summaryrefslogtreecommitdiff
path: root/src/runtime/asm_amd64p32.s
blob: 99c8569bc94c538445a6419dfd306aa6ab81c29b (plain)
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
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

#include "go_asm.h"
#include "go_tls.h"
#include "funcdata.h"
#include "textflag.h"

TEXT runtime·rt0_go(SB),NOSPLIT,$0
	// copy arguments forward on an even stack
	MOVL	argc+0(FP), AX
	MOVL	argv+4(FP), BX
	MOVL	SP, CX
	SUBL	$128, SP		// plenty of scratch
	ANDL	$~15, CX
	MOVL	CX, SP

	MOVL	AX, 16(SP)
	MOVL	BX, 24(SP)
	
	// create istack out of the given (operating system) stack.
	MOVL	$runtime·g0(SB), DI
	LEAL	(-64*1024+104)(SP), BX
	MOVL	BX, g_stackguard0(DI)
	MOVL	BX, g_stackguard1(DI)
	MOVL	BX, (g_stack+stack_lo)(DI)
	MOVL	SP, (g_stack+stack_hi)(DI)

	// find out information about the processor we're on
	MOVQ	$0, AX
	CPUID
	CMPQ	AX, $0
	JE	nocpuinfo
	MOVQ	$1, AX
	CPUID
	MOVL	CX, runtime·cpuid_ecx(SB)
	MOVL	DX, runtime·cpuid_edx(SB)
nocpuinfo:	
	
needtls:
	LEAL	runtime·tls0(SB), DI
	CALL	runtime·settls(SB)

	// store through it, to make sure it works
	get_tls(BX)
	MOVQ	$0x123, g(BX)
	MOVQ	runtime·tls0(SB), AX
	CMPQ	AX, $0x123
	JEQ 2(PC)
	MOVL	AX, 0	// abort
ok:
	// set the per-goroutine and per-mach "registers"
	get_tls(BX)
	LEAL	runtime·g0(SB), CX
	MOVL	CX, g(BX)
	LEAL	runtime·m0(SB), AX

	// save m->g0 = g0
	MOVL	CX, m_g0(AX)
	// save m0 to g0->m
	MOVL	AX, g_m(CX)

	CLD				// convention is D is always left cleared
	CALL	runtime·check(SB)

	MOVL	16(SP), AX		// copy argc
	MOVL	AX, 0(SP)
	MOVL	24(SP), AX		// copy argv
	MOVL	AX, 4(SP)
	CALL	runtime·args(SB)
	CALL	runtime·osinit(SB)
	CALL	runtime·schedinit(SB)

	// create a new goroutine to start program
	MOVL	$runtime·main·f(SB), AX	// entry
	MOVL	$0, 0(SP)
	MOVL	AX, 4(SP)
	CALL	runtime·newproc(SB)

	// start this M
	CALL	runtime·mstart(SB)

	MOVL	$0xf1, 0xf1  // crash
	RET

DATA	runtime·main·f+0(SB)/4,$runtime·main(SB)
GLOBL	runtime·main·f(SB),RODATA,$4

TEXT runtime·breakpoint(SB),NOSPLIT,$0-0
	INT $3
	RET

TEXT runtime·asminit(SB),NOSPLIT,$0-0
	// No per-thread init.
	RET

/*
 *  go-routine
 */

// void gosave(Gobuf*)
// save state in Gobuf; setjmp
TEXT runtime·gosave(SB), NOSPLIT, $0-4
	MOVL	buf+0(FP), AX	// gobuf
	LEAL	buf+0(FP), BX	// caller's SP
	MOVL	BX, gobuf_sp(AX)
	MOVL	0(SP), BX		// caller's PC
	MOVL	BX, gobuf_pc(AX)
	MOVL	$0, gobuf_ctxt(AX)
	MOVQ	$0, gobuf_ret(AX)
	get_tls(CX)
	MOVL	g(CX), BX
	MOVL	BX, gobuf_g(AX)
	RET

// void gogo(Gobuf*)
// restore state from Gobuf; longjmp
TEXT runtime·gogo(SB), NOSPLIT, $0-4
	MOVL	buf+0(FP), BX		// gobuf
	MOVL	gobuf_g(BX), DX
	MOVL	0(DX), CX		// make sure g != nil
	get_tls(CX)
	MOVL	DX, g(CX)
	MOVL	gobuf_sp(BX), SP	// restore SP
	MOVL	gobuf_ctxt(BX), DX
	MOVQ	gobuf_ret(BX), AX
	MOVL	$0, gobuf_sp(BX)	// clear to help garbage collector
	MOVQ	$0, gobuf_ret(BX)
	MOVL	$0, gobuf_ctxt(BX)
	MOVL	gobuf_pc(BX), BX
	JMP	BX

