summaryrefslogtreecommitdiffstats
path: root/fs/xfs/linux-2.6/xfs_lrw.c
blob: aa9daaea6c34b2d98697b899e73ee10e4222e1bb (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
/*
 * Copyright (c) 2000-2003 Silicon Graphics, Inc.  All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of version 2 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.
 *
 * Further, this software is distributed without any warranty that it is
 * free of the rightful claim of any third person regarding infringement
 * or the like.  Any license provided herein, whether implied or
 * otherwise, applies only to this software file.  Patent licenses, if
 * any, provided herein do not apply to combinations of this program with
 * other software, or any other product whatsoever.
 *
 * You should have received a copy of the GNU General Public License along
 * with this program; if not, write the Free Software Foundation, Inc., 59
 * Temple Place - Suite 330, Boston MA 02111-1307, USA.
 *
 * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
 * Mountain View, CA  94043, or:
 *
 * http://www.sgi.com
 *
 * For further information regarding this notice, see:
 *
 * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
 */
/*
 *  fs/xfs/linux/xfs_lrw.c (Linux Read Write stuff)
 *
 */

#include "xfs.h"

#include "xfs_fs.h"
#include "xfs_inum.h"
#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir.h"
#include "xfs_dir2.h"
#include "xfs_alloc.h"
#include "xfs_dmapi.h"
#include "xfs_quota.h"
#include "xfs_mount.h"
#include "xfs_alloc_btree.h"
#include "xfs_bmap_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_attr_sf.h"
#include "xfs_dir_sf.h"
#include "xfs_dir2_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_bmap.h"
#include "xfs_bit.h"
#include "xfs_rtalloc.h"
#include "xfs_error.h"
#include "xfs_itable.h"
#include "xfs_rw.h"
#include "xfs_acl.h"
#include "xfs_cap.h"
#include "xfs_mac.h"
#include "xfs_attr.h"
#include "xfs_inode_item.h"
#include "xfs_buf_item.h"
#include "xfs_utils.h"
#include "xfs_iomap.h"

#include <linux/capability.h>
#include <linux/writeback.h>


#if defined(XFS_RW_TRACE)
void
xfs_rw_enter_trace(
	int			tag,
	xfs_iocore_t		*io,
	void			*data,
	size_t			segs,
	loff_t			offset,
	int			ioflags)
{
	xfs_inode_t	*ip = XFS_IO_INODE(io);

	if (ip->i_rwtrace == NULL)
		return;
	ktrace_enter(ip->i_rwtrace,
		(void *)(unsigned long)tag,
		(void *)ip,
		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
		(void *)data,
		(void *)((unsigned long)segs),
		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
		(void *)((unsigned long)(offset & 0xffffffff)),
		(void *)((unsigned long)ioflags),
		(void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)),
		(void *)((unsigned long)(io->io_new_size & 0xffffffff)),
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL);
}

void
xfs_inval_cached_trace(
	xfs_iocore_t	*io,
	xfs_off_t	offset,
	xfs_off_t	len,
	xfs_off_t	first,
	xfs_off_t	last)
{
	xfs_inode_t	*ip = XFS_IO_INODE(io);

	if (ip->i_rwtrace == NULL)
		return;
	ktrace_enter(ip->i_rwtrace,
		(void *)(__psint_t)XFS_INVAL_CACHED,
		(void *)ip,
		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
		(void *)((unsigned long)(offset & 0xffffffff)),
		(void *)((unsigned long)((len >> 32) & 0xffffffff)),
		(void *)((unsigned long)(len & 0xffffffff)),
		(void *)((unsigned long)((first >> 32) & 0xffffffff)),
		(void *)((unsigned long)(first & 0xffffffff)),
		(void *)((unsigned long)((last >> 32) & 0xffffffff)),
		(void *)((unsigned long)(last & 0xffffffff)),
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL);
}
#endif

