-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfile.c
1124 lines (947 loc) · 28.7 KB
/
file.c
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
/*
* BRIEF DESCRIPTION
*
* File operations for files.
*
* Copyright 2020-2023 Jiansheng Qiu <[email protected]>
* Copyright 2015-2016 Regents of the University of California,
* UCSD Non-Volatile Systems Lab, Andiry Xu <[email protected]>
* Copyright 2012-2013 Intel Corporation
* Copyright 2009-2011 Marco Stornelli <[email protected]>
* Copyright 2003 Sony Corporation
* Copyright 2003 Matsushita Electric Industrial Co., Ltd.
* 2003-2004 (c) MontaVista Software, Inc. , Steve Longerbeam
* This file is licensed under the terms of the GNU General Public
* License version 2. This program is licensed "as is" without any
* warranty of any kind, whether express or implied.
*/
#include <linux/slab.h>
#include <linux/uio.h>
#include <linux/uaccess.h>
#include <linux/falloc.h>
#include <linux/fadvise.h>
#include <asm/mman.h>
#include "nova.h"
#include "inode.h"
DEFINE_PER_CPU(uint8_t, stream_trust_degree_per_cpu);
DEFINE_PER_CPU(struct nova_rht_entry *, last_accessed_fpentry_per_cpu);
// DEFINE_PER_CPU(struct nova_pmm_entry *, last_new_fpentry_per_cpu);
static inline int nova_can_set_blocksize_hint(struct inode *inode,
struct nova_inode *pi, loff_t new_size)
{
struct nova_inode_info *si = NOVA_I(inode);
struct nova_inode_info_header *sih = &si->header;
/* Currently, we don't deallocate data blocks till the file is deleted.
* So no changing blocksize hints once allocation is done.
*/
if (sih->i_size > 0)
return 0;
return 1;
}
int nova_set_blocksize_hint(struct super_block *sb, struct inode *inode,
struct nova_inode *pi, loff_t new_size)
{
unsigned short block_type;
unsigned long irq_flags = 0;
if (!nova_can_set_blocksize_hint(inode, pi, new_size))
return 0;
if (new_size >= 0x40000000) { /* 1G */
block_type = NOVA_BLOCK_TYPE_1G;
goto hint_set;
}
if (new_size >= 0x200000) { /* 2M */
block_type = NOVA_BLOCK_TYPE_2M;
goto hint_set;
}
/* defaulting to 4K */
block_type = NOVA_BLOCK_TYPE_4K;
hint_set:
nova_dbg_verbose(
"Hint: new_size 0x%llx, i_size 0x%llx\n",
new_size, pi->i_size);
nova_dbg_verbose("Setting the hint to 0x%x\n", block_type);
nova_memunlock_inode(sb, pi, &irq_flags);
pi->i_blk_type = block_type;
nova_memlock_inode(sb, pi, &irq_flags);
return 0;
}
static loff_t nova_llseek(struct file *file, loff_t offset, int origin)
{
struct inode *inode = file->f_path.dentry->d_inode;
int retval;
if (origin != SEEK_DATA && origin != SEEK_HOLE)
return generic_file_llseek(file, offset, origin);
inode_lock(inode);
switch (origin) {
case SEEK_DATA:
retval = nova_find_region(inode, &offset, 0);
if (retval) {
inode_unlock(inode);
return retval;
}
break;
case SEEK_HOLE:
retval = nova_find_region(inode, &offset, 1);
if (retval) {
inode_unlock(inode);
return retval;
}
break;
}
if ((offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) ||
offset > inode->i_sb->s_maxbytes) {
inode_unlock(inode);
return -ENXIO;
}
if (offset != file->f_pos) {
file->f_pos = offset;
file->f_version = 0;
}
inode_unlock(inode);
return offset;
}
/* This function is called by both msync() and fsync().
