1 /* SPDX-License-Identifier: GPL-2.0 */ <<
2 /* U1memcpy.S: UltraSPARC-I/II/IIi/IIe optimiz 1 /* U1memcpy.S: UltraSPARC-I/II/IIi/IIe optimized memcpy.
3 * 2 *
4 * Copyright (C) 1997, 2004 David S. Miller (d 3 * Copyright (C) 1997, 2004 David S. Miller (davem@redhat.com)
5 * Copyright (C) 1996, 1997, 1998, 1999 Jakub 4 * Copyright (C) 1996, 1997, 1998, 1999 Jakub Jelinek (jj@ultra.linux.cz)
6 */ 5 */
7 6
8 #ifdef __KERNEL__ 7 #ifdef __KERNEL__
9 #include <linux/export.h> <<
10 #include <linux/linkage.h> <<
11 #include <asm/visasm.h> 8 #include <asm/visasm.h>
12 #include <asm/asi.h> 9 #include <asm/asi.h>
13 #define GLOBAL_SPARE g7 10 #define GLOBAL_SPARE g7
14 #else 11 #else
15 #define GLOBAL_SPARE g5 12 #define GLOBAL_SPARE g5
16 #define ASI_BLK_P 0xf0 13 #define ASI_BLK_P 0xf0
17 #define FPRS_FEF 0x04 14 #define FPRS_FEF 0x04
18 #ifdef MEMCPY_DEBUG 15 #ifdef MEMCPY_DEBUG
19 #define VISEntry rd %fprs, %o5; wr %g0, FPRS_F 16 #define VISEntry rd %fprs, %o5; wr %g0, FPRS_FEF, %fprs; \
20 clr %g1; clr %g2; clr %g3; su 17 clr %g1; clr %g2; clr %g3; subcc %g0, %g0, %g0;
21 #define VISExit and %o5, FPRS_FEF, %o5; wr %o5 18 #define VISExit and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs
22 #else 19 #else
23 #define VISEntry rd %fprs, %o5; wr %g0, FPRS_F 20 #define VISEntry rd %fprs, %o5; wr %g0, FPRS_FEF, %fprs
24 #define VISExit and %o5, FPRS_FEF, %o5; wr %o5 21 #define VISExit and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs
25 #endif 22 #endif
26 #endif 23 #endif
27 24
28 #ifndef EX_LD 25 #ifndef EX_LD
29 #define EX_LD(x,y) x !! 26 #define EX_LD(x) x
30 #endif <<
31 #ifndef EX_LD_FP <<
32 #define EX_LD_FP(x,y) x <<
33 #endif 27 #endif
34 28
35 #ifndef EX_ST 29 #ifndef EX_ST
36 #define EX_ST(x,y) x !! 30 #define EX_ST(x) x
37 #endif 31 #endif
38 #ifndef EX_ST_FP !! 32
39 #define EX_ST_FP(x,y) x !! 33 #ifndef EX_RETVAL
>> 34 #define EX_RETVAL(x) x
40 #endif 35 #endif
41 36
42 #ifndef LOAD 37 #ifndef LOAD
43 #define LOAD(type,addr,dest) type [addr], d 38 #define LOAD(type,addr,dest) type [addr], dest
44 #endif 39 #endif
45 40
46 #ifndef LOAD_BLK 41 #ifndef LOAD_BLK
47 #define LOAD_BLK(addr,dest) ldda [addr] AS 42 #define LOAD_BLK(addr,dest) ldda [addr] ASI_BLK_P, dest
48 #endif 43 #endif
49 44
50 #ifndef STORE 45 #ifndef STORE
51 #define STORE(type,src,addr) type src, [add 46 #define STORE(type,src,addr) type src, [addr]
52 #endif 47 #endif
53 48
54 #ifndef STORE_BLK 49 #ifndef STORE_BLK
55 #define STORE_BLK(src,addr) stda src, [add 50 #define STORE_BLK(src,addr) stda src, [addr] ASI_BLK_P
56 #endif 51 #endif
57 52
58 #ifndef FUNC_NAME 53 #ifndef FUNC_NAME
59 #define FUNC_NAME memcpy 54 #define FUNC_NAME memcpy
60 #endif 55 #endif
61 56
62 #ifndef PREAMBLE 57 #ifndef PREAMBLE
63 #define PREAMBLE 58 #define PREAMBLE
64 #endif 59 #endif
65 60
66 #ifndef XCC 61 #ifndef XCC
67 #define XCC xcc 62 #define XCC xcc
68 #endif 63 #endif
69 64
70 #define FREG_FROB(f1, f2, f3, f4, f5, f6, f7, 65 #define FREG_FROB(f1, f2, f3, f4, f5, f6, f7, f8, f9) \
71 faligndata %f1, %f2, %f48 66 faligndata %f1, %f2, %f48; \
72 faligndata %f2, %f3, %f50 67 faligndata %f2, %f3, %f50; \
73 faligndata %f3, %f4, %f52 68 faligndata %f3, %f4, %f52; \
74 faligndata %f4, %f5, %f54 69 faligndata %f4, %f5, %f54; \
75 faligndata %f5, %f6, %f56 70 faligndata %f5, %f6, %f56; \
76 faligndata %f6, %f7, %f58 71 faligndata %f6, %f7, %f58; \
77 faligndata %f7, %f8, %f60 72 faligndata %f7, %f8, %f60; \
78 faligndata %f8, %f9, %f62 73 faligndata %f8, %f9, %f62;
79 74
80 #define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc !! 75 #define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, len, jmptgt) \
81 EX_LD_FP(LOAD_BLK(%src, %fdest), U1_gs !! 76 EX_LD(LOAD_BLK(%src, %fdest)); \
82 EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_g !! 77 EX_ST(STORE_BLK(%fsrc, %dest)); \
83 add %src, 0x40, %s !! 78 add %src, 0x40, %src; \
84 subcc %GLOBAL_SPARE, !! 79 subcc %len, 0x40, %len; \
85 be,pn %xcc, jmptgt; !! 80 be,pn %xcc, jmptgt; \
86 add %dest, 0x40, % !! 81 add %dest, 0x40, %dest; \
87 !! 82
88 #define LOOP_CHUNK1(src, dest, branch_dest) !! 83 #define LOOP_CHUNK1(src, dest, len, branch_dest) \
89 MAIN_LOOP_CHUNK(src, dest, f0, f48, b !! 84 MAIN_LOOP_CHUNK(src, dest, f0, f48, len, branch_dest)
90 #define LOOP_CHUNK2(src, dest, branch_dest) !! 85 #define LOOP_CHUNK2(src, dest, len, branch_dest) \
91 MAIN_LOOP_CHUNK(src, dest, f16, f48, b !! 86 MAIN_LOOP_CHUNK(src, dest, f16, f48, len, branch_dest)