/*
 *	xfs_iozero
 *
 *	xfs_iozero clears the specified range of buffer supplied,
 *	and marks all the affected blocks as valid and modified.  If
 *	an affected block is not allocated, it will be allocated.  If
 *	an affected block is not completely overwritten, and is not
 *	valid before the operation, it will be read from disk before
 *	being partially zeroed.
 */
STATIC int
xfs_iozero(
	struct inode		*ip,	/* inode			*/
	loff_t			pos,	/* offset in file		*/
	size_t			count,	/* size of data to zero		*/
	loff_t			end_size)	/* max file size to set */
{
	unsigned		bytes;
	struct page		*page;
	struct address_space	*mapping;
	char			*kaddr;
	int			status;

	mapping = ip->i_mapping;
	do {
		unsigned long index, offset;

		offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
		index = pos >> PAGE_CACHE_SHIFT;
		bytes = PAGE_CACHE_SIZE - offset;
		if (bytes > count)
			bytes = count;

		status = -ENOMEM;
		page = grab_cache_page(mapping, index);
		if (!page)
			break;

		kaddr = kmap(page);
		status = mapping->a_ops->prepare_write(NULL, page, offset,
							offset + bytes);
		if (status) {
			goto unlock;
		}

		memset((void *) (kaddr + offset), 0, bytes);
		flush_dcache_page(page);
		status = mapping->a_ops->commit_write(NULL, page, offset,
							offset + bytes);
		if (!status) {
			pos += bytes;
			count -= bytes;
			if (pos > i_size_read(ip))
				i_size_write(ip, pos < end_size ? pos : end_size);
		}

unlock:
		kunmap(page);
		unlock_page(page);
		page_cache_release(page);
		if (status)
			break;
	} while (count);

	return (-status);
}

/*
 * xfs_inval_cached_pages
 * 
 * This routine is responsible for keeping direct I/O and buffered I/O
 * somewhat coherent.  From here we make sure that we're at least
 * temporarily holding the inode I/O lock exclusively and then call
 * the page cache to flush and invalidate any cached pages.  If there
 * are no cached pages this routine will be very quick.
 */
void
xfs_inval_cached_pages(
	vnode_t		*vp,
	xfs_iocore_t	*io,
	xfs_off_t	offset,
	int		write,
	int		relock)
{
	if (VN_CACHED(vp)) {
		xfs_inval_cached_trace(io, offset, -1, ctooff(offtoct(offset)), -1);
		VOP_FLUSHINVAL_PAGES(vp, ctooff(offtoct(offset)), -1, FI_REMAPF_LOCKED);
	}

}

ssize_t			/* bytes read, or (-)  error */
xfs_read(
	bhv_desc_t		*bdp,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		segs,
	loff_t			*offset,
	int			ioflags,
	cred_t			*credp)
{
	struct file		*file = iocb->ki_filp;
	struct inode		*inode = file->f_mapping->host;
	size_t			size = 0;
	ssize_t			ret;
	xfs_fsize_t		n;
	xfs_inode_t		*ip;
	xfs_mount_t		*mp;
	vnode_t			*vp;
	unsigned long		seg;

	ip = XFS_BHVTOI(bdp);
	vp = BHV_TO_VNODE(bdp);
	mp = ip->i_mount;

	XFS_STATS_INC(xs_read_calls);

	/* START copy & waste from filemap.c */
	for (seg = 0; seg < segs; seg++) {
		const struct iovec *iv = &iovp[seg];

		/*
		 * If any segment has a negative length, or the cumulative
		 * length ever wraps negative then return -EINVAL.
		 */
		size += iv->iov_len;
		if (unlikely((ssize_t)(size|iv->iov_len) < 0))
			return XFS_ERROR(-EINVAL);
	}
	/* END copy & waste from filemap.c */

	if (unlikely(ioflags & IO_ISDIRECT)) {
		xfs_buftarg_t	*target =
			(ip->i_d.di_flags & XFS_DIFLAG_REALTIME) ?
				mp->m_rtdev_targp : mp->m_ddev_targp;
		if ((*offset & target->pbr_smask) ||
		    (size & target->pbr_smask)) {
			if (*offset == ip->i_d.di_size) {
				return (0);
			}
			return -XFS_ERROR(EINVAL);
		}
	}

	n = XFS_MAXIOFFSET(mp) - *offset;
	if ((n <= 0) || (size == 0))
		return 0;

	if (n < size)
		size = n;

