xfs_attr.c 59.2 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

L
Linus Torvalds 已提交
19
#include "xfs.h"
20
#include "xfs_fs.h"
L
Linus Torvalds 已提交
21
#include "xfs_types.h"
22
#include "xfs_bit.h"
L
Linus Torvalds 已提交
23
#include "xfs_log.h"
24
#include "xfs_inum.h"
L
Linus Torvalds 已提交
25 26 27 28 29 30
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
31
#include "xfs_da_btree.h"
L
Linus Torvalds 已提交
32
#include "xfs_bmap_btree.h"
33
#include "xfs_alloc_btree.h"
L
Linus Torvalds 已提交
34 35
#include "xfs_ialloc_btree.h"
#include "xfs_dir2_sf.h"
36
#include "xfs_attr_sf.h"
L
Linus Torvalds 已提交
37 38
#include "xfs_dinode.h"
#include "xfs_inode.h"
39 40 41
#include "xfs_alloc.h"
#include "xfs_btree.h"
#include "xfs_inode_item.h"
L
Linus Torvalds 已提交
42 43 44 45 46 47 48
#include "xfs_bmap.h"
#include "xfs_attr.h"
#include "xfs_attr_leaf.h"
#include "xfs_error.h"
#include "xfs_quota.h"
#include "xfs_trans_space.h"
#include "xfs_acl.h"
49
#include "xfs_rw.h"
50
#include "xfs_vnodeops.h"
L
Linus Torvalds 已提交
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69

/*
 * xfs_attr.c
 *
 * Provide the external interfaces to manage attribute lists.
 */

/*========================================================================
 * Function prototypes for the kernel.
 *========================================================================*/

/*
 * Internal routines when attribute list fits inside the inode.
 */
STATIC int xfs_attr_shortform_addname(xfs_da_args_t *args);

/*
 * Internal routines when attribute list is one block.
 */
70
STATIC int xfs_attr_leaf_get(xfs_da_args_t *args);
L
Linus Torvalds 已提交
71 72 73 74 75 76 77
STATIC int xfs_attr_leaf_addname(xfs_da_args_t *args);
STATIC int xfs_attr_leaf_removename(xfs_da_args_t *args);
STATIC int xfs_attr_leaf_list(xfs_attr_list_context_t *context);

/*
 * Internal routines when attribute list is more than one block.
 */
78
STATIC int xfs_attr_node_get(xfs_da_args_t *args);
L
Linus Torvalds 已提交
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
STATIC int xfs_attr_node_addname(xfs_da_args_t *args);
STATIC int xfs_attr_node_removename(xfs_da_args_t *args);
STATIC int xfs_attr_node_list(xfs_attr_list_context_t *context);
STATIC int xfs_attr_fillstate(xfs_da_state_t *state);
STATIC int xfs_attr_refillstate(xfs_da_state_t *state);

/*
 * Routines to manipulate out-of-line attribute values.
 */
STATIC int xfs_attr_rmtval_set(xfs_da_args_t *args);
STATIC int xfs_attr_rmtval_remove(xfs_da_args_t *args);

#define ATTR_RMTVALUE_MAPSIZE	1	/* # of map entries at once */

#if defined(XFS_ATTR_TRACE)
ktrace_t *xfs_attr_trace_buf;
#endif

97 98 99 100 101 102 103 104 105 106 107 108 109 110
STATIC int
xfs_attr_name_to_xname(
	struct xfs_name	*xname,
	const char	*aname)
{
	if (!aname)
		return EINVAL;
	xname->name = aname;
	xname->len = strlen(aname);
	if (xname->len >= MAXNAMELEN)
		return EFAULT;		/* match IRIX behaviour */

	return 0;
}
L
Linus Torvalds 已提交
111

112 113 114 115 116 117 118 119 120 121 122
STATIC int
xfs_inode_hasattr(
	struct xfs_inode	*ip)
{
	if (!XFS_IFORK_Q(ip) ||
	    (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
	     ip->i_d.di_anextents == 0))
		return 0;
	return 1;
}