// func mcall(fn func(*g))
// Switch to m->g0's stack, call fn(g).
// Fn must never return.  It should gogo(&g->sched)
// to keep running g.
TEXT runtime·mcall(SB), NOSPLIT, $0-4
	MOVL	fn+0(FP), DI
	
	get_tls(CX)
	MOVL	g(CX), AX	// save state in g->sched
	MOVL	0(SP), BX	// caller's PC
	MOVL	BX, (g_sched+gobuf_pc)(AX)
	LEAL	fn+0(FP), BX	// caller's SP
	MOVL	BX, (g_sched+gobuf_sp)(AX)
	MOVL	AX, (g_sched+gobuf_g)(AX)

	// switch to m->g0 & its stack, call fn
	MOVL	g(CX), BX
	MOVL	g_m(BX), BX
	MOVL	m_g0(BX), SI
	CMPL	SI, AX	// if g == m->g0 call badmcall
	JNE	3(PC)
	MOVL	$runtime·badmcall(SB), AX
	JMP	AX
	MOVL	SI, g(CX)	// g = m->g0
	MOVL	(g_sched+gobuf_sp)(SI), SP	// sp = m->g0->sched.sp
	PUSHQ	AX
	MOVL	DI, DX
	MOVL	0(DI), DI
	CALL	DI
	POPQ	AX
	MOVL	$runtime·badmcall2(SB), AX
	JMP	AX
	RET

// switchtoM is a dummy routine that onM leaves at the bottom
// of the G stack.  We need to distinguish the routine that
// lives at the bottom of the G stack from the one that lives
// at the top of the M stack because the one at the top of
// the M stack terminates the stack walk (see topofstack()).
TEXT runtime·switchtoM(SB), NOSPLIT, $0-0
	RET

// func onM_signalok(fn func())
TEXT runtime·onM_signalok(SB), NOSPLIT, $0-4
	get_tls(CX)
	MOVL	g(CX), AX	// AX = g
	MOVL	g_m(AX), BX	// BX = m
	MOVL	m_gsignal(BX), DX	// DX = gsignal
	CMPL	AX, DX
	JEQ	ongsignal
	JMP	runtime·onM(SB)

ongsignal:
	MOVL	fn+0(FP), DI	// DI = fn
	MOVL	DI, DX
	MOVL	0(DI), DI
	CALL	DI
	RET

// func onM(fn func())
TEXT runtime·onM(SB), NOSPLIT, $0-4
	MOVL	fn+0(FP), DI	// DI = fn
	get_tls(CX)
	MOVL	g(CX), AX	// AX = g
	MOVL	g_m(AX), BX	// BX = m

	MOVL	m_g0(BX), DX	// DX = g0
	CMPL	AX, DX
	JEQ	onm

	MOVL	m_curg(BX), R8
	CMPL	AX, R8
	JEQ	oncurg
	
	// Not g0, not curg. Must be gsignal, but that's not allowed.
	// Hide call from linker nosplit analysis.
	MOVL	$runtime·badonm(SB), AX
	CALL	AX

oncurg:
	// save our state in g->sched.  Pretend to
	// be switchtoM if the G stack is scanned.
	MOVL	$runtime·switchtoM(SB), SI
	MOVL	SI, (g_sched+gobuf_pc)(AX)
	MOVL	SP, (g_sched+gobuf_sp)(AX)
	MOVL	AX, (g_sched+gobuf_g)(AX)

	// switch to g0
	MOVL	DX, g(CX)
	MOVL	(g_sched+gobuf_sp)(DX), SP

	// call target function
	MOVL	DI, DX
	MOVL	0(DI), DI
	CALL	DI

	// switch back to g
	get_tls(CX)
	MOVL	g(CX), AX
	MOVL	g_m(AX), BX
	MOVL	m_curg(BX), AX
	MOVL	AX, g(CX)
	MOVL	(g_sched+gobuf_sp)(AX), SP
	MOVL	$0, (g_sched+gobuf_sp)(AX)
	RET

onm:
	// already on m stack, just call directly
	MOVL	DI, DX
	MOVL	0(DI), DI
	CALL	DI
	RET

/*
 * support for morestack
 */

// Called during function prolog when more stack is needed.
//
// The traceback routines see morestack on a g0 as being
// the top of a stack (for example, morestack calling newstack
// calling the scheduler calling newm calling gc), so we must
// record an argument size. For that purpose, it has no arguments.
TEXT runtime·morestack(SB),NOSPLIT,$0-0
	get_tls(CX)
	MOVL	g(CX), BX
	MOVL	g_m(BX), BX