* TODO: Check if we can avoid calling nova_flush_buffer() for fsync. We use
* movnti to write data to files, so we may want to avoid doing unnecessary
* nova_flush_buffer() on fsync()
*/
static int nova_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct address_space *mapping = file->f_mapping;
struct inode *inode = file->f_path.dentry->d_inode;
struct super_block *sb = inode->i_sb;
unsigned long start_pgoff, end_pgoff;
int ret = 0;
INIT_TIMING(fsync_time);
NOVA_START_TIMING(fsync_t, fsync_time);
if (datasync)
NOVA_STATS_ADD(fdatasync, 1);
/* No need to flush if the file is not mmaped */
if (!mapping_mapped(mapping))
goto persist;
start_pgoff = start >> PAGE_SHIFT;
end_pgoff = (end + 1) >> PAGE_SHIFT;
nova_dbgv("%s: msync pgoff range %lu to %lu\n",
__func__, start_pgoff, end_pgoff);
/*
* Set csum and parity.
* We do not protect data integrity during mmap, but we have to
* update csum here since msync clears dirty bit.
*/
nova_reset_mapping_csum_parity(sb, inode, mapping,
start_pgoff, end_pgoff);
ret = generic_file_fsync(file, start, end, datasync);
persist:
PERSISTENT_BARRIER();
NOVA_END_TIMING(fsync_t, fsync_time);
return ret;
}
/* This callback is called when a file is closed */
static int nova_flush(struct file *file, fl_owner_t id)
{
PERSISTENT_BARRIER();
return 0;
}
static int nova_open(struct inode *inode, struct file *filp)
{
return generic_file_open(inode, filp);
}
static long nova_fallocate(struct file *file, int mode, loff_t offset,
loff_t len)
{
struct inode *inode = file->f_path.dentry->d_inode;
struct super_block *sb = inode->i_sb;
struct nova_inode_info *si = NOVA_I(inode);
struct nova_inode_info_header *sih = &si->header;
struct nova_inode *pi;
struct nova_inode_update update;
unsigned long start_blk, num_blocks;
unsigned long blockoff;
unsigned int data_bits;
loff_t new_size;
long ret = 0;
int blocksize_mask;
INIT_TIMING(fallocate_time);
u64 epoch_id;
u32 time;
unsigned long irq_flags = 0;
/*
* Fallocate does not make much sence for CoW,
* but we still support it for DAX-mmap purpose.
*/
/* We only support the FALLOC_FL_KEEP_SIZE and FALLOC_FL_PUNCH_HOLE
* mode
*/
if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
return -EOPNOTSUPP;
if (S_ISDIR(inode->i_mode))
return -ENODEV;
new_size = len + offset;
if (!(mode & FALLOC_FL_KEEP_SIZE) && new_size > inode->i_size) {
ret = inode_newsize_ok(inode, new_size);
if (ret)
return ret;
} else {
new_size = inode->i_size;
}
nova_dbgv("%s: inode %lu, offset %lld, count %lld, mode 0x%x\n",
__func__, inode->i_ino, offset, len, mode);
NOVA_START_TIMING(fallocate_t, fallocate_time);
inode_lock(inode);
pi = nova_get_inode(sb, inode);
if (!pi) {
ret = -EACCES;
goto out;
}
epoch_id = nova_get_epoch_id(sb);
if (mode & FALLOC_FL_PUNCH_HOLE) {
if (offset % PAGE_SIZE != 0 || len != PAGE_SIZE) {
printk("offset %lld, len %lld\n", offset, len);
ret = -EOPNOTSUPP;
goto out;
}
// TODO: truncate_pagecache?
