xfs_fsops.c 18.6 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
2 3
 * Copyright (c) 2000-2005 Silicon Graphics, Inc.
 * All Rights Reserved.
L
Linus Torvalds 已提交
4
 *
5 6
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
L
Linus Torvalds 已提交
7 8
 * published by the Free Software Foundation.
 *
9 10 11 12
 * 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.
L
Linus Torvalds 已提交
13
 *
14 15 16
 * 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
L
Linus Torvalds 已提交
17 18
 */
#include "xfs.h"
19
#include "xfs_fs.h"
L
Linus Torvalds 已提交
20
#include "xfs_types.h"
21
#include "xfs_bit.h"
L
Linus Torvalds 已提交
22 23 24 25
#include "xfs_inum.h"
#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
26
#include "xfs_ag.h"
L
Linus Torvalds 已提交
27 28
#include "xfs_mount.h"
#include "xfs_bmap_btree.h"
29
#include "xfs_alloc_btree.h"
L
Linus Torvalds 已提交
30
#include "xfs_ialloc_btree.h"
31 32 33
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_inode_item.h"
L
Linus Torvalds 已提交
34 35 36 37 38 39 40 41
#include "xfs_btree.h"
#include "xfs_error.h"
#include "xfs_alloc.h"
#include "xfs_ialloc.h"
#include "xfs_fsops.h"
#include "xfs_itable.h"
#include "xfs_trans_space.h"
#include "xfs_rtalloc.h"
42
#include "xfs_rw.h"
43
#include "xfs_filestream.h"
C
Christoph Hellwig 已提交
44
#include "xfs_trace.h"
L
Linus Torvalds 已提交
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

/*
 * File system operations
 */

int
xfs_fs_geometry(
	xfs_mount_t		*mp,
	xfs_fsop_geom_t		*geo,
	int			new_version)
{
	geo->blocksize = mp->m_sb.sb_blocksize;
	geo->rtextsize = mp->m_sb.sb_rextsize;
	geo->agblocks = mp->m_sb.sb_agblocks;
	geo->agcount = mp->m_sb.sb_agcount;
	geo->logblocks = mp->m_sb.sb_logblocks;
	geo->sectsize = mp->m_sb.sb_sectsize;
	geo->inodesize = mp->m_sb.sb_inodesize;
	geo->imaxpct = mp->m_sb.sb_imax_pct;
	geo->datablocks = mp->m_sb.sb_dblocks;
	geo->rtblocks = mp->m_sb.sb_rblocks;
	geo->rtextents = mp->m_sb.sb_rextents;
	geo->logstart = mp->m_sb.sb_logstart;
	ASSERT(sizeof(geo->uuid)==sizeof(mp->m_sb.sb_uuid));
	memcpy(geo->uuid, &mp->m_sb.sb_uuid, sizeof(mp->m_sb.sb_uuid));
	if (new_version >= 2) {
		geo->sunit = mp->m_sb.sb_unit;
		geo->swidth = mp->m_sb.sb_width;
	}
	if (new_version >= 3) {
		geo->version = XFS_FSOP_GEOM_VERSION;
		geo->flags =
77
			(xfs_sb_version_hasattr(&mp->m_sb) ?
L
Linus Torvalds 已提交
78
				XFS_FSOP_GEOM_FLAGS_ATTR : 0) |
79
			(xfs_sb_version_hasnlink(&mp->m_sb) ?
L
Linus Torvalds 已提交
80
				XFS_FSOP_GEOM_FLAGS_NLINK : 0) |
81
			(xfs_sb_version_hasquota(&mp->m_sb) ?
L
Linus Torvalds 已提交
82
				XFS_FSOP_GEOM_FLAGS_QUOTA : 0) |
83
			(xfs_sb_version_hasalign(&mp->m_sb) ?
L
Linus Torvalds 已提交
84
				XFS_FSOP_GEOM_FLAGS_IALIGN : 0) |
85
			(xfs_sb_version_hasdalign(&mp->m_sb) ?
L
Linus Torvalds 已提交
86
				XFS_FSOP_GEOM_FLAGS_DALIGN : 0) |
87
			(xfs_sb_version_hasshared(&mp->m_sb) ?
L
Linus Torvalds 已提交
88
				XFS_FSOP_GEOM_FLAGS_SHARED : 0) |
89
			(xfs_sb_version_hasextflgbit(&mp->m_sb) ?
L
Linus Torvalds 已提交
90
				XFS_FSOP_GEOM_FLAGS_EXTFLG : 0) |
91
			(xfs_sb_version_hasdirv2(&mp->m_sb) ?
L
Linus Torvalds 已提交
92
				XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) |
93
			(xfs_sb_version_hassector(&mp->m_sb) ?
94
				XFS_FSOP_GEOM_FLAGS_SECTOR : 0) |
95 96
			(xfs_sb_version_hasasciici(&mp->m_sb) ?
				XFS_FSOP_GEOM_FLAGS_DIRV2CI : 0) |
D
David Chinner 已提交
97 98
			(xfs_sb_version_haslazysbcount(&mp->m_sb) ?
				XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) |
99
			(xfs_sb_version_hasattr2(&mp->m_sb) ?
100
				XFS_FSOP_GEOM_FLAGS_ATTR2 : 0);
101
		geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ?
L
Linus Torvalds 已提交
102 103 104 105 106 107
				mp->m_sb.sb_logsectsize : BBSIZE;
		geo->rtsectsize = mp->m_sb.sb_blocksize;
		geo->dirblocksize = mp->m_dirblksize;
	}
	if (new_version >= 4) {
		geo->flags |=
108
			(xfs_sb_version_haslogv2(&mp->m_sb) ?
L
Linus Torvalds 已提交
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
				XFS_FSOP_GEOM_FLAGS_LOGV2 : 0);
		geo->logsunit = mp->m_sb.sb_logsunit;
	}
	return 0;
}