	// Cannot grow scheduler stack (m->g0).
	MOVL	m_g0(BX), SI
	CMPL	g(CX), SI
	JNE	2(PC)
	MOVL	0, AX

	// Cannot grow signal stack (m->gsignal).
	MOVL	m_gsignal(BX), SI
	CMPL	g(CX), SI
	JNE	2(PC)
	MOVL	0, AX

	// Called from f.
	// Set m->morebuf to f's caller.
	MOVL	8(SP), AX	// f's caller's PC
	MOVL	AX, (m_morebuf+gobuf_pc)(BX)
	LEAL	16(SP), AX	// f's caller's SP
	MOVL	AX, (m_morebuf+gobuf_sp)(BX)
	get_tls(CX)
	MOVL	g(CX), SI
	MOVL	SI, (m_morebuf+gobuf_g)(BX)

	// Set g->sched to context in f.
	MOVL	0(SP), AX // f's PC
	MOVL	AX, (g_sched+gobuf_pc)(SI)
	MOVL	SI, (g_sched+gobuf_g)(SI)
	LEAL	8(SP), AX // f's SP
	MOVL	AX, (g_sched+gobuf_sp)(SI)
	MOVL	DX, (g_sched+gobuf_ctxt)(SI)

	// Call newstack on m->g0's stack.
	MOVL	m_g0(BX), BX
	MOVL	BX, g(CX)
	MOVL	(g_sched+gobuf_sp)(BX), SP
	CALL	runtime·newstack(SB)
	MOVL	$0, 0x1003	// crash if newstack returns
	RET

// morestack trampolines
TEXT runtime·morestack_noctxt(SB),NOSPLIT,$0
	MOVL	$0, DX
	JMP	runtime·morestack(SB)

// reflectcall: call a function with the given argument list
// func call(f *FuncVal, arg *byte, argsize, retoffset uint32).
// we don't have variable-sized frames, so we use a small number
// of constant-sized-frame functions to encode a few bits of size in the pc.
// Caution: ugly multiline assembly macros in your future!

#define DISPATCH(NAME,MAXSIZE)		\
	CMPL	CX, $MAXSIZE;		\
	JA	3(PC);			\
	MOVL	$NAME(SB), AX;		\
	JMP	AX
// Note: can't just "JMP NAME(SB)" - bad inlining results.

TEXT ·reflectcall(SB), NOSPLIT, $0-16
	MOVLQZX argsize+8(FP), CX
	DISPATCH(runtime·call16, 16)
	DISPATCH(runtime·call32, 32)
	DISPATCH(runtime·call64, 64)
	DISPATCH(runtime·call128, 128)
	DISPATCH(runtime·call256, 256)
	DISPATCH(runtime·call512, 512)
	DISPATCH(runtime·call1024, 1024)
	DISPATCH(runtime·call2048, 2048)
	DISPATCH(runtime·call4096, 4096)
	DISPATCH(runtime·call8192, 8192)
	DISPATCH(runtime·call16384, 16384)
	DISPATCH(runtime·call32768, 32768)
	DISPATCH(runtime·call65536, 65536)
	DISPATCH(runtime·call131072, 131072)
	DISPATCH(runtime·call262144, 262144)
	DISPATCH(runtime·call524288, 524288)
	DISPATCH(runtime·call1048576, 1048576)
	DISPATCH(runtime·call2097152, 2097152)
	DISPATCH(runtime·call4194304, 4194304)
	DISPATCH(runtime·call8388608, 8388608)
	DISPATCH(runtime·call16777216, 16777216)
	DISPATCH(runtime·call33554432, 33554432)
	DISPATCH(runtime·call67108864, 67108864)
	DISPATCH(runtime·call134217728, 134217728)
	DISPATCH(runtime·call268435456, 268435456)
	DISPATCH(runtime·call536870912, 536870912)
	DISPATCH(runtime·call1073741824, 1073741824)
	MOVL	$runtime·badreflectcall(SB), AX
	JMP	AX

#define CALLFN(NAME,MAXSIZE)			\
TEXT NAME(SB), WRAPPER, $MAXSIZE-16;		\
	NO_LOCAL_POINTERS;			\
	/* copy arguments to stack */		\
	MOVL	argptr+4(FP), SI;		\
	MOVL	argsize+8(FP), CX;		\
	MOVL	SP, DI;				\
	REP;MOVSB;				\
	/* call function */			\
	MOVL	f+0(FP), DX;			\
	MOVL	(DX), AX;			\
	CALL	AX;				\
	/* copy return values back */		\
	MOVL	argptr+4(FP), DI;		\
	MOVL	argsize+8(FP), CX;		\
	MOVL	retoffset+12(FP), BX;		\
	MOVL	SP, SI;				\
	ADDL	BX, DI;				\
	ADDL	BX, SI;				\
	SUBL	BX, CX;				\
	REP;MOVSB;				\
	RET