nova_truncate_file_blocks(inode, offset, offset + len,
epoch_id);
goto out;
}
inode->i_mtime = inode->i_ctime = current_time(inode);
time = current_time(inode).tv_sec;
blocksize_mask = sb->s_blocksize - 1;
start_blk = offset >> sb->s_blocksize_bits;
blockoff = offset & blocksize_mask;
num_blocks = (blockoff + len + blocksize_mask) >> sb->s_blocksize_bits;
update.tail = sih->log_tail;
update.alter_tail = sih->alter_log_tail;
data_bits = blk_type_to_shift[sih->i_blk_type];
sih->i_blocks += (num_blocks << (data_bits - sb->s_blocksize_bits));
inode->i_blocks = sih->i_blocks;
nova_dbgv("blocks: %llu, %lu\n", (u64)inode->i_blocks, sih->i_blocks);
if (ret || (mode & FALLOC_FL_KEEP_SIZE)) {
nova_memunlock_inode(sb, pi, &irq_flags);
pi->i_flags |= cpu_to_le32(NOVA_EOFBLOCKS_FL);
nova_memlock_inode(sb, pi, &irq_flags);
sih->i_flags |= cpu_to_le32(NOVA_EOFBLOCKS_FL);
}
if (!(mode & FALLOC_FL_KEEP_SIZE) && new_size > inode->i_size) {
inode->i_size = new_size;
sih->i_size = new_size;
}
nova_memunlock_inode(sb, pi, &irq_flags);
nova_update_inode_checksum(pi);
nova_update_alter_inode(sb, inode, pi);
nova_memlock_inode(sb, pi, &irq_flags);
sih->trans_id++;
out:
inode_unlock(inode);
NOVA_END_TIMING(fallocate_t, fallocate_time);
return ret;
}
static int nova_iomap_begin_nolock(struct inode *inode, loff_t offset,
loff_t length, unsigned int flags, struct iomap *iomap)
{
return nova_iomap_begin(inode, offset, length, flags, iomap, false);
}
static struct iomap_ops nova_iomap_ops_nolock = {
.iomap_begin = nova_iomap_begin_nolock,
.iomap_end = nova_iomap_end,
};
static ssize_t nova_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct inode *inode = iocb->ki_filp->f_mapping->host;
ssize_t ret;
INIT_TIMING(read_iter_time);
if (!iov_iter_count(to))
return 0;
NOVA_START_TIMING(read_iter_t, read_iter_time);
inode_lock_shared(inode);
ret = dax_iomap_rw(iocb, to, &nova_iomap_ops_nolock);
inode_unlock_shared(inode);
file_accessed(iocb->ki_filp);
NOVA_END_TIMING(read_iter_t, read_iter_time);
return ret;
}
static int nova_update_iter_csum_parity(struct super_block *sb,
struct inode *inode, loff_t offset, size_t count)
{
struct nova_inode_info *si = NOVA_I(inode);
struct nova_inode_info_header *sih = &si->header;
unsigned long start_pgoff, end_pgoff;
loff_t end;
if (data_csum == 0 && data_parity == 0)
return 0;
end = offset + count;
start_pgoff = offset >> sb->s_blocksize_bits;
end_pgoff = end >> sb->s_blocksize_bits;
if (end & (nova_inode_blk_size(sih) - 1))
end_pgoff++;
nova_reset_csum_parity_range(sb, sih, NULL, start_pgoff,
end_pgoff, 0, 0);
return 0;
}
static ssize_t nova_dax_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct super_block *sb = inode->i_sb;
struct nova_inode_info *si = NOVA_I(inode);
struct nova_inode_info_header *sih = &si->header;
loff_t offset;
size_t count;
ssize_t ret;
INIT_TIMING(write_iter_time);
NOVA_START_TIMING(write_iter_t, write_iter_time);
inode_lock(inode);
ret = generic_write_checks(iocb, from);
if (ret <= 0)
goto out_unlock;
ret = file_remove_privs(file);
if (ret)
goto out_unlock;
ret = file_update_time(file);
if (ret)
goto out_unlock;
count = iov_iter_count(from);
offset = iocb->ki_pos;
ret = dax_iomap_rw(iocb, from, &nova_iomap_ops_nolock);
if (ret > 0 && iocb->ki_pos > i_size_read(inode)) {
i_size_write(inode, iocb->ki_pos);
sih->i_size = iocb->ki_pos;
mark_inode_dirty(inode);
}
