page 132 title memcpy Copy source memory bytes to destination memcpy a

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
page ,132
title memcpy - Copy source memory bytes to destination
;***
;memcpy.asm - contains memcpy and memmove routines
;
; Copyright (c) Microsoft Corporation. All rights reserved.
;
;Purpose:
; memcpy() copies a source memory buffer to a destination buffer.
; Overlapping buffers are not treated specially, so propogation may occur.
; memmove() copies a source memory buffer to a destination buffer.
; Overlapping buffers are treated specially, to avoid propogation.
;
;*******************************************************************************
.xlist
include cruntime.inc
.list
M_EXIT macro
ret ; _cdecl return
endm ; M_EXIT
CODESEG
extrn _VEC_memcpy:near
extrn __sse2_available:dword
page
;***
;memcpy - Copy source buffer to destination buffer
;
;Purpose:
; memcpy() copies a source memory buffer to a destination memory buffer.
; This routine does NOT recognize overlapping buffers, and thus can lead
; to propogation.
; For cases where propogation must be avoided, memmove() must be used.
;
; Algorithm:
;
; Same as memmove. See Below
;
;
;memmove - Copy source buffer to destination buffer
;
;Purpose:
; memmove() copies a source memory buffer to a destination memory buffer.
; This routine recognize overlapping buffers to avoid propogation.
; For cases where propogation is not a problem, memcpy() can be used.
;
; Algorithm:
;
; void * memmove(void * dst, void * src, size_t count)
; {
; void * ret = dst;
;
; if (dst <= src || dst >= (src + count)) {
; /*
; * Non-Overlapping Buffers
; * copy from lower addresses to higher addresses
; */
; while (count--)
; *dst++ = *src++;
; }
; else {
; /*
; * Overlapping Buffers
; * copy from higher addresses to lower addresses
; */
; dst += count - 1;
; src += count - 1;
;
; while (count--)
; *dst-- = *src--;
; }
;
; return(ret);
; }
;
;
;Entry:
; void *dst = pointer to destination buffer
; const void *src = pointer to source buffer
; size_t count = number of bytes to copy
;
;Exit:
; Returns a pointer to the destination buffer in AX/DX:AX
;
;Uses:
; CX, DX
;
;Exceptions:
;*******************************************************************************
ifdef MEM_MOVE
_MEM_ equ <memmove>
else ; MEM_MOVE
_MEM_ equ <memcpy>
endif ; MEM_MOVE
% public _MEM_
_MEM_ proc \
dst:ptr byte, \
src:ptr byte, \
count:IWORD
; destination pointer
; source pointer
; number of bytes to copy
; push ebp ;U - save old frame pointer
; mov ebp, esp ;V - set new frame pointer
push edi ;U - save edi
push esi ;V - save esi
mov esi,[src] ;U - esi = source
mov ecx,[count] ;V - ecx = number of bytes to move
mov edi,[dst] ;U - edi = dest
;
; Check for overlapping buffers:
; If (dst <= src) Or (dst >= src + Count) Then
; Do normal (Upwards) Copy
; Else
; Do Downwards Copy to avoid propagation
;
mov eax,ecx ;V - eax = byte count...
mov edx,ecx ;U - edx = byte count...
add eax,esi ;V - eax = point past source end
cmp edi,esi ;U - dst <= src ?
jbe short CopyUp ;V - yes, copy toward higher addresses
cmp edi,eax ;U - dst < (src + count) ?
jb CopyDown ;V - yes, copy toward lower addresses
;
; Copy toward higher addresses.
;
CopyUp:
;
; First, see if we can use a "fast" copy SSE2 routine
; block size greater than min threshold?
cmp ecx,0100h
jb Dword_align
; SSE2 supported?
cmp DWORD PTR __sse2_available,0
je Dword_align
; alignments equal?
push edi
push esi
and edi,15
and esi,15
cmp edi,esi
pop esi
pop edi
jne Dword_align
; do fast SSE2 copy, params already on stack
pop esi
pop edi
pop ebp
jmp _VEC_memcpy
; no return
;
; The algorithm for forward moves is to align the destination to a dword
; boundary and so we can move dwords with an aligned destination. This
; occurs in 3 steps.
