-
Notifications
You must be signed in to change notification settings - Fork 17.6k
/
mheap.go
2047 lines (1841 loc) · 64.4 KB
/
mheap.go
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
// 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.
// Page heap.
//
// See malloc.go for overview.
package runtime
import (
"internal/cpu"
"runtime/internal/atomic"
"runtime/internal/sys"
"unsafe"
)
const (
// minPhysPageSize is a lower-bound on the physical page size. The
// true physical page size may be larger than this. In contrast,
// sys.PhysPageSize is an upper-bound on the physical page size.
minPhysPageSize = 4096
// maxPhysPageSize is the maximum page size the runtime supports.
maxPhysPageSize = 512 << 10
// maxPhysHugePageSize sets an upper-bound on the maximum huge page size
// that the runtime supports.
maxPhysHugePageSize = pallocChunkBytes
// pagesPerReclaimerChunk indicates how many pages to scan from the
// pageInUse bitmap at a time. Used by the page reclaimer.
//
// Higher values reduce contention on scanning indexes (such as
// h.reclaimIndex), but increase the minimum latency of the
// operation.
//
// The time required to scan this many pages can vary a lot depending
// on how many spans are actually freed. Experimentally, it can
// scan for pages at ~300 GB/ms on a 2.6GHz Core i7, but can only
// free spans at ~32 MB/ms. Using 512 pages bounds this at
// roughly 100µs.
//
// Must be a multiple of the pageInUse bitmap element size and
// must also evenly divide pagesPerArena.
pagesPerReclaimerChunk = 512
// physPageAlignedStacks indicates whether stack allocations must be
// physical page aligned. This is a requirement for MAP_STACK on
// OpenBSD.
physPageAlignedStacks = GOOS == "openbsd"
)
// Main malloc heap.
// The heap itself is the "free" and "scav" treaps,
// but all the other global data is here too.
//
// mheap must not be heap-allocated because it contains mSpanLists,
// which must not be heap-allocated.
//
//go:notinheap
type mheap struct {
// lock must only be acquired on the system stack, otherwise a g
// could self-deadlock if its stack grows with the lock held.
lock mutex
pages pageAlloc // page allocation data structure
sweepgen uint32 // sweep generation, see comment in mspan; written during STW
sweepdone uint32 // all spans are swept
sweepers uint32 // number of active sweepone calls
// allspans is a slice of all mspans ever created. Each mspan
// appears exactly once.
//
// The memory for allspans is manually managed and can be
// reallocated and move as the heap grows.
//
// In general, allspans is protected by mheap_.lock, which
// prevents concurrent access as well as freeing the backing
// store. Accesses during STW might not hold the lock, but
// must ensure that allocation cannot happen around the
// access (since that may free the backing store).
allspans []*mspan // all spans out there
_ uint32 // align uint64 fields on 32-bit for atomics
// Proportional sweep
//
// These parameters represent a linear function from heap_live
// to page sweep count. The proportional sweep system works to
// stay in the black by keeping the current page sweep count
// above this line at the current heap_live.
//
// The line has slope sweepPagesPerByte and passes through a
// basis point at (sweepHeapLiveBasis, pagesSweptBasis). At
// any given time, the system is at (memstats.heap_live,
// pagesSwept) in this space.
//
// It's important that the line pass through a point we
// control rather than simply starting at a (0,0) origin
// because that lets us adjust sweep pacing at any time while
// accounting for current progress. If we could only adjust
// the slope, it would create a discontinuity in debt if any
// progress has already been made.
pagesInUse uint64 // pages of spans in stats mSpanInUse; updated atomically
pagesSwept uint64 // pages swept this cycle; updated atomically
pagesSweptBasis uint64 // pagesSwept to use as the origin of the sweep ratio; updated atomically
sweepHeapLiveBasis uint64 // value of heap_live to use as the origin of sweep ratio; written with lock, read without
sweepPagesPerByte float64 // proportional sweep ratio; written with lock, read without
// TODO(austin): pagesInUse should be a uintptr, but the 386
// compiler can't 8-byte align fields.
// scavengeGoal is the amount of total retained heap memory (measured by
// heapRetained) that the runtime will try to maintain by returning memory
// to the OS.
scavengeGoal uint64
// Page reclaimer state
// reclaimIndex is the page index in allArenas of next page to
// reclaim. Specifically, it refers to page (i %
// pagesPerArena) of arena allArenas[i / pagesPerArena].