nova_update_iter_csum_parity(sb, inode, offset, count);
out_unlock:
inode_unlock(inode);
if (ret > 0)
ret = generic_write_sync(iocb, ret);
NOVA_END_TIMING(write_iter_t, write_iter_time);
return ret;
}
static ssize_t
do_dax_mapping_read(struct file *filp, char __user *buf,
size_t len, loff_t *ppos)
{
struct inode *inode = filp->f_mapping->host;
struct super_block *sb = inode->i_sb;
struct nova_inode_info *si = NOVA_I(inode);
struct nova_inode_info_header *sih = &si->header;
struct nova_file_write_entry *entry;
struct nova_file_write_entry *entryc, entry_copy;
pgoff_t index, end_index;
unsigned long offset;
loff_t isize, pos;
size_t copied = 0, error = 0;
INIT_TIMING(memcpy_time);
pos = *ppos;
index = pos >> PAGE_SHIFT;
offset = pos & ~PAGE_MASK;
if (!access_ok(buf, len)) {
error = -EFAULT;
goto out;
}
isize = i_size_read(inode);
if (!isize)
goto out;
nova_dbgv("%s: inode %lu, offset %lld, count %lu, size %lld\n",
__func__, inode->i_ino, pos, len, isize);
if (len > isize - pos)
len = isize - pos;
if (len <= 0)
goto out;
entryc = (metadata_csum == 0) ? entry : &entry_copy;
end_index = (isize - 1) >> PAGE_SHIFT;
do {
unsigned long nr, left;
unsigned long nvmm;
void *dax_mem = NULL;
int zero = 0;
/* nr is the maximum number of bytes to copy from this page */
if (index >= end_index) {
if (index > end_index)
goto out;
nr = ((isize - 1) & ~PAGE_MASK) + 1;
if (nr <= offset)
goto out;
}
entry = nova_get_write_entry(sb, sih, index);
if (unlikely(entry == NULL)) {
nova_dbgv("Required extent not found: pgoff %lu, inode size %lld\n",
index, isize);
nr = PAGE_SIZE;
zero = 1;
goto memcpy;
}
if (metadata_csum == 0)
entryc = entry;
else if (!nova_verify_entry_csum(sb, entry, entryc))
return -EIO;
/* Find contiguous blocks */
if (index < entryc->pgoff ||
index - entryc->pgoff >= entryc->num_pages) {
nova_err(sb, "%s ERROR: %lu, entry pgoff %llu, num %u, blocknr %llu\n",
__func__, index, entry->pgoff,
entry->num_pages, entry->block >> PAGE_SHIFT);
return -EINVAL;
}
if (entryc->reassigned == 0) {
nr = (entryc->num_pages - (index - entryc->pgoff))
* PAGE_SIZE;
} else {
nr = PAGE_SIZE;
}
nvmm = get_nvmm(sb, sih, entryc, index);
dax_mem = nova_get_block(sb, (nvmm << PAGE_SHIFT));
memcpy:
nr = nr - offset;
if (nr > len - copied)
nr = len - copied;
if ((!zero) && (data_csum > 0)) {
if (nova_find_pgoff_in_vma(inode, index))
goto skip_verify;
if (!nova_verify_data_csum(sb, sih, nvmm, offset, nr)) {
nova_err(sb, "%s: nova data checksum and recovery fail! inode %lu, offset %lu, entry pgoff %lu, %u pages, pgoff %lu\n",
__func__, inode->i_ino, offset,
entry->pgoff, entry->num_pages, index);
error = -EIO;
goto out;
}
}
skip_verify:
NOVA_START_TIMING(memcpy_r_nvmm_t, memcpy_time);
if (!zero)
left = __copy_to_user(buf + copied,
dax_mem + offset, nr);
else
left = __clear_user(buf + copied, nr);
NOVA_END_TIMING(memcpy_r_nvmm_t, memcpy_time);
if (left) {
nova_dbg("%s ERROR!: bytes %lu, left %lu\n",
__func__, nr, left);
error = -EFAULT;
goto out;
}
copied += (nr - left);
offset += (nr - left);
index += offset >> PAGE_SHIFT;
offset &= ~PAGE_MASK;
} while (copied < len);
out:
*ppos = pos + copied;
if (filp)
file_accessed(filp);
NOVA_STATS_ADD(read_bytes, copied);
nova_dbgv("%s returned %zu\n", __func__, copied);
return copied ? copied : error;
}
/*
* Wrappers. We need to use the rcu read lock to avoid
* concurrent truncate operation. No problem for write because we held
* lock.