CALLFN(·call16, 16)
CALLFN(·call32, 32)
CALLFN(·call64, 64)
CALLFN(·call128, 128)
CALLFN(·call256, 256)
CALLFN(·call512, 512)
CALLFN(·call1024, 1024)
CALLFN(·call2048, 2048)
CALLFN(·call4096, 4096)
CALLFN(·call8192, 8192)
CALLFN(·call16384, 16384)
CALLFN(·call32768, 32768)
CALLFN(·call65536, 65536)
CALLFN(·call131072, 131072)
CALLFN(·call262144, 262144)
CALLFN(·call524288, 524288)
CALLFN(·call1048576, 1048576)
CALLFN(·call2097152, 2097152)
CALLFN(·call4194304, 4194304)
CALLFN(·call8388608, 8388608)
CALLFN(·call16777216, 16777216)
CALLFN(·call33554432, 33554432)
CALLFN(·call67108864, 67108864)
CALLFN(·call134217728, 134217728)
CALLFN(·call268435456, 268435456)
CALLFN(·call536870912, 536870912)
CALLFN(·call1073741824, 1073741824)

// bool cas(int32 *val, int32 old, int32 new)
// Atomically:
//	if(*val == old){
//		*val = new;
//		return 1;
//	} else
//		return 0;
TEXT runtime·cas(SB), NOSPLIT, $0-17
	MOVL	ptr+0(FP), BX
	MOVL	old+4(FP), AX
	MOVL	new+8(FP), CX
	LOCK
	CMPXCHGL	CX, 0(BX)
	JZ 4(PC)
	MOVL	$0, AX
	MOVB	AX, ret+16(FP)
	RET
	MOVL	$1, AX
	MOVB	AX, ret+16(FP)
	RET

TEXT runtime·casuintptr(SB), NOSPLIT, $0-17
	JMP	runtime·cas(SB)

TEXT runtime·atomicloaduintptr(SB), NOSPLIT, $0-12
	JMP	runtime·atomicload(SB)

TEXT runtime·atomicloaduint(SB), NOSPLIT, $0-12
	JMP	runtime·atomicload(SB)

TEXT runtime·atomicstoreuintptr(SB), NOSPLIT, $0-12
	JMP	runtime·atomicstore(SB)

// bool	runtime·cas64(uint64 *val, uint64 old, uint64 new)
// Atomically:
//	if(*val == *old){
//		*val = new;
//		return 1;
//	} else {
//		return 0;
//	}
TEXT runtime·cas64(SB), NOSPLIT, $0-25
	MOVL	ptr+0(FP), BX
	MOVQ	old+8(FP), AX
	MOVQ	new+16(FP), CX
	LOCK
	CMPXCHGQ	CX, 0(BX)
	JNZ	cas64_fail
	MOVL	$1, AX
	MOVB	AX, ret+24(FP)
	RET
cas64_fail:
	MOVL	$0, AX
	MOVB	AX, ret+24(FP)
	RET

// bool casp(void **val, void *old, void *new)
// Atomically:
//	if(*val == old){
//		*val = new;
//		return 1;
//	} else
//		return 0;
TEXT runtime·casp1(SB), NOSPLIT, $0-17
	MOVL	ptr+0(FP), BX
	MOVL	old+4(FP), AX
	MOVL	new+8(FP), CX
	LOCK
	CMPXCHGL	CX, 0(BX)
	JZ 4(PC)
	MOVL	$0, AX
	MOVB	AX, ret+16(FP)
	RET
	MOVL	$1, AX
	MOVB	AX, ret+16(FP)
	RET

// uint32 xadd(uint32 volatile *val, int32 delta)
// Atomically:
//	*val += delta;
//	return *val;
TEXT runtime·xadd(SB), NOSPLIT, $0-12
	MOVL	ptr+0(FP), BX
	MOVL	delta+4(FP), AX
	MOVL	AX, CX
	LOCK
	XADDL	AX, 0(BX)
	ADDL	CX, AX
	MOVL	AX, ret+8(FP)
	RET

TEXT runtime·xadd64(SB), NOSPLIT, $0-24
	MOVL	ptr+0(FP), BX
	MOVQ	delta+8(FP), AX
	MOVQ	AX, CX
	LOCK
	XADDQ	AX, 0(BX)
	ADDQ	CX, AX
	MOVQ	AX, ret+16(FP)
	RET

TEXT runtime·xchg(SB), NOSPLIT, $0-12
	MOVL	ptr+0(FP), BX
	MOVL	new+4(FP), AX
	XCHGL	AX, 0(BX)
	MOVL	AX, ret+8(FP)
	RET