//
// If this is >= 1<<63, the page reclaimer is done scanning
// the page marks.
//
// This is accessed atomically.
reclaimIndex uint64
// reclaimCredit is spare credit for extra pages swept. Since
// the page reclaimer works in large chunks, it may reclaim
// more than requested. Any spare pages released go to this
// credit pool.
//
// This is accessed atomically.
reclaimCredit uintptr
// arenas is the heap arena map. It points to the metadata for
// the heap for every arena frame of the entire usable virtual
// address space.
//
// Use arenaIndex to compute indexes into this array.
//
// For regions of the address space that are not backed by the
// Go heap, the arena map contains nil.
//
// Modifications are protected by mheap_.lock. Reads can be
// performed without locking; however, a given entry can
// transition from nil to non-nil at any time when the lock
// isn't held. (Entries never transitions back to nil.)
//
// In general, this is a two-level mapping consisting of an L1
// map and possibly many L2 maps. This saves space when there
// are a huge number of arena frames. However, on many
// platforms (even 64-bit), arenaL1Bits is 0, making this
// effectively a single-level map. In this case, arenas[0]
// will never be nil.
arenas [1 << arenaL1Bits]*[1 << arenaL2Bits]*heapArena
// heapArenaAlloc is pre-reserved space for allocating heapArena
// objects. This is only used on 32-bit, where we pre-reserve
// this space to avoid interleaving it with the heap itself.
heapArenaAlloc linearAlloc
// arenaHints is a list of addresses at which to attempt to
// add more heap arenas. This is initially populated with a
// set of general hint addresses, and grown with the bounds of
// actual heap arena ranges.
arenaHints *arenaHint
// arena is a pre-reserved space for allocating heap arenas
// (the actual arenas). This is only used on 32-bit.
arena linearAlloc
// allArenas is the arenaIndex of every mapped arena. This can
// be used to iterate through the address space.
//
// Access is protected by mheap_.lock. However, since this is
// append-only and old backing arrays are never freed, it is
// safe to acquire mheap_.lock, copy the slice header, and
// then release mheap_.lock.
allArenas []arenaIdx
// sweepArenas is a snapshot of allArenas taken at the
// beginning of the sweep cycle. This can be read safely by
// simply blocking GC (by disabling preemption).
sweepArenas []arenaIdx
// markArenas is a snapshot of allArenas taken at the beginning
// of the mark cycle. Because allArenas is append-only, neither
// this slice nor its contents will change during the mark, so
// it can be read safely.
markArenas []arenaIdx
// curArena is the arena that the heap is currently growing
// into. This should always be physPageSize-aligned.
curArena struct {
base, end uintptr
}
_ uint32 // ensure 64-bit alignment of central
// central free lists for small size classes.
// the padding makes sure that the mcentrals are
// spaced CacheLinePadSize bytes apart, so that each mcentral.lock
// gets its own cache line.
// central is indexed by spanClass.
central [numSpanClasses]struct {
mcentral mcentral
pad [cpu.CacheLinePadSize - unsafe.Sizeof(mcentral{})%cpu.CacheLinePadSize]byte
}
spanalloc fixalloc // allocator for span*
cachealloc fixalloc // allocator for mcache*
specialfinalizeralloc fixalloc // allocator for specialfinalizer*
specialprofilealloc fixalloc // allocator for specialprofile*
speciallock mutex // lock for special record allocators.
arenaHintAlloc fixalloc // allocator for arenaHints
unused *specialfinalizer // never set, just here to force the specialfinalizer type into DWARF
}
var mheap_ mheap
// A heapArena stores metadata for a heap arena. heapArenas are stored
// outside of the Go heap and accessed via the mheap_.arenas index.
//
//go:notinheap
type heapArena struct {
// bitmap stores the pointer/scalar bitmap for the words in
// this arena. See mbitmap.go for a description. Use the
// heapBits type to access this.
bitmap [heapArenaBitmapBytes]byte
// spans maps from virtual address page ID within this arena to *mspan.
// For allocated spans, their pages map to the span itself.
// For free spans, only the lowest and highest pages map to the span itself.
// Internal pages map to an arbitrary span.
// For pages that have never been allocated, spans entries are nil.
