 r6 = byteop3p (r3:2, r1:0) (lo, r) ;
\r
67 r7 = byteop3p (r3:2, r1:0) (hi, r) ;
\r
69 //Dreg = A1.L + A1.H, Dreg = A0.L + A0.H ; /* (b) */
\r
71 R0 = A1.L + A1.H, R0= A0.L + A0.H ;
\r
72 R0 = A1.L + A1.H, R1= A0.L + A0.H ;
\r
73 R2 = A1.L + A1.H, R3= A0.L + A0.H ;
\r
74 R4 = A1.L + A1.H, R5= A0.L + A0.H ;
\r
75 R6 = A1.L + A1.H, R7= A0.L + A0.H ;
\r
77 /* forward byte order operands */
\r
78 //( Dreg, Dreg ) = BYTEOP16P ( Dreg_pair, Dreg_pair ) ; /* (b) */
\r
79 (r7,r0) = BYTEOP16P ( r3:2,r1:0 ) ;
\r
80 (r1,r2) = byteop16p (r3:2,r1:0) ;
\r
81 (r0,r1) = BYTEOP16P ( r3:2,r1:0 ) ;
\r
82 (r2,r3) = byteop16p (r3:2,r1:0) ;
\r
83 (r7,r0) = BYTEOP16P (r1:0, r3:2) ;
\r
84 (r1,r2) = byteop16p (r1:0,r3:2) ;
\r
85 (r0,r1) = BYTEOP16P (r1:0, r3:2) ;
\r
86 (r2,r3) = byteop16p (r1:0,r3:2) ;
\r
88 /* reverse byte order operands */
\r
89 //( Dreg, Dreg ) = BYTEOP16P ( Dreg_pair, Dreg_pair ) (R); /* (b) */
\r
90 (r7,r0) = BYTEOP16P ( r3:2,r1:0 )(r) ;
\r
91 (r1,r2) = byteop16p (r3:2,r1:0)(r) ;
\r
92 (r0,r1) = BYTEOP16P ( r3:2,r1:0 )(r) ;
\r
93 (r2,r3) = byteop16p (r3:2,r1:0)(r) ;
\r
94 (r7,r0) = BYTEOP16P (r1:0, r3:2)(r) ;
\r
95 (r1,r2) = byteop16p (r1:0,r3:2)(r) ;
\r
96 (r0,r1) = BYTEOP16P (r1:0, r3:2)(r) ;
\r
97 (r2,r3) = byteop16p (r1:0,r3:2)(r) ;
\r
99 /* forward byte order operands */
\r
100 //Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) ; /* (b) */
\r
101 //Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (T) ; /* truncated (b)*/
\r
102 /* reverse byte order operands */
\r
103 //Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (R) ; /* (b) */
\r
104 //Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (T, R) ; /* truncated (b) */
\r
106 r3 = byteop1p (r1:0, r3:2) ;
\r
107 r3 = byteop1p (r1:0, r3:2) (r) ;
\r
108 r3 = byteop1p (r1:0, r3:2) (t) ;
\r
109 r3 = byteop1p (r1:0, r3:2) (t,r) ;
\r
111 r0 = byteop1p (r3:2,r1:0);
\r
112 r1 = byteop1p (r3:2,r1:0)(r) ;
\r
113 r2 = byteop1p (r3:2,r1:0)(t) ;
\r
114 r3 = byteop1p (r3:2,r1:0)(t,r) ;
\r
116 /* forward byte order operands */
\r
117 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDL) ;
\r
118 /* round into low bytes (b) */
\r
119 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDH) ;
\r
120 /* round into high bytes (b) */
\r
121 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TL) ;
\r
122 /* truncate into low bytes (b) */
\r
123 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TH) ;
\r
124 /* truncate into high bytes (b) */
\r
125 /* reverse byte order operands */
\r
126 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDL, R) ;
\r
127 /* round into low bytes (b) */
\r
128 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDH, R) ;
\r
129 /* round into high bytes (b) */
\r
130 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TL, R) ;
\r
131 /* truncate into low bytes (b) */
\r
132 //Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TH, R) ;
\r
133 /* truncate into high bytes (b) */
\r
135 r3 = byteop2p (r1:0, r3:2) (rndl) ;
\r
136 r3 = byteop2p (r1:0, r3:2) (rndh) ;
\r
137 r3 = byteop2p (r1:0, r3:2) (tl) ;
\r
138 r3 = byteop2p (r1:0, r3:2) (th) ;
\r
139 r3 = byteop2p (r1:0, r3:2) (rndl, r) ;
\r
140 r3 = byteop2p (r1:0, r3:2) (rndh, r) ;
\r
141 r3 = byteop2p (r1:0, r3:2) (tl, r) ;
\r
142 r3 = byteop2p (r1:0, r3:2) (th, r) ;
