1017 	add	%o0, 4, %o0		! advance SRC by 4
1052 	add	%o0, 8, %o0		! update SRC
1063 	add	%o0, 2, %o0		! advance SRC by 2
1133 	add	%o0, 4, %o0		! advance SRC by 4
1147 	add	%o0, 32, %o0		! advance SRC by 32
1162 	add	%o0, 8, %o0		! advance SRC by 8
1197 	add	%o0, 16, %o0		! advance SRC by 16
1212 	add	%o0, 4, %o0		! advance SRC by 4
1244 	add	%o0, 8, %o0		! advance SRC by 8
1314 #define	SRC	%i3
1323 	! Using SRC as a tmp here
1329 	ldub	[REALSRC], SRC		! move 4 bytes per loop iteration
1330 	stb	SRC, [DST]
1332 	ldub	[REALSRC + 1], SRC
1334 	stb	SRC, [DST + 1]
1335 	ldub	[REALSRC - 2], SRC
1337 	stb	SRC, [DST - 2]
1338 	ldub	[REALSRC - 1], SRC
1340 	  stb	SRC, [DST - 1]
1345 1:	ldub	[REALSRC], SRC
1350 	  stb	SRC, [DST - 1]
1353 	andn	REALSRC, 0x7, SRC
1356 	! SRC - 8-byte aligned
1358 	prefetch [SRC], #one_read
1359 	prefetch [SRC + (1 * VIS_BLOCKSIZE)], #one_read
1360 	prefetch [SRC + (2 * VIS_BLOCKSIZE)], #one_read
1361 	prefetch [SRC + (3 * VIS_BLOCKSIZE)], #one_read
1362 	ldd	[SRC], %f0
1364 	prefetch [SRC + (4 * VIS_BLOCKSIZE)], #one_read
1366 	ldd	[SRC + 0x08], %f2
1368 	prefetch [SRC + (5 * VIS_BLOCKSIZE)], #one_read
1370 	ldd	[SRC + 0x10], %f4
1372 	prefetch [SRC + (6 * VIS_BLOCKSIZE)], #one_read
1375 	ldd	[SRC + 0x18], %f6
1377 	prefetch [SRC + (7 * VIS_BLOCKSIZE)], #one_read
1380 	ldd	[SRC + 0x20], %f8
1382 	ldd	[SRC + 0x28], %f10
1384 	ldd	[SRC + 0x30], %f12
1386 	ldd	[SRC + 0x38], %f14
1388 	ldd	[SRC + VIS_BLOCKSIZE], %f0
1390 	add	SRC, VIS_BLOCKSIZE, SRC
1396 	ldd	[SRC + 0x08], %f2
1398 	ldd	[SRC + 0x10], %f4
1401 	ldd	[SRC + 0x18], %f6
1403 	ldd	[SRC + 0x20], %f8
1405 	ldd	[SRC + 0x28], %f10
1407 	ldd	[SRC + 0x30], %f12
1409 	ldd	[SRC + 0x38], %f14
1412 	ldd	[SRC + VIS_BLOCKSIZE], %f0
1414 	prefetch [SRC + ((CHEETAH_PREFETCH) * VIS_BLOCKSIZE) + 8], #one_read
1416 	prefetch [SRC + ((CHEETAH_2ND_PREFETCH) * VIS_BLOCKSIZE)], #one_read
1420 	  add	SRC, VIS_BLOCKSIZE, SRC
1436 	ldd	[SRC + 0x08], %f2
1438 	ldd	[SRC + 0x10], %f4
1441 	ldd	[SRC + 0x18], %f6
1443 	ldd	[SRC + 0x20], %f8
1445 	ldd	[SRC + 0x28], %f10
1447 	ldd	[SRC + 0x30], %f12
1449 	ldd	[SRC + 0x38], %f14
1453 	add	SRC, VIS_BLOCKSIZE, SRC