TEXT runtime·xchg64(SB), NOSPLIT, $0-24
	MOVL	ptr+0(FP), BX
	MOVQ	new+8(FP), AX
	XCHGQ	AX, 0(BX)
	MOVQ	AX, ret+16(FP)
	RET

TEXT runtime·xchgp1(SB), NOSPLIT, $0-12
	MOVL	ptr+0(FP), BX
	MOVL	new+4(FP), AX
	XCHGL	AX, 0(BX)
	MOVL	AX, ret+8(FP)
	RET

TEXT runtime·xchguintptr(SB), NOSPLIT, $0-12
	JMP	runtime·xchg(SB)

TEXT runtime·procyield(SB),NOSPLIT,$0-0
	MOVL	cycles+0(FP), AX
again:
	PAUSE
	SUBL	$1, AX
	JNZ	again
	RET

TEXT runtime·atomicstorep1(SB), NOSPLIT, $0-8
	MOVL	ptr+0(FP), BX
	MOVL	val+4(FP), AX
	XCHGL	AX, 0(BX)
	RET

TEXT runtime·atomicstore(SB), NOSPLIT, $0-8
	MOVL	ptr+0(FP), BX
	MOVL	val+4(FP), AX
	XCHGL	AX, 0(BX)
	RET

TEXT runtime·atomicstore64(SB), NOSPLIT, $0-16
	MOVL	ptr+0(FP), BX
	MOVQ	val+8(FP), AX
	XCHGQ	AX, 0(BX)
	RET

// void	runtime·atomicor8(byte volatile*, byte);
TEXT runtime·atomicor8(SB), NOSPLIT, $0-5
	MOVL	ptr+0(FP), BX
	MOVB	val+4(FP), AX
	LOCK
	ORB	AX, 0(BX)
	RET

// void jmpdefer(fn, sp);
// called from deferreturn.
// 1. pop the caller
// 2. sub 5 bytes from the callers return
// 3. jmp to the argument
TEXT runtime·jmpdefer(SB), NOSPLIT, $0-8
	MOVL	fv+0(FP), DX
	MOVL	argp+4(FP), BX
	LEAL	-8(BX), SP	// caller sp after CALL
	SUBL	$5, (SP)	// return to CALL again
	MOVL	0(DX), BX
	JMP	BX	// but first run the deferred function

// asmcgocall(void(*fn)(void*), void *arg)
// Not implemented.
TEXT runtime·asmcgocall(SB),NOSPLIT,$0-8
	MOVL	0, AX
	RET

// asmcgocall(void(*fn)(void*), void *arg)
// Not implemented.
TEXT runtime·asmcgocall_errno(SB),NOSPLIT,$0-12
	MOVL	0, AX
	RET

// cgocallback(void (*fn)(void*), void *frame, uintptr framesize)
// Not implemented.
TEXT runtime·cgocallback(SB),NOSPLIT,$0-12
	MOVL	0, AX
	RET

// void setg(G*); set g. for use by needm.
// Not implemented.
TEXT runtime·setg(SB), NOSPLIT, $0-4
	MOVL	0, AX
	RET

// check that SP is in range [g->stack.lo, g->stack.hi)
TEXT runtime·stackcheck(SB), NOSPLIT, $0-0
	get_tls(CX)
	MOVL	g(CX), AX
	CMPL	(g_stack+stack_hi)(AX), SP
	JHI	2(PC)
	MOVL	0, AX
	CMPL	SP, (g_stack+stack_lo)(AX)
	JHI	2(PC)
	MOVL	0, AX
	RET

TEXT runtime·memclr(SB),NOSPLIT,$0-8
	MOVL	ptr+0(FP), DI
	MOVL	n+4(FP), CX
	MOVQ	CX, BX
	ANDQ	$7, BX
	SHRQ	$3, CX
	MOVQ	$0, AX
	CLD
	REP
	STOSQ
	MOVQ	BX, CX
	REP
	STOSB
	RET

TEXT runtime·getcallerpc(SB),NOSPLIT,$0-12
	MOVL	argp+0(FP),AX		// addr of first arg
	MOVL	-8(AX),AX		// get calling pc
	MOVL	AX, ret+8(FP)
	RET

TEXT runtime·gogetcallerpc(SB),NOSPLIT,$0-12
	MOVL	p+0(FP),AX		// addr of first arg
	MOVL	-8(AX),AX		// get calling pc
	MOVL	AX, ret+8(FP)
	RET

TEXT runtime·setcallerpc(SB),NOSPLIT,$0-8
	MOVL	argp+0(FP),AX		// addr of first arg
	MOVL	pc+4(FP), BX		// pc to set
	MOVQ	BX, -8(AX)		// set calling pc
	RET

