| blob | 97e4b024c69290200f90124e2de11e41100daa18 |
1 /* Copy SIZE bytes from SRC to DEST.
2 For UltraSPARC.
3 Copyright (C) 1996-2014 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, see
20 <http://www.gnu.org/licenses/>. */
22 #include <sysdep.h>
23 #include <asm/asi.h>
24 #ifndef XCC
25 #define USE_BPR
26 .register %g2, #scratch
27 .register %g3, #scratch
28 .register %g6, #scratch
29 #define XCC xcc
30 #endif
31 #define FPRS_FEF 4
33 #define FREG_FROB(f1, f2, f3, f4, f5, f6, f7, f8, f9) \
34 faligndata %f1, %f2, %f48; \
35 faligndata %f2, %f3, %f50; \
36 faligndata %f3, %f4, %f52; \
37 faligndata %f4, %f5, %f54; \
38 faligndata %f5, %f6, %f56; \
39 faligndata %f6, %f7, %f58; \
40 faligndata %f7, %f8, %f60; \
41 faligndata %f8, %f9, %f62;
43 #define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, len, jmptgt) \
44 ldda [%src] %asi, %fdest; \
45 add %src, 0x40, %src; \
46 add %dest, 0x40, %dest; \
47 subcc %len, 0x40, %len; \
48 be,pn %xcc, jmptgt; \
49 stda %fsrc, [%dest - 0x40] %asi;
51 #define LOOP_CHUNK1(src, dest, len, branch_dest) \
52 MAIN_LOOP_CHUNK(src, dest, f0, f48, len, branch_dest)
53 #define LOOP_CHUNK2(src, dest, len, branch_dest) \
54 MAIN_LOOP_CHUNK(src, dest, f16, f48, len, branch_dest)
55 #define LOOP_CHUNK3(src, dest, len, branch_dest) \
56 MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
58 #define STORE_SYNC(dest, fsrc) \
59 stda %fsrc, [%dest] %asi; \
60 add %dest, 0x40, %dest;
62 #define STORE_JUMP(dest, fsrc, target) \
63 stda %fsrc, [%dest] %asi; \
64 add %dest, 0x40, %dest; \
65 ba,pt %xcc, target;
67 #define VISLOOP_PAD nop; nop; nop; nop; \
68 nop; nop; nop; nop; \
69 nop; nop; nop; nop; \
70 nop; nop; nop;
72 #define FINISH_VISCHUNK(dest, f0, f1, left) \
73 subcc %left, 8, %left; \
74 bl,pn %xcc, 205f; \
75 faligndata %f0, %f1, %f48; \
76 std %f48, [%dest]; \
77 add %dest, 8, %dest;
79 #define UNEVEN_VISCHUNK(dest, f0, f1, left) \
80 subcc %left, 8, %left; \
81 bl,pn %xcc, 205f; \
82 fsrc2 %f0, %f1; \
83 ba,a,pt %xcc, 204f;
85 /* Macros for non-VIS memcpy code. */
86 #define MOVE_BIGCHUNK(src, dst, offset, t0, t1, t2, t3) \
87 ldx [%src + offset + 0x00], %t0; \
88 ldx [%src + offset + 0x08], %t1; \
89 ldx [%src + offset + 0x10], %t2; \
90 ldx [%src + offset + 0x18], %t3; \
91 stw %t0, [%dst + offset + 0x04]; \
92 srlx %t0, 32, %t0; \
93 stw %t0, [%dst + offset + 0x00]; \
94 stw %t1, [%dst + offset + 0x0c]; \
95 srlx %t1, 32, %t1; \
96 stw %t1, [%dst + offset + 0x08]; \
97 stw %t2, [%dst + offset + 0x14]; \
98 srlx %t2, 32, %t2; \
99 stw %t2, [%dst + offset + 0x10]; \
100 stw %t3, [%dst + offset + 0x1c]; \
101 srlx %t3, 32, %t3; \
102 stw %t3, [%dst + offset + 0x18];
104 #define MOVE_BIGALIGNCHUNK(src, dst, offset, t0, t1, t2, t3) \
105 ldx [%src + offset + 0x00], %t0; \
106 ldx [%src + offset + 0x08], %t1; \
107 ldx [%src + offset + 0x10], %t2; \
108 ldx [%src + offset + 0x18], %t3; \
109 stx %t0, [%dst + offset + 0x00]; \
110 stx %t1, [%dst + offset + 0x08]; \