	if (XFS_FORCED_SHUTDOWN(mp)) {
		return -EIO;
	}

	if (unlikely(ioflags & IO_ISDIRECT))
		down(&inode->i_sem);
	xfs_ilock(ip, XFS_IOLOCK_SHARED);

	if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) &&
	    !(ioflags & IO_INVIS)) {
		vrwlock_t locktype = VRWLOCK_READ;

		ret = -XFS_SEND_DATA(mp, DM_EVENT_READ,
					BHV_TO_VNODE(bdp), *offset, size,
					FILP_DELAY_FLAG(file), &locktype);
		if (ret) {
			xfs_iunlock(ip, XFS_IOLOCK_SHARED);
			goto unlock_isem;
		}
	}

	xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore,
				(void *)iovp, segs, *offset, ioflags);
	ret = __generic_file_aio_read(iocb, iovp, segs, offset);
	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
		ret = wait_on_sync_kiocb(iocb);
	if (ret > 0)
		XFS_STATS_ADD(xs_read_bytes, ret);

	xfs_iunlock(ip, XFS_IOLOCK_SHARED);

	if (likely(!(ioflags & IO_INVIS)))
		xfs_ichgtime(ip, XFS_ICHGTIME_ACC);

unlock_isem:
	if (unlikely(ioflags & IO_ISDIRECT))
		up(&inode->i_sem);
	return ret;
}

ssize_t
xfs_sendfile(
	bhv_desc_t		*bdp,
	struct file		*filp,
	loff_t			*offset,
	int			ioflags,
	size_t			count,
	read_actor_t		actor,
	void			*target,
	cred_t			*credp)
{
	ssize_t			ret;
	xfs_fsize_t		n;
	xfs_inode_t		*ip;
	xfs_mount_t		*mp;
	vnode_t			*vp;

	ip = XFS_BHVTOI(bdp);
	vp = BHV_TO_VNODE(bdp);
	mp = ip->i_mount;

	XFS_STATS_INC(xs_read_calls);

	n = XFS_MAXIOFFSET(mp) - *offset;
	if ((n <= 0) || (count == 0))
		return 0;

	if (n < count)
		count = n;

	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
		return -EIO;

	xfs_ilock(ip, XFS_IOLOCK_SHARED);

	if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) &&
	    (!(ioflags & IO_INVIS))) {
		vrwlock_t locktype = VRWLOCK_READ;
		int error;

		error = XFS_SEND_DATA(mp, DM_EVENT_READ, BHV_TO_VNODE(bdp), *offset, count,
				      FILP_DELAY_FLAG(filp), &locktype);
		if (error) {
			xfs_iunlock(ip, XFS_IOLOCK_SHARED);
			return -error;
		}
	}
	xfs_rw_enter_trace(XFS_SENDFILE_ENTER, &ip->i_iocore,
		   (void *)(unsigned long)target, count, *offset, ioflags);
	ret = generic_file_sendfile(filp, offset, count, actor, target);

	xfs_iunlock(ip, XFS_IOLOCK_SHARED);

	if (ret > 0)
		XFS_STATS_ADD(xs_read_bytes, ret);

	if (likely(!(ioflags & IO_INVIS)))
		xfs_ichgtime(ip, XFS_ICHGTIME_ACC);

	return ret;
}

/*
 * This routine is called to handle zeroing any space in the last
 * block of the file that is beyond the EOF.  We do this since the
 * size is being increased without writing anything to that block
 * and we don't want anyone to read the garbage on the disk.
 */
STATIC int				/* error (positive) */
xfs_zero_last_block(
	struct inode	*ip,
	xfs_iocore_t	*io,
	xfs_off_t	offset,
	xfs_fsize_t	isize,
	xfs_fsize_t	end_size)
{
	xfs_fileoff_t	last_fsb;
	xfs_mount_t	*mp;
	int		nimaps;
	int		zero_offset;
	int		zero_len;
	int		isize_fsb_offset;
	int		error = 0;
	xfs_bmbt_irec_t	imap;
	loff_t		loff;
	size_t		lsize;