L
Linus Torvalds 已提交
123 124 125 126 127
/*========================================================================
 * Overall external interface routines.
 *========================================================================*/

int
128 129
xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name,
		char *value, int *valuelenp, int flags)
L
Linus Torvalds 已提交
130 131 132 133
{
	xfs_da_args_t   args;
	int             error;

134 135
	if (!xfs_inode_hasattr(ip))
		return ENOATTR;
L
Linus Torvalds 已提交
136 137 138 139 140

	/*
	 * Fill in the arg structure for this request.
	 */
	memset((char *)&args, 0, sizeof(args));
141 142
	args.name = name->name;
	args.namelen = name->len;
L
Linus Torvalds 已提交
143 144 145 146 147 148 149 150 151 152
	args.value = value;
	args.valuelen = *valuelenp;
	args.flags = flags;
	args.hashval = xfs_da_hashname(args.name, args.namelen);
	args.dp = ip;
	args.whichfork = XFS_ATTR_FORK;

	/*
	 * Decide on what work routines to call based on the inode size.
	 */
153
	if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
L
Linus Torvalds 已提交
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
		error = xfs_attr_shortform_getvalue(&args);
	} else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK)) {
		error = xfs_attr_leaf_get(&args);
	} else {
		error = xfs_attr_node_get(&args);
	}

	/*
	 * Return the number of bytes in the value to the caller.
	 */
	*valuelenp = args.valuelen;

	if (error == EEXIST)
		error = 0;
	return(error);
}

int
172 173 174 175 176
xfs_attr_get(
	xfs_inode_t	*ip,
	const char	*name,
	char		*value,
	int		*valuelenp,
177
	int		flags)
L
Linus Torvalds 已提交
178
{
179 180
	int		error;
	struct xfs_name	xname;
L
Linus Torvalds 已提交
181 182 183 184 185 186

	XFS_STATS_INC(xs_attr_get);

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

187 188 189 190
	error = xfs_attr_name_to_xname(&xname, name);
	if (error)
		return error;

L
Linus Torvalds 已提交
191
	xfs_ilock(ip, XFS_ILOCK_SHARED);
192
	error = xfs_attr_fetch(ip, &xname, value, valuelenp, flags);
L
Linus Torvalds 已提交
193 194 195 196
	xfs_iunlock(ip, XFS_ILOCK_SHARED);
	return(error);
}

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
/*
 * Calculate how many blocks we need for the new attribute,
 */
int
xfs_attr_calc_size(
	struct xfs_inode 	*ip,
	int			namelen,
	int			valuelen,
	int			*local)
{
	struct xfs_mount 	*mp = ip->i_mount;
	int			size;
	int			nblks;

	/*
	 * Determine space new attribute will use, and if it would be
	 * "local" or "remote" (note: local != inline).
	 */
	size = xfs_attr_leaf_newentsize(namelen, valuelen,
					mp->m_sb.sb_blocksize, local);

	nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK);
	if (*local) {
		if (size > (mp->m_sb.sb_blocksize >> 1)) {
			/* Double split possible */
			nblks *= 2;
		}
	} else {
		/*
		 * Out of line attribute, cannot double split, but
		 * make room for the attribute value itself.
		 */
		uint	dblocks = XFS_B_TO_FSB(mp, valuelen);
		nblks += dblocks;
		nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK);
	}

	return nblks;
}

237 238 239
STATIC int
xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
		char *value, int valuelen, int flags)
L
Linus Torvalds 已提交
240 241 242 243 244
{
	xfs_da_args_t	args;
	xfs_fsblock_t	firstblock;
	xfs_bmap_free_t flist;
	int		error, err2, committed;
245
	xfs_mount_t	*mp = dp->i_mount;
L
Linus Torvalds 已提交
246
	int             rsvd = (flags & ATTR_ROOT) != 0;
247
	int		local;
L
Linus Torvalds 已提交
248 249 250 251 252 253 254 255 256 257 258 259

	/*
	 * Attach the dquots to the inode.
	 */
	if ((error = XFS_QM_DQATTACH(mp, dp, 0)))
		return (error);