111 stx %t2, [%dst + offset + 0x10]; \
112 stx %t3, [%dst + offset + 0x18]; \
113 ldx [%src + offset + 0x20], %t0; \
114 ldx [%src + offset + 0x28], %t1; \
115 ldx [%src + offset + 0x30], %t2; \
116 ldx [%src + offset + 0x38], %t3; \
117 stx %t0, [%dst + offset + 0x20]; \
118 stx %t1, [%dst + offset + 0x28]; \
119 stx %t2, [%dst + offset + 0x30]; \
120 stx %t3, [%dst + offset + 0x38];
122 #define MOVE_LASTCHUNK(src, dst, offset, t0, t1, t2, t3) \
123 ldx [%src - offset - 0x10], %t0; \
124 ldx [%src - offset - 0x08], %t1; \
125 stw %t0, [%dst - offset - 0x0c]; \
126 srlx %t0, 32, %t2; \
127 stw %t2, [%dst - offset - 0x10]; \
128 stw %t1, [%dst - offset - 0x04]; \
129 srlx %t1, 32, %t3; \
130 stw %t3, [%dst - offset - 0x08];
132 #define MOVE_LASTALIGNCHUNK(src, dst, offset, t0, t1) \
133 ldx [%src - offset - 0x10], %t0; \
134 ldx [%src - offset - 0x08], %t1; \
135 stx %t0, [%dst - offset - 0x10]; \
136 stx %t1, [%dst - offset - 0x08];
138 .text
139 .align 32
140 ENTRY(__memcpy_large)
141 200: be,pt %xcc, 201f /* CTI */
142 andcc %o0, 0x38, %g5 /* IEU1 Group */
143 mov 8, %g1 /* IEU0 */
144 sub %g1, %g2, %g2 /* IEU0 Group */
145 andcc %o0, 1, %g0 /* IEU1 */
146 be,pt %icc, 2f /* CTI */
147 sub %o2, %g2, %o2 /* IEU0 Group */
148 1: ldub [%o1], %o5 /* Load Group */
149 add %o1, 1, %o1 /* IEU0 */
150 add %o0, 1, %o0 /* IEU1 */
151 subcc %g2, 1, %g2 /* IEU1 Group */
152 be,pn %xcc, 3f /* CTI */
153 stb %o5, [%o0 - 1] /* Store */
154 2: ldub [%o1], %o5 /* Load Group */
155 add %o0, 2, %o0 /* IEU0 */
156 ldub [%o1 + 1], %g3 /* Load Group */
157 subcc %g2, 2, %g2 /* IEU1 Group */
158 stb %o5, [%o0 - 2] /* Store */
159 add %o1, 2, %o1 /* IEU0 */
160 bne,pt %xcc, 2b /* CTI Group */
161 stb %g3, [%o0 - 1] /* Store */
162 3: andcc %o0, 0x38, %g5 /* IEU1 Group */
163 201: be,pt %icc, 202f /* CTI */
164 mov 64, %g1 /* IEU0 */
165 fsrc2 %f0, %f2 /* FPU */
166 sub %g1, %g5, %g5 /* IEU0 Group */
167 alignaddr %o1, %g0, %g1 /* GRU Group */
168 ldd [%g1], %f4 /* Load Group */
169 sub %o2, %g5, %o2 /* IEU0 */
170 1: ldd [%g1 + 0x8], %f6 /* Load Group */
171 add %g1, 0x8, %g1 /* IEU0 Group */
172 subcc %g5, 8, %g5 /* IEU1 */
173 faligndata %f4, %f6, %f0 /* GRU Group */
174 std %f0, [%o0] /* Store */
175 add %o1, 8, %o1 /* IEU0 Group */
176 be,pn %xcc, 202f /* CTI */
177 add %o0, 8, %o0 /* IEU1 */
178 ldd [%g1 + 0x8], %f4 /* Load Group */
179 add %g1, 8, %g1 /* IEU0 */
180 subcc %g5, 8, %g5 /* IEU1 */
181 faligndata %f6, %f4, %f0 /* GRU Group */
182 std %f0, [%o0] /* Store */
183 add %o1, 8, %o1 /* IEU0 */
184 bne,pt %xcc, 1b /* CTI Group */
185 add %o0, 8, %o0 /* IEU0 */
186 202: membar #LoadStore | #StoreStore | #StoreLoad /* LSU Group */
187 wr %g0, ASI_BLK_P, %asi /* LSU Group */
188 subcc %o2, 0x40, %g6 /* IEU1 Group */
189 mov %o1, %g1 /* IEU0 */
190 andncc %g6, (0x40 - 1), %g6 /* IEU1 Group */
191 srl %g1, 3, %g2 /* IEU0 */
192 sub %o2, %g6, %g3 /* IEU0 Group */
193 andn %o1, (0x40 - 1), %o1 /* IEU1 */
194 and %g2, 7, %g2 /* IEU0 Group */