static int
xfs_growfs_data_private(
	xfs_mount_t		*mp,		/* mount point for filesystem */
	xfs_growfs_data_t	*in)		/* growfs data input struct */
{
	xfs_agf_t		*agf;
	xfs_agi_t		*agi;
	xfs_agnumber_t		agno;
	xfs_extlen_t		agsize;
	xfs_extlen_t		tmpsize;
	xfs_alloc_rec_t		*arec;
126
	struct xfs_btree_block	*block;
L
Linus Torvalds 已提交
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143
	xfs_buf_t		*bp;
	int			bucket;
	int			dpct;
	int			error;
	xfs_agnumber_t		nagcount;
	xfs_agnumber_t		nagimax = 0;
	xfs_rfsblock_t		nb, nb_mod;
	xfs_rfsblock_t		new;
	xfs_rfsblock_t		nfree;
	xfs_agnumber_t		oagcount;
	int			pct;
	xfs_trans_t		*tp;

	nb = in->newblocks;
	pct = in->imaxpct;
	if (nb < mp->m_sb.sb_dblocks || pct < 0 || pct > 100)
		return XFS_ERROR(EINVAL);
144 145
	if ((error = xfs_sb_validate_fsb_count(&mp->m_sb, nb)))
		return error;
L
Linus Torvalds 已提交
146 147 148 149 150 151 152 153 154 155 156 157 158 159
	dpct = pct - mp->m_sb.sb_imax_pct;
	error = xfs_read_buf(mp, mp->m_ddev_targp,
			XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
			XFS_FSS_TO_BB(mp, 1), 0, &bp);
	if (error)
		return error;
	ASSERT(bp);
	xfs_buf_relse(bp);

	new = nb;	/* use new as a temporary here */
	nb_mod = do_div(new, mp->m_sb.sb_agblocks);
	nagcount = new + (nb_mod != 0);
	if (nb_mod && nb_mod < XFS_MIN_AG_BLOCKS) {
		nagcount--;
160
		nb = (xfs_rfsblock_t)nagcount * mp->m_sb.sb_agblocks;
L
Linus Torvalds 已提交
161 162 163 164 165
		if (nb < mp->m_sb.sb_dblocks)
			return XFS_ERROR(EINVAL);
	}
	new = nb - mp->m_sb.sb_dblocks;
	oagcount = mp->m_sb.sb_agcount;
166

167 168 169 170 171
	/* allocate the new per-ag structures */
	if (nagcount > oagcount) {
		error = xfs_initialize_perag(mp, nagcount, &nagimax);
		if (error)
			return error;
L
Linus Torvalds 已提交
172
	}
173