//
// Modifications are protected by mheap.lock. Reads can be
// performed without locking, but ONLY from indexes that are
// known to contain in-use or stack spans. This means there
// must not be a safe-point between establishing that an
// address is live and looking it up in the spans array.
spans [pagesPerArena]*mspan
// pageInUse is a bitmap that indicates which spans are in
// state mSpanInUse. This bitmap is indexed by page number,
// but only the bit corresponding to the first page in each
// span is used.
//
// Reads and writes are atomic.
pageInUse [pagesPerArena / 8]uint8
// pageMarks is a bitmap that indicates which spans have any
// marked objects on them. Like pageInUse, only the bit
// corresponding to the first page in each span is used.
//
// Writes are done atomically during marking. Reads are
// non-atomic and lock-free since they only occur during
// sweeping (and hence never race with writes).
//
// This is used to quickly find whole spans that can be freed.
//
// TODO(austin): It would be nice if this was uint64 for
// faster scanning, but we don't have 64-bit atomic bit
// operations.
pageMarks [pagesPerArena / 8]uint8
// pageSpecials is a bitmap that indicates which spans have
// specials (finalizers or other). Like pageInUse, only the bit
// corresponding to the first page in each span is used.
//
// Writes are done atomically whenever a special is added to
// a span and whenever the last special is removed from a span.
// Reads are done atomically to find spans containing specials
// during marking.
pageSpecials [pagesPerArena / 8]uint8
// checkmarks stores the debug.gccheckmark state. It is only
// used if debug.gccheckmark > 0.
checkmarks *checkmarksMap
// zeroedBase marks the first byte of the first page in this
// arena which hasn't been used yet and is therefore already
// zero. zeroedBase is relative to the arena base.
// Increases monotonically until it hits heapArenaBytes.
//
// This field is sufficient to determine if an allocation
// needs to be zeroed because the page allocator follows an
// address-ordered first-fit policy.
//
// Read atomically and written with an atomic CAS.
zeroedBase uintptr
}
// arenaHint is a hint for where to grow the heap arenas. See
// mheap_.arenaHints.
//
//go:notinheap
type arenaHint struct {
addr uintptr
down bool
next *arenaHint
}
// An mspan is a run of pages.
//
// When a mspan is in the heap free treap, state == mSpanFree
// and heapmap(s->start) == span, heapmap(s->start+s->npages-1) == span.
// If the mspan is in the heap scav treap, then in addition to the
// above scavenged == true. scavenged == false in all other cases.
//
// When a mspan is allocated, state == mSpanInUse or mSpanManual
// and heapmap(i) == span for all s->start <= i < s->start+s->npages.
// Every mspan is in one doubly-linked list, either in the mheap's
// busy list or one of the mcentral's span lists.
// An mspan representing actual memory has state mSpanInUse,
// mSpanManual, or mSpanFree. Transitions between these states are
// constrained as follows:
//
// * A span may transition from free to in-use or manual during any GC
// phase.
//
// * During sweeping (gcphase == _GCoff), a span may transition from
// in-use to free (as a result of sweeping) or manual to free (as a
// result of stacks being freed).
//
// * During GC (gcphase != _GCoff), a span *must not* transition from
// manual or in-use to free. Because concurrent GC may read a pointer
// and then look up its span, the span state must be monotonic.
//
// Setting mspan.state to mSpanInUse or mSpanManual must be done
// atomically and only after all other span fields are valid.
// Likewise, if inspecting a span is contingent on it being
// mSpanInUse, the state should be loaded atomically and checked
// before depending on other fields. This allows the garbage collector
// to safely deal with potentially invalid pointers, since resolving
// such pointers may race with a span being allocated.
type mSpanState uint8
const (
mSpanDead mSpanState = iota
mSpanInUse // allocated for garbage collected heap
mSpanManual // allocated for manual management (e.g., stack allocator)
)
// mSpanStateNames are the names of the span states, indexed by
// mSpanState.
var mSpanStateNames = []string{
"mSpanDead",
"mSpanInUse",
"mSpanManual",
"mSpanFree",
}
// mSpanStateBox holds an mSpanState and provides atomic operations on
// it. This is a separate type to disallow accidental comparison or
// assignment with mSpanState.
type mSpanStateBox struct {
s mSpanState
}
func (b *mSpanStateBox) set(s mSpanState) {
atomic.Store8((*uint8)(&b.s), uint8(s))
}
func (b *mSpanStateBox) get() mSpanState {
return mSpanState(atomic.Load8((*uint8)(&b.s)))
}
// mSpanList heads a linked list of spans.