195 andncc %g3, 0x7, %g3 /* IEU1 */
196 fsrc2 %f0, %f2 /* FPU */
197 sub %g3, 0x10, %g3 /* IEU0 Group */
198 sub %o2, %g6, %o2 /* IEU1 */
199 alignaddr %g1, %g0, %g0 /* GRU Group */
200 add %g1, %g6, %g1 /* IEU0 Group */
201 subcc %o2, %g3, %o2 /* IEU1 */
202 ldda [%o1 + 0x00] %asi, %f0 /* LSU Group */
203 add %g1, %g3, %g1 /* IEU0 */
204 ldda [%o1 + 0x40] %asi, %f16 /* LSU Group */
205 sub %g6, 0x80, %g6 /* IEU0 */
206 ldda [%o1 + 0x80] %asi, %f32 /* LSU Group */
207 /* Clk1 Group 8-( */
208 /* Clk2 Group 8-( */
209 /* Clk3 Group 8-( */
210 /* Clk4 Group 8-( */
211 203: rd %pc, %g5 /* PDU Group 8-( */
212 addcc %g5, %lo(300f - 203b), %g5 /* IEU1 Group */
213 sll %g2, 9, %g2 /* IEU0 */
214 jmpl %g5 + %g2, %g0 /* CTI Group brk forced*/
215 addcc %o1, 0xc0, %o1 /* IEU1 Group */
217 .align 512 /* OK, here comes the fun part... */
218 300: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) LOOP_CHUNK1(o1, o0, g6, 301f)
219 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) LOOP_CHUNK2(o1, o0, g6, 302f)
220 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) LOOP_CHUNK3(o1, o0, g6, 303f)
221 b,pt %xcc, 300b+4; faligndata %f0, %f2, %f48
222 301: FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) STORE_SYNC(o0, f48) membar #Sync
223 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) STORE_JUMP(o0, f48, 400f) membar #Sync
224 302: FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0) STORE_SYNC(o0, f48) membar #Sync
225 FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) STORE_JUMP(o0, f48, 416f) membar #Sync
226 303: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16) STORE_SYNC(o0, f48) membar #Sync
227 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32) STORE_JUMP(o0, f48, 432f) membar #Sync
228 VISLOOP_PAD
229 310: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) LOOP_CHUNK1(o1, o0, g6, 311f)
230 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) LOOP_CHUNK2(o1, o0, g6, 312f)
231 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) LOOP_CHUNK3(o1, o0, g6, 313f)
232 b,pt %xcc, 310b+4; faligndata %f2, %f4, %f48
233 311: FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) STORE_SYNC(o0, f48) membar #Sync
234 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) STORE_JUMP(o0, f48, 402f) membar #Sync
235 312: FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2) STORE_SYNC(o0, f48) membar #Sync
236 FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) STORE_JUMP(o0, f48, 418f) membar #Sync
237 313: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18) STORE_SYNC(o0, f48) membar #Sync
238 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34) STORE_JUMP(o0, f48, 434f) membar #Sync
239 VISLOOP_PAD
240 320: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) LOOP_CHUNK1(o1, o0, g6, 321f)
241 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) LOOP_CHUNK2(o1, o0, g6, 322f)
242 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) LOOP_CHUNK3(o1, o0, g6, 323f)