*/
static ssize_t nova_dax_file_read(struct file *filp, char __user *buf,
size_t len, loff_t *ppos)
{
struct inode *inode = filp->f_mapping->host;
ssize_t res;
INIT_TIMING(dax_read_time);
NOVA_START_TIMING(dax_read_t, dax_read_time);
inode_lock_shared(inode);
res = do_dax_mapping_read(filp, buf, len, ppos);
inode_unlock_shared(inode);
NOVA_END_TIMING(dax_read_t, dax_read_time);
return res;
}
struct cow_write_env {
// Immutable
struct inode *inode;
struct super_block *sb;
struct nova_inode_info_header *sih;
u64 epoch_id;
u32 time;
struct nova_inode *pi;
// Mutable
struct nova_file_write_entry entry_data;
struct nova_inode_update update;
loff_t pos;
u64 begin_tail;
ssize_t written;
// For statistics
unsigned long step;
};
static int advance(struct cow_write_env *env, size_t written,
struct nova_write_para_continuous *wp)
{
u64 file_size;
struct nova_fp empty_fp = {0};
int ret;
BUG_ON(written == 0);
if (env->pos + written > env->inode->i_size)
file_size = cpu_to_le64(env->pos + written);
else
file_size = cpu_to_le64(env->inode->i_size);
nova_init_file_write_entry(env->sb, env->sih, &env->entry_data,
env->epoch_id, env->pos >> env->sb->s_blocksize_bits, wp->num,
wp->blocknr, env->time, file_size);
// incorporate fp into file write entry
if (wp->num == 1) {
// collisioned entry
if (wp->normal.last_accessed == NULL)
env->entry_data.fp = empty_fp;
else
env->entry_data.fp = wp->normal.last_accessed->fp;
} else {
env->entry_data.fp = empty_fp;
}
ret = nova_append_file_write_entry(env->sb, env->pi, env->inode,
&env->entry_data, &env->update);
if (ret) {
nova_dbg("%s: append inode entry failed\n", __func__);
return -ENOSPC;
}
if (wp->blocknr_next != 0) {
wp->blocknr = wp->blocknr_next;
wp->num = 1;
wp->blocknr_next = 0;
} else {
wp->blocknr = 0;
wp->num = 0;
}
env->pos += written;
if (env->begin_tail == 0)
env->begin_tail = env->update.curr_entry;
env->written += written;
env->step += 1;
return 0;
}
static __always_inline void nova_enter_dedup(struct light_dedup_meta *meta, struct nova_write_para_continuous *wp)
{
int cpu, ret;
int dedup_ctx = -1;
if (!wp->seq_file)
dedup_ctx = light_dedup_srcu_read_lock();
// ret = read_trylock(&meta->worker_lock);
// if (!ret) {
// } else {
// atomic64_fetch_add_relaxed(1, &meta->rcu_unprotected_worker_num);
// // lock is held, now we release it
// read_unlock(&meta->worker_lock);
// }
wp->dedup_ctx = dedup_ctx;
// NOTE: prevent access any freed entry in the section
// if call_rcu to free happens before read lock,
// num_holders = 0, we will not access it
// else call_rcu is delayed after read unlock
cpu = get_cpu();
wp->normal.first_accessed = wp->normal.last_accessed = per_cpu(last_accessed_fpentry_per_cpu, cpu);
wp->stream_trust_degree = per_cpu(stream_trust_degree_per_cpu, cpu);
put_cpu();
wp->prefetched_blocknr[0] = wp->prefetched_blocknr[1] = 0;
}
// https://stackoverflow.com/questions/4016061/why-is-inlining-considered-faster-than-a-function-call
// doing a call can flush the instruction pipeline and force the CPU to wait for memory being fetched
static __always_inline void nova_exit_dedup(struct light_dedup_meta *meta, struct nova_write_para_continuous *wp,
bool exception)
{
bool append = wp->append;
unsigned long num_blocks = wp->num_blocks;
int cpu;
if (!exception) {
cpu = get_cpu();
if (!wp->seq_file) {
per_cpu(last_accessed_fpentry_per_cpu, cpu) = wp->normal.last_accessed;
// do not protect if there is no deletion.
if (wp->normal.last_accessed != wp->normal.first_accessed) {
if (wp->normal.last_accessed)
incr_holders(wp->normal.last_accessed);
if (wp->normal.first_accessed)
decr_holders(meta, wp->normal.first_accessed);
}
}
per_cpu(stream_trust_degree_per_cpu, cpu) = wp->stream_trust_degree;
put_cpu();
}
if (!wp->seq_file)
light_dedup_srcu_read_unlock(wp->dedup_ctx);
}
/*
* Perform a COW write. Must hold the inode lock before calling.