1597 	mov	REALSRC, SRC
1599 	prefetch [SRC], #one_read
1600 	prefetch [SRC + (1 * VIS_BLOCKSIZE)], #one_read
1601 	prefetch [SRC + (2 * VIS_BLOCKSIZE)], #one_read
1602 	prefetch [SRC + (3 * VIS_BLOCKSIZE)], #one_read
1603 	ldd	[SRC], %f0
1605 	prefetch [SRC + (4 * VIS_BLOCKSIZE)], #one_read
1607 	ldd	[SRC + 0x08], %f2
1609 	prefetch [SRC + (5 * VIS_BLOCKSIZE)], #one_read
1611 	ldd	[SRC + 0x10], %f4
1613 	prefetch [SRC + (6 * VIS_BLOCKSIZE)], #one_read
1616 	ldd	[SRC + 0x18], %f6
1618 	prefetch [SRC + (7 * VIS_BLOCKSIZE)], #one_read
1621 	ldd	[SRC + 0x20], %f8
1623 	ldd	[SRC + 0x28], %f10
1625 	ldd	[SRC + 0x30], %f12
1627 	ldd	[SRC + 0x38], %f14
1629 	ldd	[SRC + VIS_BLOCKSIZE], %f0
1631 	add	SRC, VIS_BLOCKSIZE, SRC
1636 	ldd	[SRC + 0x08], %f2
1638 	ldd	[SRC + 0x10], %f4
1641 	ldd	[SRC + 0x18], %f6
1643 	ldd	[SRC + 0x20], %f8
1645 	ldd	[SRC + 0x28], %f10
1647 	ldd	[SRC + 0x30], %f12
1649 	ldd	[SRC + 0x38], %f14
1651 	ldd	[SRC + VIS_BLOCKSIZE], %f0
1653 	prefetch [SRC + ((CHEETAH_PREFETCH) * VIS_BLOCKSIZE) + 8], #one_read
1657 	prefetch [SRC + ((CHEETAH_2ND_PREFETCH) * VIS_BLOCKSIZE)], #one_read
1659 	  add	SRC, VIS_BLOCKSIZE, SRC
1662 	ldd	[SRC + 0x08], %f2
1664 	ldd	[SRC + 0x10], %f4
1667 	ldd	[SRC + 0x18], %f6
1669 	ldd	[SRC + 0x20], %f8
1671 	ldd	[SRC + 0x28], %f10
1673 	ldd	[SRC + 0x30], %f12
1675 	ldd	[SRC + 0x38], %f14
1679 	add	SRC, VIS_BLOCKSIZE, SRC
1929 	add	%o0, 4, %o0		! advance SRC by 4
1966 	add	%o0, 8, %o0		! update SRC
1980 	add	%o0, 2, %o0		! advance SRC by 2
2050 	add	%o0, 4, %o0		! advance SRC by 4
2068 	add	%o0, 32, %o0		! advance SRC by 32
2085 	add	%o0, 8, %o0		! advance SRC by 8
2123 	add	%o0, 16, %o0		! advance SRC by 16
2140 	add	%o0, 4, %o0		! advance SRC by 4
2176 	add	%o0, 8, %o0		! advance SRC by 8
2267 	! Using SRC as a tmp here
2273 	ldub	[REALSRC], SRC		! move 4 bytes per loop iteration
2274 	stba	SRC, [DST]%asi
2276 	ldub	[REALSRC + 1], SRC
2278 	stba	SRC, [DST + 1]%asi
2279 	ldub	[REALSRC - 2], SRC
2281 	stba	SRC, [DST - 2]%asi
2282 	ldub	[REALSRC - 1], SRC
2284 	  stba	SRC, [DST - 1]%asi
2289 1:	ldub	[REALSRC], SRC
2294 	  stba	SRC, [DST - 1]%asi
2297 	andn	REALSRC, 0x7, SRC