;
; - move x = ((4 - Dest & 3) & 3) bytes
; - move y = ((L-x) >> 2) dwords
; - move (L - x - y*4) bytes
;
Dword_align:
test edi,11b ;U - destination dword aligned?
jnz short CopyLeadUp ;V - if we are not dword aligned already, align
shr ecx,2 ;U - shift down to dword count
and edx,11b ;V - trailing byte count
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindUp ;V - if so, then jump
rep movsd ;N - move all of our dwords
jmp dword ptr TrailUpVec[edx*4] ;N - process trailing bytes
;
; Code to do optimal memory copies for non-dword-aligned destinations.
;
; The following length check is done for two reasons:
;
; 1. to ensure that the actual move length is greater than any possiale
; alignment move, and
;
; 2. to skip the multiple move logic for small moves where it would
; be faster to move the bytes with one instruction.
;
align @WordSize
CopyLeadUp:
mov eax,edi ;U - get destination offset
mov edx,11b ;V - prepare for mask
sub ecx,4 ;U - check for really short string - sub for adjust
jb short ByteCopyUp ;V - branch to just copy bytes
and eax,11b ;U - get offset within first dword
add ecx,eax ;V - update size after leading bytes copied
jmp dword ptr LeadUpVec[eax*4-4] ;N - process leading bytes
align @WordSize
ByteCopyUp:
jmp dword ptr TrailUpVec[ecx*4+16] ;N - process just bytes
align @WordSize
CopyUnwindUp:
jmp dword ptr UnwindUpVec[ecx*4] ;N - unwind dword copy
align @WordSize
LeadUpVec dd LeadUp1, LeadUp2, LeadUp3
align @WordSize
LeadUp1:
and edx,ecx ;U - trailing byte count
mov al,[esi] ;V - get first byte from source
mov [edi],al ;U - write second byte to destination
mov al,[esi+1] ;V - get second byte from source
mov [edi+1],al ;U - write second byte to destination
mov al,[esi+2] ;V - get third byte from source
shr ecx,2 ;U - shift down to dword count
mov [edi+2],al ;V - write third byte to destination
add esi,3 ;U - advance source pointer
add edi,3 ;V - advance destination pointer
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindUp ;V - if so, then jump
rep movsd ;N - move all of our dwords
jmp dword ptr TrailUpVec[edx*4] ;N - process trailing bytes
align @WordSize
LeadUp2:
and edx,ecx ;U - trailing byte count
mov al,[esi] ;V - get first byte from source
mov [edi],al ;U - write second byte to destination
mov al,[esi+1] ;V - get second byte from source
shr ecx,2 ;U - shift down to dword count
mov [edi+1],al ;V - write second byte to destination
add esi,2 ;U - advance source pointer
add edi,2 ;V - advance destination pointer
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindUp ;V - if so, then jump
rep movsd ;N - move all of our dwords
jmp dword ptr TrailUpVec[edx*4] ;N - process trailing bytes
align @WordSize
LeadUp3:
and edx,ecx ;U - trailing byte count
mov al,[esi] ;V - get first byte from source
mov [edi],al ;U - write second byte to destination
add esi,1 ;V - advance source pointer
shr ecx,2 ;U - shift down to dword count
add edi,1 ;V - advance destination pointer
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindUp ;V - if so, then jump
rep movsd ;N - move all of our dwords
jmp dword ptr TrailUpVec[edx*4] ;N - process trailing bytes
align @WordSize
UnwindUpVec dd UnwindUp0, UnwindUp1, UnwindUp2, UnwindUp3
dd UnwindUp4, UnwindUp5, UnwindUp6, UnwindUp7
UnwindUp7:
mov eax,[esi+ecx*4-28] ;U - get dword from source
;V - spare
mov [edi+ecx*4-28],eax ;U - put dword into destination
UnwindUp6:
mov eax,[esi+ecx*4-24] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-24],eax ;U - put dword into destination
UnwindUp5:
mov eax,[esi+ecx*4-20] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-20],eax ;U - put dword into destination
UnwindUp4:
mov eax,[esi+ecx*4-16] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-16],eax ;U - put dword into destination
UnwindUp3:
mov eax,[esi+ecx*4-12] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-12],eax ;U - put dword into destination
UnwindUp2:
mov eax,[esi+ecx*4-8] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-8],eax ;U - put dword into destination
UnwindUp1:
mov eax,[esi+ecx*4-4] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4-4],eax ;U - put dword into destination
lea eax,[ecx*4] ;V - compute update for pointer
add esi,eax ;U - update source pointer
add edi,eax ;V - update destination pointer
UnwindUp0:
jmp dword ptr TrailUpVec[edx*4] ;N - process trailing bytes
;-----------------------------------------------------------------------------
align @WordSize
TrailUpVec dd TrailUp0, TrailUp1, TrailUp2, TrailUp3
align @WordSize
TrailUp0:
mov eax,[dst] ;U - return pointer to destination
pop esi ;V - restore esi
pop edi ;U - restore edi
;V - spare
M_EXIT
align @WordSize
TrailUp1:
mov al,[esi] ;U - get byte from source
;V - spare
mov [edi],al ;U - put byte in destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
align @WordSize
TrailUp2:
mov al,[esi] ;U - get first byte from source
;V - spare
mov [edi],al ;U - put first byte into destination
mov al,[esi+1] ;V - get second byte from source
mov [edi+1],al ;U - put second byte into destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
align @WordSize
TrailUp3:
mov al,[esi] ;U - get first byte from source
;V - spare
mov [edi],al ;U - put first byte into destination
mov al,[esi+1] ;V - get second byte from source
mov [edi+1],al ;U - put second byte into destination
mov al,[esi+2] ;V - get third byte from source
mov [edi+2],al ;U - put third byte into destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
;-----------------------------------------------------------------------------
;-----------------------------------------------------------------------------
;-----------------------------------------------------------------------------
;
; Copy down to avoid propogation in overlapping buffers.
;
align @WordSize
CopyDown:
lea esi,[esi+ecx-4] ;U - point to 4 bytes before src buffer end
lea edi,[edi+ecx-4] ;V - point to 4 bytes before dest buffer end
;
; See if the destination start is dword aligned
;
test edi,11b ;U - test if dword aligned
jnz short CopyLeadDown ;V - if not, jump
shr ecx,2 ;U - shift down to dword count
and edx,11b ;V - trailing byte count
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindDown ;V - if so, then jump
std ;N - set direction flag
rep movsd ;N - move all of our dwords
cld ;N - clear direction flag back
jmp dword ptr TrailDownVec[edx*4] ;N - process trailing bytes
align @WordSize
CopyUnwindDown:
neg ecx ;U - negate dword count for table merging
;V - spare
jmp dword ptr UnwindDownVec[ecx*4+28] ;N - unwind copy
align @WordSize
CopyLeadDown:
mov eax,edi ;U - get destination offset
mov edx,11b ;V - prepare for mask
cmp ecx,4 ;U - check for really short string
jb short ByteCopyDown ;V - branch to just copy bytes
and eax,11b ;U - get offset within first dword
sub ecx,eax ;U - to update size after lead copied
jmp dword ptr LeadDownVec[eax*4-4] ;N - process leading bytes
align @WordSize
ByteCopyDown:
jmp dword ptr TrailDownVec[ecx*4] ;N - process just bytes
align @WordSize
LeadDownVec dd LeadDown1, LeadDown2, LeadDown3
align @WordSize
LeadDown1:
mov al,[esi+3] ;U - load first byte
and edx,ecx ;V - trailing byte count
mov [edi+3],al ;U - write out first byte
sub esi,1 ;V - point to last src dword
shr ecx,2 ;U - shift down to dword count
sub edi,1 ;V - point to last dest dword
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindDown ;V - if so, then jump
std ;N - set direction flag
rep movsd ;N - move all of our dwords
cld ;N - clear direction flag
jmp dword ptr TrailDownVec[edx*4] ;N - process trailing bytes
align @WordSize
LeadDown2:
mov al,[esi+3] ;U - load first byte