92 #define LOOP_CHUNK3(src, dest, branch_dest) !! 87 #define LOOP_CHUNK3(src, dest, len, branch_dest) \
93 MAIN_LOOP_CHUNK(src, dest, f32, f48, b !! 88 MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
94 89
95 #define DO_SYNC membar #Sync; 90 #define DO_SYNC membar #Sync;
96 #define STORE_SYNC(dest, fsrc) 91 #define STORE_SYNC(dest, fsrc) \
97 EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_g !! 92 EX_ST(STORE_BLK(%fsrc, %dest)); \
98 add %dest, 0x40, % 93 add %dest, 0x40, %dest; \
99 DO_SYNC 94 DO_SYNC
100 95
101 #define STORE_JUMP(dest, fsrc, target) 96 #define STORE_JUMP(dest, fsrc, target) \
102 EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_g !! 97 EX_ST(STORE_BLK(%fsrc, %dest)); \
103 add %dest, 0x40, % 98 add %dest, 0x40, %dest; \
104 ba,pt %xcc, target; 99 ba,pt %xcc, target; \
105 nop; 100 nop;
106 101
107 #define FINISH_VISCHUNK(dest, f0, f1) !! 102 #define FINISH_VISCHUNK(dest, f0, f1, left) \
108 subcc %g3, 8, %g3; !! 103 subcc %left, 8, %left;\
109 bl,pn %xcc, 95f; !! 104 bl,pn %xcc, 95f; \
110 faligndata %f0, %f1, %f48 !! 105 faligndata %f0, %f1, %f48; \
111 EX_ST_FP(STORE(std, %f48, %dest), U1_g !! 106 EX_ST(STORE(std, %f48, %dest)); \
112 add %dest, 8, %des 107 add %dest, 8, %dest;
113 108
114 #define UNEVEN_VISCHUNK_LAST(dest, f0, f1) !! 109 #define UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \
115 subcc %g3, 8, %g3; !! 110 subcc %left, 8, %left; \
116 bl,pn %xcc, 95f; !! 111 bl,pn %xcc, 95f; \
117 fsrc2 %f0, %f1; 112 fsrc2 %f0, %f1;
118 113
119 #define UNEVEN_VISCHUNK(dest, f0, f1) !! 114 #define UNEVEN_VISCHUNK(dest, f0, f1, left) \
120 UNEVEN_VISCHUNK_LAST(dest, f0, f1) !! 115 UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \
121 ba,a,pt %xcc, 93f; 116 ba,a,pt %xcc, 93f;
122 117
123 .register %g2,#scratch 118 .register %g2,#scratch
124 .register %g3,#scratch 119 .register %g3,#scratch
125 120
126 .text 121 .text
127 #ifndef EX_RETVAL <<
128 #define EX_RETVAL(x) x <<
129 ENTRY(U1_g1_1_fp) <<
130 VISExitHalf <<
131 add %g1, 1, %g1 <<
132 add %g1, %g2, %g1 <<
133 retl <<
134 add %g1, %o2, %o0 <<
135 ENDPROC(U1_g1_1_fp) <<
136 ENTRY(U1_g2_0_fp) <<
137 VISExitHalf <<
138 retl <<
139 add %g2, %o2, %o0 <<
140 ENDPROC(U1_g2_0_fp) <<
141 ENTRY(U1_g2_8_fp) <<
142 VISExitHalf <<
143 add %g2, 8, %g2 <<
144 retl <<
145 add %g2, %o2, %o0 <<
146 ENDPROC(U1_g2_8_fp) <<
147 ENTRY(U1_gs_0_fp) <<
148 VISExitHalf <<
149 add %GLOBAL_SPARE, %g3, %o <<
150 retl <<
151 add %o0, %o2, %o0 <<
152 ENDPROC(U1_gs_0_fp) <<
153 ENTRY(U1_gs_80_fp) <<
154 VISExitHalf <<
155 add %GLOBAL_SPARE, 0x80, % <<
156 add %GLOBAL_SPARE, %g3, %o <<
157 retl <<
158 add %o0, %o2, %o0 <<
159 ENDPROC(U1_gs_80_fp) <<
160 ENTRY(U1_gs_40_fp) <<
161 VISExitHalf <<
162 add %GLOBAL_SPARE, 0x40, % <<
163 add %GLOBAL_SPARE, %g3, %o <<
164 retl <<
165 add %o0, %o2, %o0 <<
166 ENDPROC(U1_gs_40_fp) <<
167 ENTRY(U1_g3_0_fp) <<
168 VISExitHalf <<
169 retl <<
170 add %g3, %o2, %o0 <<
171 ENDPROC(U1_g3_0_fp) <<
172 ENTRY(U1_g3_8_fp) <<
173 VISExitHalf <<
174 add %g3, 8, %g3 <<
175 retl <<
176 add %g3, %o2, %o0 <<
177 ENDPROC(U1_g3_8_fp) <<
178 ENTRY(U1_o2_0_fp) <<
179 VISExitHalf <<
180 retl <<
181 mov %o2, %o0 <<
182 ENDPROC(U1_o2_0_fp) <<
183 ENTRY(U1_o2_1_fp) <<
184 VISExitHalf <<
185 retl <<
186 add %o2, 1, %o0 <<
187 ENDPROC(U1_o2_1_fp) <<
188 ENTRY(U1_gs_0) <<
189 VISExitHalf <<
190 retl <<
191 add %GLOBAL_SPARE, %o2, %o <<
192 ENDPROC(U1_gs_0) <<
193 ENTRY(U1_gs_8) <<
194 VISExitHalf <<
195 add %GLOBAL_SPARE, %o2, %G <<
196 retl <<
197 add %GLOBAL_SPARE, 0x8, %o <<
198 ENDPROC(U1_gs_8) <<
199 ENTRY(U1_gs_10) <<
200 VISExitHalf <<
201 add %GLOBAL_SPARE, %o2, %G <<
202 retl <<
203 add %GLOBAL_SPARE, 0x10, % <<
204 ENDPROC(U1_gs_10) <<
205 ENTRY(U1_o2_0) <<
206 retl <<
207 mov %o2, %o0 <<
208 ENDPROC(U1_o2_0) <<
209 ENTRY(U1_o2_8) <<
210 retl <<
211 add %o2, 8, %o0 <<
212 ENDPROC(U1_o2_8) <<
213 ENTRY(U1_o2_4) <<
214 retl <<
215 add %o2, 4, %o0 <<
216 ENDPROC(U1_o2_4) <<
217 ENTRY(U1_o2_1) <<
218 retl <<
219 add %o2, 1, %o0 <<
220 ENDPROC(U1_o2_1) <<
221 ENTRY(U1_g1_0) <<
222 retl <<
223 add %g1, %o2, %o0 <<
224 ENDPROC(U1_g1_0) <<
225 ENTRY(U1_g1_1) <<
226 add %g1, 1, %g1 <<
227 retl <<
228 add %g1, %o2, %o0 <<
229 ENDPROC(U1_g1_1) <<
230 ENTRY(U1_gs_0_o2_adj) <<
231 and %o2, 7, %o2 <<
232 retl <<
233 add %GLOBAL_SPARE, %o2, %o <<
234 ENDPROC(U1_gs_0_o2_adj) <<
235 ENTRY(U1_gs_8_o2_adj) <<
236 and %o2, 7, %o2 <<
237 add %GLOBAL_SPARE, 8, %GLO <<
238 retl <<
239 add %GLOBAL_SPARE, %o2, %o <<
240 ENDPROC(U1_gs_8_o2_adj) <<
241 #endif <<