2300 	! SRC - 8-byte aligned
2302 	prefetch [SRC], #one_read
2303 	prefetch [SRC + (1 * VIS_BLOCKSIZE)], #one_read
2304 	prefetch [SRC + (2 * VIS_BLOCKSIZE)], #one_read
2305 	prefetch [SRC + (3 * VIS_BLOCKSIZE)], #one_read
2306 	ldd	[SRC], %f16
2308 	prefetch [SRC + (4 * VIS_BLOCKSIZE)], #one_read
2310 	ldd	[SRC + 0x08], %f18
2312 	prefetch [SRC + (5 * VIS_BLOCKSIZE)], #one_read
2314 	ldd	[SRC + 0x10], %f20
2316 	prefetch [SRC + (6 * VIS_BLOCKSIZE)], #one_read
2319 	ldd	[SRC + 0x18], %f22
2321 	prefetch [SRC + (7 * VIS_BLOCKSIZE)], #one_read
2324 	ldd	[SRC + 0x20], %f24
2326 	ldd	[SRC + 0x28], %f26
2328 	ldd	[SRC + 0x30], %f28
2330 	ldd	[SRC + 0x38], %f30
2332 	ldd	[SRC + VIS_BLOCKSIZE], %f16
2334 	add	SRC, VIS_BLOCKSIZE, SRC
2340 	ldd	[SRC + 0x08], %f18
2342 	ldd	[SRC + 0x10], %f20
2345 	ldd	[SRC + 0x18], %f22
2347 	ldd	[SRC + 0x20], %f24
2349 	ldd	[SRC + 0x28], %f26
2351 	ldd	[SRC + 0x30], %f28
2353 	ldd	[SRC + 0x38], %f30
2356 	ldd	[SRC + VIS_BLOCKSIZE], %f16
2358 	prefetch [SRC + ((CHEETAH_PREFETCH) * VIS_BLOCKSIZE) + 8], #one_read
2360 	prefetch [SRC + ((CHEETAH_2ND_PREFETCH) * VIS_BLOCKSIZE)], #one_read
2364 	  add	SRC, VIS_BLOCKSIZE, SRC
2380 	ldd	[SRC + 0x08], %f18
2382 	ldd	[SRC + 0x10], %f20
2385 	ldd	[SRC + 0x18], %f22
2387 	ldd	[SRC + 0x20], %f24
2389 	ldd	[SRC + 0x28], %f26
2391 	ldd	[SRC + 0x30], %f28
2393 	ldd	[SRC + 0x38], %f30
2397 	add	SRC, VIS_BLOCKSIZE, SRC
2722 	add	%o0, 1, %o0		! advance SRC by 1
2724 	add	%o0, 1, %o0		! advance SRC by 1
2728 	add	%o0, 1, %o0		! advance SRC by 1
2731 	add	%o0, 1, %o0		! advance SRC by 1
2761 	add	%o0, 4, %o0		! update SRC
2764 	add	%o0, 4, %o0		! update SRC
2775 	add	%o0, 2, %o0		! advance SRC by 2
2845 	add	%o0, 4, %o0		! advance SRC by 4
2858 	add	%o0, 8, %o0		! advance SRC by 8
2860 	add	%o0, 8, %o0		! advance SRC by 8
2864 	add	%o0, 8, %o0		! advance SRC by 8
2867 	add	%o0, 8, %o0		! advance SRC by 8
2877 	add	%o0, 8, %o0		! advance SRC by 8
2913 	add	%o0, 4, %o0		! advance SRC by 4
2915 	add	%o0, 4, %o0		! advance SRC by 4
2919 	add	%o0, 4, %o0		! advance SRC by 4
2922 	add	%o0, 4, %o0		! advance SRC by 4
2932 	add	%o0, 4, %o0		! advance SRC by 4