TEXT runtime·getcallersp(SB),NOSPLIT,$0-12
	MOVL	argp+0(FP), AX
	MOVL	AX, ret+8(FP)
	RET

// func gogetcallersp(p unsafe.Pointer) uintptr
TEXT runtime·gogetcallersp(SB),NOSPLIT,$0-12
	MOVL	p+0(FP),AX		// addr of first arg
	MOVL	AX, ret+8(FP)
	RET

// int64 runtime·cputicks(void)
TEXT runtime·cputicks(SB),NOSPLIT,$0-0
	RDTSC
	SHLQ	$32, DX
	ADDQ	DX, AX
	MOVQ	AX, ret+0(FP)
	RET

// hash function using AES hardware instructions
// For now, our one amd64p32 system (NaCl) does not
// support using AES instructions, so have not bothered to
// write the implementations. Can copy and adjust the ones
// in asm_amd64.s when the time comes.

TEXT runtime·aeshash(SB),NOSPLIT,$0-20
	MOVL	AX, ret+16(FP)
	RET

TEXT runtime·aeshashstr(SB),NOSPLIT,$0-20
	MOVL	AX, ret+16(FP)
	RET

TEXT runtime·aeshash32(SB),NOSPLIT,$0-20
	MOVL	AX, ret+16(FP)
	RET

TEXT runtime·aeshash64(SB),NOSPLIT,$0-20
	MOVL	AX, ret+16(FP)
	RET

TEXT runtime·memeq(SB),NOSPLIT,$0-17
	MOVL	a+0(FP), SI
	MOVL	b+4(FP), DI
	MOVL	size+8(FP), BX
	CALL	runtime·memeqbody(SB)
	MOVB	AX, ret+16(FP)
	RET

// eqstring tests whether two strings are equal.
// See runtime_test.go:eqstring_generic for
// equivalent Go code.
TEXT runtime·eqstring(SB),NOSPLIT,$0-17
	MOVL	s1len+4(FP), AX
	MOVL	s2len+12(FP), BX
	CMPL	AX, BX
	JNE	different
	MOVL	s1str+0(FP), SI
	MOVL	s2str+8(FP), DI
	CMPL	SI, DI
	JEQ	same
	CALL	runtime·memeqbody(SB)
	MOVB	AX, v+16(FP)
	RET
same:
	MOVB	$1, v+16(FP)
	RET
different:
	MOVB	$0, v+16(FP)
	RET

// a in SI
// b in DI
// count in BX
TEXT runtime·memeqbody(SB),NOSPLIT,$0-0
	XORQ	AX, AX

	CMPQ	BX, $8
	JB	small
	
	// 64 bytes at a time using xmm registers
hugeloop:
	CMPQ	BX, $64
	JB	bigloop
	MOVOU	(SI), X0
	MOVOU	(DI), X1
	MOVOU	16(SI), X2
	MOVOU	16(DI), X3
	MOVOU	32(SI), X4
	MOVOU	32(DI), X5
	MOVOU	48(SI), X6
	MOVOU	48(DI), X7
	PCMPEQB	X1, X0
	PCMPEQB	X3, X2
	PCMPEQB	X5, X4
	PCMPEQB	X7, X6
	PAND	X2, X0
	PAND	X6, X4
	PAND	X4, X0
	PMOVMSKB X0, DX
	ADDQ	$64, SI
	ADDQ	$64, DI
	SUBQ	$64, BX
	CMPL	DX, $0xffff
	JEQ	hugeloop
	RET

	// 8 bytes at a time using 64-bit register
bigloop:
	CMPQ	BX, $8
	JBE	leftover
	MOVQ	(SI), CX
	MOVQ	(DI), DX
	ADDQ	$8, SI
	ADDQ	$8, DI
	SUBQ	$8, BX
	CMPQ	CX, DX
	JEQ	bigloop
	RET

	// remaining 0-8 bytes
leftover:
	ADDQ	BX, SI
	ADDQ	BX, DI
	MOVQ	-8(SI), CX
	MOVQ	-8(DI), DX
	CMPQ	CX, DX
	SETEQ	AX
	RET

small:
	CMPQ	BX, $0
	JEQ	equal

	LEAQ	0(BX*8), CX
	NEGQ	CX

	CMPB	SI, $0xf8
	JA	si_high

	// load at SI won't cross a page boundary.
	MOVQ	(SI), SI
	JMP	si_finish
si_high:
	// address ends in 11111xxx.  Load up to bytes we want, move to correct position.
	MOVQ	BX, DX
	ADDQ	SI, DX
	MOVQ	-8(DX), SI
	SHRQ	CX, SI
si_finish:

