diff options
author | Ralf Baechle <ralf@linux-mips.org> | 2008-01-29 05:14:59 -0500 |
---|---|---|
committer | Ralf Baechle <ralf@linux-mips.org> | 2008-01-29 05:14:59 -0500 |
commit | c5ec1983e45d25446a023e98207e30ab1bf2311a (patch) | |
tree | 68124f5166dc284d7d1c6280f4642a6719a6c554 /arch/mips/lib/memcpy.S | |
parent | f21d8508080f05a8b05c99fce57cd76ac43666fa (diff) |
[MIPS] Eleminate local symbols from the symbol table.
These symbols appear in oprofile output, stacktraces and similar but only
make the output harder to read. Many identical symbol names such as
"both_aligned" were also being used in multiple source files making it
impossible to see which file actually was meant. So let's get rid of them.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Diffstat (limited to 'arch/mips/lib/memcpy.S')
-rw-r--r-- | arch/mips/lib/memcpy.S | 182 |
1 files changed, 91 insertions, 91 deletions
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S index 01e450b1ebc9..c06cccf60bec 100644 --- a/arch/mips/lib/memcpy.S +++ b/arch/mips/lib/memcpy.S | |||
@@ -191,7 +191,7 @@ | |||
191 | .align 5 | 191 | .align 5 |
192 | LEAF(memcpy) /* a0=dst a1=src a2=len */ | 192 | LEAF(memcpy) /* a0=dst a1=src a2=len */ |
193 | move v0, dst /* return value */ | 193 | move v0, dst /* return value */ |
194 | __memcpy: | 194 | .L__memcpy: |
195 | FEXPORT(__copy_user) | 195 | FEXPORT(__copy_user) |
196 | /* | 196 | /* |
197 | * Note: dst & src may be unaligned, len may be 0 | 197 | * Note: dst & src may be unaligned, len may be 0 |
@@ -213,45 +213,45 @@ FEXPORT(__copy_user) | |||
213 | and t1, dst, ADDRMASK | 213 | and t1, dst, ADDRMASK |
214 | PREF( 0, 1*32(src) ) | 214 | PREF( 0, 1*32(src) ) |
215 | PREF( 1, 1*32(dst) ) | 215 | PREF( 1, 1*32(dst) ) |
216 | bnez t2, copy_bytes_checklen | 216 | bnez t2, .Lcopy_bytes_checklen |
217 | and t0, src, ADDRMASK | 217 | and t0, src, ADDRMASK |
218 | PREF( 0, 2*32(src) ) | 218 | PREF( 0, 2*32(src) ) |
219 | PREF( 1, 2*32(dst) ) | 219 | PREF( 1, 2*32(dst) ) |
220 | bnez t1, dst_unaligned | 220 | bnez t1, .Ldst_unaligned |
221 | nop | 221 | nop |
222 | bnez t0, src_unaligned_dst_aligned | 222 | bnez t0, .Lsrc_unaligned_dst_aligned |
223 | /* | 223 | /* |
224 | * use delay slot for fall-through | 224 | * use delay slot for fall-through |
225 | * src and dst are aligned; need to compute rem | 225 | * src and dst are aligned; need to compute rem |
226 | */ | 226 | */ |
227 | both_aligned: | 227 | .Lboth_aligned: |
228 | SRL t0, len, LOG_NBYTES+3 # +3 for 8 units/iter | 228 | SRL t0, len, LOG_NBYTES+3 # +3 for 8 units/iter |
229 | beqz t0, cleanup_both_aligned # len < 8*NBYTES | 229 | beqz t0, .Lcleanup_both_aligned # len < 8*NBYTES |
230 | and rem, len, (8*NBYTES-1) # rem = len % (8*NBYTES) | 230 | and rem, len, (8*NBYTES-1) # rem = len % (8*NBYTES) |