	ASSERT(ismrlocked(io->io_lock, MR_UPDATE) != 0);
	ASSERT(offset > isize);

	mp = io->io_mount;

	isize_fsb_offset = XFS_B_FSB_OFFSET(mp, isize);
	if (isize_fsb_offset == 0) {
		/*
		 * There are no extra bytes in the last block on disk to
		 * zero, so return.
		 */
		return 0;
	}

	last_fsb = XFS_B_TO_FSBT(mp, isize);
	nimaps = 1;
	error = XFS_BMAPI(mp, NULL, io, last_fsb, 1, 0, NULL, 0, &imap,
			  &nimaps, NULL);
	if (error) {
		return error;
	}
	ASSERT(nimaps > 0);
	/*
	 * If the block underlying isize is just a hole, then there
	 * is nothing to zero.
	 */
	if (imap.br_startblock == HOLESTARTBLOCK) {
		return 0;
	}
	/*
	 * Zero the part of the last block beyond the EOF, and write it
	 * out sync.  We need to drop the ilock while we do this so we
	 * don't deadlock when the buffer cache calls back to us.
	 */
	XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL| XFS_EXTSIZE_RD);
	loff = XFS_FSB_TO_B(mp, last_fsb);
	lsize = XFS_FSB_TO_B(mp, 1);

	zero_offset = isize_fsb_offset;
	zero_len = mp->m_sb.sb_blocksize - isize_fsb_offset;

	error = xfs_iozero(ip, loff + zero_offset, zero_len, end_size);

	XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
	ASSERT(error >= 0);
	return error;
}

/*
 * Zero any on disk space between the current EOF and the new,
 * larger EOF.  This handles the normal case of zeroing the remainder
 * of the last block in the file and the unusual case of zeroing blocks
 * out beyond the size of the file.  This second case only happens
 * with fixed size extents and when the system crashes before the inode
 * size was updated but after blocks were allocated.  If fill is set,
 * then any holes in the range are filled and zeroed.  If not, the holes
 * are left alone as holes.
 */

int					/* error (positive) */
xfs_zero_eof(
	vnode_t		*vp,
	xfs_iocore_t	*io,
	xfs_off_t	offset,		/* starting I/O offset */
	xfs_fsize_t	isize,		/* current inode size */
	xfs_fsize_t	end_size)	/* terminal inode size */
{
	struct inode	*ip = LINVFS_GET_IP(vp);
	xfs_fileoff_t	start_zero_fsb;
	xfs_fileoff_t	end_zero_fsb;
	xfs_fileoff_t	prev_zero_fsb;
	xfs_fileoff_t	zero_count_fsb;
	xfs_fileoff_t	last_fsb;
	xfs_extlen_t	buf_len_fsb;
	xfs_extlen_t	prev_zero_count;
	xfs_mount_t	*mp;
	int		nimaps;
	int		error = 0;
	xfs_bmbt_irec_t	imap;
	loff_t		loff;
	size_t		lsize;

	ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
	ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));

	mp = io->io_mount;

	/*
	 * First handle zeroing the block on which isize resides.
	 * We only zero a part of that block so it is handled specially.
	 */
	error = xfs_zero_last_block(ip, io, offset, isize, end_size);
	if (error) {
		ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
		ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
		return error;
	}

	/*
	 * Calculate the range between the new size and the old
	 * where blocks needing to be zeroed may exist.  To get the
	 * block where the last byte in the file currently resides,
	 * we need to subtract one from the size and truncate back
	 * to a block boundary.  We subtract 1 in case the size is
	 * exactly on a block boundary.
	 */
	last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
	start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
	end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
	ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
	if (last_fsb == end_zero_fsb) {
		/*
		 * The size was only incremented on its last block.
		 * We took care of that above, so just return.
		 */
		return 0;
	}

