aboutsummaryrefslogtreecommitdiff
path: root/fs/xfs/xfs_rw.c
blob: 4c199d18f8505a774da899dcbb097599ec0e08f7 (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
/*
 * Copyright (c) 2000-2006 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_fs.h"
#include "xfs_types.h"
#include "xfs_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_bmap_btree.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_dir2_sf.h"
#include "xfs_attr_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_inode_item.h"
#include "xfs_itable.h"
#include "xfs_btree.h"
#include "xfs_alloc.h"
#include "xfs_ialloc.h"
#include "xfs_attr.h"
#include "xfs_bmap.h"
#include "xfs_error.h"
#include "xfs_buf_item.h"
#include "xfs_rw.h"

/*
 * This is a subroutine for xfs_write() and other writers (xfs_ioctl)
 * which clears the setuid and setgid bits when a file is written.
 */
int
xfs_write_clear_setuid(
	xfs_inode_t	*ip)
{
	xfs_mount_t	*mp;
	xfs_trans_t	*tp;
	int		error;

	mp = ip->i_mount;
	tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
	if ((error = xfs_trans_reserve(tp, 0,
				      XFS_WRITEID_LOG_RES(mp),
				      0, 0, 0))) {
		xfs_trans_cancel(tp, 0);
		return error;
	}
	xfs_ilock(ip, XFS_ILOCK_EXCL);
	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
	xfs_trans_ihold(tp, ip);
	ip->i_d.di_mode &= ~S_ISUID;

	/*
	 * Note that we don't have to worry about mandatory
	 * file locking being disabled here because we only
	 * clear the S_ISGID bit if the Group execute bit is
	 * on, but if it was on then mandatory locking wouldn't
	 * have been enabled.
	 */
	if (ip->i_d.di_mode & S_IXGRP) {
		ip->i_d.di_mode &= ~S_ISGID;
	}
	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
	xfs_trans_set_sync(tp);
	error = xfs_trans_commit(tp, 0);
	xfs_iunlock(ip, XFS_ILOCK_EXCL);
	return 0;
}

/*
 * Force a shutdown of the filesystem instantly while keeping
 * the filesystem consistent. We don't do an unmount here; just shutdown
 * the shop, make sure that absolutely nothing persistent happens to
 * this filesystem after this point.
 */
void
xfs_do_force_shutdown(
	xfs_mount_t	*mp,
	int		flags,
	char		*fname,
	int		lnnum)
{
	int		logerror;

	logerror = flags & SHUTDOWN_LOG_IO_ERROR;

	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
		cmn_err(CE_NOTE, "xfs_force_shutdown(%s,0x%x) called from "
				 "line %d of file %s.  Return address = 0x%p",
			mp->m_fsname, flags, lnnum, fname, __return_address);
	}
	/*
	 * No need to duplicate efforts.
	 */
	if (XFS_FORCED_SHUTDOWN(mp) && !logerror)
		return;

	/*
	 * This flags XFS_MOUNT_FS_SHUTDOWN, makes sure that we don't
	 * queue up anybody new on the log reservations, and wakes up
	 * everybody who's sleeping on log reservations to tell them
	 * the bad news.
	 */
	if (xfs_log_force_umount(mp, logerror))
		return;

	if (flags & SHUTDOWN_CORRUPT_INCORE) {
		xfs_cmn_err(XFS_PTAG_SHUTDOWN_CORRUPT, CE_ALERT, mp,
    "Corruption of in-memory data detected.  Shutting down filesystem: %s",
			mp->m_fsname);
		if (XFS_ERRLEVEL_HIGH <= xfs_error_level) {
			xfs_stack_trace();
		}
	} else if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
		if (logerror) {
			xfs_cmn_err(XFS_PTAG_SHUTDOWN_LOGERROR, CE_ALERT, mp,
		"Log I/O Error Detected.  Shutting down filesystem: %s",
				mp->m_fsname);
		} else if (flags & SHUTDOWN_DEVICE_REQ) {
			xfs_cmn_err(XFS_PTAG_SHUTDOWN_IOERROR, CE_ALERT, mp,
		"All device paths lost.  Shutting down filesystem: %s",
				mp->m_fsname);
		} else if (!(flags & SHUTDOWN_REMOTE_REQ)) {
			xfs_cmn_err(XFS_PTAG_SHUTDOWN_IOERROR, CE_ALERT, mp,
		"I/O Error Detected.  Shutting down filesystem: %s",
				mp->m_fsname);
		}
	}
	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
		cmn_err(CE_ALERT, "Please umount the filesystem, "
				  "and rectify the problem(s)");
	}
}


/*
 * Called when we want to stop a buffer from getting written or read.
 * We attach the EIO error, muck with its flags, and call biodone
 * so that the proper iodone callbacks get called.
 */
int
xfs_bioerror(
	xfs_buf_t *bp)
{

#ifdef XFSERRORDEBUG
	ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone);
#endif

	/*
	 * No need to wait until the buffer is unpinned.
	 * We aren't flushing it.
	 */
	xfs_buftrace("XFS IOERROR", bp);
	XFS_BUF_ERROR(bp, EIO);
	/*
	 * We're calling biodone, so delete B_DONE flag. Either way
	 * we have to call the iodone callback, and calling biodone
	 * probably is the best way since it takes care of
	 * GRIO as well.
	 */
	XFS_BUF_UNREAD(bp);
	XFS_BUF_UNDELAYWRITE(bp);
	XFS_BUF_UNDONE(bp);
	XFS_BUF_STALE(bp);