	/*
	 * If the inode doesn't have an attribute fork, add one.
	 * (inode must not be locked when we call this routine)
	 */
	if (XFS_IFORK_Q(dp) == 0) {
260
		int sf_size = sizeof(xfs_attr_sf_hdr_t) +
261
			      XFS_ATTR_SF_ENTSIZE_BYNAME(name->len, valuelen);
262 263

		if ((error = xfs_bmap_add_attrfork(dp, sf_size, rsvd)))
L
Linus Torvalds 已提交
264 265 266 267 268 269 270
			return(error);
	}

	/*
	 * Fill in the arg structure for this request.
	 */
	memset((char *)&args, 0, sizeof(args));
271 272
	args.name = name->name;
	args.namelen = name->len;
L
Linus Torvalds 已提交
273 274 275 276 277 278 279 280
	args.value = value;
	args.valuelen = valuelen;
	args.flags = flags;
	args.hashval = xfs_da_hashname(args.name, args.namelen);
	args.dp = dp;
	args.firstblock = &firstblock;
	args.flist = &flist;
	args.whichfork = XFS_ATTR_FORK;
281
	args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
L
Linus Torvalds 已提交
282 283

	/* Size is now blocks for attribute data */
284
	args.total = xfs_attr_calc_size(dp, name->len, valuelen, &local);
L
Linus Torvalds 已提交
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305

	/*
	 * Start our first transaction of the day.
	 *
	 * All future transactions during this code must be "chained" off
	 * this one via the trans_dup() call.  All transactions will contain
	 * the inode, and the inode will always be marked with trans_ihold().
	 * Since the inode will be locked in all transactions, we must log
	 * the inode in every transaction to let it float upward through
	 * the log.
	 */
	args.trans = xfs_trans_alloc(mp, XFS_TRANS_ATTR_SET);

	/*
	 * Root fork attributes can use reserved data blocks for this
	 * operation if necessary
	 */

	if (rsvd)
		args.trans->t_flags |= XFS_TRANS_RESERVE;

306 307 308
	if ((error = xfs_trans_reserve(args.trans, args.total,
			XFS_ATTRSET_LOG_RES(mp, args.total), 0,
			XFS_TRANS_PERM_LOG_RES, XFS_ATTRSET_LOG_COUNT))) {
L
Linus Torvalds 已提交
309 310 311 312 313
		xfs_trans_cancel(args.trans, 0);
		return(error);
	}
	xfs_ilock(dp, XFS_ILOCK_EXCL);

314 315 316
	error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, args.total, 0,
				rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
				       XFS_QMOPT_RES_REGBLKS);
L
Linus Torvalds 已提交
317 318 319 320 321 322 323 324 325 326
	if (error) {
		xfs_iunlock(dp, XFS_ILOCK_EXCL);
		xfs_trans_cancel(args.trans, XFS_TRANS_RELEASE_LOG_RES);
		return (error);
	}

	xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
	xfs_trans_ihold(args.trans, dp);

	/*
327
	 * If the attribute list is non-existent or a shortform list,
L
Linus Torvalds 已提交
328 329 330 331 332 333 334 335 336 337
	 * upgrade it to a single-leaf-block attribute list.
	 */
	if ((dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) ||
	    ((dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS) &&
	     (dp->i_d.di_anextents == 0))) {

		/*
		 * Build initial attribute list (if required).
		 */
		if (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS)
338
			xfs_attr_shortform_create(&args);
L
Linus Torvalds 已提交
339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360

		/*
		 * Try to add the attr to the attribute list in
		 * the inode.
		 */
		error = xfs_attr_shortform_addname(&args);
		if (error != ENOSPC) {
			/*
			 * Commit the shortform mods, and we're done.
			 * NOTE: this is also the error path (EEXIST, etc).
			 */
			ASSERT(args.trans != NULL);

			/*
			 * If this is a synchronous mount, make sure that
			 * the transaction goes to disk before returning
			 * to the user.
			 */
			if (mp->m_flags & XFS_MOUNT_WSYNC) {
				xfs_trans_set_sync(args.trans);
			}
			err2 = xfs_trans_commit(args.trans,
361
						 XFS_TRANS_RELEASE_LOG_RES);
L
Linus Torvalds 已提交
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
			xfs_iunlock(dp, XFS_ILOCK_EXCL);