2965 	add	%o0, 2, %o0		! advance SRC by 2
2967 	add	%o0, 2, %o0		! advance SRC by 2
2971 	add	%o0, 2, %o0		! advance SRC by 2
2974 	add	%o0, 2, %o0		! advance SRC by 2
3055 	! Using SRC as a tmp here
3061 	lduba	[REALSRC]%asi, SRC	! move 4 bytes per loop iteration
3062 	stb	SRC, [DST]
3064 	lduba	[REALSRC + 1]%asi, SRC
3066 	stb	SRC, [DST + 1]
3067 	lduba	[REALSRC - 2]%asi, SRC
3069 	stb	SRC, [DST - 2]
3070 	lduba	[REALSRC - 1]%asi, SRC
3072 	  stb	SRC, [DST - 1]
3077 1:	lduba	[REALSRC]%asi, SRC
3082 	  stb	SRC, [DST - 1]
3085 	andn	REALSRC, 0x7, SRC
3088 	! SRC - 8-byte aligned
3090 	prefetcha [SRC]%asi, #one_read
3091 	prefetcha [SRC + (1 * VIS_BLOCKSIZE)]%asi, #one_read
3092 	prefetcha [SRC + (2 * VIS_BLOCKSIZE)]%asi, #one_read
3093 	prefetcha [SRC + (3 * VIS_BLOCKSIZE)]%asi, #one_read
3094 	ldda	[SRC]%asi, %f16
3096 	prefetcha [SRC + (4 * VIS_BLOCKSIZE)]%asi, #one_read
3098 	ldda	[SRC + 0x08]%asi, %f18
3100 	prefetcha [SRC + (5 * VIS_BLOCKSIZE)]%asi, #one_read
3102 	ldda	[SRC + 0x10]%asi, %f20
3104 	prefetcha [SRC + (6 * VIS_BLOCKSIZE)]%asi, #one_read
3107 	ldda	[SRC + 0x18]%asi, %f22
3109 	prefetcha [SRC + (7 * VIS_BLOCKSIZE)]%asi, #one_read
3112 	ldda	[SRC + 0x20]%asi, %f24
3114 	ldda	[SRC + 0x28]%asi, %f26
3116 	ldda	[SRC + 0x30]%asi, %f28
3118 	ldda	[SRC + 0x38]%asi, %f30
3120 	ldda	[SRC + VIS_BLOCKSIZE]%asi, %f16
3122 	add	SRC, VIS_BLOCKSIZE, SRC
3128 	ldda	[SRC + 0x08]%asi, %f18
3130 	ldda	[SRC + 0x10]%asi, %f20
3133 	ldda	[SRC + 0x18]%asi, %f22
3135 	ldda	[SRC + 0x20]%asi, %f24
3137 	ldda	[SRC + 0x28]%asi, %f26
3139 	ldda	[SRC + 0x30]%asi, %f28
3141 	ldda	[SRC + 0x38]%asi, %f30
3144 	ldda	[SRC + VIS_BLOCKSIZE]%asi, %f16
3146 	prefetcha [SRC + ((CHEETAH_PREFETCH) * VIS_BLOCKSIZE) + 8]%asi, #one_read
3148 	prefetcha [SRC + ((CHEETAH_2ND_PREFETCH) * VIS_BLOCKSIZE)]%asi, #one_read
3152 	  add	SRC, VIS_BLOCKSIZE, SRC
3168 	ldda	[SRC + 0x08]%asi, %f18
3170 	ldda	[SRC + 0x10]%asi, %f20
3173 	ldda	[SRC + 0x18]%asi, %f22
3175 	ldda	[SRC + 0x20]%asi, %f24
3177 	ldda	[SRC + 0x28]%asi, %f26
3179 	ldda	[SRC + 0x30]%asi, %f28
3181 	ldda	[SRC + 0x38]%asi, %f30
3185 	add	SRC, VIS_BLOCKSIZE, SRC

Indexes created Tue Jul 24 14:28:13 CEST 2018