//
//go:notinheap
type mSpanList struct {
first *mspan // first span in list, or nil if none
last *mspan // last span in list, or nil if none
}
//go:notinheap
type mspan struct {
next *mspan // next span in list, or nil if none
prev *mspan // previous span in list, or nil if none
list *mSpanList // For debugging. TODO: Remove.
startAddr uintptr // address of first byte of span aka s.base()
npages uintptr // number of pages in span
manualFreeList gclinkptr // list of free objects in mSpanManual spans
// freeindex is the slot index between 0 and nelems at which to begin scanning
// for the next free object in this span.
// Each allocation scans allocBits starting at freeindex until it encounters a 0
// indicating a free object. freeindex is then adjusted so that subsequent scans begin
// just past the newly discovered free object.
//
// If freeindex == nelem, this span has no free objects.
//
// allocBits is a bitmap of objects in this span.
// If n >= freeindex and allocBits[n/8] & (1<<(n%8)) is 0
// then object n is free;
// otherwise, object n is allocated. Bits starting at nelem are
// undefined and should never be referenced.
//
// Object n starts at address n*elemsize + (start << pageShift).
freeindex uintptr
// TODO: Look up nelems from sizeclass and remove this field if it
// helps performance.
nelems uintptr // number of object in the span.
// Cache of the allocBits at freeindex. allocCache is shifted
// such that the lowest bit corresponds to the bit freeindex.
// allocCache holds the complement of allocBits, thus allowing
// ctz (count trailing zero) to use it directly.
// allocCache may contain bits beyond s.nelems; the caller must ignore
// these.
allocCache uint64
// allocBits and gcmarkBits hold pointers to a span's mark and
// allocation bits. The pointers are 8 byte aligned.
// There are three arenas where this data is held.
// free: Dirty arenas that are no longer accessed
// and can be reused.
// next: Holds information to be used in the next GC cycle.
// current: Information being used during this GC cycle.
// previous: Information being used during the last GC cycle.
// A new GC cycle starts with the call to finishsweep_m.
// finishsweep_m moves the previous arena to the free arena,
// the current arena to the previous arena, and
// the next arena to the current arena.
// The next arena is populated as the spans request
// memory to hold gcmarkBits for the next GC cycle as well
// as allocBits for newly allocated spans.
//
// The pointer arithmetic is done "by hand" instead of using
// arrays to avoid bounds checks along critical performance
// paths.
// The sweep will free the old allocBits and set allocBits to the
// gcmarkBits. The gcmarkBits are replaced with a fresh zeroed
// out memory.
allocBits *gcBits
gcmarkBits *gcBits
// sweep generation:
// if sweepgen == h->sweepgen - 2, the span needs sweeping
// if sweepgen == h->sweepgen - 1, the span is currently being swept
// if sweepgen == h->sweepgen, the span is swept and ready to use
// if sweepgen == h->sweepgen + 1, the span was cached before sweep began and is still cached, and needs sweeping
// if sweepgen == h->sweepgen + 3, the span was swept and then cached and is still cached
// h->sweepgen is incremented by 2 after every GC
sweepgen uint32
divMul uint16 // for divide by elemsize - divMagic.mul
baseMask uint16 // if non-0, elemsize is a power of 2, & this will get object allocation base
allocCount uint16 // number of allocated objects
spanclass spanClass // size class and noscan (uint8)
state mSpanStateBox // mSpanInUse etc; accessed atomically (get/set methods)
needzero uint8 // needs to be zeroed before allocation
divShift uint8 // for divide by elemsize - divMagic.shift
divShift2 uint8 // for divide by elemsize - divMagic.shift2
elemsize uintptr // computed from sizeclass or from npages
limit uintptr // end of data in span
speciallock mutex // guards specials list
specials *special // linked list of special records sorted by offset.
}
func (s *mspan) base() uintptr {
return s.startAddr
}
func (s *mspan) layout() (size, n, total uintptr) {
total = s.npages << _PageShift
size = s.elemsize
if size > 0 {
n = total / size
}
return
}
// recordspan adds a newly allocated span to h.allspans.