	ASSERT(start_zero_fsb <= end_zero_fsb);
	prev_zero_fsb = NULLFILEOFF;
	prev_zero_count = 0;
	while (start_zero_fsb <= end_zero_fsb) {
		nimaps = 1;
		zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
		error = XFS_BMAPI(mp, NULL, io, start_zero_fsb, zero_count_fsb,
				  0, NULL, 0, &imap, &nimaps, NULL);
		if (error) {
			ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
			ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
			return error;
		}
		ASSERT(nimaps > 0);

		if (imap.br_state == XFS_EXT_UNWRITTEN ||
		    imap.br_startblock == HOLESTARTBLOCK) {
			/*
			 * This loop handles initializing pages that were
			 * partially initialized by the code below this
			 * loop. It basically zeroes the part of the page
			 * that sits on a hole and sets the page as P_HOLE
			 * and calls remapf if it is a mapped file.
			 */
			prev_zero_fsb = NULLFILEOFF;
			prev_zero_count = 0;
			start_zero_fsb = imap.br_startoff +
					 imap.br_blockcount;
			ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
			continue;
		}

		/*
		 * There are blocks in the range requested.
		 * Zero them a single write at a time.  We actually
		 * don't zero the entire range returned if it is
		 * too big and simply loop around to get the rest.
		 * That is not the most efficient thing to do, but it
		 * is simple and this path should not be exercised often.
		 */
		buf_len_fsb = XFS_FILBLKS_MIN(imap.br_blockcount,
					      mp->m_writeio_blocks << 8);
		/*
		 * Drop the inode lock while we're doing the I/O.
		 * We'll still have the iolock to protect us.
		 */
		XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);

		loff = XFS_FSB_TO_B(mp, start_zero_fsb);
		lsize = XFS_FSB_TO_B(mp, buf_len_fsb);

		error = xfs_iozero(ip, loff, lsize, end_size);

		if (error) {
			goto out_lock;
		}

		prev_zero_fsb = start_zero_fsb;
		prev_zero_count = buf_len_fsb;
		start_zero_fsb = imap.br_startoff + buf_len_fsb;
		ASSERT(start_zero_fsb <= (end_zero_fsb + 1));

		XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
	}

	return 0;

out_lock:

	XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
	ASSERT(error >= 0);
	return error;
}

ssize_t				/* bytes written, or (-) error */
xfs_write(
	bhv_desc_t		*bdp,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		nsegs,
	loff_t			*offset,
	int			ioflags,
	cred_t			*credp)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	unsigned long		segs = nsegs;
	xfs_inode_t		*xip;
	xfs_mount_t		*mp;
	ssize_t			ret = 0, error = 0;
	xfs_fsize_t		isize, new_size;
	xfs_iocore_t		*io;
	vnode_t			*vp;
	unsigned long		seg;
	int			iolock;
	int			eventsent = 0;
	vrwlock_t		locktype;
	size_t			ocount = 0, count;
	loff_t			pos;
	int			need_isem = 1, need_flush = 0;

	XFS_STATS_INC(xs_write_calls);

	vp = BHV_TO_VNODE(bdp);
	xip = XFS_BHVTOI(bdp);

	for (seg = 0; seg < segs; seg++) {
		const struct iovec *iv = &iovp[seg];

		/*
		 * If any segment has a negative length, or the cumulative
		 * length ever wraps negative then return -EINVAL.
		 */
		ocount += iv->iov_len;
		if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
			return -EINVAL;
		if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
			continue;
		if (seg == 0)
			return -EFAULT;
		segs = seg;
		ocount -= iv->iov_len;  /* This segment is no good */
		break;
	}

	count = ocount;
	pos = *offset;

	if (count == 0)
		return 0;

	io = &xip->i_iocore;
	mp = io->io_mount;

	if (XFS_FORCED_SHUTDOWN(mp))
		return -EIO;

	fs_check_frozen(vp->v_vfsp, SB_FREEZE_WRITE);

	if (ioflags & IO_ISDIRECT) {
		xfs_buftarg_t	*target =
			(xip->i_d.di_flags & XFS_DIFLAG_REALTIME) ?
				mp->m_rtdev_targp : mp->m_ddev_targp;

		if (ioflags & IO_ISAIO)
			return XFS_ERROR(-ENOSYS);

		if ((pos & target->pbr_smask) || (count & target->pbr_smask))
			return XFS_ERROR(-EINVAL);

		if (!VN_CACHED(vp) && pos < i_size_read(inode))
			need_isem = 0;

		if (VN_CACHED(vp))
			need_flush = 1;
	}

relock:
	if (need_isem) {
		iolock = XFS_IOLOCK_EXCL;
		locktype = VRWLOCK_WRITE;