L
Linus Torvalds 已提交
174
	tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS);
175
	tp->t_flags |= XFS_TRANS_RESERVE;
L
Linus Torvalds 已提交
176 177 178 179 180 181
	if ((error = xfs_trans_reserve(tp, XFS_GROWFS_SPACE_RES(mp),
			XFS_GROWDATA_LOG_RES(mp), 0, 0, 0))) {
		xfs_trans_cancel(tp, 0);
		return error;
	}

182 183 184 185 186
	/*
	 * Write new AG headers to disk. Non-transactional, but written
	 * synchronously so they are completed prior to the growfs transaction
	 * being logged.
	 */
L
Linus Torvalds 已提交
187 188 189 190 191 192
	nfree = 0;
	for (agno = nagcount - 1; agno >= oagcount; agno--, new -= agsize) {
		/*
		 * AG freelist header block
		 */
		bp = xfs_buf_get(mp->m_ddev_targp,
193 194
				 XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
L
Linus Torvalds 已提交
195 196
		agf = XFS_BUF_TO_AGF(bp);
		memset(agf, 0, mp->m_sb.sb_sectsize);
197 198 199
		agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC);
		agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION);
		agf->agf_seqno = cpu_to_be32(agno);
L
Linus Torvalds 已提交
200 201 202 203 204 205
		if (agno == nagcount - 1)
			agsize =
				nb -
				(agno * (xfs_rfsblock_t)mp->m_sb.sb_agblocks);
		else
			agsize = mp->m_sb.sb_agblocks;
206 207 208 209 210
		agf->agf_length = cpu_to_be32(agsize);
		agf->agf_roots[XFS_BTNUM_BNOi] = cpu_to_be32(XFS_BNO_BLOCK(mp));
		agf->agf_roots[XFS_BTNUM_CNTi] = cpu_to_be32(XFS_CNT_BLOCK(mp));
		agf->agf_levels[XFS_BTNUM_BNOi] = cpu_to_be32(1);
		agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1);
L
Linus Torvalds 已提交
211
		agf->agf_flfirst = 0;
212
		agf->agf_fllast = cpu_to_be32(XFS_AGFL_SIZE(mp) - 1);
L
Linus Torvalds 已提交
213 214
		agf->agf_flcount = 0;
		tmpsize = agsize - XFS_PREALLOC_BLOCKS(mp);
215 216
		agf->agf_freeblks = cpu_to_be32(tmpsize);
		agf->agf_longest = cpu_to_be32(tmpsize);
L
Linus Torvalds 已提交
217 218 219 220 221 222 223 224
		error = xfs_bwrite(mp, bp);
		if (error) {
			goto error0;
		}
		/*
		 * AG inode header block
		 */
		bp = xfs_buf_get(mp->m_ddev_targp,
225 226
				 XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
L
Linus Torvalds 已提交
227 228
		agi = XFS_BUF_TO_AGI(bp);
		memset(agi, 0, mp->m_sb.sb_sectsize);
229 230 231 232
		agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC);
		agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION);
		agi->agi_seqno = cpu_to_be32(agno);
		agi->agi_length = cpu_to_be32(agsize);
L
Linus Torvalds 已提交
233
		agi->agi_count = 0;
234 235
		agi->agi_root = cpu_to_be32(XFS_IBT_BLOCK(mp));
		agi->agi_level = cpu_to_be32(1);
L
Linus Torvalds 已提交
236
		agi->agi_freecount = 0;
237 238
		agi->agi_newino = cpu_to_be32(NULLAGINO);
		agi->agi_dirino = cpu_to_be32(NULLAGINO);
L
Linus Torvalds 已提交
239
		for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++)
240
			agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO);
L
Linus Torvalds 已提交
241 242 243 244 245 246 247 248
		error = xfs_bwrite(mp, bp);
		if (error) {
			goto error0;
		}
		/*
		 * BNO btree root block
		 */
		bp = xfs_buf_get(mp->m_ddev_targp,
249 250 251
				 XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
				 BTOBB(mp->m_sb.sb_blocksize),
				 XBF_LOCK | XBF_MAPPED);
252
		block = XFS_BUF_TO_BLOCK(bp);
L