231 | PREF( 0, 3*32(src) ) | 231 | PREF( 0, 3*32(src) ) |
232 | PREF( 1, 3*32(dst) ) | 232 | PREF( 1, 3*32(dst) ) |
233 | .align 4 | 233 | .align 4 |
234 | 1: | 234 | 1: |
235 | R10KCBARRIER(0(ra)) | 235 | R10KCBARRIER(0(ra)) |
236 | EXC( LOAD t0, UNIT(0)(src), l_exc) | 236 | EXC( LOAD t0, UNIT(0)(src), .Ll_exc) |
237 | EXC( LOAD t1, UNIT(1)(src), l_exc_copy) | 237 | EXC( LOAD t1, UNIT(1)(src), .Ll_exc_copy) |
238 | EXC( LOAD t2, UNIT(2)(src), l_exc_copy) | 238 | EXC( LOAD t2, UNIT(2)(src), .Ll_exc_copy) |
239 | EXC( LOAD t3, UNIT(3)(src), l_exc_copy) | 239 | EXC( LOAD t3, UNIT(3)(src), .Ll_exc_copy) |
240 | SUB len, len, 8*NBYTES | 240 | SUB len, len, 8*NBYTES |
241 | EXC( LOAD t4, UNIT(4)(src), l_exc_copy) | 241 | EXC( LOAD t4, UNIT(4)(src), .Ll_exc_copy) |
242 | EXC( LOAD t7, UNIT(5)(src), l_exc_copy) | 242 | EXC( LOAD t7, UNIT(5)(src), .Ll_exc_copy) |
243 | EXC( STORE t0, UNIT(0)(dst), s_exc_p8u) | 243 | EXC( STORE t0, UNIT(0)(dst), .Ls_exc_p8u) |
244 | EXC( STORE t1, UNIT(1)(dst), s_exc_p7u) | 244 | EXC( STORE t1, UNIT(1)(dst), .Ls_exc_p7u) |
245 | EXC( LOAD t0, UNIT(6)(src), l_exc_copy) | 245 | EXC( LOAD t0, UNIT(6)(src), .Ll_exc_copy) |
246 | EXC( LOAD t1, UNIT(7)(src), l_exc_copy) | 246 | EXC( LOAD t1, UNIT(7)(src), .Ll_exc_copy) |
247 | ADD src, src, 8*NBYTES | 247 | ADD src, src, 8*NBYTES |
248 | ADD dst, dst, 8*NBYTES | 248 | ADD dst, dst, 8*NBYTES |
249 | EXC( STORE t2, UNIT(-6)(dst), s_exc_p6u) | 249 | EXC( STORE t2, UNIT(-6)(dst), .Ls_exc_p6u) |
250 | EXC( STORE t3, UNIT(-5)(dst), s_exc_p5u) | 250 | EXC( STORE t3, UNIT(-5)(dst), .Ls_exc_p5u) |
251 | EXC( STORE t4, UNIT(-4)(dst), s_exc_p4u) | 251 | EXC( STORE t4, UNIT(-4)(dst), .Ls_exc_p4u) |
252 | EXC( STORE t7, UNIT(-3)(dst), s_exc_p3u) | 252 | EXC( STORE t7, UNIT(-3)(dst), .Ls_exc_p3u) |
253 | EXC( STORE t0, UNIT(-2)(dst), s_exc_p2u) | 253 | EXC( STORE t0, UNIT(-2)(dst), .Ls_exc_p2u) |
254 | EXC( STORE t1, UNIT(-1)(dst), s_exc_p1u) | 254 | EXC( STORE t1, UNIT(-1)(dst), .Ls_exc_p1u) |
255 | PREF( 0, 8*32(src) ) | 255 | PREF( 0, 8*32(src) ) |
256 | PREF( 1, 8*32(dst) ) | 256 | PREF( 1, 8*32(dst) ) |
257 | bne len, rem, 1b | 257 | bne len, rem, 1b |
@@ -260,41 +260,41 @@ EXC( STORE t1, UNIT(-1)(dst), s_exc_p1u) | |||
260 | /* | 260 | /* |
261 | * len == rem == the number of bytes left to copy < 8*NBYTES | 261 | * len == rem == the number of bytes left to copy < 8*NBYTES |
262 | */ | 262 | */ |
263 | cleanup_both_aligned: | 263 | .Lcleanup_both_aligned: |
264 | beqz len, done | 264 | beqz len, .Ldone |
265 | sltu t0, len, 4*NBYTES | 265 | sltu t0, len, 4*NBYTES |
266 | bnez t0, less_than_4units | 266 | bnez t0, .Lless_than_4units |
267 | and rem, len, (NBYTES-1) # rem = len % NBYTES | 267 | and rem, len, (NBYTES-1) # rem = len % NBYTES |
268 | /* | 268 | /* |