243 b,pt %xcc, 320b+4; faligndata %f4, %f6, %f48
244 321: FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) STORE_SYNC(o0, f48) membar #Sync
245 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) STORE_JUMP(o0, f48, 404f) membar #Sync
246 322: FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4) STORE_SYNC(o0, f48) membar #Sync
247 FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) STORE_JUMP(o0, f48, 420f) membar #Sync
248 323: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20) STORE_SYNC(o0, f48) membar #Sync
249 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36) STORE_JUMP(o0, f48, 436f) membar #Sync
250 VISLOOP_PAD
251 330: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) LOOP_CHUNK1(o1, o0, g6, 331f)
252 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) LOOP_CHUNK2(o1, o0, g6, 332f)
253 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) LOOP_CHUNK3(o1, o0, g6, 333f)
254 b,pt %xcc, 330b+4; faligndata %f6, %f8, %f48
255 331: FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) STORE_SYNC(o0, f48) membar #Sync
256 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) STORE_JUMP(o0, f48, 406f) membar #Sync
257 332: FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6) STORE_SYNC(o0, f48) membar #Sync
258 FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) STORE_JUMP(o0, f48, 422f) membar #Sync
259 333: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22) STORE_SYNC(o0, f48) membar #Sync
260 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38) STORE_JUMP(o0, f48, 438f) membar #Sync
261 VISLOOP_PAD
262 340: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) LOOP_CHUNK1(o1, o0, g6, 341f)
263 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) LOOP_CHUNK2(o1, o0, g6, 342f)
264 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) LOOP_CHUNK3(o1, o0, g6, 343f)
265 b,pt %xcc, 340b+4; faligndata %f8, %f10, %f48
266 341: FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) STORE_SYNC(o0, f48) membar #Sync
267 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) STORE_JUMP(o0, f48, 408f) membar #Sync
268 342: FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8) STORE_SYNC(o0, f48) membar #Sync
269 FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) STORE_JUMP(o0, f48, 424f) membar #Sync
270 343: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24) STORE_SYNC(o0, f48) membar #Sync
271 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40) STORE_JUMP(o0, f48, 440f) membar #Sync
272 VISLOOP_PAD
273 350: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) LOOP_CHUNK1(o1, o0, g6, 351f)
274 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) LOOP_CHUNK2(o1, o0, g6, 352f)
275 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) LOOP_CHUNK3(o1, o0, g6, 353f)
276 b,pt %xcc, 350b+4; faligndata %f10, %f12, %f48
277 351: FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) STORE_SYNC(o0, f48) membar #Sync