	XFS_BUF_CLR_BDSTRAT_FUNC(bp);
	xfs_biodone(bp);

	return (EIO);
}

/*
 * Same as xfs_bioerror, except that we are releasing the buffer
 * here ourselves, and avoiding the biodone call.
 * This is meant for userdata errors; metadata bufs come with
 * iodone functions attached, so that we can track down errors.
 */
int
xfs_bioerror_relse(
	xfs_buf_t *bp)
{
	int64_t fl;

	ASSERT(XFS_BUF_IODONE_FUNC(bp) != xfs_buf_iodone_callbacks);
	ASSERT(XFS_BUF_IODONE_FUNC(bp) != xlog_iodone);

	xfs_buftrace("XFS IOERRELSE", bp);
	fl = XFS_BUF_BFLAGS(bp);
	/*
	 * No need to wait until the buffer is unpinned.
	 * We aren't flushing it.
	 *
	 * chunkhold expects B_DONE to be set, whether
	 * we actually finish the I/O or not. We don't want to
	 * change that interface.
	 */
	XFS_BUF_UNREAD(bp);
	XFS_BUF_UNDELAYWRITE(bp);
	XFS_BUF_DONE(bp);
	XFS_BUF_STALE(bp);
	XFS_BUF_CLR_IODONE_FUNC(bp);
	XFS_BUF_CLR_BDSTRAT_FUNC(bp);
	if (!(fl & XFS_B_ASYNC)) {
		/*
		 * Mark b_error and B_ERROR _both_.
		 * Lot's of chunkcache code assumes that.
		 * There's no reason to mark error for
		 * ASYNC buffers.
		 */
		XFS_BUF_ERROR(bp, EIO);
		XFS_BUF_FINISH_IOWAIT(bp);
	} else {
		xfs_buf_relse(bp);
	}
	return (EIO);
}

/*
 * Prints out an ALERT message about I/O error.
 */
void
xfs_ioerror_alert(
	char			*func,
	struct xfs_mount	*mp,
	xfs_buf_t		*bp,
	xfs_daddr_t		blkno)
{
	cmn_err(CE_ALERT,
 "I/O error in filesystem (\"%s\") meta-data dev %s block 0x%llx"
 "       (\"%s\") error %d buf count %zd",
		(!mp || !mp->m_fsname) ? "(fs name not set)" : mp->m_fsname,
		XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
		(__uint64_t)blkno, func,
		XFS_BUF_GETERROR(bp), XFS_BUF_COUNT(bp));
}

/*
 * This isn't an absolute requirement, but it is
 * just a good idea to call xfs_read_buf instead of
 * directly doing a read_buf call. For one, we shouldn't
 * be doing this disk read if we are in SHUTDOWN state anyway,
 * so this stops that from happening. Secondly, this does all
 * the error checking stuff and the brelse if appropriate for
 * the caller, so the code can be a little leaner.
 */

int
xfs_read_buf(
	struct xfs_mount *mp,
	xfs_buftarg_t	 *target,
	xfs_daddr_t	 blkno,
	int              len,
	uint             flags,
	xfs_buf_t	 **bpp)
{
	xfs_buf_t	 *bp;
	int		 error;

	if (!flags)
		flags = XBF_LOCK | XBF_MAPPED;

	bp = xfs_buf_read(target, blkno, len, flags);
	if (!bp)
		return XFS_ERROR(EIO);
	error = XFS_BUF_GETERROR(bp);
	if (bp && !error && !XFS_FORCED_SHUTDOWN(mp)) {
		*bpp = bp;
	} else {
		*bpp = NULL;
		if (error) {
			xfs_ioerror_alert("xfs_read_buf", mp, bp, XFS_BUF_ADDR(bp));
		} else {
			error = XFS_ERROR(EIO);
		}
		if (bp) {
			XFS_BUF_UNDONE(bp);
			XFS_BUF_UNDELAYWRITE(bp);
			XFS_BUF_STALE(bp);
			/*
			 * brelse clears B_ERROR and b_error
			 */
			xfs_buf_relse(bp);
		}
	}
	return (error);
}

/*
 * Wrapper around bwrite() so that we can trap
 * write errors, and act accordingly.
 */
int
xfs_bwrite(
	struct xfs_mount *mp,
	struct xfs_buf	 *bp)
{
	int	error;

	/*
	 * XXXsup how does this work for quotas.
	 */
	XFS_BUF_SET_BDSTRAT_FUNC(bp, xfs_bdstrat_cb);
	bp->b_mount = mp;
	XFS_BUF_WRITE(bp);

	if ((error = XFS_bwrite(bp))) {
		ASSERT(mp);
		/*
		 * Cannot put a buftrace here since if the buffer is not
		 * B_HOLD then we will brelse() the buffer before returning
		 * from bwrite and we could be tracing a buffer that has
		 * been reused.
		 */
		xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
	}
	return (error);
}

/*
 * helper function to extract extent size hint from inode
 */
xfs_extlen_t
xfs_get_extsz_hint(
	struct xfs_inode	*ip)
{
	xfs_extlen_t		extsz;

	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
				? ip->i_d.di_extsize
				: ip->i_mount->m_sb.sb_rextsize;
		ASSERT(extsz);
	} else {
		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
				? ip->i_d.di_extsize : 0;
	}

	return extsz;
}