269 | * len >= 4*NBYTES | 269 | * len >= 4*NBYTES |
270 | */ | 270 | */ |
271 | EXC( LOAD t0, UNIT(0)(src), l_exc) | 271 | EXC( LOAD t0, UNIT(0)(src), .Ll_exc) |
272 | EXC( LOAD t1, UNIT(1)(src), l_exc_copy) | 272 | EXC( LOAD t1, UNIT(1)(src), .Ll_exc_copy) |
273 | EXC( LOAD t2, UNIT(2)(src), l_exc_copy) | 273 | EXC( LOAD t2, UNIT(2)(src), .Ll_exc_copy) |
274 | EXC( LOAD t3, UNIT(3)(src), l_exc_copy) | 274 | EXC( LOAD t3, UNIT(3)(src), .Ll_exc_copy) |
275 | SUB len, len, 4*NBYTES | 275 | SUB len, len, 4*NBYTES |
276 | ADD src, src, 4*NBYTES | 276 | ADD src, src, 4*NBYTES |
277 | R10KCBARRIER(0(ra)) | 277 | R10KCBARRIER(0(ra)) |
278 | EXC( STORE t0, UNIT(0)(dst), s_exc_p4u) | 278 | EXC( STORE t0, UNIT(0)(dst), .Ls_exc_p4u) |
279 | EXC( STORE t1, UNIT(1)(dst), s_exc_p3u) | 279 | EXC( STORE t1, UNIT(1)(dst), .Ls_exc_p3u) |
280 | EXC( STORE t2, UNIT(2)(dst), s_exc_p2u) | 280 | EXC( STORE t2, UNIT(2)(dst), .Ls_exc_p2u) |
281 | EXC( STORE t3, UNIT(3)(dst), s_exc_p1u) | 281 | EXC( STORE t3, UNIT(3)(dst), .Ls_exc_p1u) |
282 | .set reorder /* DADDI_WAR */ | 282 | .set reorder /* DADDI_WAR */ |
283 | ADD dst, dst, 4*NBYTES | 283 | ADD dst, dst, 4*NBYTES |
284 | beqz len, done | 284 | beqz len, .Ldone |
285 | .set noreorder | 285 | .set noreorder |
286 | less_than_4units: | 286 | .Lless_than_4units: |
287 | /* | 287 | /* |
288 | * rem = len % NBYTES | 288 | * rem = len % NBYTES |
289 | */ | 289 | */ |
290 | beq rem, len, copy_bytes | 290 | beq rem, len, .Lcopy_bytes |
291 | nop | 291 | nop |
292 | 1: | 292 | 1: |
293 | R10KCBARRIER(0(ra)) | 293 | R10KCBARRIER(0(ra)) |
294 | EXC( LOAD t0, 0(src), l_exc) | 294 | EXC( LOAD t0, 0(src), .Ll_exc) |
295 | ADD src, src, NBYTES | 295 | ADD src, src, NBYTES |
296 | SUB len, len, NBYTES | 296 | SUB len, len, NBYTES |
297 | EXC( STORE t0, 0(dst), s_exc_p1u) | 297 | EXC( STORE t0, 0(dst), .Ls_exc_p1u) |
298 | .set reorder /* DADDI_WAR */ | 298 | .set reorder /* DADDI_WAR */ |
299 | ADD dst, dst, NBYTES | 299 | ADD dst, dst, NBYTES |
300 | bne rem, len, 1b | 300 | bne rem, len, 1b |
@@ -312,17 +312,17 @@ EXC( STORE t0, 0(dst), s_exc_p1u) | |||
312 | * more instruction-level parallelism. | 312 | * more instruction-level parallelism. |
313 | */ | 313 | */ |
314 | #define bits t2 | 314 | #define bits t2 |
315 | beqz len, done | 315 | beqz len, .Ldone |
316 | ADD t1, dst, len # t1 is just past last byte of dst | 316 | ADD t1, dst, len # t1 is just past last byte of dst |
317 | li bits, 8*NBYTES | 317 | li bits, 8*NBYTES |
318 | SLL rem, len, 3 # rem = number of bits to keep | 318 | SLL rem, len, 3 # rem = number of bits to keep |
319 | EXC( LOAD t0, 0(src), l_exc) | 319 | EXC( LOAD t0, 0(src), .Ll_exc) |
320 | SUB bits, bits, rem # bits = number of bits to discard | 320 | SUB bits, bits, rem # bits = number of bits to discard |