Linus Torvalds 已提交
253
		memset(block, 0, mp->m_sb.sb_blocksize);
254
		block->bb_magic = cpu_to_be32(XFS_ABTB_MAGIC);
L
Linus Torvalds 已提交
255
		block->bb_level = 0;
256
		block->bb_numrecs = cpu_to_be16(1);
257 258
		block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
		block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
259
		arec = XFS_ALLOC_REC_ADDR(mp, block, 1);
260 261 262
		arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
		arec->ar_blockcount = cpu_to_be32(
			agsize - be32_to_cpu(arec->ar_startblock));
L
Linus Torvalds 已提交
263 264 265 266 267 268 269 270
		error = xfs_bwrite(mp, bp);
		if (error) {
			goto error0;
		}
		/*
		 * CNT btree root block
		 */
		bp = xfs_buf_get(mp->m_ddev_targp,
271 272 273
				 XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
				 BTOBB(mp->m_sb.sb_blocksize),
				 XBF_LOCK | XBF_MAPPED);
274
		block = XFS_BUF_TO_BLOCK(bp);
L
Linus Torvalds 已提交
275
		memset(block, 0, mp->m_sb.sb_blocksize);
276
		block->bb_magic = cpu_to_be32(XFS_ABTC_MAGIC);
L
Linus Torvalds 已提交
277
		block->bb_level = 0;
278
		block->bb_numrecs = cpu_to_be16(1);
279 280
		block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
		block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
281
		arec = XFS_ALLOC_REC_ADDR(mp, block, 1);
282 283 284 285
		arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
		arec->ar_blockcount = cpu_to_be32(
			agsize - be32_to_cpu(arec->ar_startblock));
		nfree += be32_to_cpu(arec->ar_blockcount);
L
Linus Torvalds 已提交
286 287 288 289 290 291 292 293
		error = xfs_bwrite(mp, bp);
		if (error) {
			goto error0;
		}
		/*
		 * INO btree root block
		 */
		bp = xfs_buf_get(mp->m_ddev_targp,
294 295 296
				 XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
				 BTOBB(mp->m_sb.sb_blocksize),
				 XBF_LOCK | XBF_MAPPED);
297
		block = XFS_BUF_TO_BLOCK(bp);
L
Linus Torvalds 已提交
298
		memset(block, 0, mp->m_sb.sb_blocksize);
299
		block->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
L
Linus Torvalds 已提交
300 301
		block->bb_level = 0;
		block->bb_numrecs = 0;
302 303
		block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
		block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
L
Linus Torvalds 已提交
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
		error = xfs_bwrite(mp, bp);
		if (error) {
			goto error0;
		}
	}
	xfs_trans_agblocks_delta(tp, nfree);
	/*
	 * There are new blocks in the old last a.g.
	 */
	if (new) {
		/*
		 * Change the agi length.
		 */
		error = xfs_ialloc_read_agi(mp, tp, agno, &bp);
		if (error) {
			goto error0;
		}
		ASSERT(bp);
		agi = XFS_BUF_TO_AGI(bp);
323
		be32_add_cpu(&agi->agi_length, new);
L
Linus Torvalds 已提交
324
		ASSERT(nagcount == oagcount ||
325
		       be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks);
L
Linus Torvalds 已提交
326 327 328 329 330 331 332 333 334 335
		xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH);
		/*
		 * Change agf length.
		 */
		error = xfs_alloc_read_agf(mp, tp, agno, 0, &bp);
		if (error) {
			goto error0;
		}
		ASSERT(bp);
		agf = XFS_BUF_TO_AGF(bp);
336
		be32_add_cpu(&agf->agf_length, new);
337 338
		ASSERT(be32_to_cpu(agf->agf_length) ==
		       be32_to_cpu(agi->agi_length));
C
Christoph Hellwig 已提交
339