//
// This only happens the first time a span is allocated from
// mheap.spanalloc (it is not called when a span is reused).
//
// Write barriers are disallowed here because it can be called from
// gcWork when allocating new workbufs. However, because it's an
// indirect call from the fixalloc initializer, the compiler can't see
// this.
//
// The heap lock must be held.
//
//go:nowritebarrierrec
func recordspan(vh unsafe.Pointer, p unsafe.Pointer) {
h := (*mheap)(vh)
s := (*mspan)(p)
assertLockHeld(&h.lock)
if len(h.allspans) >= cap(h.allspans) {
n := 64 * 1024 / sys.PtrSize
if n < cap(h.allspans)*3/2 {
n = cap(h.allspans) * 3 / 2
}
var new []*mspan
sp := (*slice)(unsafe.Pointer(&new))
sp.array = sysAlloc(uintptr(n)*sys.PtrSize, &memstats.other_sys)
if sp.array == nil {
throw("runtime: cannot allocate memory")
}
sp.len = len(h.allspans)
sp.cap = n
if len(h.allspans) > 0 {
copy(new, h.allspans)
}
oldAllspans := h.allspans
*(*notInHeapSlice)(unsafe.Pointer(&h.allspans)) = *(*notInHeapSlice)(unsafe.Pointer(&new))
if len(oldAllspans) != 0 {
sysFree(unsafe.Pointer(&oldAllspans[0]), uintptr(cap(oldAllspans))*unsafe.Sizeof(oldAllspans[0]), &memstats.other_sys)
}
}
h.allspans = h.allspans[:len(h.allspans)+1]
h.allspans[len(h.allspans)-1] = s
}
// A spanClass represents the size class and noscan-ness of a span.
//
// Each size class has a noscan spanClass and a scan spanClass. The
// noscan spanClass contains only noscan objects, which do not contain
// pointers and thus do not need to be scanned by the garbage
// collector.
type spanClass uint8
const (
numSpanClasses = _NumSizeClasses << 1
tinySpanClass = spanClass(tinySizeClass<<1 | 1)
)
func makeSpanClass(sizeclass uint8, noscan bool) spanClass {
return spanClass(sizeclass<<1) | spanClass(bool2int(noscan))
}
func (sc spanClass) sizeclass() int8 {
return int8(sc >> 1)
}
func (sc spanClass) noscan() bool {
return sc&1 != 0
}
// arenaIndex returns the index into mheap_.arenas of the arena
// containing metadata for p. This index combines of an index into the
// L1 map and an index into the L2 map and should be used as
// mheap_.arenas[ai.l1()][ai.l2()].
//
// If p is outside the range of valid heap addresses, either l1() or
// l2() will be out of bounds.
//
// It is nosplit because it's called by spanOf and several other
// nosplit functions.
//
//go:nosplit
func arenaIndex(p uintptr) arenaIdx {
return arenaIdx((p - arenaBaseOffset) / heapArenaBytes)
}
// arenaBase returns the low address of the region covered by heap
// arena i.
func arenaBase(i arenaIdx) uintptr {
return uintptr(i)*heapArenaBytes + arenaBaseOffset
}
type arenaIdx uint
func (i arenaIdx) l1() uint {
if arenaL1Bits == 0 {
// Let the compiler optimize this away if there's no
// L1 map.
return 0
} else {
return uint(i) >> arenaL1Shift
}
}
func (i arenaIdx) l2() uint {
if arenaL1Bits == 0 {
return uint(i)
} else {
return uint(i) & (1<<arenaL2Bits - 1)
}
}
// inheap reports whether b is a pointer into a (potentially dead) heap object.
// It returns false for pointers into mSpanManual spans.
// Non-preemptible because it is used by write barriers.
//go:nowritebarrier
//go:nosplit
func inheap(b uintptr) bool {
return spanOfHeap(b) != nil
}
// inHeapOrStack is a variant of inheap that returns true for pointers
// into any allocated heap span.
//
//go:nowritebarrier
//go:nosplit
func inHeapOrStack(b uintptr) bool {
s := spanOf(b)
if s == nil || b < s.base() {
return false
}
switch s.state.get() {
case mSpanInUse, mSpanManual:
return b < s.limit
default:
return false
}
}
// spanOf returns the span of p. If p does not point into the heap
// arena or no span has ever contained p, spanOf returns nil.