		down(&inode->i_sem);
	} else {
		iolock = XFS_IOLOCK_SHARED;
		locktype = VRWLOCK_WRITE_DIRECT;
	}

	xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);

	isize = i_size_read(inode);

	if (file->f_flags & O_APPEND)
		*offset = isize;

start:
	error = -generic_write_checks(file, &pos, &count,
					S_ISBLK(inode->i_mode));
	if (error) {
		xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
		goto out_unlock_isem;
	}

	new_size = pos + count;
	if (new_size > isize)
		io->io_new_size = new_size;

	if ((DM_EVENT_ENABLED(vp->v_vfsp, xip, DM_EVENT_WRITE) &&
	    !(ioflags & IO_INVIS) && !eventsent)) {
		loff_t		savedsize = pos;
		int		dmflags = FILP_DELAY_FLAG(file);

		if (need_isem)
			dmflags |= DM_FLAGS_ISEM;

		xfs_iunlock(xip, XFS_ILOCK_EXCL);
		error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, vp,
				      pos, count,
				      dmflags, &locktype);
		if (error) {
			xfs_iunlock(xip, iolock);
			goto out_unlock_isem;
		}
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		eventsent = 1;

		/*
		 * The iolock was dropped and reaquired in XFS_SEND_DATA
		 * so we have to recheck the size when appending.
		 * We will only "goto start;" once, since having sent the
		 * event prevents another call to XFS_SEND_DATA, which is
		 * what allows the size to change in the first place.
		 */
		if ((file->f_flags & O_APPEND) && savedsize != isize) {
			pos = isize = xip->i_d.di_size;
			goto start;
		}
	}

	/*
	 * On Linux, generic_file_write updates the times even if
	 * no data is copied in so long as the write had a size.
	 *
	 * We must update xfs' times since revalidate will overcopy xfs.
	 */
	if (!(ioflags & IO_INVIS)) {
		xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
		inode_update_time(inode, 1);
	}

	/*
	 * If the offset is beyond the size of the file, we have a couple
	 * of things to do. First, if there is already space allocated
	 * we need to either create holes or zero the disk or ...
	 *
	 * If there is a page where the previous size lands, we need
	 * to zero it out up to the new size.
	 */

	if (pos > isize) {
		error = xfs_zero_eof(BHV_TO_VNODE(bdp), io, pos,
					isize, pos + count);
		if (error) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			goto out_unlock_isem;
		}
	}
	xfs_iunlock(xip, XFS_ILOCK_EXCL);

	/*
	 * If we're writing the file then make sure to clear the
	 * setuid and setgid bits if the process is not being run
	 * by root.  This keeps people from modifying setuid and
	 * setgid binaries.
	 */

	if (((xip->i_d.di_mode & S_ISUID) ||
	    ((xip->i_d.di_mode & (S_ISGID | S_IXGRP)) ==
		(S_ISGID | S_IXGRP))) &&
	     !capable(CAP_FSETID)) {
		error = xfs_write_clear_setuid(xip);
		if (likely(!error))
			error = -remove_suid(file->f_dentry);
		if (unlikely(error)) {
			xfs_iunlock(xip, iolock);
			goto out_unlock_isem;
		}
	}

retry:
	/* We can write back this queue in page reclaim */
	current->backing_dev_info = mapping->backing_dev_info;

	if ((ioflags & IO_ISDIRECT)) {
		if (need_flush) {
			xfs_inval_cached_trace(io, pos, -1,
					ctooff(offtoct(pos)), -1);
			VOP_FLUSHINVAL_PAGES(vp, ctooff(offtoct(pos)),
					-1, FI_REMAPF_LOCKED);
		}

		if (need_isem) {
			/* demote the lock now the cached pages are gone */
			XFS_ILOCK_DEMOTE(mp, io, XFS_IOLOCK_EXCL);
			up(&inode->i_sem);

			iolock = XFS_IOLOCK_SHARED;
			locktype = VRWLOCK_WRITE_DIRECT;
			need_isem = 0;
		}

 		xfs_rw_enter_trace(XFS_DIOWR_ENTER, io, (void *)iovp, segs,
				*offset, ioflags);
		ret = generic_file_direct_write(iocb, iovp,
				&segs, pos, offset, count, ocount);

