| blob | 3742573248d89577ca9ff50f4a6119879900c620 |
1 /* Copy SIZE bytes from SRC to DEST.
2 For UltraSPARC.
3 Copyright (C) 1996, 97, 98, 99, 2003 Free Software Foundation, Inc.
4 This file is part of the GNU C Library.
5 Contributed by David S. Miller (davem@caip.rutgers.edu) and
6 Jakub Jelinek (jakub@redhat.com).
8 The GNU C Library is free software; you can redistribute it and/or
9 modify it under the terms of the GNU Lesser General Public
10 License as published by the Free Software Foundation; either
11 version 2.1 of the License, or (at your option) any later version.
13 The GNU C Library is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 Lesser General Public License for more details.
18 You should have received a copy of the GNU Lesser General Public
19 License along with the GNU C Library; if not, write to the Free
20 Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
21 02111-1307 USA. */
23 #include <sysdep.h>
24 #include <asm/asi.h>
25 #ifndef XCC
26 #define USE_BPR
27 .register %g2, #scratch
28 .register %g3, #scratch
29 .register %g6, #scratch
30 #define XCC xcc
31 #endif
32 #define FPRS_FEF 4
34 #define FREG_FROB(f1, f2, f3, f4, f5, f6, f7, f8, f9) \
35 faligndata %f1, %f2, %f48; \
36 faligndata %f2, %f3, %f50; \
37 faligndata %f3, %f4, %f52; \
38 faligndata %f4, %f5, %f54; \
39 faligndata %f5, %f6, %f56; \
40 faligndata %f6, %f7, %f58; \
41 faligndata %f7, %f8, %f60; \
42 faligndata %f8, %f9, %f62;
44 #define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, len, jmptgt) \
45 ldda [%src] %asi, %fdest; \
46 add %src, 0x40, %src; \
47 add %dest, 0x40, %dest; \
48 subcc %len, 0x40, %len; \
49 be,pn %xcc, jmptgt; \
50 stda %fsrc, [%dest - 0x40] %asi;
52 #define LOOP_CHUNK1(src, dest, len, branch_dest) \
53 MAIN_LOOP_CHUNK(src, dest, f0, f48, len, branch_dest)
54 #define LOOP_CHUNK2(src, dest, len, branch_dest) \
55 MAIN_LOOP_CHUNK(src, dest, f16, f48, len, branch_dest)
56 #define LOOP_CHUNK3(src, dest, len, branch_dest) \
57 MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
59 #define STORE_SYNC(dest, fsrc) \
60 stda %fsrc, [%dest] %asi; \
61 add %dest, 0x40, %dest;
63 #define STORE_JUMP(dest, fsrc, target) \
64 stda %fsrc, [%dest] %asi; \
65 add %dest, 0x40, %dest; \
66 ba,pt %xcc, target;
68 #define VISLOOP_PAD nop; nop; nop; nop; \
69 nop; nop; nop; nop; \
70 nop; nop; nop; nop; \
71 nop; nop; nop;
73 #define FINISH_VISCHUNK(dest, f0, f1, left) \
74 subcc %left, 8, %left; \
75 bl,pn %xcc, 205f; \
76 faligndata %f0, %f1, %f48; \
77 std %f48, [%dest]; \
78 add %dest, 8, %dest;
80 #define UNEVEN_VISCHUNK(dest, f0, f1, left) \
81 subcc %left, 8, %left; \
82 bl,pn %xcc, 205f; \
83 fsrc1 %f0, %f1; \
84 ba,a,pt %xcc, 204f;
86 /* Macros for non-VIS memcpy code. */
87 #define MOVE_BIGCHUNK(src, dst, offset, t0, t1, t2, t3) \
88 ldx [%src + offset + 0x00], %t0; \
89 ldx [%src + offset + 0x08], %t1; \
90 ldx [%src + offset + 0x10], %t2; \
91 ldx [%src + offset + 0x18], %t3; \
92 stw %t0, [%dst + offset + 0x04]; \
93 srlx %t0, 32, %t0; \
94 stw %t0, [%dst + offset + 0x00]; \
95 stw %t1, [%dst + offset + 0x0c]; \
96 srlx %t1, 32, %t1; \
97 stw %t1, [%dst + offset + 0x08]; \
98 stw %t2, [%dst + offset + 0x14]; \
99 srlx %t2, 32, %t2; \
100 stw %t2, [%dst + offset + 0x10]; \
101 stw %t3, [%dst + offset + 0x1c]; \
102 srlx %t3, 32, %t3; \
103 stw %t3, [%dst + offset + 0x18];
105 #define MOVE_BIGALIGNCHUNK(src, dst, offset, t0, t1, t2, t3) \
106 ldx [%src + offset + 0x00], %t0; \
107 ldx [%src + offset + 0x08], %t1; \
108 ldx [%src + offset + 0x10], %t2; \
109 ldx [%src + offset + 0x18], %t3; \
110 stx %t0, [%dst + offset + 0x00]; \
111 stx %t1, [%dst + offset + 0x08]; \
112 stx %t2, [%dst + offset + 0x10]; \
113 stx %t3, [%dst + offset + 0x18]; \
114 ldx [%src + offset + 0x20], %t0; \
115 ldx [%src + offset + 0x28], %t1; \
116 ldx [%src + offset + 0x30], %t2; \
117 ldx [%src + offset + 0x38], %t3; \
118 stx %t0, [%dst + offset + 0x20]; \
119 stx %t1, [%dst + offset + 0x28]; \
120 stx %t2, [%dst + offset + 0x30]; \
121 stx %t3, [%dst + offset + 0x38];
123 #define MOVE_LASTCHUNK(src, dst, offset, t0, t1, t2, t3) \
124 ldx [%src - offset - 0x10], %t0; \
125 ldx [%src - offset - 0x08], %t1; \
126 stw %t0, [%dst - offset - 0x0c]; \
127 srlx %t0, 32, %t2; \
128 stw %t2, [%dst - offset - 0x10]; \
129 stw %t1, [%dst - offset - 0x04]; \
130 srlx %t1, 32, %t3; \
131 stw %t3, [%dst - offset - 0x08];
133 #define MOVE_LASTALIGNCHUNK(src, dst, offset, t0, t1) \
134 ldx [%src - offset - 0x10], %t0; \