//
// If p does not point to allocated memory, this may return a non-nil
// span that does *not* contain p. If this is a possibility, the
// caller should either call spanOfHeap or check the span bounds
// explicitly.
//
// Must be nosplit because it has callers that are nosplit.
//
//go:nosplit
func spanOf(p uintptr) *mspan {
// This function looks big, but we use a lot of constant
// folding around arenaL1Bits to get it under the inlining
// budget. Also, many of the checks here are safety checks
// that Go needs to do anyway, so the generated code is quite
// short.
ri := arenaIndex(p)
if arenaL1Bits == 0 {
// If there's no L1, then ri.l1() can't be out of bounds but ri.l2() can.
if ri.l2() >= uint(len(mheap_.arenas[0])) {
return nil
}
} else {
// If there's an L1, then ri.l1() can be out of bounds but ri.l2() can't.
if ri.l1() >= uint(len(mheap_.arenas)) {
return nil
}
}
l2 := mheap_.arenas[ri.l1()]
if arenaL1Bits != 0 && l2 == nil { // Should never happen if there's no L1.
return nil
}
ha := l2[ri.l2()]
if ha == nil {
return nil
}
return ha.spans[(p/pageSize)%pagesPerArena]
}
// spanOfUnchecked is equivalent to spanOf, but the caller must ensure
// that p points into an allocated heap arena.
//
// Must be nosplit because it has callers that are nosplit.
//
//go:nosplit
func spanOfUnchecked(p uintptr) *mspan {
ai := arenaIndex(p)
return mheap_.arenas[ai.l1()][ai.l2()].spans[(p/pageSize)%pagesPerArena]
}
// spanOfHeap is like spanOf, but returns nil if p does not point to a
// heap object.
//
// Must be nosplit because it has callers that are nosplit.
//
//go:nosplit
func spanOfHeap(p uintptr) *mspan {
s := spanOf(p)
// s is nil if it's never been allocated. Otherwise, we check
// its state first because we don't trust this pointer, so we
// have to synchronize with span initialization. Then, it's
// still possible we picked up a stale span pointer, so we
// have to check the span's bounds.
if s == nil || s.state.get() != mSpanInUse || p < s.base() || p >= s.limit {
return nil
}
return s
}
// pageIndexOf returns the arena, page index, and page mask for pointer p.
// The caller must ensure p is in the heap.
func pageIndexOf(p uintptr) (arena *heapArena, pageIdx uintptr, pageMask uint8) {
ai := arenaIndex(p)
arena = mheap_.arenas[ai.l1()][ai.l2()]
pageIdx = ((p / pageSize) / 8) % uintptr(len(arena.pageInUse))
pageMask = byte(1 << ((p / pageSize) % 8))
return
}
// Initialize the heap.
func (h *mheap) init() {
lockInit(&h.lock, lockRankMheap)
lockInit(&h.speciallock, lockRankMheapSpecial)
h.spanalloc.init(unsafe.Sizeof(mspan{}), recordspan, unsafe.Pointer(h), &memstats.mspan_sys)
h.cachealloc.init(unsafe.Sizeof(mcache{}), nil, nil, &memstats.mcache_sys)
h.specialfinalizeralloc.init(unsafe.Sizeof(specialfinalizer{}), nil, nil, &memstats.other_sys)
h.specialprofilealloc.init(unsafe.Sizeof(specialprofile{}), nil, nil, &memstats.other_sys)
h.arenaHintAlloc.init(unsafe.Sizeof(arenaHint{}), nil, nil, &memstats.other_sys)
// Don't zero mspan allocations. Background sweeping can
// inspect a span concurrently with allocating it, so it's
// important that the span's sweepgen survive across freeing
// and re-allocating a span to prevent background sweeping
// from improperly cas'ing it from 0.
//
// This is safe because mspan contains no heap pointers.
h.spanalloc.zero = false
// h->mapcache needs no init
for i := range h.central {
h.central[i].mcentral.init(spanClass(i))
}
h.pages.init(&h.lock, &memstats.gcMiscSys)
}
// reclaim sweeps and reclaims at least npage pages into the heap.
// It is called before allocating npage pages to keep growth in check.
//
// reclaim implements the page-reclaimer half of the sweeper.