278 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) STORE_JUMP(o0, f48, 410f) membar #Sync
279 352: FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10) STORE_SYNC(o0, f48) membar #Sync
280 FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) STORE_JUMP(o0, f48, 426f) membar #Sync
281 353: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26) STORE_SYNC(o0, f48) membar #Sync
282 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42) STORE_JUMP(o0, f48, 442f) membar #Sync
283 VISLOOP_PAD
284 360: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) LOOP_CHUNK1(o1, o0, g6, 361f)
285 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) LOOP_CHUNK2(o1, o0, g6, 362f)
286 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) LOOP_CHUNK3(o1, o0, g6, 363f)
287 b,pt %xcc, 360b+4; faligndata %f12, %f14, %f48
288 361: FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) STORE_SYNC(o0, f48) membar #Sync
289 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) STORE_JUMP(o0, f48, 412f) membar #Sync
290 362: FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12) STORE_SYNC(o0, f48) membar #Sync
291 FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) STORE_JUMP(o0, f48, 428f) membar #Sync
292 363: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28) STORE_SYNC(o0, f48) membar #Sync
293 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44) STORE_JUMP(o0, f48, 444f) membar #Sync
294 VISLOOP_PAD
295 370: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) LOOP_CHUNK1(o1, o0, g6, 371f)
296 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) LOOP_CHUNK2(o1, o0, g6, 372f)
297 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) LOOP_CHUNK3(o1, o0, g6, 373f)
298 b,pt %xcc, 370b+4; faligndata %f14, %f16, %f48
299 371: FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) STORE_SYNC(o0, f48) membar #Sync
300 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) STORE_JUMP(o0, f48, 414f) membar #Sync
301 372: FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14) STORE_SYNC(o0, f48) membar #Sync
302 FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) STORE_JUMP(o0, f48, 430f) membar #Sync
303 373: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30) STORE_SYNC(o0, f48) membar #Sync
304 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46) STORE_JUMP(o0, f48, 446f) membar #Sync
305 VISLOOP_PAD
306 400: FINISH_VISCHUNK(o0, f0, f2, g3)
307 402: FINISH_VISCHUNK(o0, f2, f4, g3)
308 404: FINISH_VISCHUNK(o0, f4, f6, g3)
309 406: FINISH_VISCHUNK(o0, f6, f8, g3)
310 408: FINISH_VISCHUNK(o0, f8, f10, g3)
311 410: FINISH_VISCHUNK(o0, f10, f12, g3)
312 412: FINISH_VISCHUNK(o0, f12, f14, g3)
313 414: UNEVEN_VISCHUNK(o0, f14, f0, g3)
314 416: FINISH_VISCHUNK(o0, f16, f18, g3)
315 418: FINISH_VISCHUNK(o0, f18, f20, g3)
316 420: FINISH_VISCHUNK(o0, f20, f22, g3)
317 422: FINISH_VISCHUNK(o0, f22, f24, g3)
318 424: FINISH_VISCHUNK(o0, f24, f26, g3)
319 426: FINISH_VISCHUNK(o0, f26, f28, g3)