*/
static ssize_t do_nova_cow_file_write(struct file *filp,
const char __user *buf, size_t len, loff_t *ppos)
{
struct address_space *mapping = filp->f_mapping;
struct cow_write_env env;
struct nova_sb_info *sbi;
struct light_dedup_meta *meta;
struct nova_inode inode_copy;
size_t offset, bytes;
unsigned long num_blocks;
struct llist_node *kbuf_node;
struct kbuf_obj *kbuf_obj;
struct nova_write_para_continuous wp;
int cpu, idx;
unsigned long irq_flags = 0;
ssize_t ret;
INIT_TIMING(cow_write_time);
INIT_TIMING(copy_from_user_time);
if (len == 0)
return 0;
NOVA_START_TIMING(do_cow_write_t, cow_write_time);
if (!access_ok(buf, len)) {
ret = -EFAULT;
goto err_out0;
}
env.inode = mapping->host;
env.sb = env.inode->i_sb;
env.sih = &NOVA_I(env.inode)->header;
/* nova_inode tail pointer will be updated and we make sure all other
* inode fields are good before checksumming the whole structure
*/
if (nova_check_inode_integrity(env.sb, env.sih->ino, env.sih->pi_addr,
env.sih->alter_pi_addr, &inode_copy, 0) < 0) {
ret = -EIO;
goto err_out0;
}
env.pos = *ppos;
if (filp->f_flags & O_APPEND)
env.pos = i_size_read(env.inode);
offset = env.pos & (env.sb->s_blocksize - 1);
num_blocks = ((len + offset - 1) >> env.sb->s_blocksize_bits) + 1;
if (nova_check_overlap_vmas(env.sb, env.sih,
env.pos >> env.sb->s_blocksize_bits, num_blocks)) {
nova_dbgv("COW write overlaps with vma: inode %lu, pgoff %lu, %lu blocks\n",
env.inode->i_ino,
(unsigned long)(env.pos >> env.sb->s_blocksize_bits),
num_blocks);
NOVA_STATS_ADD(cow_overlap_mmap, 1);
NOVA_END_TIMING(do_cow_write_t, cow_write_time);
return do_nova_inplace_file_write(filp, buf, len, ppos);
}
// nova_info("fadvise, sequential?: %d, set_advise: %d\n", filp->f_mode & FMODE_RANDOM, env.sih->set_advise);
// FIXME: seqeuntial overwrite is not supported
wp.seq_file = ((filp->f_mode & FMODE_RANDOM) && env.sih->set_advise) ? false : true;
// nova_info("seq file: %d\n", wp.seq_file);
ret = file_remove_privs(filp);
if (ret)
goto err_out0;
env.epoch_id = nova_get_epoch_id(env.sb);
env.inode->i_ctime = env.inode->i_mtime = current_time(env.inode);
env.time = current_time(env.inode).tv_sec;
env.pi = nova_get_block(env.sb, env.sih->pi_addr);
env.update.tail = env.sih->log_tail;
env.update.alter_tail = env.sih->alter_log_tail;
env.begin_tail = 0;
env.written = 0;
env.step = 0;
sbi = NOVA_SB(env.sb);
meta = &sbi->light_dedup_meta;
atomic64_fetch_add_relaxed(1, &meta->thread_num);
wp.ubuf = buf;
wp.len = len;
wp.blocknr = 0;
wp.num = 0;
wp.blocknr_next = 0;
kbuf_node = generic_cache_alloc(&meta->kbuf_cache, GFP_KERNEL);
if (kbuf_node == NULL) {
ret = -ENOMEM;
goto err_out1;
}
kbuf_obj = container_of(kbuf_node, struct kbuf_obj, node);
wp.kbuf = kbuf_obj->kbuf;
if (env.pos == i_size_read(env.inode))
wp.append = true;
else
wp.append = false;
wp.num_blocks = num_blocks;
nova_dbgv("%s: inode %lu, offset %lld, count %lu\n",
__func__, env.inode->i_ino, env.pos, len);
nova_enter_dedup(meta, &wp);
if (offset != 0) {
bytes = env.sb->s_blocksize - offset;
if (bytes > wp.len)
bytes = wp.len;
ret = nova_handle_head_tail_blocks(env.sb, env.inode, env.pos,
bytes, wp.kbuf);