//
// h.lock must NOT be held.
func (h *mheap) reclaim(npage uintptr) {
// TODO(austin): Half of the time spent freeing spans is in
// locking/unlocking the heap (even with low contention). We
// could make the slow path here several times faster by
// batching heap frees.
// Bail early if there's no more reclaim work.
if atomic.Load64(&h.reclaimIndex) >= 1<<63 {
return
}
// Disable preemption so the GC can't start while we're
// sweeping, so we can read h.sweepArenas, and so
// traceGCSweepStart/Done pair on the P.
mp := acquirem()
if trace.enabled {
traceGCSweepStart()
}
arenas := h.sweepArenas
locked := false
for npage > 0 {
// Pull from accumulated credit first.
if credit := atomic.Loaduintptr(&h.reclaimCredit); credit > 0 {
take := credit
if take > npage {
// Take only what we need.
take = npage
}
if atomic.Casuintptr(&h.reclaimCredit, credit, credit-take) {
npage -= take
}
continue
}
// Claim a chunk of work.
idx := uintptr(atomic.Xadd64(&h.reclaimIndex, pagesPerReclaimerChunk) - pagesPerReclaimerChunk)
if idx/pagesPerArena >= uintptr(len(arenas)) {
// Page reclaiming is done.
atomic.Store64(&h.reclaimIndex, 1<<63)
break
}
if !locked {
// Lock the heap for reclaimChunk.
lock(&h.lock)
locked = true
}
// Scan this chunk.
nfound := h.reclaimChunk(arenas, idx, pagesPerReclaimerChunk)
if nfound <= npage {
npage -= nfound
} else {
// Put spare pages toward global credit.
atomic.Xadduintptr(&h.reclaimCredit, nfound-npage)
npage = 0
}
}
if locked {
unlock(&h.lock)
}
if trace.enabled {
traceGCSweepDone()
}
releasem(mp)
}
// reclaimChunk sweeps unmarked spans that start at page indexes [pageIdx, pageIdx+n).
// It returns the number of pages returned to the heap.
//
// h.lock must be held and the caller must be non-preemptible. Note: h.lock may be
// temporarily unlocked and re-locked in order to do sweeping or if tracing is
// enabled.
func (h *mheap) reclaimChunk(arenas []arenaIdx, pageIdx, n uintptr) uintptr {
// The heap lock must be held because this accesses the
// heapArena.spans arrays using potentially non-live pointers.
// In particular, if a span were freed and merged concurrently
// with this probing heapArena.spans, it would be possible to
// observe arbitrary, stale span pointers.
assertLockHeld(&h.lock)
n0 := n
var nFreed uintptr
sg := h.sweepgen
for n > 0 {
ai := arenas[pageIdx/pagesPerArena]
ha := h.arenas[ai.l1()][ai.l2()]
// Get a chunk of the bitmap to work on.
arenaPage := uint(pageIdx % pagesPerArena)
inUse := ha.pageInUse[arenaPage/8:]
marked := ha.pageMarks[arenaPage/8:]
if uintptr(len(inUse)) > n/8 {
inUse = inUse[:n/8]
marked = marked[:n/8]
}
// Scan this bitmap chunk for spans that are in-use
// but have no marked objects on them.
for i := range inUse {
inUseUnmarked := atomic.Load8(&inUse[i]) &^ marked[i]
if inUseUnmarked == 0 {
continue
}
for j := uint(0); j < 8; j++ {
if inUseUnmarked&(1<<j) != 0 {
s := ha.spans[arenaPage+uint(i)*8+j]
if atomic.Load(&s.sweepgen) == sg-2 && atomic.Cas(&s.sweepgen, sg-2, sg-1) {
npages := s.npages
unlock(&h.lock)
if s.sweep(false) {
nFreed += npages
}
lock(&h.lock)
// Reload inUse. It's possible nearby
// spans were freed when we dropped the
// lock and we don't want to get stale
// pointers from the spans array.
inUseUnmarked = atomic.Load8(&inUse[i]) &^ marked[i]
}
}
}
}
// Advance.
pageIdx += uintptr(len(inUse) * 8)
n -= uintptr(len(inUse) * 8)
}
if trace.enabled {
unlock(&h.lock)
// Account for pages scanned but not reclaimed.
traceGCSweepSpan((n0 - nFreed) * pageSize)
lock(&h.lock)
}
assertLockHeld(&h.lock) // Must be locked on return.
return nFreed
}
// spanAllocType represents the type of allocation to make, or
// the type of allocation to be freed.
type spanAllocType uint8
const (
spanAllocHeap spanAllocType = iota // heap span
spanAllocStack // stack span
spanAllocPtrScalarBits // unrolled GC prog bitmap span
spanAllocWorkBuf // work buf span
)
// manual returns true if the span allocation is manually managed.
func (s spanAllocType) manual() bool {
return s != spanAllocHeap
}
// alloc allocates a new span of npage pages from the GC'd heap.