135 ldx [%src - offset - 0x08], %t1; \
136 stx %t0, [%dst - offset - 0x10]; \
137 stx %t1, [%dst - offset - 0x08];
139 /* Macros for non-VIS memmove code. */
140 #define RMOVE_BIGCHUNK(src, dst, offset, t0, t1, t2, t3) \
141 ldx [%src - offset - 0x20], %t0; \
142 ldx [%src - offset - 0x18], %t1; \
143 ldx [%src - offset - 0x10], %t2; \
144 ldx [%src - offset - 0x08], %t3; \
145 stw %t0, [%dst - offset - 0x1c]; \
146 srlx %t0, 32, %t0; \
147 stw %t0, [%dst - offset - 0x20]; \
148 stw %t1, [%dst - offset - 0x14]; \
149 srlx %t1, 32, %t1; \
150 stw %t1, [%dst - offset - 0x18]; \
151 stw %t2, [%dst - offset - 0x0c]; \
152 srlx %t2, 32, %t2; \
153 stw %t2, [%dst - offset - 0x10]; \
154 stw %t3, [%dst - offset - 0x04]; \
155 srlx %t3, 32, %t3; \
156 stw %t3, [%dst - offset - 0x08];
158 #define RMOVE_BIGALIGNCHUNK(src, dst, offset, t0, t1, t2, t3) \
159 ldx [%src - offset - 0x20], %t0; \
160 ldx [%src - offset - 0x18], %t1; \
161 ldx [%src - offset - 0x10], %t2; \
162 ldx [%src - offset - 0x08], %t3; \
163 stx %t0, [%dst - offset - 0x20]; \
164 stx %t1, [%dst - offset - 0x18]; \
165 stx %t2, [%dst - offset - 0x10]; \
166 stx %t3, [%dst - offset - 0x08]; \
167 ldx [%src - offset - 0x40], %t0; \
168 ldx [%src - offset - 0x38], %t1; \
169 ldx [%src - offset - 0x30], %t2; \
170 ldx [%src - offset - 0x28], %t3; \
171 stx %t0, [%dst - offset - 0x40]; \
172 stx %t1, [%dst - offset - 0x38]; \
173 stx %t2, [%dst - offset - 0x30]; \
174 stx %t3, [%dst - offset - 0x28];
176 #define RMOVE_LASTCHUNK(src, dst, offset, t0, t1, t2, t3) \
177 ldx [%src + offset + 0x00], %t0; \
178 ldx [%src + offset + 0x08], %t1; \
179 stw %t0, [%dst + offset + 0x04]; \
180 srlx %t0, 32, %t2; \
181 stw %t2, [%dst + offset + 0x00]; \
182 stw %t1, [%dst + offset + 0x0c]; \
183 srlx %t1, 32, %t3; \
184 stw %t3, [%dst + offset + 0x08];
186 #define RMOVE_LASTALIGNCHUNK(src, dst, offset, t0, t1) \
187 ldx [%src + offset + 0x00], %t0; \
188 ldx [%src + offset + 0x08], %t1; \
189 stx %t0, [%dst + offset + 0x00]; \
190 stx %t1, [%dst + offset + 0x08];
192 .text
193 .align 32
195 ENTRY(bcopy)
196 sub %o1, %o0, %o4 /* IEU0 Group */
197 mov %o0, %g3 /* IEU1 */
198 cmp %o4, %o2 /* IEU1 Group */
199 mov %o1, %o0 /* IEU0 */
200 bgeu,pt %XCC, 210f /* CTI */
201 mov %g3, %o1 /* IEU0 Group */
202 #ifndef USE_BPR
203 srl %o2, 0, %o2 /* IEU1 */
204 #endif
205 brnz,pn %o2, 220f /* CTI Group */
206 add %o0, %o2, %o0 /* IEU0 */
207 retl
208 nop
209 END(bcopy)
211 .align 32
212 200: be,pt %xcc, 201f /* CTI */
213 andcc %o0, 0x38, %g5 /* IEU1 Group */
214 mov 8, %g1 /* IEU0 */
215 sub %g1, %g2, %g2 /* IEU0 Group */
216 andcc %o0, 1, %g0 /* IEU1 */
217 be,pt %icc, 2f /* CTI */
218 sub %o2, %g2, %o2 /* IEU0 Group */
219 1: ldub [%o1], %o5 /* Load Group */
220 add %o1, 1, %o1 /* IEU0 */
221 add %o0, 1, %o0 /* IEU1 */
222 subcc %g2, 1, %g2 /* IEU1 Group */
223 be,pn %xcc, 3f /* CTI */
224 stb %o5, [%o0 - 1] /* Store */
225 2: ldub [%o1], %o5 /* Load Group */
226 add %o0, 2, %o0 /* IEU0 */
227 ldub [%o1 + 1], %g3 /* Load Group */
228 subcc %g2, 2, %g2 /* IEU1 Group */
229 stb %o5, [%o0 - 2] /* Store */
230 add %o1, 2, %o1 /* IEU0 */
231 bne,pt %xcc, 2b /* CTI Group */
232 stb %g3, [%o0 - 1] /* Store */
233 3: andcc %o0, 0x38, %g5 /* IEU1 Group */
234 201: be,pt %icc, 202f /* CTI */
235 mov 64, %g1 /* IEU0 */
236 fmovd %f0, %f2 /* FPU */
237 sub %g1, %g5, %g5 /* IEU0 Group */
238 alignaddr %o1, %g0, %g1 /* GRU Group */
239 ldd [%g1], %f4 /* Load Group */
240 sub %o2, %g5, %o2 /* IEU0 */
241 1: ldd [%g1 + 0x8], %f6 /* Load Group */
242 add %g1, 0x8, %g1 /* IEU0 Group */
243 subcc %g5, 8, %g5 /* IEU1 */
244 faligndata %f4, %f6, %f0 /* GRU Group */
245 std %f0, [%o0] /* Store */
246 add %o1, 8, %o1 /* IEU0 Group */
247 be,pn %xcc, 202f /* CTI */
248 add %o0, 8, %o0 /* IEU1 */
249 ldd [%g1 + 0x8], %f4 /* Load Group */
250 add %g1, 8, %g1 /* IEU0 */
251 subcc %g5, 8, %g5 /* IEU1 */
252 faligndata %f6, %f4, %f0 /* GRU Group */
253 std %f0, [%o0] /* Store */
254 add %o1, 8, %o1 /* IEU0 */
255 bne,pt %xcc, 1b /* CTI Group */
256 add %o0, 8, %o0 /* IEU0 */
257 202: membar #LoadStore | #StoreStore | #StoreLoad /* LSU Group */
258 wr %g0, ASI_BLK_P, %asi /* LSU Group */
259 subcc %o2, 0x40, %g6 /* IEU1 Group */
260 mov %o1, %g1 /* IEU0 */
261 andncc %g6, (0x40 - 1), %g6 /* IEU1 Group */
262 srl %g1, 3, %g2 /* IEU0 */
263 sub %o2, %g6, %g3 /* IEU0 Group */
264 andn %o1, (0x40 - 1), %o1 /* IEU1 */
265 and %g2, 7, %g2 /* IEU0 Group */
266 andncc %g3, 0x7, %g3 /* IEU1 */
