1
2.EXTERN MY_LABEL2;
3.section .text;
4
5//
6//13 VIDEO PIXEL OPERATIONS
7//
8
9//Dreg = ALIGN8 ( Dreg, Dreg ) ; /* overlay 1 byte (b) */
10R0 = ALIGN8(R0, R0);
11R0 = ALIGN8(R0, R1);
12R0 = ALIGN8(R1, R0);
13R0 = ALIGN8(R1, R1);
14R0 = ALIGN8(R1, R2);
15R3 = ALIGN8(R4, R5);
16R6 = ALIGN8(R7, R0);
17R1 = ALIGN8(R2, R3);
18R4 = ALIGN8(R5, R6);
19R7 = ALIGN8(R0, R1);
20R2 = ALIGN8(R3, R4);
21R5 = ALIGN8(R6, R7);
22
23//Dreg = ALIGN16 ( Dreg, Dreg ) ; /* overlay 2 bytes (b) */
24R0 = ALIGN16(R0, R0);
25R0 = ALIGN16(R0, R1);
26R0 = ALIGN16(R1, R0);
27R0 = ALIGN16(R1, R1);
28R0 = ALIGN16(R1, R2);
29R3 = ALIGN16(R4, R5);
30R6 = ALIGN16(R7, R0);
31R1 = ALIGN16(R2, R3);
32R4 = ALIGN16(R5, R6);
33R7 = ALIGN16(R0, R1);
34R2 = ALIGN16(R3, R4);
35R5 = ALIGN16(R6, R7);
36
37//Dreg = ALIGN24 ( Dreg, Dreg ) ; /* overlay 3 bytes (b) */
38R0 = ALIGN24(R0, R0);
39R0 = ALIGN24(R0, R1);
40R0 = ALIGN24(R1, R0);
41R0 = ALIGN24(R1, R1);
42R0 = ALIGN24(R1, R2);
43R3 = ALIGN24(R4, R5);
44R6 = ALIGN24(R7, R0);
45R1 = ALIGN24(R2, R3);
46R4 = ALIGN24(R5, R6);
47R7 = ALIGN24(R0, R1);
48R2 = ALIGN24(R3, R4);
49R5 = ALIGN24(R6, R7);
50
51DISALGNEXCPT ; /* (b) */
52
53/* forward byte order operands */
54//Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (LO) ; /* sum into low bytes (b) */
55//Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (HI) ; /* sum into high bytes (b) */
56/* reverse byte order operands */
57//Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (LO, R) ; /* sum into low bytes (b) */
58//Dreg = BYTEOP3P (Dreg_pair, Dreg_pair) (HI, R) ; /* sum into high bytes (b) */
59
60r0 = byteop3p (r1:0, r3:2) (lo) ;
61r1 = byteop3p (r1:0, r3:2) (hi) ;
62r2 = byteop3p (r1:0, r3:2) (lo, r) ;
63r3 = byteop3p (r1:0, r3:2) (hi, r) ;
64r4 = byteop3p (r3:2, r1:0) (lo) ;
65r5 = byteop3p (r3:2, r1:0) (hi) ;
66r6 = byteop3p (r3:2, r1:0) (lo, r) ;
67r7 = byteop3p (r3:2, r1:0) (hi, r) ;
68
69//Dreg = A1.L + A1.H, Dreg = A0.L + A0.H ; /* (b) */
70
71R0 = A1.L + A1.H, R1= A0.L + A0.H ;
72R2 = A1.L + A1.H, R3= A0.L + A0.H ;
73R4 = A1.L + A1.H, R5= A0.L + A0.H ;
74R6 = A1.L + A1.H, R7= A0.L + A0.H ;
75
76/* forward byte order operands */
77//( Dreg, Dreg ) = BYTEOP16P ( Dreg_pair, Dreg_pair ) ; /* (b) */
78(r7,r0) = BYTEOP16P ( r3:2,r1:0 ) ;
79(r1,r2) = byteop16p (r3:2,r1:0) ;
80(r0,r1) = BYTEOP16P ( r3:2,r1:0 ) ;
81(r2,r3) = byteop16p (r3:2,r1:0) ;
82(r7,r0) = BYTEOP16P (r1:0, r3:2) ;
83(r1,r2) = byteop16p (r1:0,r3:2) ;
84(r0,r1) = BYTEOP16P (r1:0, r3:2) ;
85(r2,r3) = byteop16p (r1:0,r3:2) ;
86
87/* reverse byte order operands */
88//( Dreg, Dreg ) = BYTEOP16P ( Dreg_pair, Dreg_pair ) (R); /* (b) */