242 <<
243 .align 64 122 .align 64
244 123
245 .globl FUNC_NAME 124 .globl FUNC_NAME
246 .type FUNC_NAME,#function 125 .type FUNC_NAME,#function
247 FUNC_NAME: /* %o0=dst, %o1=src, % 126 FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
248 srlx %o2, 31, %g2 127 srlx %o2, 31, %g2
249 cmp %g2, 0 128 cmp %g2, 0
250 tne %xcc, 5 129 tne %xcc, 5
251 PREAMBLE 130 PREAMBLE
252 mov %o0, %o4 131 mov %o0, %o4
253 cmp %o2, 0 132 cmp %o2, 0
254 be,pn %XCC, 85f 133 be,pn %XCC, 85f
255 or %o0, %o1, %o3 134 or %o0, %o1, %o3
256 cmp %o2, 16 135 cmp %o2, 16
257 blu,a,pn %XCC, 80f 136 blu,a,pn %XCC, 80f
258 or %o3, %o2, %o3 137 or %o3, %o2, %o3
259 138
260 cmp %o2, (5 * 64) 139 cmp %o2, (5 * 64)
261 blu,pt %XCC, 70f 140 blu,pt %XCC, 70f
262 andcc %o3, 0x7, %g0 141 andcc %o3, 0x7, %g0
263 142
264 /* Clobbers o5/g1/g2/g3/g7/icc/xcc. * 143 /* Clobbers o5/g1/g2/g3/g7/icc/xcc. */
265 VISEntry 144 VISEntry
266 145
267 /* Is 'dst' already aligned on an 64-b 146 /* Is 'dst' already aligned on an 64-byte boundary? */
268 andcc %o0, 0x3f, %g2 147 andcc %o0, 0x3f, %g2
269 be,pt %XCC, 2f 148 be,pt %XCC, 2f
270 149
271 /* Compute abs((dst & 0x3f) - 0x40) in 150 /* Compute abs((dst & 0x3f) - 0x40) into %g2. This is the number
272 * of bytes to copy to make 'dst' 64-b 151 * of bytes to copy to make 'dst' 64-byte aligned. We pre-
273 * subtract this from 'len'. 152 * subtract this from 'len'.
274 */ 153 */
275 sub %o0, %o1, %GLOBAL_SPAR 154 sub %o0, %o1, %GLOBAL_SPARE
276 sub %g2, 0x40, %g2 155 sub %g2, 0x40, %g2
277 sub %g0, %g2, %g2 156 sub %g0, %g2, %g2
278 sub %o2, %g2, %o2 157 sub %o2, %g2, %o2
279 andcc %g2, 0x7, %g1 158 andcc %g2, 0x7, %g1
280 be,pt %icc, 2f 159 be,pt %icc, 2f
281 and %g2, 0x38, %g2 160 and %g2, 0x38, %g2
282 161
283 1: subcc %g1, 0x1, %g1 162 1: subcc %g1, 0x1, %g1
284 EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3), !! 163 EX_LD(LOAD(ldub, %o1 + 0x00, %o3))
285 EX_ST_FP(STORE(stb, %o3, %o1 + %GLOBAL !! 164 EX_ST(STORE(stb, %o3, %o1 + %GLOBAL_SPARE))
286 bgu,pt %XCC, 1b 165 bgu,pt %XCC, 1b
287 add %o1, 0x1, %o1 166 add %o1, 0x1, %o1
288 167
289 add %o1, %GLOBAL_SPARE, %o 168 add %o1, %GLOBAL_SPARE, %o0
290 169
291 2: cmp %g2, 0x0 170 2: cmp %g2, 0x0
292 and %o1, 0x7, %g1 171 and %o1, 0x7, %g1
293 be,pt %icc, 3f 172 be,pt %icc, 3f
294 alignaddr %o1, %g0, %o1 173 alignaddr %o1, %g0, %o1
295 174
296 EX_LD_FP(LOAD(ldd, %o1, %f4), U1_g2_0_ !! 175 EX_LD(LOAD(ldd, %o1, %f4))
297 1: EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6), U1 !! 176 1: EX_LD(LOAD(ldd, %o1 + 0x8, %f6))
298 add %o1, 0x8, %o1 177 add %o1, 0x8, %o1
299 subcc %g2, 0x8, %g2 178 subcc %g2, 0x8, %g2
300 faligndata %f4, %f6, %f0 179 faligndata %f4, %f6, %f0
301 EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8 !! 180 EX_ST(STORE(std, %f0, %o0))
302 be,pn %icc, 3f 181 be,pn %icc, 3f
303 add %o0, 0x8, %o0 182 add %o0, 0x8, %o0
304 183
305 EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4), U1 !! 184 EX_LD(LOAD(ldd, %o1 + 0x8, %f4))
306 add %o1, 0x8, %o1 185 add %o1, 0x8, %o1
307 subcc %g2, 0x8, %g2 186 subcc %g2, 0x8, %g2
308 faligndata %f6, %f4, %f0 187 faligndata %f6, %f4, %f0
309 EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8 !! 188 EX_ST(STORE(std, %f0, %o0))
310 bne,pt %icc, 1b 189 bne,pt %icc, 1b
311 add %o0, 0x8, %o0 190 add %o0, 0x8, %o0
312 191
313 /* Destination is 64-byte aligned. */ 192 /* Destination is 64-byte aligned. */
314 3: 193 3:
315 membar #LoadStore | #StoreS 194 membar #LoadStore | #StoreStore | #StoreLoad
316 195
317 subcc %o2, 0x40, %GLOBAL_SPA 196 subcc %o2, 0x40, %GLOBAL_SPARE
318 add %o1, %g1, %g1 197 add %o1, %g1, %g1
319 andncc %GLOBAL_SPARE, (0x40 - 198 andncc %GLOBAL_SPARE, (0x40 - 1), %GLOBAL_SPARE
320 srl %g1, 3, %g2 199 srl %g1, 3, %g2
321 sub %o2, %GLOBAL_SPARE, %g 200 sub %o2, %GLOBAL_SPARE, %g3
322 andn %o1, (0x40 - 1), %o1 201 andn %o1, (0x40 - 1), %o1
323 and %g2, 7, %g2 202 and %g2, 7, %g2
324 andncc %g3, 0x7, %g3 203 andncc %g3, 0x7, %g3
325 fsrc2 %f0, %f2 204 fsrc2 %f0, %f2
326 sub %g3, 0x8, %g3 205 sub %g3, 0x8, %g3
327 sub %o2, %GLOBAL_SPARE, %o 206 sub %o2, %GLOBAL_SPARE, %o2
328 207
329 add %g1, %GLOBAL_SPARE, %g 208 add %g1, %GLOBAL_SPARE, %g1
330 subcc %o2, %g3, %o2 209 subcc %o2, %g3, %o2
331 210
332 EX_LD_FP(LOAD_BLK(%o1, %f0), U1_gs_0_f !! 211 EX_LD(LOAD_BLK(%o1, %f0))
333 add %o1, 0x40, %o1 212 add %o1, 0x40, %o1
334 add %g1, %g3, %g1 213 add %g1, %g3, %g1
335 EX_LD_FP(LOAD_BLK(%o1, %f16), U1_gs_0_ !! 214 EX_LD(LOAD_BLK(%o1, %f16))
336 add %o1, 0x40, %o1 215 add %o1, 0x40, %o1