if (ret)
goto err_out2;
/* Now copy from user buf */
NOVA_START_TIMING(copy_from_user_t, copy_from_user_time);
if (copy_from_user(wp.kbuf + offset, wp.ubuf, bytes)) {
NOVA_END_TIMING(copy_from_user_t, copy_from_user_time);
ret = -EFAULT;
goto err_out2;
}
NOVA_END_TIMING(copy_from_user_t, copy_from_user_time);
ret = light_dedup_incr_ref(meta, offset, bytes, wp.kbuf, wp.ubuf, &wp.normal);
if (ret < 0)
goto err_out2;
wp.ubuf += bytes;
wp.len -= bytes;
wp.blocknr = wp.normal.blocknr;
wp.num = 1;
ret = advance(&env, bytes, &wp);
if (ret < 0)
goto err_out2;
}
while (wp.len >= env.sb->s_blocksize) {
if (wp.normal.last_accessed) {
if (atomic64_read(&wp.normal.last_accessed->refcount) == 0)
wp.normal.last_accessed = NULL;
}
ret = light_dedup_incr_ref_continuous(sbi, &wp);
if (ret < 0)
goto err_out2;
ret = advance(&env, wp.num * env.sb->s_blocksize, &wp);
if (ret < 0)
goto err_out2;
}
if (wp.num != 0) {
ret = advance(&env, wp.num * env.sb->s_blocksize, &wp);
if (ret < 0)
goto err_out2;
}
BUG_ON(wp.num != 0);
if (wp.len != 0) {
bytes = wp.len;
ret = nova_handle_head_tail_blocks(env.sb, env.inode, env.pos, bytes,
wp.kbuf);
if (ret)
goto err_out2;
/* Now copy from user buf */
NOVA_START_TIMING(copy_from_user_t, copy_from_user_time);
if (copy_from_user(wp.kbuf, wp.ubuf, bytes)) {
NOVA_END_TIMING(copy_from_user_t, copy_from_user_time);
ret = -EFAULT;
goto err_out2;
}
NOVA_END_TIMING(copy_from_user_t, copy_from_user_time);
ret = light_dedup_incr_ref(meta, 0, bytes, wp.kbuf, wp.ubuf, &wp.normal);
if (ret < 0)
goto err_out2;
wp.ubuf += bytes;
wp.len -= bytes;
wp.blocknr = wp.normal.blocknr;
wp.num = 1;
ret = advance(&env, bytes, &wp);
if (ret < 0)
goto err_out2;
}
// nova_flush_entry_if_not_null(wp.normal.last_ref_entries[0], false);
// nova_flush_entry_if_not_null(wp.normal.last_ref_entries[1], false);
nova_exit_dedup(meta, &wp, false);
// if (!in_the_same_cacheline(wp.normal.last_new_entries[0],
// wp.normal.last_new_entries[1]))
// nova_flush_entry_if_not_null(wp.normal.last_new_entries[1],
// false);
// if (wp.normal.dirty_map_blocknr_to_pentry != NULL)
// nova_flush_cacheline(wp.normal.dirty_map_blocknr_to_pentry,
// false);
env.sih->i_blocks += (num_blocks <<
(blk_type_to_shift[env.sih->i_blk_type] -
env.sb->s_blocksize_bits));
nova_memunlock_inode(env.sb, env.pi, &irq_flags);
nova_update_inode(env.sb, env.inode, env.pi, &env.update, 1);
nova_memlock_inode(env.sb, env.pi, &irq_flags);
/* Free the overlap blocks after the write is committed */
// ret = nova_reassign_file_tree(env.sb, env.sih, env.begin_tail);
// if (ret)
// goto err_out2;
env.inode->i_blocks = env.sih->i_blocks;
NOVA_STATS_ADD(cow_write_breaks, env.step);
nova_dbgv("blocks: %llu, %lu\n",
(u64)env.inode->i_blocks, env.sih->i_blocks);
*ppos = env.pos;
if (env.pos > env.inode->i_size) {
i_size_write(env.inode, env.pos);
env.sih->i_size = env.pos;
}
env.sih->trans_id++;
NOVA_STATS_ADD(cow_write_bytes, env.written);
generic_cache_free(&meta->kbuf_cache, kbuf_node);
atomic64_fetch_sub_relaxed(1, &meta->thread_num);
return env.written;
err_out2:
BUG_ON(1);
// TODO: Make sure that the clean up does not fail.