340
		xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH);
L
Linus Torvalds 已提交
341 342 343 344
		/*
		 * Free the new space.
		 */
		error = xfs_free_extent(tp, XFS_AGB_TO_FSB(mp, agno,
345
			be32_to_cpu(agf->agf_length) - new), new);
L
Linus Torvalds 已提交
346 347 348 349
		if (error) {
			goto error0;
		}
	}
350 351 352 353 354 355

	/*
	 * Update changed superblock fields transactionally. These are not
	 * seen by the rest of the world until the transaction commit applies
	 * them atomically to the superblock.
	 */
L
Linus Torvalds 已提交
356 357 358 359 360 361 362 363 364
	if (nagcount > oagcount)
		xfs_trans_mod_sb(tp, XFS_TRANS_SB_AGCOUNT, nagcount - oagcount);
	if (nb > mp->m_sb.sb_dblocks)
		xfs_trans_mod_sb(tp, XFS_TRANS_SB_DBLOCKS,
				 nb - mp->m_sb.sb_dblocks);
	if (nfree)
		xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, nfree);
	if (dpct)
		xfs_trans_mod_sb(tp, XFS_TRANS_SB_IMAXPCT, dpct);
365
	error = xfs_trans_commit(tp, 0);
366
	if (error)
L
Linus Torvalds 已提交
367
		return error;
368

L
Linus Torvalds 已提交
369 370 371 372 373 374 375 376 377
	/* New allocation groups fully initialized, so update mount struct */
	if (nagimax)
		mp->m_maxagi = nagimax;
	if (mp->m_sb.sb_imax_pct) {
		__uint64_t icount = mp->m_sb.sb_dblocks * mp->m_sb.sb_imax_pct;
		do_div(icount, 100);
		mp->m_maxicount = icount << mp->m_sb.sb_inopblog;
	} else
		mp->m_maxicount = 0;
378 379

	/* update secondary superblocks. */
L
Linus Torvalds 已提交
380 381 382 383 384 385 386 387 388 389
	for (agno = 1; agno < nagcount; agno++) {
		error = xfs_read_buf(mp, mp->m_ddev_targp,
				  XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)),
				  XFS_FSS_TO_BB(mp, 1), 0, &bp);
		if (error) {
			xfs_fs_cmn_err(CE_WARN, mp,
			"error %d reading secondary superblock for ag %d",
				error, agno);
			break;
		}
390
		xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb, XFS_SB_ALL_BITS);
L
Linus Torvalds 已提交
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
		/*
		 * If we get an error writing out the alternate superblocks,
		 * just issue a warning and continue.  The real work is
		 * already done and committed.
		 */
		if (!(error = xfs_bwrite(mp, bp))) {
			continue;
		} else {
			xfs_fs_cmn_err(CE_WARN, mp,
		"write error %d updating secondary superblock for ag %d",
				error, agno);
			break; /* no point in continuing */
		}
	}
	return 0;

 error0:
	xfs_trans_cancel(tp, XFS_TRANS_ABORT);
	return error;
}

static int
xfs_growfs_log_private(
	xfs_mount_t		*mp,	/* mount point for filesystem */
	xfs_growfs_log_t	*in)	/* growfs log input struct */
{
	xfs_extlen_t		nb;

	nb = in->newblocks;
	if (nb < XFS_MIN_LOG_BLOCKS || nb < XFS_B_TO_FSB(mp, XFS_MIN_LOG_BYTES))
		return XFS_ERROR(EINVAL);
	if (nb == mp->m_sb.sb_logblocks &&
	    in->isint == (mp->m_sb.sb_logstart != 0))
		return XFS_ERROR(EINVAL);
	/*
	 * Moving the log is hard, need new interfaces to sync
	 * the log first, hold off all activity while moving it.
	 * Can have shorter or longer log in the same space,
	 * or transform internal to external log or vice versa.
	 */
	return XFS_ERROR(ENOSYS);
}

/*
 * protected versions of growfs function acquire and release locks on the mount
 * point - exported through ioctls: XFS_IOC_FSGROWFSDATA, XFS_IOC_FSGROWFSLOG,
 * XFS_IOC_FSGROWFSRT
 */


int
xfs_growfs_data(
	xfs_mount_t		*mp,
	xfs_growfs_data_t	*in)
{
	int error;
447 448 449

	if (!capable(CAP_SYS_ADMIN))
		return XFS_ERROR(EPERM);
450
	if (!mutex_trylock(&mp->m_growlock))
L
Linus Torvalds 已提交
451 452
		return XFS_ERROR(EWOULDBLOCK);
	error = xfs_growfs_data_private(mp, in);
453
	mutex_unlock(&mp->m_growlock);
L
Linus Torvalds 已提交
454 455 456 457 458 459 460 461 462
	return error;
}

int
xfs_growfs_log(
	xfs_mount_t		*mp,
	xfs_growfs_log_t	*in)
{
	int error;
463 464 465

	if (!capable(CAP_SYS_ADMIN))
		return XFS_ERROR(EPERM);
466
	if (!mutex_trylock(&mp->m_growlock))
L
Linus Torvalds 已提交
467 468
		return XFS_ERROR(EWOULDBLOCK);
	error = xfs_growfs_log_private(mp, in);
469
	mutex_unlock(&mp->m_growlock);
L
Linus Torvalds 已提交
470 471 472 473 474 475 476 477 478 479 480 481
	return error;
}