337 sub %GLOBAL_SPARE, 0x80, % 216 sub %GLOBAL_SPARE, 0x80, %GLOBAL_SPARE
338 EX_LD_FP(LOAD_BLK(%o1, %f32), U1_gs_80 !! 217 EX_LD(LOAD_BLK(%o1, %f32))
339 add %o1, 0x40, %o1 218 add %o1, 0x40, %o1
340 219
341 /* There are 8 instances of the unroll 220 /* There are 8 instances of the unrolled loop,
342 * one for each possible alignment of 221 * one for each possible alignment of the
343 * source buffer. Each loop instance 222 * source buffer. Each loop instance is 452
344 * bytes. 223 * bytes.
345 */ 224 */
346 sll %g2, 3, %o3 225 sll %g2, 3, %o3
347 sub %o3, %g2, %o3 226 sub %o3, %g2, %o3
348 sllx %o3, 4, %o3 227 sllx %o3, 4, %o3
349 add %o3, %g2, %o3 228 add %o3, %g2, %o3
350 sllx %o3, 2, %g2 229 sllx %o3, 2, %g2
351 1: rd %pc, %o3 230 1: rd %pc, %o3
352 add %o3, %lo(1f - 1b), %o3 231 add %o3, %lo(1f - 1b), %o3
353 jmpl %o3 + %g2, %g0 232 jmpl %o3 + %g2, %g0
354 nop 233 nop
355 234
356 .align 64 235 .align 64
357 1: FREG_FROB(f0, f2, f4, f6, f8, f10,f12, 236 1: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
358 LOOP_CHUNK1(o1, o0, 1f) !! 237 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
359 FREG_FROB(f16,f18,f20,f22,f24,f26,f28, 238 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
360 LOOP_CHUNK2(o1, o0, 2f) !! 239 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
361 FREG_FROB(f32,f34,f36,f38,f40,f42,f44, 240 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
362 LOOP_CHUNK3(o1, o0, 3f) !! 241 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
363 ba,pt %xcc, 1b+4 242 ba,pt %xcc, 1b+4
364 faligndata %f0, %f2, %f48 243 faligndata %f0, %f2, %f48
365 1: FREG_FROB(f16,f18,f20,f22,f24,f26,f28, 244 1: FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
366 STORE_SYNC(o0, f48) 245 STORE_SYNC(o0, f48)
367 FREG_FROB(f32,f34,f36,f38,f40,f42,f44, 246 FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
368 STORE_JUMP(o0, f48, 40f) 247 STORE_JUMP(o0, f48, 40f)
369 2: FREG_FROB(f32,f34,f36,f38,f40,f42,f44, 248 2: FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
370 STORE_SYNC(o0, f48) 249 STORE_SYNC(o0, f48)
371 FREG_FROB(f0, f2, f4, f6, f8, f10,f12, 250 FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
372 STORE_JUMP(o0, f48, 48f) 251 STORE_JUMP(o0, f48, 48f)
373 3: FREG_FROB(f0, f2, f4, f6, f8, f10,f12, 252 3: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
374 STORE_SYNC(o0, f48) 253 STORE_SYNC(o0, f48)
375 FREG_FROB(f16,f18,f20,f22,f24,f26,f28, 254 FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
376 STORE_JUMP(o0, f48, 56f) 255 STORE_JUMP(o0, f48, 56f)
377 256
378 1: FREG_FROB(f2, f4, f6, f8, f10,f12,f14, 257 1: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
379 LOOP_CHUNK1(o1, o0, 1f) !! 258 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
380 FREG_FROB(f18,f20,f22,f24,f26,f28,f30, 259 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
381 LOOP_CHUNK2(o1, o0, 2f) !! 260 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
382 FREG_FROB(f34,f36,f38,f40,f42,f44,f46, 261 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
383 LOOP_CHUNK3(o1, o0, 3f) !! 262 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
384 ba,pt %xcc, 1b+4 263 ba,pt %xcc, 1b+4
385 faligndata %f2, %f4, %f48 264 faligndata %f2, %f4, %f48
386 1: FREG_FROB(f18,f20,f22,f24,f26,f28,f30, 265 1: FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
387 STORE_SYNC(o0, f48) 266 STORE_SYNC(o0, f48)
388 FREG_FROB(f34,f36,f38,f40,f42,f44,f46, 267 FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
389 STORE_JUMP(o0, f48, 41f) 268 STORE_JUMP(o0, f48, 41f)
390 2: FREG_FROB(f34,f36,f38,f40,f42,f44,f46, 269 2: FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
391 STORE_SYNC(o0, f48) 270 STORE_SYNC(o0, f48)
392 FREG_FROB(f2, f4, f6, f8, f10,f12,f14, 271 FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
393 STORE_JUMP(o0, f48, 49f) 272 STORE_JUMP(o0, f48, 49f)
394 3: FREG_FROB(f2, f4, f6, f8, f10,f12,f14, 273 3: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
395 STORE_SYNC(o0, f48) 274 STORE_SYNC(o0, f48)
396 FREG_FROB(f18,f20,f22,f24,f26,f28,f30, 275 FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
397 STORE_JUMP(o0, f48, 57f) 276 STORE_JUMP(o0, f48, 57f)
398 277
399 1: FREG_FROB(f4, f6, f8, f10,f12,f14,f16, 278 1: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
400 LOOP_CHUNK1(o1, o0, 1f) !! 279 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
401 FREG_FROB(f20,f22,f24,f26,f28,f30,f32, 280 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