and edx,ecx ;V - trailing byte count
mov [edi+3],al ;U - write out first byte
mov al,[esi+2] ;V - get second byte from source
shr ecx,2 ;U - shift down to dword count
mov [edi+2],al ;V - write second byte to destination
sub esi,2 ;U - point to last src dword
sub edi,2 ;V - point to last dest dword
cmp ecx,8 ;U - test if small enough for unwind copy
jb short CopyUnwindDown ;V - if so, then jump
std ;N - set direction flag
rep movsd ;N - move all of our dwords
cld ;N - clear direction flag
jmp dword ptr TrailDownVec[edx*4] ;N - process trailing bytes
align @WordSize
LeadDown3:
mov al,[esi+3] ;U - load first byte
and edx,ecx ;V - trailing byte count
mov [edi+3],al ;U - write out first byte
mov al,[esi+2] ;V - get second byte from source
mov [edi+2],al ;U - write second byte to destination
mov al,[esi+1] ;V - get third byte from source
shr ecx,2 ;U - shift down to dword count
mov [edi+1],al ;V - write third byte to destination
sub esi,3 ;U - point to last src dword
sub edi,3 ;V - point to last dest dword
cmp ecx,8 ;U - test if small enough for unwind copy
jb CopyUnwindDown ;V - if so, then jump
std ;N - set direction flag
rep movsd ;N - move all of our dwords
cld ;N - clear direction flag
jmp dword ptr TrailDownVec[edx*4] ;N - process trailing bytes
;------------------------------------------------------------------
align @WordSize
UnwindDownVec dd UnwindDown7, UnwindDown6, UnwindDown5, UnwindDown4
dd UnwindDown3, UnwindDown2, UnwindDown1, UnwindDown0
UnwindDown7:
mov eax,[esi+ecx*4+28] ;U - get dword from source
;V - spare
mov [edi+ecx*4+28],eax ;U - put dword into destination
UnwindDown6:
mov eax,[esi+ecx*4+24] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+24],eax ;U - put dword into destination
UnwindDown5:
mov eax,[esi+ecx*4+20] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+20],eax ;U - put dword into destination
UnwindDown4:
mov eax,[esi+ecx*4+16] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+16],eax ;U - put dword into destination
UnwindDown3:
mov eax,[esi+ecx*4+12] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+12],eax ;U - put dword into destination
UnwindDown2:
mov eax,[esi+ecx*4+8] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+8],eax ;U - put dword into destination
UnwindDown1:
mov eax,[esi+ecx*4+4] ;U(entry)/V(not) - get dword from source
;V(entry) - spare
mov [edi+ecx*4+4],eax ;U - put dword into destination
lea eax,[ecx*4] ;V - compute update for pointer
add esi,eax ;U - update source pointer
add edi,eax ;V - update destination pointer
UnwindDown0:
jmp dword ptr TrailDownVec[edx*4] ;N - process trailing bytes
;-----------------------------------------------------------------------------
align @WordSize
TrailDownVec dd TrailDown0, TrailDown1, TrailDown2, TrailDown3
align @WordSize
TrailDown0:
mov eax,[dst] ;U - return pointer to destination
;V - spare
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
align @WordSize
TrailDown1:
mov al,[esi+3] ;U - get byte from source
;V - spare
mov [edi+3],al ;U - put byte in destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
align @WordSize
TrailDown2:
mov al,[esi+3] ;U - get first byte from source
;V - spare
mov [edi+3],al ;U - put first byte into destination
mov al,[esi+2] ;V - get second byte from source
mov [edi+2],al ;U - put second byte into destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
align @WordSize
TrailDown3:
mov al,[esi+3] ;U - get first byte from source
;V - spare
mov [edi+3],al ;U - put first byte into destination
mov al,[esi+2] ;V - get second byte from source
mov [edi+2],al ;U - put second byte into destination
mov al,[esi+1] ;V - get third byte from source
mov [edi+1],al ;U - put third byte into destination
mov eax,[dst] ;V - return pointer to destination
pop esi ;U - restore esi
pop edi ;V - restore edi
M_EXIT
_MEM_ endp
end