321 | SHIFT_DISCARD t0, t0, bits | 321 | SHIFT_DISCARD t0, t0, bits |
322 | EXC( STREST t0, -1(t1), s_exc) | 322 | EXC( STREST t0, -1(t1), .Ls_exc) |
323 | jr ra | 323 | jr ra |
324 | move len, zero | 324 | move len, zero |
325 | dst_unaligned: | 325 | .Ldst_unaligned: |
326 | /* | 326 | /* |
327 | * dst is unaligned | 327 | * dst is unaligned |
328 | * t0 = src & ADDRMASK | 328 | * t0 = src & ADDRMASK |
@@ -333,23 +333,23 @@ dst_unaligned: | |||
333 | * Set match = (src and dst have same alignment) | 333 | * Set match = (src and dst have same alignment) |
334 | */ | 334 | */ |
335 | #define match rem | 335 | #define match rem |
336 | EXC( LDFIRST t3, FIRST(0)(src), l_exc) | 336 | EXC( LDFIRST t3, FIRST(0)(src), .Ll_exc) |
337 | ADD t2, zero, NBYTES | 337 | ADD t2, zero, NBYTES |
338 | EXC( LDREST t3, REST(0)(src), l_exc_copy) | 338 | EXC( LDREST t3, REST(0)(src), .Ll_exc_copy) |
339 | SUB t2, t2, t1 # t2 = number of bytes copied | 339 | SUB t2, t2, t1 # t2 = number of bytes copied |
340 | xor match, t0, t1 | 340 | xor match, t0, t1 |
341 | R10KCBARRIER(0(ra)) | 341 | R10KCBARRIER(0(ra)) |
342 | EXC( STFIRST t3, FIRST(0)(dst), s_exc) | 342 | EXC( STFIRST t3, FIRST(0)(dst), .Ls_exc) |
343 | beq len, t2, done | 343 | beq len, t2, .Ldone |
344 | SUB len, len, t2 | 344 | SUB len, len, t2 |
345 | ADD dst, dst, t2 | 345 | ADD dst, dst, t2 |
346 | beqz match, both_aligned | 346 | beqz match, .Lboth_aligned |
347 | ADD src, src, t2 | 347 | ADD src, src, t2 |
348 | 348 | ||
349 | src_unaligned_dst_aligned: | 349 | .Lsrc_unaligned_dst_aligned: |
350 | SRL t0, len, LOG_NBYTES+2 # +2 for 4 units/iter | 350 | SRL t0, len, LOG_NBYTES+2 # +2 for 4 units/iter |
351 | PREF( 0, 3*32(src) ) | 351 | PREF( 0, 3*32(src) ) |
352 | beqz t0, cleanup_src_unaligned | 352 | beqz t0, .Lcleanup_src_unaligned |
353 | and rem, len, (4*NBYTES-1) # rem = len % 4*NBYTES | 353 | and rem, len, (4*NBYTES-1) # rem = len % 4*NBYTES |
354 | PREF( 1, 3*32(dst) ) | 354 | PREF( 1, 3*32(dst) ) |
355 | 1: | 355 | 1: |
@@ -360,58 +360,58 @@ src_unaligned_dst_aligned: | |||
360 | * are to the same unit (unless src is aligned, but it's not). | 360 | * are to the same unit (unless src is aligned, but it's not). |
361 | */ | 361 | */ |
362 | R10KCBARRIER(0(ra)) | 362 | R10KCBARRIER(0(ra)) |
363 | EXC( LDFIRST t0, FIRST(0)(src), l_exc) | 363 | EXC( LDFIRST t0, FIRST(0)(src), .Ll_exc) |
364 | EXC( LDFIRST t1, FIRST(1)(src), l_exc_copy) | 364 | EXC( LDFIRST t1, FIRST(1)(src), .Ll_exc_copy) |
365 | SUB len, len, 4*NBYTES | 365 | SUB len, len, 4*NBYTES |
366 | EXC( LDREST t0, REST(0)(src), l_exc_copy) | 366 | EXC( LDREST t0, REST(0)(src), .Ll_exc_copy) |
367 | EXC( LDREST t1, REST(1)(src), l_exc_copy) | 367 | EXC( LDREST t1, REST(1)(src), .Ll_exc_copy) |