320 428: FINISH_VISCHUNK(o0, f28, f30, g3)
321 430: UNEVEN_VISCHUNK(o0, f30, f0, g3)
322 432: FINISH_VISCHUNK(o0, f32, f34, g3)
323 434: FINISH_VISCHUNK(o0, f34, f36, g3)
324 436: FINISH_VISCHUNK(o0, f36, f38, g3)
325 438: FINISH_VISCHUNK(o0, f38, f40, g3)
326 440: FINISH_VISCHUNK(o0, f40, f42, g3)
327 442: FINISH_VISCHUNK(o0, f42, f44, g3)
328 444: FINISH_VISCHUNK(o0, f44, f46, g3)
329 446: UNEVEN_VISCHUNK(o0, f46, f0, g3)
330 204: ldd [%o1], %f2 /* Load Group */
331 add %o1, 8, %o1 /* IEU0 */
332 subcc %g3, 8, %g3 /* IEU1 */
333 faligndata %f0, %f2, %f8 /* GRU Group */
334 std %f8, [%o0] /* Store */
335 bl,pn %xcc, 205f /* CTI */
336 add %o0, 8, %o0 /* IEU0 Group */
337 ldd [%o1], %f0 /* Load Group */
338 add %o1, 8, %o1 /* IEU0 */
339 subcc %g3, 8, %g3 /* IEU1 */
340 faligndata %f2, %f0, %f8 /* GRU Group */
341 std %f8, [%o0] /* Store */
342 bge,pt %xcc, 204b /* CTI */
343 add %o0, 8, %o0 /* IEU0 Group */
344 205: brz,pt %o2, 207f /* CTI Group */
345 mov %g1, %o1 /* IEU0 */
346 206: ldub [%o1], %g5 /* LOAD */
347 add %o1, 1, %o1 /* IEU0 */
348 add %o0, 1, %o0 /* IEU1 */
349 subcc %o2, 1, %o2 /* IEU1 */
350 bne,pt %xcc, 206b /* CTI */
351 stb %g5, [%o0 - 1] /* Store Group */
352 207: membar #StoreLoad | #StoreStore /* LSU Group */
353 wr %g0, FPRS_FEF, %fprs
354 retl
355 mov %g4, %o0
357 208: andcc %o2, 1, %g0 /* IEU1 Group */
358 be,pt %icc, 2f+4 /* CTI */
359 1: ldub [%o1], %g5 /* LOAD Group */
360 add %o1, 1, %o1 /* IEU0 */
361 add %o0, 1, %o0 /* IEU1 */
362 subcc %o2, 1, %o2 /* IEU1 Group */
363 be,pn %xcc, 209f /* CTI */
364 stb %g5, [%o0 - 1] /* Store */
365 2: ldub [%o1], %g5 /* LOAD Group */
366 add %o0, 2, %o0 /* IEU0 */
367 ldub [%o1 + 1], %o5 /* LOAD Group */
368 add %o1, 2, %o1 /* IEU0 */
369 subcc %o2, 2, %o2 /* IEU1 Group */
370 stb %g5, [%o0 - 2] /* Store */
371 bne,pt %xcc, 2b /* CTI */
372 stb %o5, [%o0 - 1] /* Store */
373 209: retl
374 mov %g4, %o0
375 END(__memcpy_large)
377 ENTRY(__mempcpy)
378 ba,pt %xcc, 210f
379 add %o0, %o2, %g4
380 END(__mempcpy)
382 .align 32
383 ENTRY(memcpy)
384 mov %o0, %g4 /* IEU0 Group */
385 210:
386 #ifndef USE_BPR
387 srl %o2, 0, %o2 /* IEU1 */
388 #endif
389 brz,pn %o2, 209b /* CTI Group */
390 218: cmp %o2, 15 /* IEU1 */
391 bleu,pn %xcc, 208b /* CTI Group */
392 cmp %o2, (64 * 6) /* IEU1 */
393 bgeu,pn %xcc, 200b /* CTI Group */
394 andcc %o0, 7, %g2 /* IEU1 */
395 sub %o0, %o1, %g5 /* IEU0 */
396 andcc %g5, 3, %o5 /* IEU1 Group */
397 bne,pn %xcc, 212f /* CTI */
398 andcc %o1, 3, %g0 /* IEU1 Group */
399 be,a,pt %xcc, 216f /* CTI */
400 andcc %o1, 4, %g0 /* IEU1 Group */
401 andcc %o1, 1, %g0 /* IEU1 Group */
402 be,pn %xcc, 4f /* CTI */
403 andcc %o1, 2, %g0 /* IEU1 Group */
404 ldub [%o1], %g2 /* Load Group */
405 add %o1, 1, %o1 /* IEU0 */
406 add %o0, 1, %o0 /* IEU1 */
407 sub %o2, 1, %o2 /* IEU0 Group */
408 bne,pn %xcc, 5f /* CTI Group */
409 stb %g2, [%o0 - 1] /* Store */
410 4: lduh [%o1], %g2 /* Load Group */
411 add %o1, 2, %o1 /* IEU0 */
412 add %o0, 2, %o0 /* IEU1 */