402 LOOP_CHUNK2(o1, o0, 2f) !! 281 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
403 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, 282 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
404 LOOP_CHUNK3(o1, o0, 3f) !! 283 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
405 ba,pt %xcc, 1b+4 284 ba,pt %xcc, 1b+4
406 faligndata %f4, %f6, %f48 285 faligndata %f4, %f6, %f48
407 1: FREG_FROB(f20,f22,f24,f26,f28,f30,f32, 286 1: FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
408 STORE_SYNC(o0, f48) 287 STORE_SYNC(o0, f48)
409 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, 288 FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
410 STORE_JUMP(o0, f48, 42f) 289 STORE_JUMP(o0, f48, 42f)
411 2: FREG_FROB(f36,f38,f40,f42,f44,f46,f0, 290 2: FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
412 STORE_SYNC(o0, f48) 291 STORE_SYNC(o0, f48)
413 FREG_FROB(f4, f6, f8, f10,f12,f14,f16, 292 FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
414 STORE_JUMP(o0, f48, 50f) 293 STORE_JUMP(o0, f48, 50f)
415 3: FREG_FROB(f4, f6, f8, f10,f12,f14,f16, 294 3: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
416 STORE_SYNC(o0, f48) 295 STORE_SYNC(o0, f48)
417 FREG_FROB(f20,f22,f24,f26,f28,f30,f32, 296 FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
418 STORE_JUMP(o0, f48, 58f) 297 STORE_JUMP(o0, f48, 58f)
419 298
420 1: FREG_FROB(f6, f8, f10,f12,f14,f16,f18, 299 1: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
421 LOOP_CHUNK1(o1, o0, 1f) !! 300 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
422 FREG_FROB(f22,f24,f26,f28,f30,f32,f34, 301 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
423 LOOP_CHUNK2(o1, o0, 2f) !! 302 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
424 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, 303 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
425 LOOP_CHUNK3(o1, o0, 3f) !! 304 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
426 ba,pt %xcc, 1b+4 305 ba,pt %xcc, 1b+4
427 faligndata %f6, %f8, %f48 306 faligndata %f6, %f8, %f48
428 1: FREG_FROB(f22,f24,f26,f28,f30,f32,f34, 307 1: FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
429 STORE_SYNC(o0, f48) 308 STORE_SYNC(o0, f48)
430 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, 309 FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
431 STORE_JUMP(o0, f48, 43f) 310 STORE_JUMP(o0, f48, 43f)
432 2: FREG_FROB(f38,f40,f42,f44,f46,f0, f2, 311 2: FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
433 STORE_SYNC(o0, f48) 312 STORE_SYNC(o0, f48)
434 FREG_FROB(f6, f8, f10,f12,f14,f16,f18, 313 FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
435 STORE_JUMP(o0, f48, 51f) 314 STORE_JUMP(o0, f48, 51f)
436 3: FREG_FROB(f6, f8, f10,f12,f14,f16,f18, 315 3: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
437 STORE_SYNC(o0, f48) 316 STORE_SYNC(o0, f48)
438 FREG_FROB(f22,f24,f26,f28,f30,f32,f34, 317 FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
439 STORE_JUMP(o0, f48, 59f) 318 STORE_JUMP(o0, f48, 59f)
440 319
441 1: FREG_FROB(f8, f10,f12,f14,f16,f18,f20, 320 1: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
442 LOOP_CHUNK1(o1, o0, 1f) !! 321 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
443 FREG_FROB(f24,f26,f28,f30,f32,f34,f36, 322 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
444 LOOP_CHUNK2(o1, o0, 2f) !! 323 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
445 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, 324 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
446 LOOP_CHUNK3(o1, o0, 3f) !! 325 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
447 ba,pt %xcc, 1b+4 326 ba,pt %xcc, 1b+4
448 faligndata %f8, %f10, %f48 327 faligndata %f8, %f10, %f48
449 1: FREG_FROB(f24,f26,f28,f30,f32,f34,f36, 328 1: FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
450 STORE_SYNC(o0, f48) 329 STORE_SYNC(o0, f48)
451 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, 330 FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
452 STORE_JUMP(o0, f48, 44f) 331 STORE_JUMP(o0, f48, 44f)
453 2: FREG_FROB(f40,f42,f44,f46,f0, f2, f4, 332 2: FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
454 STORE_SYNC(o0, f48) 333 STORE_SYNC(o0, f48)
455 FREG_FROB(f8, f10,f12,f14,f16,f18,f20, 334 FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
456 STORE_JUMP(o0, f48, 52f) 335 STORE_JUMP(o0, f48, 52f)
457 3: FREG_FROB(f8, f10,f12,f14,f16,f18,f20, 336 3: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