//
// spanclass indicates the span's size class and scannability.
//
// If needzero is true, the memory for the returned span will be zeroed.
func (h *mheap) alloc(npages uintptr, spanclass spanClass, needzero bool) *mspan {
// Don't do any operations that lock the heap on the G stack.
// It might trigger stack growth, and the stack growth code needs
// to be able to allocate heap.
var s *mspan
systemstack(func() {
// To prevent excessive heap growth, before allocating n pages
// we need to sweep and reclaim at least n pages.
if h.sweepdone == 0 {
h.reclaim(npages)
}
s = h.allocSpan(npages, spanAllocHeap, spanclass)
})
if s != nil {
if needzero && s.needzero != 0 {
memclrNoHeapPointers(unsafe.Pointer(s.base()), s.npages<<_PageShift)
}
s.needzero = 0
}
return s
}
// allocManual allocates a manually-managed span of npage pages.
// allocManual returns nil if allocation fails.
//
// allocManual adds the bytes used to *stat, which should be a
// memstats in-use field. Unlike allocations in the GC'd heap, the
// allocation does *not* count toward heap_inuse or heap_sys.
//
// The memory backing the returned span may not be zeroed if
// span.needzero is set.
//
// allocManual must be called on the system stack because it may
// acquire the heap lock via allocSpan. See mheap for details.
//
// If new code is written to call allocManual, do NOT use an
// existing spanAllocType value and instead declare a new one.
//
//go:systemstack
func (h *mheap) allocManual(npages uintptr, typ spanAllocType) *mspan {
if !typ.manual() {
throw("manual span allocation called with non-manually-managed type")
}
return h.allocSpan(npages, typ, 0)
}
// setSpans modifies the span map so [spanOf(base), spanOf(base+npage*pageSize))
// is s.
func (h *mheap) setSpans(base, npage uintptr, s *mspan) {
p := base / pageSize
ai := arenaIndex(base)
ha := h.arenas[ai.l1()][ai.l2()]
for n := uintptr(0); n < npage; n++ {
i := (p + n) % pagesPerArena
if i == 0 {
ai = arenaIndex(base + n*pageSize)
ha = h.arenas[ai.l1()][ai.l2()]
}
ha.spans[i] = s
}
}
// allocNeedsZero checks if the region of address space [base, base+npage*pageSize),
// assumed to be allocated, needs to be zeroed, updating heap arena metadata for
// future allocations.
//
// This must be called each time pages are allocated from the heap, even if the page
// allocator can otherwise prove the memory it's allocating is already zero because
// they're fresh from the operating system. It updates heapArena metadata that is
// critical for future page allocations.
//
// There are no locking constraints on this method.
func (h *mheap) allocNeedsZero(base, npage uintptr) (needZero bool) {
for npage > 0 {
ai := arenaIndex(base)
ha := h.arenas[ai.l1()][ai.l2()]
zeroedBase := atomic.Loaduintptr(&ha.zeroedBase)
arenaBase := base % heapArenaBytes
if arenaBase < zeroedBase {
// We extended into the non-zeroed part of the
// arena, so this region needs to be zeroed before use.
//
// zeroedBase is monotonically increasing, so if we see this now then
// we can be sure we need to zero this memory region.
//
// We still need to update zeroedBase for this arena, and
// potentially more arenas.
needZero = true
}
// We may observe arenaBase > zeroedBase if we're racing with one or more
// allocations which are acquiring memory directly before us in the address
// space. But, because we know no one else is acquiring *this* memory, it's
// still safe to not zero.
// Compute how far into the arena we extend into, capped
// at heapArenaBytes.
arenaLimit := arenaBase + npage*pageSize
if arenaLimit > heapArenaBytes {
arenaLimit = heapArenaBytes
}