413 sub %o2, 2, %o2 /* IEU0 */
414 sth %g2, [%o0 - 2] /* Store Group + bubble */
415 5: andcc %o1, 4, %g0 /* IEU1 */
416 216: be,a,pn %xcc, 2f /* CTI */
417 andcc %o2, -128, %g6 /* IEU1 Group */
418 lduw [%o1], %g5 /* Load Group */
419 add %o1, 4, %o1 /* IEU0 */
420 add %o0, 4, %o0 /* IEU1 */
421 sub %o2, 4, %o2 /* IEU0 Group */
422 stw %g5, [%o0 - 4] /* Store */
423 andcc %o2, -128, %g6 /* IEU1 Group */
424 2: be,pn %xcc, 215f /* CTI */
425 andcc %o0, 4, %g0 /* IEU1 Group */
426 be,pn %xcc, 82f + 4 /* CTI Group */
427 5: MOVE_BIGCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
428 MOVE_BIGCHUNK(o1, o0, 0x20, g1, g3, g5, o5)
429 MOVE_BIGCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
430 MOVE_BIGCHUNK(o1, o0, 0x60, g1, g3, g5, o5)
431 35: subcc %g6, 128, %g6 /* IEU1 Group */
432 add %o1, 128, %o1 /* IEU0 */
433 bne,pt %xcc, 5b /* CTI */
434 add %o0, 128, %o0 /* IEU0 Group */
435 215: andcc %o2, 0x70, %g6 /* IEU1 Group */
436 41: be,pn %xcc, 80f /* CTI */
437 andcc %o2, 8, %g0 /* IEU1 Group */
438 /* Clk1 8-( */
439 /* Clk2 8-( */
440 /* Clk3 8-( */
441 /* Clk4 8-( */
442 79: rd %pc, %o5 /* PDU Group */
443 sll %g6, 1, %g5 /* IEU0 Group */
444 add %o1, %g6, %o1 /* IEU1 */
445 sub %o5, %g5, %o5 /* IEU0 Group */
446 jmpl %o5 + %lo(80f - 79b), %g0 /* CTI Group brk forced*/
447 add %o0, %g6, %o0 /* IEU0 Group */
448 36: MOVE_LASTCHUNK(o1, o0, 0x60, g2, g3, g5, o5)
449 MOVE_LASTCHUNK(o1, o0, 0x50, g2, g3, g5, o5)
450 MOVE_LASTCHUNK(o1, o0, 0x40, g2, g3, g5, o5)
451 MOVE_LASTCHUNK(o1, o0, 0x30, g2, g3, g5, o5)
452 MOVE_LASTCHUNK(o1, o0, 0x20, g2, g3, g5, o5)
453 MOVE_LASTCHUNK(o1, o0, 0x10, g2, g3, g5, o5)
454 MOVE_LASTCHUNK(o1, o0, 0x00, g2, g3, g5, o5)
455 80: be,pt %xcc, 81f /* CTI */
456 andcc %o2, 4, %g0 /* IEU1 */
457 ldx [%o1], %g2 /* Load Group */
458 add %o0, 8, %o0 /* IEU0 */
459 stw %g2, [%o0 - 0x4] /* Store Group */
460 add %o1, 8, %o1 /* IEU1 */
461 srlx %g2, 32, %g2 /* IEU0 Group */
462 stw %g2, [%o0 - 0x8] /* Store */
463 81: be,pt %xcc, 1f /* CTI */
464 andcc %o2, 2, %g0 /* IEU1 Group */
465 lduw [%o1], %g2 /* Load Group */
466 add %o1, 4, %o1 /* IEU0 */
467 stw %g2, [%o0] /* Store Group */
468 add %o0, 4, %o0 /* IEU0 */
469 1: be,pt %xcc, 1f /* CTI */
470 andcc %o2, 1, %g0 /* IEU1 Group */
471 lduh [%o1], %g2 /* Load Group */
472 add %o1, 2, %o1 /* IEU0 */
473 sth %g2, [%o0] /* Store Group */
474 add %o0, 2, %o0 /* IEU0 */
475 1: be,pt %xcc, 211f /* CTI */
476 nop /* IEU1 */
477 ldub [%o1], %g2 /* Load Group */
478 stb %g2, [%o0] /* Store Group + bubble */
479 211: retl
480 mov %g4, %o0
482 82: MOVE_BIGALIGNCHUNK(o1, o0, 0x00, g1, g3, g5, o5)
483 MOVE_BIGALIGNCHUNK(o1, o0, 0x40, g1, g3, g5, o5)
484 37: subcc %g6, 128, %g6 /* IEU1 Group */
485 add %o1, 128, %o1 /* IEU0 */
486 bne,pt %xcc, 82b /* CTI */
487 add %o0, 128, %o0 /* IEU0 Group */
488 andcc %o2, 0x70, %g6 /* IEU1 */
489 be,pn %xcc, 84f /* CTI */
490 andcc %o2, 8, %g0 /* IEU1 Group */
491 /* Clk1 8-( */
492 /* Clk2 8-( */