89(r7,r0) = BYTEOP16P ( r3:2,r1:0 )(r) ;
90(r1,r2) = byteop16p (r3:2,r1:0)(r) ;
91(r0,r1) = BYTEOP16P ( r3:2,r1:0 )(r) ;
92(r2,r3) = byteop16p (r3:2,r1:0)(r) ;
93(r7,r0) = BYTEOP16P (r1:0, r3:2)(r) ;
94(r1,r2) = byteop16p (r1:0,r3:2)(r) ;
95(r0,r1) = BYTEOP16P (r1:0, r3:2)(r) ;
96(r2,r3) = byteop16p (r1:0,r3:2)(r) ;
97
98/* forward byte order operands */
99//Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) ; /* (b) */
100//Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (T) ; /* truncated (b)*/
101/* reverse byte order operands */
102//Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (R) ; /* (b) */
103//Dreg = BYTEOP1P (Dreg_pair, Dreg_pair) (T, R) ; /* truncated (b) */
104
105r3 = byteop1p (r1:0, r3:2) ;
106r3 = byteop1p (r1:0, r3:2) (r) ;
107r3 = byteop1p (r1:0, r3:2) (t) ;
108r3 = byteop1p (r1:0, r3:2) (t,r) ;
109
110r0 = byteop1p (r3:2,r1:0);
111r1 = byteop1p (r3:2,r1:0)(r) ;
112r2 = byteop1p (r3:2,r1:0)(t) ;
113r3 = byteop1p (r3:2,r1:0)(t,r) ;
114
115/* forward byte order operands */
116//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDL) ;
117/* round into low bytes (b) */
118//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDH) ;
119/* round into high bytes (b) */
120//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TL) ;
121/* truncate into low bytes (b) */
122//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TH) ;
123/* truncate into high bytes (b) */
124/* reverse byte order operands */
125//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDL, R) ;
126/* round into low bytes (b) */
127//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (RNDH, R) ;
128/* round into high bytes (b) */
129//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TL, R) ;
130/* truncate into low bytes (b) */
131//Dreg = BYTEOP2P (Dreg_pair, Dreg_pair) (TH, R) ;
132/* truncate into high bytes (b) */
133
134r3 = byteop2p (r1:0, r3:2) (rndl) ;
135r3 = byteop2p (r1:0, r3:2) (rndh) ;
136r3 = byteop2p (r1:0, r3:2) (tl) ;
137r3 = byteop2p (r1:0, r3:2) (th) ;
138r3 = byteop2p (r1:0, r3:2) (rndl, r) ;
139r3 = byteop2p (r1:0, r3:2) (rndh, r) ;
140r3 = byteop2p (r1:0, r3:2) (tl, r) ;
141r3 = byteop2p (r1:0, r3:2) (th, r) ;
142
143r0 = byteop2p (r1:0, r3:2) (rndl) ;
144r1 = byteop2p (r1:0, r3:2) (rndh) ;
145r2 = byteop2p (r1:0, r3:2) (tl) ;
146r3 = byteop2p (r1:0, r3:2) (th) ;
147r4 = byteop2p (r1:0, r3:2) (rndl, r) ;
148r5 = byteop2p (r1:0, r3:2) (rndh, r) ;
149r6 = byteop2p (r1:0, r3:2) (tl, r) ;
150r7 = byteop2p (r1:0, r3:2) (th, r) ;
151
152r0 = byteop2p (r3:2, r3:2) (rndl) ;
153r1 = byteop2p (r3:2, r3:2) (rndh) ;
154r2 = byteop2p (r3:2, r3:2) (tl) ;
