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
|
/*
* Copyright (c) 2006-2007 Silicon Graphics, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
#include "xfs_bmap_btree.h"
#include "xfs_inum.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_ag.h"
#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_mount.h"
#include "xfs_bmap.h"
#include "xfs_alloc.h"
#include "xfs_utils.h"
#include "xfs_mru_cache.h"
#include "xfs_filestream.h"
#include "xfs_trace.h"
#ifdef XFS_FILESTREAMS_TRACE
ktrace_t *xfs_filestreams_trace_buf;
STATIC void
xfs_filestreams_trace(
xfs_mount_t *mp, /* mount point */
int type, /* type of trace */
const char *func, /* source function */
int line, /* source line number */
__psunsigned_t arg0,
__psunsigned_t arg1,
__psunsigned_t arg2,
__psunsigned_t arg3,
__psunsigned_t arg4,
__psunsigned_t arg5)
{
ktrace_enter(xfs_filestreams_trace_buf,
(void *)(__psint_t)(type | (line << 16)),
(void *)func,
(void *)(__psunsigned_t)current_pid(),
(void *)mp,
(void *)(__psunsigned_t)arg0,
(void *)(__psunsigned_t)arg1,
(void *)(__psunsigned_t)arg2,
(void *)(__psunsigned_t)arg3,
(void *)(__psunsigned_t)arg4,
(void *)(__psunsigned_t)arg5,
NULL, NULL, NULL, NULL, NULL, NULL);
}
#define TRACE0(mp,t) TRACE6(mp,t,0,0,0,0,0,0)
#define TRACE1(mp,t,a0) TRACE6(mp,t,a0,0,0,0,0,0)
#define TRACE2(mp,t,a0,a1) TRACE6(mp,t,a0,a1,0,0,0,0)
#define TRACE3(mp,t,a0,a1,a2) TRACE6(mp,t,a0,a1,a2,0,0,0)
#define TRACE4(mp,t,a0,a1,a2,a3) TRACE6(mp,t,a0,a1,a2,a3,0,0)
#define TRACE5(mp,t,a0,a1,a2,a3,a4) TRACE6(mp,t,a0,a1,a2,a3,a4,0)
#define TRACE6(mp,t,a0,a1,a2,a3,a4,a5) \
xfs_filestreams_trace(mp, t, __func__, __LINE__, \
(__psunsigned_t)a0, (__psunsigned_t)a1, \
(__psunsigned_t)a2, (__psunsigned_t)a3, \
(__psunsigned_t)a4, (__psunsigned_t)a5)
#define TRACE_AG_SCAN(mp, ag, ag2) \
TRACE2(mp, XFS_FSTRM_KTRACE_AGSCAN, ag, ag2);
#define TRACE_AG_PICK1(mp, max_ag, maxfree) \
TRACE2(mp, XFS_FSTRM_KTRACE_AGPICK1, max_ag, maxfree);
#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag) \
TRACE6(mp, XFS_FSTRM_KTRACE_AGPICK2, ag, ag2, \
cnt, free, scan, flag)
#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2) \
TRACE5(mp, XFS_FSTRM_KTRACE_UPDATE, ip, ag, cnt, ag2, cnt2)
#define TRACE_FREE(mp, ip, pip, ag, cnt) \
TRACE4(mp, XFS_FSTRM_KTRACE_FREE, ip, pip, ag, cnt)
#define TRACE_LOOKUP(mp, ip, pip, ag, cnt) \
TRACE4(mp, XFS_FSTRM_KTRACE_ITEM_LOOKUP, ip, pip, ag, cnt)
#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt) \
TRACE4(mp, XFS_FSTRM_KTRACE_ASSOCIATE, ip, pip, ag, cnt)
#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt) \
TRACE6(mp, XFS_FSTRM_KTRACE_MOVEAG, ip, pip, oag, ocnt, nag, ncnt)
#define TRACE_ORPHAN(mp, ip, ag) \
TRACE2(mp, XFS_FSTRM_KTRACE_ORPHAN, ip, ag);
#else
#define TRACE_AG_SCAN(mp, ag, ag2)
#define TRACE_AG_PICK1(mp, max_ag, maxfree)
#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag)
#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2)
#define TRACE_FREE(mp, ip, pip, ag, cnt)
#define TRACE_LOOKUP(mp, ip, pip, ag, cnt)
#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt)
#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt)
#define TRACE_ORPHAN(mp, ip, ag)
#endif
static kmem_zone_t *item_zone;
/*
* Structure for associating a file or a directory with an allocation group.