267 fmovd %f0, %f2 /* FPU */
268 sub %g3, 0x10, %g3 /* IEU0 Group */
269 sub %o2, %g6, %o2 /* IEU1 */
270 alignaddr %g1, %g0, %g0 /* GRU Group */
271 add %g1, %g6, %g1 /* IEU0 Group */
272 subcc %o2, %g3, %o2 /* IEU1 */
273 ldda [%o1 + 0x00] %asi, %f0 /* LSU Group */
274 add %g1, %g3, %g1 /* IEU0 */
275 ldda [%o1 + 0x40] %asi, %f16 /* LSU Group */
276 sub %g6, 0x80, %g6 /* IEU0 */
277 ldda [%o1 + 0x80] %asi, %f32 /* LSU Group */
278 /* Clk1 Group 8-( */
279 /* Clk2 Group 8-( */
280 /* Clk3 Group 8-( */
281 /* Clk4 Group 8-( */
282 203: rd %pc, %g5 /* PDU Group 8-( */
283 addcc %g5, %lo(300f - 203b), %g5 /* IEU1 Group */
284 sll %g2, 9, %g2 /* IEU0 */
285 jmpl %g5 + %g2, %g0 /* CTI Group brk forced*/
286 addcc %o1, 0xc0, %o1 /* IEU1 Group */
288 .align 512 /* OK, here comes the fun part... */
289 300: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) LOOP_CHUNK1(o1, o0, g6, 301f)
290 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) LOOP_CHUNK2(o1, o0, g6, 302f)
291 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) LOOP_CHUNK3(o1, o0, g6, 303f)
292 b,pt %xcc, 300b+4; faligndata %f0, %f2, %f48
293 301: FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) STORE_SYNC(o0, f48) membar #Sync
294 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) STORE_JUMP(o0, f48, 400f) membar #Sync
295 302: FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) STORE_SYNC(o0, f48) membar #Sync
296 FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) STORE_JUMP(o0, f48, 416f) membar #Sync
297 303: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) STORE_SYNC(o0, f48) membar #Sync
298 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) STORE_JUMP(o0, f48, 432f) membar #Sync
299 VISLOOP_PAD
300 310: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) LOOP_CHUNK1(o1, o0, g6, 311f)
301 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) LOOP_CHUNK2(o1, o0, g6, 312f)
302 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) LOOP_CHUNK3(o1, o0, g6, 313f)
303 b,pt %xcc, 310b+4; faligndata %f2, %f4, %f48
304 311: FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) STORE_SYNC(o0, f48) membar #Sync
305 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) STORE_JUMP(o0, f48, 402f) membar #Sync
306 312: FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) STORE_SYNC(o0, f48) membar #Sync
307 FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) STORE_JUMP(o0, f48, 418f) membar #Sync
308 313: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) STORE_SYNC(o0, f48) membar #Sync
309 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) STORE_JUMP(o0, f48, 434f) membar #Sync
310 VISLOOP_PAD
311 320: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) LOOP_CHUNK1(o1, o0, g6, 321f)
312 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) LOOP_CHUNK2(o1, o0, g6, 322f)
313 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) LOOP_CHUNK3(o1, o0, g6, 323f)
314 b,pt %xcc, 320b+4; faligndata %f4, %f6, %f48
315 321: FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) STORE_SYNC(o0, f48) membar #Sync
316 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) STORE_JUMP(o0, f48, 404f) membar #Sync
317 322: FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) STORE_SYNC(o0, f48) membar #Sync
318 FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) STORE_JUMP(o0, f48, 420f) membar #Sync
319 323: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) STORE_SYNC(o0, f48) membar #Sync
320 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) STORE_JUMP(o0, f48, 436f) membar #Sync
321 VISLOOP_PAD
322 330: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) LOOP_CHUNK1(o1, o0, g6, 331f)
323 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) LOOP_CHUNK2(o1, o0, g6, 332f)
324 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) LOOP_CHUNK3(o1, o0, g6, 333f)
325 b,pt %xcc, 330b+4; faligndata %f6, %f8, %f48
326 331: FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) STORE_SYNC(o0, f48) membar #Sync
327 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) STORE_JUMP(o0, f48, 406f) membar #Sync
328 332: FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) STORE_SYNC(o0, f48) membar #Sync
329 FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) STORE_JUMP(o0, f48, 422f) membar #Sync
330 333: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) STORE_SYNC(o0, f48) membar #Sync
331 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) STORE_JUMP(o0, f48, 438f) membar #Sync
332 VISLOOP_PAD