155r3 = byteop2p (r3:2, r3:2) (th) ;
156r4 = byteop2p (r3:2, r3:2) (rndl, r) ;
157r5 = byteop2p (r3:2, r3:2) (rndh, r) ;
158r6 = byteop2p (r3:2, r3:2) (tl, r) ;
159r7 = byteop2p (r3:2, r3:2) (th, r) ;
160
161//Dreg = BYTEPACK ( Dreg, Dreg ) ; /* (b) */
162r0 = bytepack (r0,r0) ;
163r1 = bytepack (r2,r3) ;
164r4 = bytepack (r5,r6) ;
165r7 = bytepack (r0,r1) ;
166r2 = bytepack (r3,r4) ;
167r5 = bytepack (r6,r7) ;
168
169/* forward byte order operands */
170//(Dreg, Dreg) = BYTEOP16M (Dreg_pair, Dreg_pair) ; /* (b */)
171/* reverse byte order operands */
172//(Dreg, Dreg) = BYTEOP16M (Dreg-pair, Dreg-pair) (R) ; /* (b) */
173
174(r1,r2)= byteop16m (r3:2,r1:0) ;
175(r1,r2)= byteop16m (r3:2,r1:0) (r) ;
176(r0,r1)= byteop16m (r3:2,r1:0) ;
177(r2,r3)= byteop16m (r3:2,r1:0) (r) ;
178(r3,r5)= byteop16m (r3:2,r1:0) ;
179(r6,r7)= byteop16m (r3:2,r1:0) (r) ;
180
181(r1,r2)= byteop16m (r1:0,r1:0) ;
182(r1,r2)= byteop16m (r1:0,r1:0) (r) ;
183(r0,r1)= byteop16m (r1:0,r1:0) ;
184(r2,r3)= byteop16m (r1:0,r1:0) (r) ;
185(r3,r5)= byteop16m (r1:0,r1:0) ;
186(r6,r7)= byteop16m (r1:0,r1:0) (r) ;
187
188(r1,r2)= byteop16m (r1:0,r3:2) ;
189(r1,r2)= byteop16m (r1:0,r3:2) (r) ;
190(r0,r1)= byteop16m (r1:0,r3:2) ;
191(r2,r3)= byteop16m (r1:0,r3:2) (r) ;
192(r3,r5)= byteop16m (r1:0,r3:2) ;
193(r6,r7)= byteop16m (r1:0,r3:2) (r) ;
194
195(r1,r2)= byteop16m (r3:2,r3:2) ;
196(r1,r2)= byteop16m (r3:2,r3:2) (r) ;
197(r0,r1)= byteop16m (r3:2,r3:2) ;
198(r2,r3)= byteop16m (r3:2,r3:2) (r) ;
199(r3,r5)= byteop16m (r3:2,r3:2) ;
200(r6,r7)= byteop16m (r3:2,r3:2) (r) ;
201
202//SAA (Dreg_pair, Dreg_pair) ; /* forward byte order operands (b) */
203//SAA (Dreg_pair, Dreg_pair) (R) ; /* reverse byte order operands (b) */
204
205saa(r1:0, r3:2) || r0 = [i0++] || r2 = [i1++] ; /* parallel fill instructions */
206saa (r1:0, r3:2) (R) || r1 = [i0++] || r3 = [i1++] ; /* reverse, parallel fill instructions */
207saa (r1:0, r3:2) ; /* last SAA in a loop, no more fill required */
208
209//( Dreg , Dreg ) = BYTEUNPACK Dreg_pair ; /* (b) */
210//( Dreg , Dreg ) = BYTEUNPACK Dreg_pair (R) ; /* reverse source order (b) */
211
212(r6,r5) = byteunpack r1:0 ; /* non-reversing sources */
213(r6,r5) = byteunpack r1:0 (R) ; /* reversing sources case */
214(r6,r5) = byteunpack r3:2 ; /* non-reversing sources */
215(r6,r5) = byteunpack r3:2 (R) ; /* reversing sources case */
216(r0,r1) = byteunpack r1:0 ; /* non-reversing sources */
217(r2,r3) = byteunpack r1:0 (R) ; /* reversing sources case */
218(r4,r5) = byteunpack r3:2 ; /* non-reversing sources */
219(r6,r7) = byteunpack r3:2 (R) ; /* reversing sources case */
220