if (wp.blocknr_next != 0) {
nova_deref_blocks(env.sb, wp.blocknr_next, 1);
}
BUG_ON(nova_cleanup_incomplete_write(env.sb, env.sih,
wp.blocknr, wp.num, env.begin_tail, env.update.tail) < 0);
generic_cache_free(&meta->kbuf_cache, kbuf_node);
err_out1:
atomic64_fetch_sub_relaxed(1, &meta->thread_num);
err_out0:
nova_exit_dedup(meta, &wp, true);
NOVA_END_TIMING(do_cow_write_t, cow_write_time);
return ret;
}
/*
* Acquire locks and perform COW write.
*/
ssize_t nova_cow_file_write(struct file *filp,
const char __user *buf, size_t len, loff_t *ppos)
{
struct address_space *mapping = filp->f_mapping;
struct inode *inode = mapping->host;
int ret;
INIT_TIMING(time);
if (len == 0)
return 0;
NOVA_START_TIMING(cow_write_t, time);
sb_start_write(inode->i_sb);
inode_lock(inode);
ret = do_nova_cow_file_write(filp, buf, len, ppos);
inode_unlock(inode);
sb_end_write(inode->i_sb);
NOVA_END_TIMING(cow_write_t, time);
return ret;
}
static ssize_t nova_dax_file_write(struct file *filp, const char __user *buf,
size_t len, loff_t *ppos)
{
struct address_space *mapping = filp->f_mapping;
struct inode *inode = mapping->host;
if (test_opt(inode->i_sb, DATA_COW))
return nova_cow_file_write(filp, buf, len, ppos);
else
return nova_inplace_file_write(filp, buf, len, ppos);
}
static ssize_t do_nova_dax_file_write(struct file *filp, const char __user *buf,
size_t len, loff_t *ppos)
{
struct address_space *mapping = filp->f_mapping;
struct inode *inode = mapping->host;
if (test_opt(inode->i_sb, DATA_COW))
return do_nova_cow_file_write(filp, buf, len, ppos);
else
return do_nova_inplace_file_write(filp, buf, len, ppos);
}
static int nova_dax_file_mmap(struct file *file, struct vm_area_struct *vma)
{
struct inode *inode = file->f_mapping->host;
file_accessed(file);
vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
vma->vm_ops = &nova_dax_vm_ops;
nova_insert_write_vma(vma);
nova_dbg_mmap4k("[%s:%d] inode %lu, MMAP 4KPAGE vm_start(0x%lx), vm_end(0x%lx), vm pgoff %lu, %lu blocks, vm_flags(0x%lx), vm_page_prot(0x%lx)\n",
__func__, __LINE__,
inode->i_ino, vma->vm_start, vma->vm_end,
vma->vm_pgoff,
(vma->vm_end - vma->vm_start) >> PAGE_SHIFT,
vma->vm_flags,
pgprot_val(vma->vm_page_prot));
return 0;
}
static int nova_fadvise(struct file *file, loff_t offset, loff_t len, int advice)
{
struct inode *inode = file->f_mapping->host;
struct nova_inode_info_header *sih = &NOVA_I(inode)->header;