333 340: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) LOOP_CHUNK1(o1, o0, g6, 341f)
334 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) LOOP_CHUNK2(o1, o0, g6, 342f)
335 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) LOOP_CHUNK3(o1, o0, g6, 343f)
336 b,pt %xcc, 340b+4; faligndata %f8, %f10, %f48
337 341: FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) STORE_SYNC(o0, f48) membar #Sync
338 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) STORE_JUMP(o0, f48, 408f) membar #Sync
339 342: FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) STORE_SYNC(o0, f48) membar #Sync
340 FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) STORE_JUMP(o0, f48, 424f) membar #Sync
341 343: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) STORE_SYNC(o0, f48) membar #Sync
342 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) STORE_JUMP(o0, f48, 440f) membar #Sync
343 VISLOOP_PAD
344 350: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) LOOP_CHUNK1(o1, o0, g6, 351f)
345 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) LOOP_CHUNK2(o1, o0, g6, 352f)
346 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) LOOP_CHUNK3(o1, o0, g6, 353f)
347 b,pt %xcc, 350b+4; faligndata %f10, %f12, %f48
348 351: FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) STORE_SYNC(o0, f48) membar #Sync
349 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) STORE_JUMP(o0, f48, 410f) membar #Sync
350 352: FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) STORE_SYNC(o0, f48) membar #Sync
351 FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) STORE_JUMP(o0, f48, 426f) membar #Sync
352 353: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) STORE_SYNC(o0, f48) membar #Sync
353 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) STORE_JUMP(o0, f48, 442f) membar #Sync
354 VISLOOP_PAD
355 360: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) LOOP_CHUNK1(o1, o0, g6, 361f)
356 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) LOOP_CHUNK2(o1, o0, g6, 362f)
357 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) LOOP_CHUNK3(o1, o0, g6, 363f)
358 b,pt %xcc, 360b+4; faligndata %f12, %f14, %f48
359 361: FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) STORE_SYNC(o0, f48) membar #Sync
360 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) STORE_JUMP(o0, f48, 412f) membar #Sync
361 362: FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) STORE_SYNC(o0, f48) membar #Sync
362 FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) STORE_JUMP(o0, f48, 428f) membar #Sync
363 363: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) STORE_SYNC(o0, f48) membar #Sync
364 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) STORE_JUMP(o0, f48, 444f) membar #Sync
365 VISLOOP_PAD
366 370: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) LOOP_CHUNK1(o1, o0, g6, 371f)
367 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) LOOP_CHUNK2(o1, o0, g6, 372f)
368 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) LOOP_CHUNK3(o1, o0, g6, 373f)
369 b,pt %xcc, 370b+4; faligndata %f14, %f16, %f48
370 371: FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) STORE_SYNC(o0, f48) membar #Sync
371 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) STORE_JUMP(o0, f48, 414f) membar #Sync
372 372: FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) STORE_SYNC(o0, f48) membar #Sync
373 FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) STORE_JUMP(o0, f48, 430f) membar #Sync
374 373: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) STORE_SYNC(o0, f48) membar #Sync
375 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) STORE_JUMP(o0, f48, 446f) membar #Sync
376 VISLOOP_PAD
377 400: FINISH_VISCHUNK(o0, f0, f2, g3)
378 402: FINISH_VISCHUNK(o0, f2, f4, g3)
379 404: FINISH_VISCHUNK(o0, f4, f6, g3)
380 406: FINISH_VISCHUNK(o0, f6, f8, g3)
381 408: FINISH_VISCHUNK(o0, f8, f10, g3)
382 410: FINISH_VISCHUNK(o0, f10, f12, g3)
383 412: FINISH_VISCHUNK(o0, f12, f14, g3)
384 414: UNEVEN_VISCHUNK(o0, f14, f0, g3)
385 416: FINISH_VISCHUNK(o0, f16, f18, g3)
386 418: FINISH_VISCHUNK(o0, f18, f20, g3)
387 420: FINISH_VISCHUNK(o0, f20, f22, g3)
388 422: FINISH_VISCHUNK(o0, f22, f24, g3)
389 424: FINISH_VISCHUNK(o0, f24, f26, g3)
390 426: FINISH_VISCHUNK(o0, f26, f28, g3)
391 428: FINISH_VISCHUNK(o0, f28, f30, g3)
392 430: UNEVEN_VISCHUNK(o0, f30, f0, g3)
393 432: FINISH_VISCHUNK(o0, f32, f34, g3)
394 434: FINISH_VISCHUNK(o0, f34, f36, g3)