	// same for DI.
	CMPB	DI, $0xf8
	JA	di_high
	MOVQ	(DI), DI
	JMP	di_finish
di_high:
	MOVQ	BX, DX
	ADDQ	DI, DX
	MOVQ	-8(DX), DI
	SHRQ	CX, DI
di_finish:

	SUBQ	SI, DI
	SHLQ	CX, DI
equal:
	SETEQ	AX
	RET

TEXT runtime·cmpstring(SB),NOSPLIT,$0-20
	MOVL	s1_base+0(FP), SI
	MOVL	s1_len+4(FP), BX
	MOVL	s2_base+8(FP), DI
	MOVL	s2_len+12(FP), DX
	CALL	runtime·cmpbody(SB)
	MOVL	AX, ret+16(FP)
	RET

TEXT runtime·cmpbytes(SB),NOSPLIT,$0-28
	MOVL	s1+0(FP), SI
	MOVL	s1+4(FP), BX
	MOVL	s2+12(FP), DI
	MOVL	s2+16(FP), DX
	CALL	runtime·cmpbody(SB)
	MOVQ	AX, res+24(FP)
	RET

// input:
//   SI = a
//   DI = b
//   BX = alen
//   DX = blen
// output:
//   AX = 1/0/-1
TEXT runtime·cmpbody(SB),NOSPLIT,$0-0
	CMPQ	SI, DI
	JEQ	cmp_allsame
	CMPQ	BX, DX
	MOVQ	DX, R8
	CMOVQLT	BX, R8 // R8 = min(alen, blen) = # of bytes to compare
	CMPQ	R8, $8
	JB	cmp_small

cmp_loop:
	CMPQ	R8, $16
	JBE	cmp_0through16
	MOVOU	(SI), X0
	MOVOU	(DI), X1
	PCMPEQB X0, X1
	PMOVMSKB X1, AX
	XORQ	$0xffff, AX	// convert EQ to NE
	JNE	cmp_diff16	// branch if at least one byte is not equal
	ADDQ	$16, SI
	ADDQ	$16, DI
	SUBQ	$16, R8
	JMP	cmp_loop
	
	// AX = bit mask of differences
cmp_diff16:
	BSFQ	AX, BX	// index of first byte that differs
	XORQ	AX, AX
	ADDQ	BX, SI
	MOVB	(SI), CX
	ADDQ	BX, DI
	CMPB	CX, (DI)
	SETHI	AX
	LEAQ	-1(AX*2), AX	// convert 1/0 to +1/-1
	RET

	// 0 through 16 bytes left, alen>=8, blen>=8
cmp_0through16:
	CMPQ	R8, $8
	JBE	cmp_0through8
	MOVQ	(SI), AX
	MOVQ	(DI), CX
	CMPQ	AX, CX
	JNE	cmp_diff8
cmp_0through8:
	ADDQ	R8, SI
	ADDQ	R8, DI
	MOVQ	-8(SI), AX
	MOVQ	-8(DI), CX
	CMPQ	AX, CX
	JEQ	cmp_allsame

	// AX and CX contain parts of a and b that differ.
cmp_diff8:
	BSWAPQ	AX	// reverse order of bytes
	BSWAPQ	CX
	XORQ	AX, CX
	BSRQ	CX, CX	// index of highest bit difference
	SHRQ	CX, AX	// move a's bit to bottom
	ANDQ	$1, AX	// mask bit
	LEAQ	-1(AX*2), AX // 1/0 => +1/-1
	RET

	// 0-7 bytes in common
cmp_small:
	LEAQ	(R8*8), CX	// bytes left -> bits left
	NEGQ	CX		//  - bits lift (== 64 - bits left mod 64)
	JEQ	cmp_allsame

	// load bytes of a into high bytes of AX
	CMPB	SI, $0xf8
	JA	cmp_si_high
	MOVQ	(SI), SI
	JMP	cmp_si_finish
cmp_si_high:
	ADDQ	R8, SI
	MOVQ	-8(SI), SI
	SHRQ	CX, SI
cmp_si_finish:
	SHLQ	CX, SI

	// load bytes of b in to high bytes of BX
	CMPB	DI, $0xf8
	JA	cmp_di_high
	MOVQ	(DI), DI
	JMP	cmp_di_finish
cmp_di_high:
	ADDQ	R8, DI
	MOVQ	-8(DI), DI
	SHRQ	CX, DI
cmp_di_finish:
	SHLQ	CX, DI