\r
144 r0 = byteop2p (r1:0, r3:2) (rndl) ;
\r
145 r1 = byteop2p (r1:0, r3:2) (rndh) ;
\r
146 r2 = byteop2p (r1:0, r3:2) (tl) ;
\r
147 r3 = byteop2p (r1:0, r3:2) (th) ;
\r
148 r4 = byteop2p (r1:0, r3:2) (rndl, r) ;
\r
149 r5 = byteop2p (r1:0, r3:2) (rndh, r) ;
\r
150 r6 = byteop2p (r1:0, r3:2) (tl, r) ;
\r
151 r7 = byteop2p (r1:0, r3:2) (th, r) ;
\r
153 r0 = byteop2p (r3:2, r3:2) (rndl) ;
\r
154 r1 = byteop2p (r3:2, r3:2) (rndh) ;
\r
155 r2 = byteop2p (r3:2, r3:2) (tl) ;
\r
156 r3 = byteop2p (r3:2, r3:2) (th) ;
\r
157 r4 = byteop2p (r3:2, r3:2) (rndl, r) ;
\r
158 r5 = byteop2p (r3:2, r3:2) (rndh, r) ;
\r
159 r6 = byteop2p (r3:2, r3:2) (tl, r) ;
\r
160 r7 = byteop2p (r3:2, r3:2) (th, r) ;
\r
162 //Dreg = BYTEPACK ( Dreg, Dreg ) ; /* (b) */
\r
163 r0 = bytepack (r0,r0) ;
\r
164 r1 = bytepack (r2,r3) ;
\r
165 r4 = bytepack (r5,r6) ;
\r
166 r7 = bytepack (r0,r1) ;
\r
167 r2 = bytepack (r3,r4) ;
\r
168 r5 = bytepack (r6,r7) ;
\r
170 /* forward byte order operands */
\r
171 //(Dreg, Dreg) = BYTEOP16M (Dreg_pair, Dreg_pair) ; /* (b */)
\r
172 /* reverse byte order operands */
\r
173 //(Dreg, Dreg) = BYTEOP16M (Dreg-pair, Dreg-pair) (R) ; /* (b) */
\r
175 (r1,r2)= byteop16m (r3:2,r1:0) ;
\r
176 (r1,r2)= byteop16m (r3:2,r1:0) (r) ;
\r
177 (r0,r1)= byteop16m (r3:2,r1:0) ;
\r
178 (r2,r3)= byteop16m (r3:2,r1:0) (r) ;
\r
179 (r3,r5)= byteop16m (r3:2,r1:0) ;
\r
180 (r6,r7)= byteop16m (r3:2,r1:0) (r) ;
\r
182 (r1,r2)= byteop16m (r1:0,r1:0) ;
\r
183 (r1,r2)= byteop16m (r1:0,r1:0) (r) ;
\r
184 (r0,r1)= byteop16m (r1:0,r1:0) ;
\r
185 (r2,r3)= byteop16m (r1:0,r1:0) (r) ;
\r
186 (r3,r5)= byteop16m (r1:0,r1:0) ;
\r
187 (r6,r7)= byteop16m (r1:0,r1:0) (r) ;
\r
189 (r1,r2)= byteop16m (r1:0,r3:2) ;
\r
190 (r1,r2)= byteop16m (r1:0,r3:2) (r) ;
\r
191 (r0,r1)= byteop16m (r1:0,r3:2) ;
\r
192 (r2,r3)= byteop16m (r1:0,r3:2) (r) ;
\r
193 (r3,r5)= byteop16m (r1:0,r3:2) ;
\r
194 (r6,r7)= byteop16m (r1:0,r3:2) (r) ;
\r
196 (r1,r2)= byteop16m (r3:2,r3:2) ;
\r
197 (r1,r2)= byteop16m (r3:2,r3:2) (r) ;
\r
198 (r0,r1)= byteop16m (r3:2,r3:2) ;
\r
199 (r2,r3)= byteop16m (r3:2,r3:2) (r) ;
\r
200 (r3,r5)= byteop16m (r3:2,r3:2) ;
\r
201 (r6,r7)= byteop16m (r3:2,r3:2) (r) ;
\r
203 //SAA (Dreg_pair, Dreg_pair) ; /* forward byte order operands (b) */
\r
204 //SAA (Dreg_pair, Dreg_pair) (R) ; /* reverse byte order operands (b) */
\r
206 saa(r1:0, r3:2) || r0 = [i0++] || r2 = [i1++] ; /* parallel fill instructions */
\r
207 saa (r1:0, r3:2) (R) || r1 = [i0++] || r3 = [i1++] ; /* reverse, parallel fill instructions */
\r
208 saa (r1:0, r3:2) ; /* last SAA in a loop, no more fill required */
\r
210 //( Dreg , Dreg ) = BYTEUNPACK Dreg_pair ; /* (b) */
\r
211 //( Dreg , Dreg ) = BYTEUNPACK Dreg_pair (R) ; /* reverse source order (b) */
\r
213 (r6,r5) = byteunpack r1:0 ; /* non-reversing sources */
\r
214 (r6,r5) = byteunpack r1:0 (R) ; /* reversing sources case */
\r
215 (r6,r5) = byteunpack r3:2 ; /* non-reversing sources */
\r
216 (r6,r5) = byteunpack r3:2 (R) ; /* reversing sources case */
\r
217 (r0,r1) = byteunpack r1:0 ; /* non-reversing sources */
\r
218 (r2,r3) = byteunpack r1:0 (R) ; /* reversing sources case */
\r
219 (r4,r5) = byteunpack r3:2 ; /* non-reversing sources */
\r
220 (r6,r7) = byteunpack r3:2 (R) ; /* reversing sources case */
\r