493 /* Clk3 8-( */
494 /* Clk4 8-( */
495 83: rd %pc, %o5 /* PDU Group */
496 add %o1, %g6, %o1 /* IEU0 Group */
497 sub %o5, %g6, %o5 /* IEU1 */
498 jmpl %o5 + %lo(84f - 83b), %g0 /* CTI Group brk forced*/
499 add %o0, %g6, %o0 /* IEU0 Group */
500 38: MOVE_LASTALIGNCHUNK(o1, o0, 0x60, g2, g3)
501 MOVE_LASTALIGNCHUNK(o1, o0, 0x50, g2, g3)
502 MOVE_LASTALIGNCHUNK(o1, o0, 0x40, g2, g3)
503 MOVE_LASTALIGNCHUNK(o1, o0, 0x30, g2, g3)
504 MOVE_LASTALIGNCHUNK(o1, o0, 0x20, g2, g3)
505 MOVE_LASTALIGNCHUNK(o1, o0, 0x10, g2, g3)
506 MOVE_LASTALIGNCHUNK(o1, o0, 0x00, g2, g3)
507 84: be,pt %xcc, 85f /* CTI Group */
508 andcc %o2, 4, %g0 /* IEU1 */
509 ldx [%o1], %g2 /* Load Group */
510 add %o0, 8, %o0 /* IEU0 */
511 add %o1, 8, %o1 /* IEU0 Group */
512 stx %g2, [%o0 - 0x8] /* Store */
513 85: be,pt %xcc, 1f /* CTI */
514 andcc %o2, 2, %g0 /* IEU1 Group */
515 lduw [%o1], %g2 /* Load Group */
516 add %o0, 4, %o0 /* IEU0 */
517 add %o1, 4, %o1 /* IEU0 Group */
518 stw %g2, [%o0 - 0x4] /* Store */
519 1: be,pt %xcc, 1f /* CTI */
520 andcc %o2, 1, %g0 /* IEU1 Group */
521 lduh [%o1], %g2 /* Load Group */
522 add %o0, 2, %o0 /* IEU0 */
523 add %o1, 2, %o1 /* IEU0 Group */
524 sth %g2, [%o0 - 0x2] /* Store */
525 1: be,pt %xcc, 1f /* CTI */
526 nop /* IEU0 Group */
527 ldub [%o1], %g2 /* Load Group */
528 stb %g2, [%o0] /* Store Group + bubble */
529 1: retl
530 mov %g4, %o0
532 212: brz,pt %g2, 2f /* CTI Group */
533 mov 8, %g1 /* IEU0 */
534 sub %g1, %g2, %g2 /* IEU0 Group */
535 sub %o2, %g2, %o2 /* IEU0 Group */
536 1: ldub [%o1], %g5 /* Load Group */
537 add %o1, 1, %o1 /* IEU0 */
538 add %o0, 1, %o0 /* IEU1 */
539 subcc %g2, 1, %g2 /* IEU1 Group */
540 bne,pt %xcc, 1b /* CTI */
541 stb %g5, [%o0 - 1] /* Store */
542 2: andn %o2, 7, %g5 /* IEU0 Group */
543 and %o2, 7, %o2 /* IEU1 */
544 fsrc2 %f0, %f2 /* FPU */
545 alignaddr %o1, %g0, %g1 /* GRU Group */
546 ldd [%g1], %f4 /* Load Group */
547 1: ldd [%g1 + 0x8], %f6 /* Load Group */
548 add %g1, 0x8, %g1 /* IEU0 Group */
549 subcc %g5, 8, %g5 /* IEU1 */
550 faligndata %f4, %f6, %f0 /* GRU Group */
551 std %f0, [%o0] /* Store */
552 add %o1, 8, %o1 /* IEU0 Group */
553 be,pn %xcc, 213f /* CTI */
554 add %o0, 8, %o0 /* IEU1 */
555 ldd [%g1 + 0x8], %f4 /* Load Group */
556 add %g1, 8, %g1 /* IEU0 */
557 subcc %g5, 8, %g5 /* IEU1 */
558 faligndata %f6, %f4, %f0 /* GRU Group */
559 std %f0, [%o0] /* Store */
560 add %o1, 8, %o1 /* IEU0 */
561 bne,pn %xcc, 1b /* CTI Group */
562 add %o0, 8, %o0 /* IEU0 */
563 213: brz,pn %o2, 214f /* CTI Group */
564 nop /* IEU0 */
565 ldub [%o1], %g5 /* LOAD */
566 add %o1, 1, %o1 /* IEU0 */
567 add %o0, 1, %o0 /* IEU1 */
568 subcc %o2, 1, %o2 /* IEU1 */
569 bne,pt %xcc, 206b /* CTI */
570 stb %g5, [%o0 - 1] /* Store Group */
571 214: wr %g0, FPRS_FEF, %fprs
572 retl
573 mov %g4, %o0
574 END(memcpy)
576 libc_hidden_builtin_def (memcpy)
578 libc_hidden_def (__mempcpy)
579 weak_alias (__mempcpy, mempcpy)
580 libc_hidden_builtin_def (mempcpy)