/*
 * exported through ioctl XFS_IOC_FSCOUNTS
 */

int
xfs_fs_counts(
	xfs_mount_t		*mp,
	xfs_fsop_counts_t	*cnt)
{
482
	xfs_icsb_sync_counters(mp, XFS_ICSB_LAZY_COUNT);
E
Eric Sandeen 已提交
483
	spin_lock(&mp->m_sb_lock);
484
	cnt->freedata = mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp);
L
Linus Torvalds 已提交
485 486 487
	cnt->freertx = mp->m_sb.sb_frextents;
	cnt->freeino = mp->m_sb.sb_ifree;
	cnt->allocino = mp->m_sb.sb_icount;
E
Eric Sandeen 已提交
488
	spin_unlock(&mp->m_sb_lock);
L
Linus Torvalds 已提交
489 490 491 492 493 494 495 496
	return 0;
}

/*
 * exported through ioctl XFS_IOC_SET_RESBLKS & XFS_IOC_GET_RESBLKS
 *
 * xfs_reserve_blocks is called to set m_resblks
 * in the in-core mount table. The number of unused reserved blocks
497
 * is kept in m_resblks_avail.
L
Linus Torvalds 已提交
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
 *
 * Reserve the requested number of blocks if available. Otherwise return
 * as many as possible to satisfy the request. The actual number
 * reserved are returned in outval
 *
 * A null inval pointer indicates that only the current reserved blocks
 * available  should  be returned no settings are changed.
 */

int
xfs_reserve_blocks(
	xfs_mount_t             *mp,
	__uint64_t              *inval,
	xfs_fsop_resblks_t      *outval)
{
513
	__int64_t		lcounter, delta, fdblks_delta;
L
Linus Torvalds 已提交
514 515 516 517
	__uint64_t		request;

	/* If inval is null, report current values and return */
	if (inval == (__uint64_t *)NULL) {
518 519
		if (!outval)
			return EINVAL;
L
Linus Torvalds 已提交
520 521
		outval->resblks = mp->m_resblks;
		outval->resblks_avail = mp->m_resblks_avail;
522
		return 0;
L
Linus Torvalds 已提交
523 524 525
	}

	request = *inval;
526 527 528 529 530 531

	/*
	 * With per-cpu counters, this becomes an interesting
	 * problem. we needto work out if we are freeing or allocation
	 * blocks first, then we can do the modification as necessary.
	 *
E
Eric Sandeen 已提交
532
	 * We do this under the m_sb_lock so that if we are near
533 534 535 536 537 538 539 540 541 542
	 * ENOSPC, we will hold out any changes while we work out
	 * what to do. This means that the amount of free space can
	 * change while we do this, so we need to retry if we end up
	 * trying to reserve more space than is available.
	 *
	 * We also use the xfs_mod_incore_sb() interface so that we
	 * don't have to care about whether per cpu counter are
	 * enabled, disabled or even compiled in....
	 */
retry:
E
Eric Sandeen 已提交
543
	spin_lock(&mp->m_sb_lock);
544
	xfs_icsb_sync_counters_locked(mp, 0);
L
Linus Torvalds 已提交
545 546 547 548 549

	/*
	 * If our previous reservation was larger than the current value,
	 * then move any unused blocks back to the free pool.
	 */
550
	fdblks_delta = 0;
L
Linus Torvalds 已提交
551 552 553
	if (mp->m_resblks > request) {
		lcounter = mp->m_resblks_avail - request;
		if (lcounter  > 0) {		/* release unused blocks */
554
			fdblks_delta = lcounter;
L
Linus Torvalds 已提交
555 556 557 558
			mp->m_resblks_avail -= lcounter;
		}
		mp->m_resblks = request;
	} else {
559 560 561
		__int64_t	free;

		free =  mp->m_sb.sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp);
562 563 564
		if (!free)
			goto out; /* ENOSPC and fdblks_delta = 0 */