368 | EXC( LDFIRST t2, FIRST(2)(src), l_exc_copy) | 368 | EXC( LDFIRST t2, FIRST(2)(src), .Ll_exc_copy) |
369 | EXC( LDFIRST t3, FIRST(3)(src), l_exc_copy) | 369 | EXC( LDFIRST t3, FIRST(3)(src), .Ll_exc_copy) |
370 | EXC( LDREST t2, REST(2)(src), l_exc_copy) | 370 | EXC( LDREST t2, REST(2)(src), .Ll_exc_copy) |
371 | EXC( LDREST t3, REST(3)(src), l_exc_copy) | 371 | EXC( LDREST t3, REST(3)(src), .Ll_exc_copy) |
372 | PREF( 0, 9*32(src) ) # 0 is PREF_LOAD (not streamed) | 372 | PREF( 0, 9*32(src) ) # 0 is PREF_LOAD (not streamed) |
373 | ADD src, src, 4*NBYTES | 373 | ADD src, src, 4*NBYTES |
374 | #ifdef CONFIG_CPU_SB1 | 374 | #ifdef CONFIG_CPU_SB1 |
375 | nop # improves slotting | 375 | nop # improves slotting |
376 | #endif | 376 | #endif |
377 | EXC( STORE t0, UNIT(0)(dst), s_exc_p4u) | 377 | EXC( STORE t0, UNIT(0)(dst), .Ls_exc_p4u) |
378 | EXC( STORE t1, UNIT(1)(dst), s_exc_p3u) | 378 | EXC( STORE t1, UNIT(1)(dst), .Ls_exc_p3u) |
379 | EXC( STORE t2, UNIT(2)(dst), s_exc_p2u) | 379 | EXC( STORE t2, UNIT(2)(dst), .Ls_exc_p2u) |
380 | EXC( STORE t3, UNIT(3)(dst), s_exc_p1u) | 380 | EXC( STORE t3, UNIT(3)(dst), .Ls_exc_p1u) |
381 | PREF( 1, 9*32(dst) ) # 1 is PREF_STORE (not streamed) | 381 | PREF( 1, 9*32(dst) ) # 1 is PREF_STORE (not streamed) |
382 | .set reorder /* DADDI_WAR */ | 382 | .set reorder /* DADDI_WAR */ |
383 | ADD dst, dst, 4*NBYTES | 383 | ADD dst, dst, 4*NBYTES |
384 | bne len, rem, 1b | 384 | bne len, rem, 1b |
385 | .set noreorder | 385 | .set noreorder |
386 | 386 | ||
387 | cleanup_src_unaligned: | 387 | .Lcleanup_src_unaligned: |
388 | beqz len, done | 388 | beqz len, .Ldone |
389 | and rem, len, NBYTES-1 # rem = len % NBYTES | 389 | and rem, len, NBYTES-1 # rem = len % NBYTES |
390 | beq rem, len, copy_bytes | 390 | beq rem, len, .Lcopy_bytes |
391 | nop | 391 | nop |
392 | 1: | 392 | 1: |
393 | R10KCBARRIER(0(ra)) | 393 | R10KCBARRIER(0(ra)) |
394 | EXC( LDFIRST t0, FIRST(0)(src), l_exc) | 394 | EXC( LDFIRST t0, FIRST(0)(src), .Ll_exc) |
395 | EXC( LDREST t0, REST(0)(src), l_exc_copy) | 395 | EXC( LDREST t0, REST(0)(src), .Ll_exc_copy) |
396 | ADD src, src, NBYTES | 396 | ADD src, src, NBYTES |
397 | SUB len, len, NBYTES | 397 | SUB len, len, NBYTES |
398 | EXC( STORE t0, 0(dst), s_exc_p1u) | 398 | EXC( STORE t0, 0(dst), .Ls_exc_p1u) |
399 | .set reorder /* DADDI_WAR */ | 399 | .set reorder /* DADDI_WAR */ |
400 | ADD dst, dst, NBYTES | 400 | ADD dst, dst, NBYTES |
401 | bne len, rem, 1b | 401 | bne len, rem, 1b |
402 | .set noreorder | 402 | .set noreorder |
403 | 403 | ||
404 | copy_bytes_checklen: | 404 | .Lcopy_bytes_checklen: |
405 | beqz len, done | 405 | beqz len, .Ldone |
406 | nop | 406 | nop |
407 | copy_bytes: | 407 | .Lcopy_bytes: |
408 | /* 0 < len < NBYTES */ | 408 | /* 0 < len < NBYTES */ |
409 | R10KCBARRIER(0(ra)) | 409 | R10KCBARRIER(0(ra)) |