395 436: FINISH_VISCHUNK(o0, f36, f38, g3)
396 438: FINISH_VISCHUNK(o0, f38, f40, g3)
397 440: FINISH_VISCHUNK(o0, f40, f42, g3)
398 442: FINISH_VISCHUNK(o0, f42, f44, g3)
399 444: FINISH_VISCHUNK(o0, f44, f46, g3)
400 446: UNEVEN_VISCHUNK(o0, f46, f0, g3)
401 204: ldd [%o1], %f2 /* Load Group */
402 add %o1, 8, %o1 /* IEU0 */
403 subcc %g3, 8, %g3 /* IEU1 */
404 faligndata %f0, %f2, %f8 /* GRU Group */
405 std %f8, [%o0] /* Store */
406 bl,pn %xcc, 205f /* CTI */
407 add %o0, 8, %o0 /* IEU0 Group */
408 ldd [%o1], %f0 /* Load Group */
409 add %o1, 8, %o1 /* IEU0 */
410 subcc %g3, 8, %g3 /* IEU1 */
411 faligndata %f2, %f0, %f8 /* GRU Group */
412 std %f8, [%o0] /* Store */
413 bge,pt %xcc, 204b /* CTI */
414 add %o0, 8, %o0 /* IEU0 Group */
415 205: brz,pt %o2, 207f /* CTI Group */
416 mov %g1, %o1 /* IEU0 */
417 206: ldub [%o1], %g5 /* LOAD */
418 add %o1, 1, %o1 /* IEU0 */
419 add %o0, 1, %o0 /* IEU1 */
420 subcc %o2, 1, %o2 /* IEU1 */
421 bne,pt %xcc, 206b /* CTI */
422 stb %g5, [%o0 - 1] /* Store Group */
423 207: membar #StoreLoad | #StoreStore /* LSU Group */
424 wr %g0, FPRS_FEF, %fprs
425 retl
426 mov %g4, %o0
428 208: andcc %o2, 1, %g0 /* IEU1 Group */
429 be,pt %icc, 2f+4 /* CTI */
430 1: ldub [%o1], %g5 /* LOAD Group */
431 add %o1, 1, %o1 /* IEU0 */
432 add %o0, 1, %o0 /* IEU1 */
433 subcc %o2, 1, %o2 /* IEU1 Group */
434 be,pn %xcc, 209f /* CTI */
435 stb %g5, [%o0 - 1] /* Store */
436 2: ldub [%o1], %g5 /* LOAD Group */
437 add %o0, 2, %o0 /* IEU0 */
438 ldub [%o1 + 1], %o5 /* LOAD Group */
439 add %o1, 2, %o1 /* IEU0 */
440 subcc %o2, 2, %o2 /* IEU1 Group */
441 stb %g5, [%o0 - 2] /* Store */
442 bne,pt %xcc, 2b /* CTI */
443 stb %o5, [%o0 - 1] /* Store */
444 209: retl
445 mov %g4, %o0
447 #ifdef USE_BPR
449 /* void *__align_cpy_4(void *dest, void *src, size_t n)
450 * SPARC v9 SYSV ABI
451 * Like memcpy, but results are undefined if (!n || ((dest | src | n) & 3))
452 */
454 .align 32
455 ENTRY(__align_cpy_4)
456 mov %o0, %g4 /* IEU0 Group */
457 cmp %o2, 15 /* IEU1 */
458 bleu,pn %xcc, 208b /* CTI */
459 cmp %o2, (64 * 6) /* IEU1 Group */
460 bgeu,pn %xcc, 200b /* CTI */
461 andcc %o0, 7, %g2 /* IEU1 Group */
462 ba,pt %xcc, 216f /* CTI */
463 andcc %o1, 4, %g0 /* IEU1 Group */
464 END(__align_cpy_4)
466 /* void *__align_cpy_8(void *dest, void *src, size_t n)
467 * SPARC v9 SYSV ABI
468 * Like memcpy, but results are undefined if (!n || ((dest | src | n) & 7))
469 */
471 .align 32
472 ENTRY(__align_cpy_8)
473 mov %o0, %g4 /* IEU0 Group */
474 cmp %o2, 15 /* IEU1 */
475 bleu,pn %xcc, 208b /* CTI */
476 cmp %o2, (64 * 6) /* IEU1 Group */
477 bgeu,pn %xcc, 201b /* CTI */
478 andcc %o0, 0x38, %g5 /* IEU1 Group */
479 andcc %o2, -128, %g6 /* IEU1 Group */
480 bne,a,pt %xcc, 82f + 4 /* CTI */
481 ldx [%o1], %g1 /* Load */
482 ba,pt %xcc, 41f /* CTI Group */
483 andcc %o2, 0x70, %g6 /* IEU1 */
484 END(__align_cpy_8)
486 /* void *__align_cpy_16(void *dest, void *src, size_t n)
487 * SPARC v9 SYSV ABI
488 * Like memcpy, but results are undefined if (!n || ((dest | src | n) & 15))
489 */
491 .align 32
492 ENTRY(__align_cpy_16)
493 mov %o0, %g4 /* IEU0 Group */
494 cmp %o2, (64 * 6) /* IEU1 */
495 bgeu,pn %xcc, 201b /* CTI */
496 andcc %o0, 0x38, %g5 /* IEU1 Group */
497 andcc %o2, -128, %g6 /* IEU1 Group */
498 bne,a,pt %xcc, 82f + 4 /* CTI */
499 ldx [%o1], %g1 /* Load */
500 ba,pt %xcc, 41f /* CTI Group */
501 andcc %o2, 0x70, %g6 /* IEU1 */
502 END(__align_cpy_16)
504 #endif
506 .align 32
507 ENTRY(memcpy)
508 210:
509 #ifndef USE_BPR
510 srl %o2, 0, %o2 /* IEU1 Group */
511 #endif
512 brz,pn %o2, 209b /* CTI Group */
513 mov %o0, %g4 /* IEU0 */
514 218: cmp %o2, 15 /* IEU1 Group */
515 bleu,pn %xcc, 208b /* CTI */
516 cmp %o2, (64 * 6) /* IEU1 Group */
517 bgeu,pn %xcc, 200b /* CTI */
518 andcc %o0, 7, %g2 /* IEU1 Group */
519 sub %o0, %o1, %g5 /* IEU0 */
520 andcc %g5, 3, %o5 /* IEU1 Group */
521 bne,pn %xcc, 212f /* CTI */
522 andcc %o1, 3, %g0 /* IEU1 Group */
523 be,a,pt %xcc, 216f /* CTI */
524 andcc %o1, 4, %g0 /* IEU1 Group */
525 andcc %o1, 1, %g0 /* IEU1 Group */
526 be,pn %xcc, 4f /* CTI */
527 andcc %o1, 2, %g0 /* IEU1 Group */
528 ldub [%o1], %g2 /* Load Group */
529 add %o1, 1, %o1 /* IEU0 */
530 add %o0, 1, %o0 /* IEU1 */
531 sub %o2, 1, %o2 /* IEU0 Group */
532 bne,pn %xcc, 5f /* CTI Group */
533 stb %g2, [%o0 - 1] /* Store */
534 4: lduh [%o1], %g2 /* Load Group */
535 add %o1, 2, %o1 /* IEU0 */
536 add %o0, 2, %o0 /* IEU1 */
537 sub %o2, 2, %o2 /* IEU0 */
538 sth %g2, [%o0 - 2] /* Store Group + bubble */