458 STORE_SYNC(o0, f48) 337 STORE_SYNC(o0, f48)
459 FREG_FROB(f24,f26,f28,f30,f32,f34,f36, 338 FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
460 STORE_JUMP(o0, f48, 60f) 339 STORE_JUMP(o0, f48, 60f)
461 340
462 1: FREG_FROB(f10,f12,f14,f16,f18,f20,f22, 341 1: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
463 LOOP_CHUNK1(o1, o0, 1f) !! 342 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
464 FREG_FROB(f26,f28,f30,f32,f34,f36,f38, 343 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
465 LOOP_CHUNK2(o1, o0, 2f) !! 344 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
466 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, 345 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
467 LOOP_CHUNK3(o1, o0, 3f) !! 346 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
468 ba,pt %xcc, 1b+4 347 ba,pt %xcc, 1b+4
469 faligndata %f10, %f12, %f48 348 faligndata %f10, %f12, %f48
470 1: FREG_FROB(f26,f28,f30,f32,f34,f36,f38, 349 1: FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
471 STORE_SYNC(o0, f48) 350 STORE_SYNC(o0, f48)
472 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, 351 FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
473 STORE_JUMP(o0, f48, 45f) 352 STORE_JUMP(o0, f48, 45f)
474 2: FREG_FROB(f42,f44,f46,f0, f2, f4, f6, 353 2: FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
475 STORE_SYNC(o0, f48) 354 STORE_SYNC(o0, f48)
476 FREG_FROB(f10,f12,f14,f16,f18,f20,f22, 355 FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
477 STORE_JUMP(o0, f48, 53f) 356 STORE_JUMP(o0, f48, 53f)
478 3: FREG_FROB(f10,f12,f14,f16,f18,f20,f22, 357 3: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
479 STORE_SYNC(o0, f48) 358 STORE_SYNC(o0, f48)
480 FREG_FROB(f26,f28,f30,f32,f34,f36,f38, 359 FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
481 STORE_JUMP(o0, f48, 61f) 360 STORE_JUMP(o0, f48, 61f)
482 361
483 1: FREG_FROB(f12,f14,f16,f18,f20,f22,f24, 362 1: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
484 LOOP_CHUNK1(o1, o0, 1f) !! 363 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
485 FREG_FROB(f28,f30,f32,f34,f36,f38,f40, 364 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
486 LOOP_CHUNK2(o1, o0, 2f) !! 365 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
487 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, 366 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
488 LOOP_CHUNK3(o1, o0, 3f) !! 367 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
489 ba,pt %xcc, 1b+4 368 ba,pt %xcc, 1b+4
490 faligndata %f12, %f14, %f48 369 faligndata %f12, %f14, %f48
491 1: FREG_FROB(f28,f30,f32,f34,f36,f38,f40, 370 1: FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
492 STORE_SYNC(o0, f48) 371 STORE_SYNC(o0, f48)
493 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, 372 FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
494 STORE_JUMP(o0, f48, 46f) 373 STORE_JUMP(o0, f48, 46f)
495 2: FREG_FROB(f44,f46,f0, f2, f4, f6, f8, 374 2: FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
496 STORE_SYNC(o0, f48) 375 STORE_SYNC(o0, f48)
497 FREG_FROB(f12,f14,f16,f18,f20,f22,f24, 376 FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
498 STORE_JUMP(o0, f48, 54f) 377 STORE_JUMP(o0, f48, 54f)
499 3: FREG_FROB(f12,f14,f16,f18,f20,f22,f24, 378 3: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
500 STORE_SYNC(o0, f48) 379 STORE_SYNC(o0, f48)
501 FREG_FROB(f28,f30,f32,f34,f36,f38,f40, 380 FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
502 STORE_JUMP(o0, f48, 62f) 381 STORE_JUMP(o0, f48, 62f)
503 382
504 1: FREG_FROB(f14,f16,f18,f20,f22,f24,f26, 383 1: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
505 LOOP_CHUNK1(o1, o0, 1f) !! 384 LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
506 FREG_FROB(f30,f32,f34,f36,f38,f40,f42, 385 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
507 LOOP_CHUNK2(o1, o0, 2f) !! 386 LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
508 FREG_FROB(f46,f0, f2, f4, f6, f8, f10, 387 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
509 LOOP_CHUNK3(o1, o0, 3f) !! 388 LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
510 ba,pt %xcc, 1b+4 389 ba,pt %xcc, 1b+4
511 faligndata %f14, %f16, %f48 390 faligndata %f14, %f16, %f48
512 1: FREG_FROB(f30,f32,f34,f36,f38,f40,f42, 391 1: FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
513 STORE_SYNC(o0, f48) 392 STORE_SYNC(o0, f48)
514 FREG_FROB(f46,f0, f2, f4, f6, f8, f10, 393 FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