L
Linus Torvalds 已提交
565
		delta = request - mp->m_resblks;
566
		lcounter = free - delta;
L
Linus Torvalds 已提交
567 568
		if (lcounter < 0) {
			/* We can't satisfy the request, just get what we can */
569 570
			mp->m_resblks += free;
			mp->m_resblks_avail += free;
571
			fdblks_delta = -free;
L
Linus Torvalds 已提交
572
		} else {
573
			fdblks_delta = -delta;
L
Linus Torvalds 已提交
574 575 576 577
			mp->m_resblks = request;
			mp->m_resblks_avail += delta;
		}
	}
578
out:
579 580 581 582
	if (outval) {
		outval->resblks = mp->m_resblks;
		outval->resblks_avail = mp->m_resblks_avail;
	}
E
Eric Sandeen 已提交
583
	spin_unlock(&mp->m_sb_lock);
584 585 586 587

	if (fdblks_delta) {
		/*
		 * If we are putting blocks back here, m_resblks_avail is
M
Malcolm Parsons 已提交
588
		 * already at its max so this will put it in the free pool.
589 590 591 592 593 594 595 596 597 598 599 600 601 602 603
		 *
		 * If we need space, we'll either succeed in getting it
		 * from the free block count or we'll get an enospc. If
		 * we get a ENOSPC, it means things changed while we were
		 * calculating fdblks_delta and so we should try again to
		 * see if there is anything left to reserve.
		 *
		 * Don't set the reserved flag here - we don't want to reserve
		 * the extra reserve blocks from the reserve.....
		 */
		int error;
		error = xfs_mod_incore_sb(mp, XFS_SBS_FDBLOCKS, fdblks_delta, 0);
		if (error == ENOSPC)
			goto retry;
	}
604
	return 0;
L
Linus Torvalds 已提交
605 606
}

607 608 609 610 611 612 613 614 615
/*
 * Dump a transaction into the log that contains no real change. This is needed
 * to be able to make the log dirty or stamp the current tail LSN into the log
 * during the covering operation.
 *
 * We cannot use an inode here for this - that will push dirty state back up
 * into the VFS and then periodic inode flushing will prevent log covering from
 * making progress. Hence we log a field in the superblock instead.
 */
616
int
617
xfs_fs_log_dummy(
618 619
	xfs_mount_t	*mp,
	int		flags)
620
{
621
	xfs_trans_t	*tp;
622
	int		error;
623

624
	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP);
625 626
	error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
					XFS_DEFAULT_LOG_COUNT);
627
	if (error) {
628
		xfs_trans_cancel(tp, 0);
629
		return error;
630 631
	}

632 633 634 635 636
	/* log the UUID because it is an unchanging field */
	xfs_mod_sb(tp, XFS_SB_UUID);
	if (flags & SYNC_WAIT)
		xfs_trans_set_sync(tp);
	return xfs_trans_commit(tp, 0);
637 638
}

L
Linus Torvalds 已提交
639 640 641 642 643 644 645
int
xfs_fs_goingdown(
	xfs_mount_t	*mp,
	__uint32_t	inflags)
{
	switch (inflags) {
	case XFS_FSOP_GOING_FLAGS_DEFAULT: {
C
Christoph Hellwig 已提交
646
		struct super_block *sb = freeze_bdev(mp->m_super->s_bdev);
L
Linus Torvalds 已提交
647

648
		if (sb && !IS_ERR(sb)) {
649
			xfs_force_shutdown(mp, SHUTDOWN_FORCE_UMOUNT);
L
Linus Torvalds 已提交
650 651
			thaw_bdev(sb->s_bdev, sb);
		}
652

L
Linus Torvalds 已提交
653 654 655
		break;
	}
	case XFS_FSOP_GOING_FLAGS_LOGFLUSH:
656
		xfs_force_shutdown(mp, SHUTDOWN_FORCE_UMOUNT);
L
Linus Torvalds 已提交
657 658
		break;
	case XFS_FSOP_GOING_FLAGS_NOLOGFLUSH:
659 660
		xfs_force_shutdown(mp,
				SHUTDOWN_FORCE_UMOUNT | SHUTDOWN_LOG_IO_ERROR);
L
Linus Torvalds 已提交
661 662 663 664 665 666 667
		break;
	default:
		return XFS_ERROR(EINVAL);
	}

	return 0;
}