539 5: andcc %o1, 4, %g0 /* IEU1 */
540 216: be,a,pn %xcc, 2f /* CTI */
541 andcc %o2, -128, %g6 /* IEU1 Group */
542 lduw [%o1], %g5 /* Load Group */
543 add %o1, 4, %o1 /* IEU0 */
544 add %o0, 4, %o0 /* IEU1 */
545 sub %o2, 4, %o2 /* IEU0 Group */
546 stw %g5, [%o0 - 4] /* Store */
547 andcc %o2, -128, %g6 /* IEU1 Group */
548 2: be,pn %xcc, 215f /* CTI */
549 andcc %o0, 4, %g0 /* IEU1 Group */
550 be,pn %xcc, 82f + 4 /* CTI Group */
551 5: MOVE_BIGCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
552 MOVE_BIGCHUNK(o1, o0, 0x20, g1, g3, g5, o5)
553 MOVE_BIGCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
554 MOVE_BIGCHUNK(o1, o0, 0x60, g1, g3, g5, o5)
555 35: subcc %g6, 128, %g6 /* IEU1 Group */
556 add %o1, 128, %o1 /* IEU0 */
557 bne,pt %xcc, 5b /* CTI */
558 add %o0, 128, %o0 /* IEU0 Group */
559 215: andcc %o2, 0x70, %g6 /* IEU1 Group */
560 41: be,pn %xcc, 80f /* CTI */
561 andcc %o2, 8, %g0 /* IEU1 Group */
562 /* Clk1 8-( */
563 /* Clk2 8-( */
564 /* Clk3 8-( */
565 /* Clk4 8-( */
566 79: rd %pc, %o5 /* PDU Group */
567 sll %g6, 1, %g5 /* IEU0 Group */
568 add %o1, %g6, %o1 /* IEU1 */
569 sub %o5, %g5, %o5 /* IEU0 Group */
570 jmpl %o5 + %lo(80f - 79b), %g0 /* CTI Group brk forced*/
571 add %o0, %g6, %o0 /* IEU0 Group */
572 36: MOVE_LASTCHUNK(o1, o0, 0x60, g2, g3, g5, o5)
573 MOVE_LASTCHUNK(o1, o0, 0x50, g2, g3, g5, o5)
574 MOVE_LASTCHUNK(o1, o0, 0x40, g2, g3, g5, o5)
575 MOVE_LASTCHUNK(o1, o0, 0x30, g2, g3, g5, o5)
576 MOVE_LASTCHUNK(o1, o0, 0x20, g2, g3, g5, o5)
577 MOVE_LASTCHUNK(o1, o0, 0x10, g2, g3, g5, o5)
578 MOVE_LASTCHUNK(o1, o0, 0x00, g2, g3, g5, o5)
579 80: be,pt %xcc, 81f /* CTI */
580 andcc %o2, 4, %g0 /* IEU1 */
581 ldx [%o1], %g2 /* Load Group */
582 add %o0, 8, %o0 /* IEU0 */
583 stw %g2, [%o0 - 0x4] /* Store Group */
584 add %o1, 8, %o1 /* IEU1 */
585 srlx %g2, 32, %g2 /* IEU0 Group */
586 stw %g2, [%o0 - 0x8] /* Store */
587 81: be,pt %xcc, 1f /* CTI */
588 andcc %o2, 2, %g0 /* IEU1 Group */
589 lduw [%o1], %g2 /* Load Group */
590 add %o1, 4, %o1 /* IEU0 */
591 stw %g2, [%o0] /* Store Group */
592 add %o0, 4, %o0 /* IEU0 */
593 1: be,pt %xcc, 1f /* CTI */
594 andcc %o2, 1, %g0 /* IEU1 Group */
595 lduh [%o1], %g2 /* Load Group */
596 add %o1, 2, %o1 /* IEU0 */
597 sth %g2, [%o0] /* Store Group */
598 add %o0, 2, %o0 /* IEU0 */
599 1: be,pt %xcc, 211f /* CTI */
600 nop /* IEU1 */
601 ldub [%o1], %g2 /* Load Group */
602 stb %g2, [%o0] /* Store Group + bubble */
603 211: retl
604 mov %g4, %o0
606 82: MOVE_BIGALIGNCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
607 MOVE_BIGALIGNCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
608 37: subcc %g6, 128, %g6 /* IEU1 Group */
609 add %o1, 128, %o1 /* IEU0 */
610 bne,pt %xcc, 82b /* CTI */
611 add %o0, 128, %o0 /* IEU0 Group */
612 andcc %o2, 0x70, %g6 /* IEU1 */
613 be,pn %xcc, 84f /* CTI */
614 andcc %o2, 8, %g0 /* IEU1 Group */
615 /* Clk1 8-( */
616 /* Clk2 8-( */
617 /* Clk3 8-( */
618 /* Clk4 8-( */
619 83: rd %pc, %o5 /* PDU Group */
620 add %o1, %g6, %o1 /* IEU0 Group */
621 sub %o5, %g6, %o5 /* IEU1 */
622 jmpl %o5 + %lo(84f - 83b), %g0 /* CTI Group brk forced*/
623 add %o0, %g6, %o0 /* IEU0 Group */
624 38: MOVE_LASTALIGNCHUNK(o1, o0, 0x60, g2, g3)
625 MOVE_LASTALIGNCHUNK(o1, o0, 0x50, g2, g3)
626 MOVE_LASTALIGNCHUNK(o1, o0, 0x40, g2, g3)
627 MOVE_LASTALIGNCHUNK(o1, o0, 0x30, g2, g3)
628 MOVE_LASTALIGNCHUNK(o1, o0, 0x20, g2, g3)
629 MOVE_LASTALIGNCHUNK(o1, o0, 0x10, g2, g3)
630 MOVE_LASTALIGNCHUNK(o1, o0, 0x00, g2, g3)
631 84: be,pt %xcc, 85f /* CTI Group */
632 andcc %o2, 4, %g0 /* IEU1 */
633 ldx [%o1], %g2 /* Load Group */
634 add %o0, 8, %o0 /* IEU0 */
635 add %o1, 8, %o1 /* IEU0 Group */
636 stx %g2, [%o0 - 0x8] /* Store */
637 85: be,pt %xcc, 1f /* CTI */
638 andcc %o2, 2, %g0 /* IEU1 Group */
639 lduw [%o1], %g2 /* Load Group */
640 add %o0, 4, %o0 /* IEU0 */
641 add %o1, 4, %o1 /* IEU0 Group */
642 stw %g2, [%o0 - 0x4] /* Store */
643 1: be,pt %xcc, 1f /* CTI */
644 andcc %o2, 1, %g0 /* IEU1 Group */
645 lduh [%o1], %g2 /* Load Group */
646 add %o0, 2, %o0 /* IEU0 */
647 add %o1, 2, %o1 /* IEU0 Group */
648 sth %g2, [%o0 - 0x2] /* Store */
649 1: be,pt %xcc, 1f /* CTI */
650 nop /* IEU0 Group */
651 ldub [%o1], %g2 /* Load Group */
652 stb %g2, [%o0] /* Store Group + bubble */
653 1: retl
654 mov %g4, %o0
656 212: brz,pt %g2, 2f /* CTI Group */
657 mov 8, %g1 /* IEU0 */
658 sub %g1, %g2, %g2 /* IEU0 Group */
659 sub %o2, %g2, %o2 /* IEU0 Group */
660 1: ldub [%o1], %g5 /* Load Group */