* The parent directory pointer is only needed for files, but since there will
* generally be vastly more files than directories in the cache, using the same
* data structure simplifies the code with very little memory overhead.
*/
typedef struct fstrm_item
{
xfs_agnumber_t ag; /* AG currently in use for the file/directory. */
xfs_inode_t *ip; /* inode self-pointer. */
xfs_inode_t *pip; /* Parent directory inode pointer. */
} fstrm_item_t;
/*
* Allocation group filestream associations are tracked with per-ag atomic
* counters. These counters allow _xfs_filestream_pick_ag() to tell whether a
* particular AG already has active filestreams associated with it. The mount
* point's m_peraglock is used to protect these counters from per-ag array
* re-allocation during a growfs operation. When xfs_growfs_data_private() is
* about to reallocate the array, it calls xfs_filestream_flush() with the
* m_peraglock held in write mode.
*
* Since xfs_mru_cache_flush() guarantees that all the free functions for all
* the cache elements have finished executing before it returns, it's safe for
* the free functions to use the atomic counters without m_peraglock protection.
* This allows the implementation of xfs_fstrm_free_func() to be agnostic about
* whether it was called with the m_peraglock held in read mode, write mode or
* not held at all. The race condition this addresses is the following:
*
* - The work queue scheduler fires and pulls a filestream directory cache
* element off the LRU end of the cache for deletion, then gets pre-empted.
* - A growfs operation grabs the m_peraglock in write mode, flushes all the
* remaining items from the cache and reallocates the mount point's per-ag
* array, resetting all the counters to zero.
* - The work queue thread resumes and calls the free function for the element
* it started cleaning up earlier. In the process it decrements the
* filestreams counter for an AG that now has no references.
*
* With a shrinkfs feature, the above scenario could panic the system.
*
* All other uses of the following macros should be protected by either the
* m_peraglock held in read mode, or the cache's internal locking exposed by the
* interval between a call to xfs_mru_cache_lookup() and a call to
* xfs_mru_cache_done(). In addition, the m_peraglock must be held in read mode
* when new elements are added to the cache.
*
* Combined, these locking rules ensure that no associations will ever exist in
* the cache that reference per-ag array elements that have since been
* reallocated.
*/
static int
xfs_filestream_peek_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
struct xfs_perag *pag;
int ret;
pag = xfs_perag_get(mp, agno);
ret = atomic_read(&pag->pagf_fstrms);
xfs_perag_put(pag);
return ret;
}
static int
xfs_filestream_get_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
struct xfs_perag *pag;
int ret;
pag = xfs_perag_get(mp, agno);
ret = atomic_inc_return(&pag->pagf_fstrms);
xfs_perag_put(pag);
return ret;
}
static void
xfs_filestream_put_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
struct xfs_perag *pag;
pag = xfs_perag_get(mp, agno);
atomic_dec(&pag->pagf_fstrms);
xfs_perag_put(pag);
}
/*
* Scan the AGs starting at startag looking for an AG that isn't in use and has
* at least minlen blocks free.
*/
static int
_xfs_filestream_pick_ag(
xfs_mount_t *mp,
xfs_agnumber_t startag,
xfs_agnumber_t *agp,
int flags,
xfs_extlen_t minlen)
{
int streams, max_streams;
int err, trylock, nscan;
xfs_extlen_t longest, free, minfree, maxfree = 0;
xfs_agnumber_t ag, max_ag = NULLAGNUMBER;
struct xfs_perag *pag;
/* 2% of an AG's blocks must be free for it to be chosen. */
minfree = mp->m_sb.sb_agblocks / 50;
ag = startag;
*agp = NULLAGNUMBER;
/* For the first pass, don't sleep trying to init the per-AG. */
trylock = XFS_ALLOC_FLAG_TRYLOCK;
for (nscan = 0; 1; nscan++) {
pag = xfs_perag_get(mp, ag);
TRACE_AG_SCAN(mp, ag, atomic_read(&pag->pagf_fstrms));
if (!pag->pagf_init) {
err = xfs_alloc_pagf_init(mp, NULL, ag, trylock);
if (err && !trylock) {
xfs_perag_put(pag);
return err;
}
}
/* Might fail sometimes during the 1st pass with trylock set. */
if (!pag->pagf_init)
goto next_ag;
/* Keep track of the AG with the most free blocks. */
if (pag->pagf_freeblks > maxfree) {
maxfree = pag->pagf_freeblks;
max_streams = atomic_read(&pag->pagf_fstrms);
max_ag = ag;
}
/*
* The AG reference count does two t
|