410 | #define COPY_BYTE(N) \ | 410 | #define COPY_BYTE(N) \ |
411 | EXC( lb t0, N(src), l_exc); \ | 411 | EXC( lb t0, N(src), .Ll_exc); \ |
412 | SUB len, len, 1; \ | 412 | SUB len, len, 1; \ |
413 | beqz len, done; \ | 413 | beqz len, .Ldone; \ |
414 | EXC( sb t0, N(dst), s_exc_p1) | 414 | EXC( sb t0, N(dst), .Ls_exc_p1) |
415 | 415 | ||
416 | COPY_BYTE(0) | 416 | COPY_BYTE(0) |
417 | COPY_BYTE(1) | 417 | COPY_BYTE(1) |
@@ -421,16 +421,16 @@ EXC( sb t0, N(dst), s_exc_p1) | |||
421 | COPY_BYTE(4) | 421 | COPY_BYTE(4) |
422 | COPY_BYTE(5) | 422 | COPY_BYTE(5) |
423 | #endif | 423 | #endif |
424 | EXC( lb t0, NBYTES-2(src), l_exc) | 424 | EXC( lb t0, NBYTES-2(src), .Ll_exc) |
425 | SUB len, len, 1 | 425 | SUB len, len, 1 |
426 | jr ra | 426 | jr ra |
427 | EXC( sb t0, NBYTES-2(dst), s_exc_p1) | 427 | EXC( sb t0, NBYTES-2(dst), .Ls_exc_p1) |
428 | done: | 428 | .Ldone: |
429 | jr ra | 429 | jr ra |
430 | nop | 430 | nop |
431 | END(memcpy) | 431 | END(memcpy) |
432 | 432 | ||
433 | l_exc_copy: | 433 | .Ll_exc_copy: |
434 | /* | 434 | /* |
435 | * Copy bytes from src until faulting load address (or until a | 435 | * Copy bytes from src until faulting load address (or until a |
436 | * lb faults) | 436 | * lb faults) |
@@ -445,14 +445,14 @@ l_exc_copy: | |||
445 | nop | 445 | nop |
446 | LOAD t0, THREAD_BUADDR(t0) | 446 | LOAD t0, THREAD_BUADDR(t0) |
447 | 1: | 447 | 1: |
448 | EXC( lb t1, 0(src), l_exc) | 448 | EXC( lb t1, 0(src), .Ll_exc) |
449 | ADD src, src, 1 | 449 | ADD src, src, 1 |
450 | sb t1, 0(dst) # can't fault -- we're copy_from_user | 450 | sb t1, 0(dst) # can't fault -- we're copy_from_user |
451 | .set reorder /* DADDI_WAR */ | 451 | .set reorder /* DADDI_WAR */ |
452 | ADD dst, dst, 1 | 452 | ADD dst, dst, 1 |
453 | bne src, t0, 1b | 453 | bne src, t0, 1b |
454 | .set noreorder | 454 | .set noreorder |
455 | l_exc: | 455 | .Ll_exc: |
456 | LOAD t0, TI_TASK($28) | 456 | LOAD t0, TI_TASK($28) |
457 | nop | 457 | nop |
458 | LOAD t0, THREAD_BUADDR(t0) # t0 is just past last good address | 458 | LOAD t0, THREAD_BUADDR(t0) # t0 is just past last good address |
@@ -471,7 +471,7 @@ l_exc: | |||
471 | */ | 471 | */ |
472 | .set reorder /* DADDI_WAR */ | 472 | .set reorder /* DADDI_WAR */ |
473 | SUB src, len, 1 | 473 | SUB src, len, 1 |
474 | beqz len, done | 474 | beqz len, .Ldone |
475 | .set noreorder | 475 | .set noreorder |
476 | 1: sb zero, 0(dst) | 476 | 1: sb zero, 0(dst) |
477 | ADD dst, dst, 1 | 477 | ADD dst, dst, 1 |
@@ -492,7 +492,7 @@ l_exc: | |||
492 | 492 | ||
493 | #define SEXC(n) \ | 493 | #define SEXC(n) \ |
494 | .set reorder; /* DADDI_WAR */ \ | 494 | .set reorder; /* DADDI_WAR */ \ |
495 | s_exc_p ## n ## u: \ | 495 | .Ls_exc_p ## n ## u: \ |
496 | ADD len, len, n*NBYTES; \ | 496 | ADD len, len, n*NBYTES; \ |