661 add %o1, 1, %o1 /* IEU0 */
662 add %o0, 1, %o0 /* IEU1 */
663 subcc %g2, 1, %g2 /* IEU1 Group */
664 bne,pt %xcc, 1b /* CTI */
665 stb %g5, [%o0 - 1] /* Store */
666 2: andn %o2, 7, %g5 /* IEU0 Group */
667 and %o2, 7, %o2 /* IEU1 */
668 fmovd %f0, %f2 /* FPU */
669 alignaddr %o1, %g0, %g1 /* GRU Group */
670 ldd [%g1], %f4 /* Load Group */
671 1: ldd [%g1 + 0x8], %f6 /* Load Group */
672 add %g1, 0x8, %g1 /* IEU0 Group */
673 subcc %g5, 8, %g5 /* IEU1 */
674 faligndata %f4, %f6, %f0 /* GRU Group */
675 std %f0, [%o0] /* Store */
676 add %o1, 8, %o1 /* IEU0 Group */
677 be,pn %xcc, 213f /* CTI */
678 add %o0, 8, %o0 /* IEU1 */
679 ldd [%g1 + 0x8], %f4 /* Load Group */
680 add %g1, 8, %g1 /* IEU0 */
681 subcc %g5, 8, %g5 /* IEU1 */
682 faligndata %f6, %f4, %f0 /* GRU Group */
683 std %f0, [%o0] /* Store */
684 add %o1, 8, %o1 /* IEU0 */
685 bne,pn %xcc, 1b /* CTI Group */
686 add %o0, 8, %o0 /* IEU0 */
687 213: brz,pn %o2, 214f /* CTI Group */
688 nop /* IEU0 */
689 ldub [%o1], %g5 /* LOAD */
690 add %o1, 1, %o1 /* IEU0 */
691 add %o0, 1, %o0 /* IEU1 */
692 subcc %o2, 1, %o2 /* IEU1 */
693 bne,pt %xcc, 206b /* CTI */
694 stb %g5, [%o0 - 1] /* Store Group */
695 214: wr %g0, FPRS_FEF, %fprs
696 retl
697 mov %g4, %o0
698 END(memcpy)
700 .align 32
701 228: andcc %o2, 1, %g0 /* IEU1 Group */
702 be,pt %icc, 2f+4 /* CTI */
703 1: ldub [%o1 - 1], %o5 /* LOAD Group */
704 sub %o1, 1, %o1 /* IEU0 */
705 sub %o0, 1, %o0 /* IEU1 */
706 subcc %o2, 1, %o2 /* IEU1 Group */
707 be,pn %xcc, 229f /* CTI */
708 stb %o5, [%o0] /* Store */
709 2: ldub [%o1 - 1], %o5 /* LOAD Group */
710 sub %o0, 2, %o0 /* IEU0 */
711 ldub [%o1 - 2], %g5 /* LOAD Group */
712 sub %o1, 2, %o1 /* IEU0 */
713 subcc %o2, 2, %o2 /* IEU1 Group */
714 stb %o5, [%o0 + 1] /* Store */
715 bne,pt %xcc, 2b /* CTI */
716 stb %g5, [%o0] /* Store */
717 229: retl
718 mov %g4, %o0
719 219: retl
720 nop
722 .align 32
723 ENTRY(memmove)
724 #ifndef USE_BPR
725 srl %o2, 0, %o2 /* IEU1 Group */
726 #endif
727 brz,pn %o2, 219b /* CTI Group */
728 sub %o0, %o1, %o4 /* IEU0 */
729 cmp %o4, %o2 /* IEU1 Group */
730 bgeu,pt %XCC, 218b /* CTI */
731 mov %o0, %g4 /* IEU0 */
732 add %o0, %o2, %o0 /* IEU0 Group */
733 220: add %o1, %o2, %o1 /* IEU1 */
734 cmp %o2, 15 /* IEU1 Group */
735 bleu,pn %xcc, 228b /* CTI */
736 andcc %o0, 7, %g2 /* IEU1 Group */
737 sub %o0, %o1, %g5 /* IEU0 */
738 andcc %g5, 3, %o5 /* IEU1 Group */
739 bne,pn %xcc, 232f /* CTI */
740 andcc %o1, 3, %g0 /* IEU1 Group */
741 be,a,pt %xcc, 236f /* CTI */
742 andcc %o1, 4, %g0 /* IEU1 Group */
743 andcc %o1, 1, %g0 /* IEU1 Group */
744 be,pn %xcc, 4f /* CTI */
745 andcc %o1, 2, %g0 /* IEU1 Group */
746 ldub [%o1 - 1], %g2 /* Load Group */
747 sub %o1, 1, %o1 /* IEU0 */
748 sub %o0, 1, %o0 /* IEU1 */
749 sub %o2, 1, %o2 /* IEU0 Group */
750 be,pn %xcc, 5f /* CTI Group */
751 stb %g2, [%o0] /* Store */
752 4: lduh [%o1 - 2], %g2 /* Load Group */
753 sub %o1, 2, %o1 /* IEU0 */
754 sub %o0, 2, %o0 /* IEU1 */
755 sub %o2, 2, %o2 /* IEU0 */
756 sth %g2, [%o0] /* Store Group + bubble */
757 5: andcc %o1, 4, %g0 /* IEU1 */
758 236: be,a,pn %xcc, 2f /* CTI */
759 andcc %o2, -128, %g6 /* IEU1 Group */
760 lduw [%o1 - 4], %g5 /* Load Group */
761 sub %o1, 4, %o1 /* IEU0 */
762 sub %o0, 4, %o0 /* IEU1 */
763 sub %o2, 4, %o2 /* IEU0 Group */
764 stw %g5, [%o0] /* Store */
765 andcc %o2, -128, %g6 /* IEU1 Group */
766 2: be,pn %xcc, 235f /* CTI */
767 andcc %o0, 4, %g0 /* IEU1 Group */
768 be,pn %xcc, 282f + 4 /* CTI Group */
769 5: RMOVE_BIGCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
770 RMOVE_BIGCHUNK(o1, o0, 0x20, g1, g3, g5, o5)
771 RMOVE_BIGCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
772 RMOVE_BIGCHUNK(o1, o0, 0x60, g1, g3, g5, o5)
773 subcc %g6, 128, %g6 /* IEU1 Group */
774 sub %o1, 128, %o1 /* IEU0 */
775 bne,pt %xcc, 5b /* CTI */
776 sub %o0, 128, %o0 /* IEU0 Group */
777 235: andcc %o2, 0x70, %g6 /* IEU1 Group */
778 41: be,pn %xcc, 280f /* CTI */
779 andcc %o2, 8, %g0 /* IEU1 Group */
780 /* Clk1 8-( */
781 /* Clk2 8-( */
782 /* Clk3 8-( */
783 /* Clk4 8-( */
784 279: rd %pc, %o5 /* PDU Group */
785 sll %g6, 1, %g5 /* IEU0 Group */
786 sub %o1, %g6, %o1 /* IEU1 */
787 sub %o5, %g5, %o5 /* IEU0 Group */
788 jmpl %o5 + %lo(280f - 279b), %g0 /* CTI Group brk forced*/
789 sub %o0, %g6, %o0 /* IEU0 Group */
790 RMOVE_LASTCHUNK(o1, o0, 0x60, g2, g3, g5, o5)
791 RMOVE_LASTCHUNK(o1, o0, 0x50, g2, g3, g5, o5)
792 RMOVE_LASTCHUNK(o1, o0, 0x40, g2, g3, g5, o5)