		/*
		 * direct-io write to a hole: fall through to buffered I/O
		 * for completing the rest of the request.
		 */
		if (ret >= 0 && ret != count) {
			XFS_STATS_ADD(xs_write_bytes, ret);

			pos += ret;
			count -= ret;

			need_isem = 1;
			ioflags &= ~IO_ISDIRECT;
			xfs_iunlock(xip, iolock);
			goto relock;
		}
	} else {
		xfs_rw_enter_trace(XFS_WRITE_ENTER, io, (void *)iovp, segs,
				*offset, ioflags);
		ret = generic_file_buffered_write(iocb, iovp, segs,
				pos, offset, count, ret);
	}

	current->backing_dev_info = NULL;

	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
		ret = wait_on_sync_kiocb(iocb);

	if ((ret == -ENOSPC) &&
	    DM_EVENT_ENABLED(vp->v_vfsp, xip, DM_EVENT_NOSPACE) &&
	    !(ioflags & IO_INVIS)) {

		xfs_rwunlock(bdp, locktype);
		error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, vp,
				DM_RIGHT_NULL, vp, DM_RIGHT_NULL, NULL, NULL,
				0, 0, 0); /* Delay flag intentionally  unused */
		if (error)
			goto out_unlock_isem;
		xfs_rwlock(bdp, locktype);
		pos = xip->i_d.di_size;
		ret = 0;
		goto retry;
	}

	if (*offset > xip->i_d.di_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		if (*offset > xip->i_d.di_size) {
			xip->i_d.di_size = *offset;
			i_size_write(inode, *offset);
			xip->i_update_core = 1;
			xip->i_update_size = 1;
		}
		xfs_iunlock(xip, XFS_ILOCK_EXCL);
	}

	error = -ret;
	if (ret <= 0)
		goto out_unlock_internal;

	XFS_STATS_ADD(xs_write_bytes, ret);

	/* Handle various SYNC-type writes */
	if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) {
		/*
		 * If we're treating this as O_DSYNC and we have not updated the
		 * size, force the log.
		 */
		if (!(mp->m_flags & XFS_MOUNT_OSYNCISOSYNC) &&
		    !(xip->i_update_size)) {
			xfs_inode_log_item_t	*iip = xip->i_itemp;

			/*
			 * If an allocation transaction occurred
			 * without extending the size, then we have to force
			 * the log up the proper point to ensure that the
			 * allocation is permanent.  We can't count on
			 * the fact that buffered writes lock out direct I/O
			 * writes - the direct I/O write could have extended
			 * the size nontransactionally, then finished before
			 * we started.  xfs_write_file will think that the file
			 * didn't grow but the update isn't safe unless the
			 * size change is logged.
			 *
			 * Force the log if we've committed a transaction
			 * against the inode or if someone else has and
			 * the commit record hasn't gone to disk (e.g.
			 * the inode is pinned).  This guarantees that
			 * all changes affecting the inode are permanent
			 * when we return.
			 */
			if (iip && iip->ili_last_lsn) {
				xfs_log_force(mp, iip->ili_last_lsn,
						XFS_LOG_FORCE | XFS_LOG_SYNC);
			} else if (xfs_ipincount(xip) > 0) {
				xfs_log_force(mp, (xfs_lsn_t)0,
						XFS_LOG_FORCE | XFS_LOG_SYNC);
			}

		} else {
			xfs_trans_t	*tp;