515 STORE_JUMP(o0, f48, 47f) 394 STORE_JUMP(o0, f48, 47f)
516 2: FREG_FROB(f46,f0, f2, f4, f6, f8, f10, 395 2: FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
517 STORE_SYNC(o0, f48) 396 STORE_SYNC(o0, f48)
518 FREG_FROB(f14,f16,f18,f20,f22,f24,f26, 397 FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
519 STORE_JUMP(o0, f48, 55f) 398 STORE_JUMP(o0, f48, 55f)
520 3: FREG_FROB(f14,f16,f18,f20,f22,f24,f26, 399 3: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
521 STORE_SYNC(o0, f48) 400 STORE_SYNC(o0, f48)
522 FREG_FROB(f30,f32,f34,f36,f38,f40,f42, 401 FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
523 STORE_JUMP(o0, f48, 63f) 402 STORE_JUMP(o0, f48, 63f)
524 403
525 40: FINISH_VISCHUNK(o0, f0, f2) !! 404 40: FINISH_VISCHUNK(o0, f0, f2, g3)
526 41: FINISH_VISCHUNK(o0, f2, f4) !! 405 41: FINISH_VISCHUNK(o0, f2, f4, g3)
527 42: FINISH_VISCHUNK(o0, f4, f6) !! 406 42: FINISH_VISCHUNK(o0, f4, f6, g3)
528 43: FINISH_VISCHUNK(o0, f6, f8) !! 407 43: FINISH_VISCHUNK(o0, f6, f8, g3)
529 44: FINISH_VISCHUNK(o0, f8, f10) !! 408 44: FINISH_VISCHUNK(o0, f8, f10, g3)
530 45: FINISH_VISCHUNK(o0, f10, f12) !! 409 45: FINISH_VISCHUNK(o0, f10, f12, g3)
531 46: FINISH_VISCHUNK(o0, f12, f14) !! 410 46: FINISH_VISCHUNK(o0, f12, f14, g3)
532 47: UNEVEN_VISCHUNK(o0, f14, f0) !! 411 47: UNEVEN_VISCHUNK(o0, f14, f0, g3)
533 48: FINISH_VISCHUNK(o0, f16, f18) !! 412 48: FINISH_VISCHUNK(o0, f16, f18, g3)
534 49: FINISH_VISCHUNK(o0, f18, f20) !! 413 49: FINISH_VISCHUNK(o0, f18, f20, g3)
535 50: FINISH_VISCHUNK(o0, f20, f22) !! 414 50: FINISH_VISCHUNK(o0, f20, f22, g3)
536 51: FINISH_VISCHUNK(o0, f22, f24) !! 415 51: FINISH_VISCHUNK(o0, f22, f24, g3)
537 52: FINISH_VISCHUNK(o0, f24, f26) !! 416 52: FINISH_VISCHUNK(o0, f24, f26, g3)
538 53: FINISH_VISCHUNK(o0, f26, f28) !! 417 53: FINISH_VISCHUNK(o0, f26, f28, g3)
539 54: FINISH_VISCHUNK(o0, f28, f30) !! 418 54: FINISH_VISCHUNK(o0, f28, f30, g3)
540 55: UNEVEN_VISCHUNK(o0, f30, f0) !! 419 55: UNEVEN_VISCHUNK(o0, f30, f0, g3)
541 56: FINISH_VISCHUNK(o0, f32, f34) !! 420 56: FINISH_VISCHUNK(o0, f32, f34, g3)
542 57: FINISH_VISCHUNK(o0, f34, f36) !! 421 57: FINISH_VISCHUNK(o0, f34, f36, g3)
543 58: FINISH_VISCHUNK(o0, f36, f38) !! 422 58: FINISH_VISCHUNK(o0, f36, f38, g3)
544 59: FINISH_VISCHUNK(o0, f38, f40) !! 423 59: FINISH_VISCHUNK(o0, f38, f40, g3)
545 60: FINISH_VISCHUNK(o0, f40, f42) !! 424 60: FINISH_VISCHUNK(o0, f40, f42, g3)
546 61: FINISH_VISCHUNK(o0, f42, f44) !! 425 61: FINISH_VISCHUNK(o0, f42, f44, g3)
547 62: FINISH_VISCHUNK(o0, f44, f46) !! 426 62: FINISH_VISCHUNK(o0, f44, f46, g3)
548 63: UNEVEN_VISCHUNK_LAST(o0, f46, f0) !! 427 63: UNEVEN_VISCHUNK_LAST(o0, f46, f0, g3)
549 428
550 93: EX_LD_FP(LOAD(ldd, %o1, %f2), U1_g3_0_ !! 429 93: EX_LD(LOAD(ldd, %o1, %f2))
551 add %o1, 8, %o1 430 add %o1, 8, %o1
552 subcc %g3, 8, %g3 431 subcc %g3, 8, %g3
553 faligndata %f0, %f2, %f8 432 faligndata %f0, %f2, %f8
554 EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8 !! 433 EX_ST(STORE(std, %f8, %o0))
555 bl,pn %xcc, 95f 434 bl,pn %xcc, 95f
556 add %o0, 8, %o0 435 add %o0, 8, %o0
557 EX_LD_FP(LOAD(ldd, %o1, %f0), U1_g3_0_ !! 436 EX_LD(LOAD(ldd, %o1, %f0))
558 add %o1, 8, %o1 437 add %o1, 8, %o1
559 subcc %g3, 8, %g3 438 subcc %g3, 8, %g3
560 faligndata %f2, %f0, %f8 439 faligndata %f2, %f0, %f8
561 EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8 !! 440 EX_ST(STORE(std, %f8, %o0))
562 bge,pt %xcc, 93b 441 bge,pt %xcc, 93b
563 add %o0, 8, %o0 442 add %o0, 8, %o0
564 443
565 95: brz,pt %o2, 2f 444 95: brz,pt %o2, 2f
566 mov %g1, %o1 445 mov %g1, %o1
567 446
568 1: EX_LD_FP(LOAD(ldub, %o1, %o3), U1_o2_0 !! 447 1: EX_LD(LOAD(ldub, %o1, %o3))
569 add %o1, 1, %o1 448 add %o1, 1, %o1
570 subcc %o2, 1, %o2 449 subcc %o2, 1, %o2
571 EX_ST_FP(STORE(stb, %o3, %o0), U1_o2_1 !! 450 EX_ST(STORE(stb, %o3, %o0))
572 bne,pt %xcc, 1b 451 bne,pt %xcc, 1b
573 add %o0, 1, %o0 452 add %o0, 1, %o0
574 453
575 2: membar #StoreLoad | #StoreSto 454 2: membar #StoreLoad | #StoreStore
576 VISExit 455 VISExit
577 retl 456 retl
578 mov EX_RETVAL(%o4), %o0 457 mov EX_RETVAL(%o4), %o0
579 458
580 .align 64 459 .align 64
581 70: /* 16 < len <= (5 * 64) */ 460 70: /* 16 < len <= (5 * 64) */
582 bne,pn %XCC, 75f 461 bne,pn %XCC, 75f
583 sub %o0, %o1, %o3 462 sub %o0, %o1, %o3
584 463
585 72: andn %o2, 0xf, %GLOBAL_SPAR 464 72: andn %o2, 0xf, %GLOBAL_SPARE
586 and %o2, 0xf, %o2 465 and %o2, 0xf, %o2
587 1: EX_LD(LOAD(ldx, %o1 + 0x00, %o5), U1_g !! 466 1: EX_LD(LOAD(ldx, %o1 + 0x00, %o5))