	BSWAPQ	SI	// reverse order of bytes
	BSWAPQ	DI
	XORQ	SI, DI	// find bit differences
	JEQ	cmp_allsame
	BSRQ	DI, CX	// index of highest bit difference
	SHRQ	CX, SI	// move a's bit to bottom
	ANDQ	$1, SI	// mask bit
	LEAQ	-1(SI*2), AX // 1/0 => +1/-1
	RET

cmp_allsame:
	XORQ	AX, AX
	XORQ	CX, CX
	CMPQ	BX, DX
	SETGT	AX	// 1 if alen > blen
	SETEQ	CX	// 1 if alen == blen
	LEAQ	-1(CX)(AX*2), AX	// 1,0,-1 result
	RET

TEXT bytes·IndexByte(SB),NOSPLIT,$0
	MOVL s+0(FP), SI
	MOVL s_len+4(FP), BX
	MOVB c+12(FP), AL
	CALL runtime·indexbytebody(SB)
	MOVL AX, ret+16(FP)
	RET

TEXT strings·IndexByte(SB),NOSPLIT,$0
	MOVL s+0(FP), SI
	MOVL s_len+4(FP), BX
	MOVB c+8(FP), AL
	CALL runtime·indexbytebody(SB)
	MOVL AX, ret+16(FP)
	RET

// input:
//   SI: data
//   BX: data len
//   AL: byte sought
// output:
//   AX
TEXT runtime·indexbytebody(SB),NOSPLIT,$0
	MOVL SI, DI

	CMPL BX, $16
	JLT indexbyte_small

	// round up to first 16-byte boundary
	TESTL $15, SI
	JZ aligned
	MOVL SI, CX
	ANDL $~15, CX
	ADDL $16, CX

	// search the beginning
	SUBL SI, CX
	REPN; SCASB
	JZ success

// DI is 16-byte aligned; get ready to search using SSE instructions
aligned:
	// round down to last 16-byte boundary
	MOVL BX, R11
	ADDL SI, R11
	ANDL $~15, R11

	// shuffle X0 around so that each byte contains c
	MOVD AX, X0
	PUNPCKLBW X0, X0
	PUNPCKLBW X0, X0
	PSHUFL $0, X0, X0
	JMP condition

sse:
	// move the next 16-byte chunk of the buffer into X1
	MOVO (DI), X1
	// compare bytes in X0 to X1
	PCMPEQB X0, X1
	// take the top bit of each byte in X1 and put the result in DX
	PMOVMSKB X1, DX
	TESTL DX, DX
	JNZ ssesuccess
	ADDL $16, DI

condition:
	CMPL DI, R11
	JLT sse

	// search the end
	MOVL SI, CX
	ADDL BX, CX
	SUBL R11, CX
	// if CX == 0, the zero flag will be set and we'll end up
	// returning a false success
	JZ failure
	REPN; SCASB
	JZ success

failure:
	MOVL $-1, AX
	RET

// handle for lengths < 16
indexbyte_small:
	MOVL BX, CX
	REPN; SCASB
	JZ success
	MOVL $-1, AX
	RET

// we've found the chunk containing the byte
// now just figure out which specific byte it is
ssesuccess:
	// get the index of the least significant set bit
	BSFW DX, DX
	SUBL SI, DI
	ADDL DI, DX
	MOVL DX, AX
	RET

success:
	SUBL SI, DI
	SUBL $1, DI
	MOVL DI, AX
	RET

TEXT bytes·Equal(SB),NOSPLIT,$0-25
	MOVL	a_len+4(FP), BX
	MOVL	b_len+16(FP), CX
	XORL	AX, AX
	CMPL	BX, CX
	JNE	eqret
	MOVL	a+0(FP), SI
	MOVL	b+12(FP), DI
	CALL	runtime·memeqbody(SB)
eqret:
	MOVB	AX, ret+24(FP)
	RET

TEXT runtime·fastrand1(SB), NOSPLIT, $0-4
	get_tls(CX)
	MOVL	g(CX), AX
	MOVL	g_m(AX), AX
	MOVL	m_fastrand(AX), DX
	ADDL	DX, DX
	MOVL	DX, BX
	XORL	$0x88888eef, DX
	CMOVLMI	BX, DX
	MOVL	DX, m_fastrand(AX)
	MOVL	DX, ret+0(FP)
	RET

TEXT runtime·return0(SB), NOSPLIT, $0
	MOVL	$0, AX
	RET

// The top-most function running on a goroutine
// returns to goexit+PCQuantum.
TEXT runtime·goexit(SB),NOSPLIT,$0-0
	BYTE	$0x90	// NOP
	CALL	runtime·goexit1(SB)	// does not return

TEXT runtime·getg(SB),NOSPLIT,$0-4
	get_tls(CX)
	MOVL	g(CX), AX
	MOVL	AX, ret+0(FP)
	RET