497 | jr ra; \ | 497 | jr ra; \ |
498 | .set noreorder | 498 | .set noreorder |
@@ -506,12 +506,12 @@ SEXC(3) | |||
506 | SEXC(2) | 506 | SEXC(2) |
507 | SEXC(1) | 507 | SEXC(1) |
508 | 508 | ||
509 | s_exc_p1: | 509 | .Ls_exc_p1: |
510 | .set reorder /* DADDI_WAR */ | 510 | .set reorder /* DADDI_WAR */ |
511 | ADD len, len, 1 | 511 | ADD len, len, 1 |
512 | jr ra | 512 | jr ra |
513 | .set noreorder | 513 | .set noreorder |
514 | s_exc: | 514 | .Ls_exc: |
515 | jr ra | 515 | jr ra |
516 | nop | 516 | nop |
517 | 517 | ||
@@ -522,20 +522,20 @@ LEAF(memmove) | |||
522 | sltu t0, a1, t0 # dst + len <= src -> memcpy | 522 | sltu t0, a1, t0 # dst + len <= src -> memcpy |
523 | sltu t1, a0, t1 # dst >= src + len -> memcpy | 523 | sltu t1, a0, t1 # dst >= src + len -> memcpy |
524 | and t0, t1 | 524 | and t0, t1 |
525 | beqz t0, __memcpy | 525 | beqz t0, .L__memcpy |
526 | move v0, a0 /* return value */ | 526 | move v0, a0 /* return value */ |
527 | beqz a2, r_out | 527 | beqz a2, .Lr_out |
528 | END(memmove) | 528 | END(memmove) |
529 | 529 | ||
530 | /* fall through to __rmemcpy */ | 530 | /* fall through to __rmemcpy */ |
531 | LEAF(__rmemcpy) /* a0=dst a1=src a2=len */ | 531 | LEAF(__rmemcpy) /* a0=dst a1=src a2=len */ |
532 | sltu t0, a1, a0 | 532 | sltu t0, a1, a0 |
533 | beqz t0, r_end_bytes_up # src >= dst | 533 | beqz t0, .Lr_end_bytes_up # src >= dst |
534 | nop | 534 | nop |
535 | ADD a0, a2 # dst = dst + len | 535 | ADD a0, a2 # dst = dst + len |
536 | ADD a1, a2 # src = src + len | 536 | ADD a1, a2 # src = src + len |
537 | 537 | ||
538 | r_end_bytes: | 538 | .Lr_end_bytes: |
539 | R10KCBARRIER(0(ra)) | 539 | R10KCBARRIER(0(ra)) |
540 | lb t0, -1(a1) | 540 | lb t0, -1(a1) |
541 | SUB a2, a2, 0x1 | 541 | SUB a2, a2, 0x1 |
@@ -543,14 +543,14 @@ r_end_bytes: | |||
543 | SUB a1, a1, 0x1 | 543 | SUB a1, a1, 0x1 |
544 | .set reorder /* DADDI_WAR */ | 544 | .set reorder /* DADDI_WAR */ |
545 | SUB a0, a0, 0x1 | 545 | SUB a0, a0, 0x1 |
546 | bnez a2, r_end_bytes | 546 | bnez a2, .Lr_end_bytes |
547 | .set noreorder | 547 | .set noreorder |
548 | 548 | ||
549 | r_out: | 549 | .Lr_out: |
550 | jr ra | 550 | jr ra |
551 | move a2, zero | 551 | move a2, zero |
552 | 552 | ||
553 | r_end_bytes_up: | 553 | .Lr_end_bytes_up: |
554 | R10KCBARRIER(0(ra)) | 554 | R10KCBARRIER(0(ra)) |
555 | lb t0, (a1) | 555 | lb t0, (a1) |
556 | SUB a2, a2, 0x1 | 556 | SUB a2, a2, 0x1 |
@@ -558,7 +558,7 @@ r_end_bytes_up: | |||
558 | ADD a1, a1, 0x1 | 558 | ADD a1, a1, 0x1 |
559 | .set reorder /* DADDI_WAR */ | 559 | .set reorder /* DADDI_WAR */ |
560 | ADD a0, a0, 0x1 | 560 | ADD a0, a0, 0x1 |
561 | bnez a2, r_end_bytes_up | 561 | bnez a2, .Lr_end_bytes_up |
562 | .set noreorder | 562 | .set noreorder |
563 | 563 | ||
564 | jr ra | 564 | jr ra |