588 EX_LD(LOAD(ldx, %o1 + 0x08, %g1), U1_g !! 467 EX_LD(LOAD(ldx, %o1 + 0x08, %g1))
589 subcc %GLOBAL_SPARE, 0x10, % 468 subcc %GLOBAL_SPARE, 0x10, %GLOBAL_SPARE
590 EX_ST(STORE(stx, %o5, %o1 + %o3), U1_g !! 469 EX_ST(STORE(stx, %o5, %o1 + %o3))
591 add %o1, 0x8, %o1 470 add %o1, 0x8, %o1
592 EX_ST(STORE(stx, %g1, %o1 + %o3), U1_g !! 471 EX_ST(STORE(stx, %g1, %o1 + %o3))
593 bgu,pt %XCC, 1b 472 bgu,pt %XCC, 1b
594 add %o1, 0x8, %o1 473 add %o1, 0x8, %o1
595 73: andcc %o2, 0x8, %g0 474 73: andcc %o2, 0x8, %g0
596 be,pt %XCC, 1f 475 be,pt %XCC, 1f
597 nop 476 nop
598 EX_LD(LOAD(ldx, %o1, %o5), U1_o2_0) !! 477 EX_LD(LOAD(ldx, %o1, %o5))
599 sub %o2, 0x8, %o2 478 sub %o2, 0x8, %o2
600 EX_ST(STORE(stx, %o5, %o1 + %o3), U1_o !! 479 EX_ST(STORE(stx, %o5, %o1 + %o3))
601 add %o1, 0x8, %o1 480 add %o1, 0x8, %o1
602 1: andcc %o2, 0x4, %g0 481 1: andcc %o2, 0x4, %g0
603 be,pt %XCC, 1f 482 be,pt %XCC, 1f
604 nop 483 nop
605 EX_LD(LOAD(lduw, %o1, %o5), U1_o2_0) !! 484 EX_LD(LOAD(lduw, %o1, %o5))
606 sub %o2, 0x4, %o2 485 sub %o2, 0x4, %o2
607 EX_ST(STORE(stw, %o5, %o1 + %o3), U1_o !! 486 EX_ST(STORE(stw, %o5, %o1 + %o3))
608 add %o1, 0x4, %o1 487 add %o1, 0x4, %o1
609 1: cmp %o2, 0 488 1: cmp %o2, 0
610 be,pt %XCC, 85f 489 be,pt %XCC, 85f
611 nop 490 nop
612 ba,pt %xcc, 90f 491 ba,pt %xcc, 90f
613 nop 492 nop
614 493
615 75: andcc %o0, 0x7, %g1 494 75: andcc %o0, 0x7, %g1
616 sub %g1, 0x8, %g1 495 sub %g1, 0x8, %g1
617 be,pn %icc, 2f 496 be,pn %icc, 2f
618 sub %g0, %g1, %g1 497 sub %g0, %g1, %g1
619 sub %o2, %g1, %o2 498 sub %o2, %g1, %o2
620 499
621 1: EX_LD(LOAD(ldub, %o1, %o5), U1_g1_0) !! 500 1: EX_LD(LOAD(ldub, %o1, %o5))
622 subcc %g1, 1, %g1 501 subcc %g1, 1, %g1
623 EX_ST(STORE(stb, %o5, %o1 + %o3), U1_g !! 502 EX_ST(STORE(stb, %o5, %o1 + %o3))
624 bgu,pt %icc, 1b 503 bgu,pt %icc, 1b
625 add %o1, 1, %o1 504 add %o1, 1, %o1
626 505
627 2: add %o1, %o3, %o0 506 2: add %o1, %o3, %o0
628 andcc %o1, 0x7, %g1 507 andcc %o1, 0x7, %g1
629 bne,pt %icc, 8f 508 bne,pt %icc, 8f
630 sll %g1, 3, %g1 509 sll %g1, 3, %g1
631 510
632 cmp %o2, 16 511 cmp %o2, 16
633 bgeu,pt %icc, 72b 512 bgeu,pt %icc, 72b
634 nop 513 nop
635 ba,a,pt %xcc, 73b 514 ba,a,pt %xcc, 73b
636 515
637 8: mov 64, %o3 516 8: mov 64, %o3
638 andn %o1, 0x7, %o1 517 andn %o1, 0x7, %o1
639 EX_LD(LOAD(ldx, %o1, %g2), U1_o2_0) !! 518 EX_LD(LOAD(ldx, %o1, %g2))
640 sub %o3, %g1, %o3 519 sub %o3, %g1, %o3
641 andn %o2, 0x7, %GLOBAL_SPAR 520 andn %o2, 0x7, %GLOBAL_SPARE
642 sllx %g2, %g1, %g2 521 sllx %g2, %g1, %g2
643 1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3), U1_gs !! 522 1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3))
644 subcc %GLOBAL_SPARE, 0x8, %G 523 subcc %GLOBAL_SPARE, 0x8, %GLOBAL_SPARE
645 add %o1, 0x8, %o1 524 add %o1, 0x8, %o1
646 srlx %g3, %o3, %o5 525 srlx %g3, %o3, %o5
647 or %o5, %g2, %o5 526 or %o5, %g2, %o5
648 EX_ST(STORE(stx, %o5, %o0), U1_gs_8_o2 !! 527 EX_ST(STORE(stx, %o5, %o0))
649 add %o0, 0x8, %o0 528 add %o0, 0x8, %o0
650 bgu,pt %icc, 1b 529 bgu,pt %icc, 1b
651 sllx %g3, %g1, %g2 530 sllx %g3, %g1, %g2
652 531
653 srl %g1, 3, %g1 532 srl %g1, 3, %g1
654 andcc %o2, 0x7, %o2 533 andcc %o2, 0x7, %o2
655 be,pn %icc, 85f 534 be,pn %icc, 85f
656 add %o1, %g1, %o1 535 add %o1, %g1, %o1
657 ba,pt %xcc, 90f 536 ba,pt %xcc, 90f
658 sub %o0, %o1, %o3 537 sub %o0, %o1, %o3
659 538
660 .align 64 539 .align 64
661 80: /* 0 < len <= 16 */ 540 80: /* 0 < len <= 16 */
662 andcc %o3, 0x3, %g0 541 andcc %o3, 0x3, %g0
663 bne,pn %XCC, 90f 542 bne,pn %XCC, 90f
664 sub %o0, %o1, %o3 543 sub %o0, %o1, %o3
665 544
666 1: EX_LD(LOAD(lduw, %o1, %g1), U1_o2_0) !! 545 1: EX_LD(LOAD(lduw, %o1, %g1))
667 subcc %o2, 4, %o2 546 subcc %o2, 4, %o2
668 EX_ST(STORE(stw, %g1, %o1 + %o3), U1_o !! 547 EX_ST(STORE(stw, %g1, %o1 + %o3))
669 bgu,pt %XCC, 1b 548 bgu,pt %XCC, 1b
670 add %o1, 4, %o1 549 add %o1, 4, %o1
671 550
672 85: retl 551 85: retl
673 mov EX_RETVAL(%o4), %o0 552 mov EX_RETVAL(%o4), %o0
674 553
675 .align 32 554 .align 32
676 90: EX_LD(LOAD(ldub, %o1, %g1), U1_o2_0) !! 555 90: EX_LD(LOAD(ldub, %o1, %g1))
677 subcc %o2, 1, %o2 556 subcc %o2, 1, %o2
678 EX_ST(STORE(stb, %g1, %o1 + %o3), U1_o !! 557 EX_ST(STORE(stb, %g1, %o1 + %o3))
679 bgu,pt %XCC, 90b 558 bgu,pt %XCC, 90b
680 add %o1, 1, %o1 559 add %o1, 1, %o1
681 retl 560 retl
682 mov EX_RETVAL(%o4), %o0 561 mov EX_RETVAL(%o4), %o0
683 562
684 .size FUNC_NAME, .-FUNC_NAME 563 .size FUNC_NAME, .-FUNC_NAME
685 EXPORT_SYMBOL(FUNC_NAME) <<
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.