793 RMOVE_LASTCHUNK(o1, o0, 0x30, g2, g3, g5, o5)
794 RMOVE_LASTCHUNK(o1, o0, 0x20, g2, g3, g5, o5)
795 RMOVE_LASTCHUNK(o1, o0, 0x10, g2, g3, g5, o5)
796 RMOVE_LASTCHUNK(o1, o0, 0x00, g2, g3, g5, o5)
797 280: be,pt %xcc, 281f /* CTI */
798 andcc %o2, 4, %g0 /* IEU1 */
799 ldx [%o1 - 8], %g2 /* Load Group */
800 sub %o0, 8, %o0 /* IEU0 */
801 stw %g2, [%o0 + 4] /* Store Group */
802 sub %o1, 8, %o1 /* IEU1 */
803 srlx %g2, 32, %g2 /* IEU0 Group */
804 stw %g2, [%o0] /* Store */
805 281: be,pt %xcc, 1f /* CTI */
806 andcc %o2, 2, %g0 /* IEU1 Group */
807 lduw [%o1 - 4], %g2 /* Load Group */
808 sub %o1, 4, %o1 /* IEU0 */
809 stw %g2, [%o0 - 4] /* Store Group */
810 sub %o0, 4, %o0 /* IEU0 */
811 1: be,pt %xcc, 1f /* CTI */
812 andcc %o2, 1, %g0 /* IEU1 Group */
813 lduh [%o1 - 2], %g2 /* Load Group */
814 sub %o1, 2, %o1 /* IEU0 */
815 sth %g2, [%o0 - 2] /* Store Group */
816 sub %o0, 2, %o0 /* IEU0 */
817 1: be,pt %xcc, 211f /* CTI */
818 nop /* IEU1 */
819 ldub [%o1 - 1], %g2 /* Load Group */
820 stb %g2, [%o0 - 1] /* Store Group + bubble */
821 211: retl
822 mov %g4, %o0
824 282: RMOVE_BIGALIGNCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
825 RMOVE_BIGALIGNCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
826 subcc %g6, 128, %g6 /* IEU1 Group */
827 sub %o1, 128, %o1 /* IEU0 */
828 bne,pt %xcc, 282b /* CTI */
829 sub %o0, 128, %o0 /* IEU0 Group */
830 andcc %o2, 0x70, %g6 /* IEU1 */
831 be,pn %xcc, 284f /* CTI */
832 andcc %o2, 8, %g0 /* IEU1 Group */
833 /* Clk1 8-( */
834 /* Clk2 8-( */
835 /* Clk3 8-( */
836 /* Clk4 8-( */
837 283: rd %pc, %o5 /* PDU Group */
838 sub %o1, %g6, %o1 /* IEU0 Group */
839 sub %o5, %g6, %o5 /* IEU1 */
840 jmpl %o5 + %lo(284f - 283b), %g0 /* CTI Group brk forced*/
841 sub %o0, %g6, %o0 /* IEU0 Group */
842 RMOVE_LASTALIGNCHUNK(o1, o0, 0x60, g2, g3)
843 RMOVE_LASTALIGNCHUNK(o1, o0, 0x50, g2, g3)
844 RMOVE_LASTALIGNCHUNK(o1, o0, 0x40, g2, g3)
845 RMOVE_LASTALIGNCHUNK(o1, o0, 0x30, g2, g3)
846 RMOVE_LASTALIGNCHUNK(o1, o0, 0x20, g2, g3)
847 RMOVE_LASTALIGNCHUNK(o1, o0, 0x10, g2, g3)
848 RMOVE_LASTALIGNCHUNK(o1, o0, 0x00, g2, g3)
849 284: be,pt %xcc, 285f /* CTI Group */
850 andcc %o2, 4, %g0 /* IEU1 */
851 ldx [%o1 - 8], %g2 /* Load Group */
852 sub %o0, 8, %o0 /* IEU0 */
853 sub %o1, 8, %o1 /* IEU0 Group */
854 stx %g2, [%o0] /* Store */
855 285: be,pt %xcc, 1f /* CTI */
856 andcc %o2, 2, %g0 /* IEU1 Group */
857 lduw [%o1 - 4], %g2 /* Load Group */
858 sub %o0, 4, %o0 /* IEU0 */
859 sub %o1, 4, %o1 /* IEU0 Group */
860 stw %g2, [%o0] /* Store */
861 1: be,pt %xcc, 1f /* CTI */
862 andcc %o2, 1, %g0 /* IEU1 Group */
863 lduh [%o1 - 2], %g2 /* Load Group */
864 sub %o0, 2, %o0 /* IEU0 */
865 sub %o1, 2, %o1 /* IEU0 Group */
866 sth %g2, [%o0] /* Store */
867 1: be,pt %xcc, 1f /* CTI */
868 nop /* IEU0 Group */
869 ldub [%o1 - 1], %g2 /* Load Group */
870 stb %g2, [%o0 - 1] /* Store Group + bubble */
871 1: retl
872 mov %g4, %o0
874 232: brz,pt %g2, 2f /* CTI Group */
875 sub %o2, %g2, %o2 /* IEU0 Group */
876 1: ldub [%o1 - 1], %g5 /* Load Group */
877 sub %o1, 1, %o1 /* IEU0 */
878 sub %o0, 1, %o0 /* IEU1 */
879 subcc %g2, 1, %g2 /* IEU1 Group */
880 bne,pt %xcc, 1b /* CTI */
881 stb %g5, [%o0] /* Store */
882 2: andn %o2, 7, %g5 /* IEU0 Group */
883 and %o2, 7, %o2 /* IEU1 */
884 fmovd %f0, %f2 /* FPU */
885 alignaddr %o1, %g0, %g1 /* GRU Group */
886 ldd [%g1], %f4 /* Load Group */
887 1: ldd [%g1 - 8], %f6 /* Load Group */
888 sub %g1, 8, %g1 /* IEU0 Group */
889 subcc %g5, 8, %g5 /* IEU1 */
890 faligndata %f6, %f4, %f0 /* GRU Group */
891 std %f0, [%o0 - 8] /* Store */
892 sub %o1, 8, %o1 /* IEU0 Group */
893 be,pn %xcc, 233f /* CTI */
894 sub %o0, 8, %o0 /* IEU1 */
895 ldd [%g1 - 8], %f4 /* Load Group */
896 sub %g1, 8, %g1 /* IEU0 */
897 subcc %g5, 8, %g5 /* IEU1 */
898 faligndata %f4, %f6, %f0 /* GRU Group */
899 std %f0, [%o0 - 8] /* Store */
900 sub %o1, 8, %o1 /* IEU0 */
901 bne,pn %xcc, 1b /* CTI Group */
902 sub %o0, 8, %o0 /* IEU0 */
903 233: brz,pn %o2, 234f /* CTI Group */
904 nop /* IEU0 */
905 237: ldub [%o1 - 1], %g5 /* LOAD */
906 sub %o1, 1, %o1 /* IEU0 */
907 sub %o0, 1, %o0 /* IEU1 */
908 subcc %o2, 1, %o2 /* IEU1 */
909 bne,pt %xcc, 237b /* CTI */
910 stb %g5, [%o0] /* Store Group */
911 234: wr %g0, FPRS_FEF, %fprs
912 retl
913 mov %g4, %o0
914 END(memmove)
916 #ifdef USE_BPR
917 weak_alias(memcpy, __align_cpy_1)
918 weak_alias(memcpy, __align_cpy_2)
919 #endif
920 libc_hidden_builtin_def (memcpy)
921 libc_hidden_builtin_def (memmove)