			/*
			 * O_SYNC or O_DSYNC _with_ a size update are handled
			 * the same way.
			 *
			 * If the write was synchronous then we need to make
			 * sure that the inode modification time is permanent.
			 * We'll have updated the timestamp above, so here
			 * we use a synchronous transaction to log the inode.
			 * It's not fast, but it's necessary.
			 *
			 * If this a dsync write and the size got changed
			 * non-transactionally, then we need to ensure that
			 * the size change gets logged in a synchronous
			 * transaction.
			 */

			tp = xfs_trans_alloc(mp, XFS_TRANS_WRITE_SYNC);
			if ((error = xfs_trans_reserve(tp, 0,
						      XFS_SWRITE_LOG_RES(mp),
						      0, 0, 0))) {
				/* Transaction reserve failed */
				xfs_trans_cancel(tp, 0);
			} else {
				/* Transaction reserve successful */
				xfs_ilock(xip, XFS_ILOCK_EXCL);
				xfs_trans_ijoin(tp, xip, XFS_ILOCK_EXCL);
				xfs_trans_ihold(tp, xip);
				xfs_trans_log_inode(tp, xip, XFS_ILOG_CORE);
				xfs_trans_set_sync(tp);
				error = xfs_trans_commit(tp, 0, NULL);
				xfs_iunlock(xip, XFS_ILOCK_EXCL);
			}
			if (error)
				goto out_unlock_internal;
		}
	
		xfs_rwunlock(bdp, locktype);
		if (need_isem)
			up(&inode->i_sem);

		error = sync_page_range(inode, mapping, pos, ret);
		if (!error)
			error = ret;
		return error;
	}

 out_unlock_internal:
	xfs_rwunlock(bdp, locktype);
 out_unlock_isem:
	if (need_isem)
		up(&inode->i_sem);
	return -error;
}

/*
 * All xfs metadata buffers except log state machine buffers
 * get this attached as their b_bdstrat callback function.
 * This is so that we can catch a buffer
 * after prematurely unpinning it to forcibly shutdown the filesystem.
 */
int
xfs_bdstrat_cb(struct xfs_buf *bp)
{
	xfs_mount_t	*mp;

	mp = XFS_BUF_FSPRIVATE3(bp, xfs_mount_t *);
	if (!XFS_FORCED_SHUTDOWN(mp)) {
		pagebuf_iorequest(bp);
		return 0;
	} else {
		xfs_buftrace("XFS__BDSTRAT IOERROR", bp);
		/*
		 * Metadata write that didn't get logged but
		 * written delayed anyway. These aren't associated
		 * with a transaction, and can be ignored.
		 */
		if (XFS_BUF_IODONE_FUNC(bp) == NULL &&
		    (XFS_BUF_ISREAD(bp)) == 0)
			return (xfs_bioerror_relse(bp));
		else
			return (xfs_bioerror(bp));
	}
}


int
xfs_bmap(bhv_desc_t	*bdp,
	xfs_off_t	offset,
	ssize_t		count,
	int		flags,
	xfs_iomap_t	*iomapp,
	int		*niomaps)
{
	xfs_inode_t	*ip = XFS_BHVTOI(bdp);
	xfs_iocore_t	*io = &ip->i_iocore;

	ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
	ASSERT(((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != 0) ==
	       ((ip->i_iocore.io_flags & XFS_IOCORE_RT) != 0));

	return xfs_iomap(io, offset, count, flags, iomapp, niomaps);
}

/*
 * Wrapper around bdstrat so that we can stop data
 * from going to disk in case we are shutting down the filesystem.
 * Typically user data goes thru this path; one of the exceptions
 * is the superblock.
 */
int
xfsbdstrat(
	struct xfs_mount	*mp,
	struct xfs_buf		*bp)
{
	ASSERT(mp);
	if (!XFS_FORCED_SHUTDOWN(mp)) {
		/* Grio redirection would go here
		 * if (XFS_BUF_IS_GRIO(bp)) {
		 */

		pagebuf_iorequest(bp);
		return 0;
	}

	xfs_buftrace("XFSBDSTRAT IOERROR", bp);
	return (xfs_bioerror_relse(bp));
}

/*
 * If the underlying (data/log/rt) device is readonly, there are some
 * operations that cannot proceed.
 */
int
xfs_dev_is_read_only(
	xfs_mount_t		*mp,
	char			*message)
{
	if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
	    xfs_readonly_buftarg(mp->m_logdev_targp) ||
	    (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
		cmn_err(CE_NOTE,
			"XFS: %s required on read-only device.", message);
		cmn_err(CE_NOTE,
			"XFS: write access unavailable, cannot proceed.");
		return EROFS;
	}
	return 0;
}