			/*
			 * Hit the inode change time.
			 */
			if (!error && (flags & ATTR_KERNOTIME) == 0) {
				xfs_ichgtime(dp, XFS_ICHGTIME_CHG);
			}
			return(error == 0 ? err2 : error);
		}

		/*
		 * It won't fit in the shortform, transform to a leaf block.
		 * GROT: another possible req'mt for a double-split btree op.
		 */
		XFS_BMAP_INIT(args.flist, args.firstblock);
		error = xfs_attr_shortform_to_leaf(&args);
		if (!error) {
			error = xfs_bmap_finish(&args.trans, args.flist,
381
						&committed);
L
Linus Torvalds 已提交
382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
		}
		if (error) {
			ASSERT(committed);
			args.trans = NULL;
			xfs_bmap_cancel(&flist);
			goto out;
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args.trans, dp);
		}

		/*
		 * Commit the leaf transformation.  We'll need another (linked)
		 * transaction to add the new attribute to the leaf.
		 */
403 404 405

		error = xfs_trans_roll(&args.trans, dp);
		if (error)
L
Linus Torvalds 已提交
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
			goto out;

	}

	if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
		error = xfs_attr_leaf_addname(&args);
	} else {
		error = xfs_attr_node_addname(&args);
	}
	if (error) {
		goto out;
	}

	/*
	 * If this is a synchronous mount, make sure that the
	 * transaction goes to disk before returning to the user.
	 */
	if (mp->m_flags & XFS_MOUNT_WSYNC) {
		xfs_trans_set_sync(args.trans);
	}

	/*
	 * Commit the last in the sequence of transactions.
	 */
	xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
431
	error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
L
Linus Torvalds 已提交
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
	xfs_iunlock(dp, XFS_ILOCK_EXCL);

	/*
	 * Hit the inode change time.
	 */
	if (!error && (flags & ATTR_KERNOTIME) == 0) {
		xfs_ichgtime(dp, XFS_ICHGTIME_CHG);
	}

	return(error);

out:
	if (args.trans)
		xfs_trans_cancel(args.trans,
			XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
	xfs_iunlock(dp, XFS_ILOCK_EXCL);
	return(error);
}

451
int
452 453 454 455 456 457
xfs_attr_set(
	xfs_inode_t	*dp,
	const char	*name,
	char		*value,
	int		valuelen,
	int		flags)
L
Linus Torvalds 已提交
458
{
459 460
	int             error;
	struct xfs_name	xname;
L
Linus Torvalds 已提交
461

462
	XFS_STATS_INC(xs_attr_set);
L
Linus Torvalds 已提交
463

464
	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
L
Linus Torvalds 已提交
465 466
		return (EIO);

467 468 469 470 471
	error = xfs_attr_name_to_xname(&xname, name);
	if (error)
		return error;

	return xfs_attr_set_int(dp, &xname, value, valuelen, flags);
472 473 474 475 476 477
}

/*
 * Generic handler routine to remove a name from an attribute list.
 * Transitions attribute list from Btree to shortform as necessary.
 */
478 479
STATIC int
xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
480 481 482 483 484 485 486
{
	xfs_da_args_t	args;
	xfs_fsblock_t	firstblock;
	xfs_bmap_free_t	flist;
	int		error;
	xfs_mount_t	*mp = dp->i_mount;

L
Linus Torvalds 已提交
487 488 489 490
	/*
	 * Fill in the arg structure for this request.
	 */
	memset((char *)&args, 0, sizeof(args));
491 492
	args.name = name->name;
	args.namelen = name->len;
L
Linus Torvalds 已提交
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
	args.flags = flags;
	args.hashval = xfs_da_hashname(args.name, args.namelen);
	args.dp = dp;
	args.firstblock = &firstblock;
	args.flist = &flist;
	args.total = 0;
	args.whichfork = XFS_ATTR_FORK;

	/*
	 * Attach the dquots to the inode.
	 */
	if ((error = XFS_QM_DQATTACH(mp, dp, 0)))
		return (error);

	/*
	 * Start our first transaction of the day.
	 *
	 * All future transactions during this code must be "chained" off
	 * this one via the trans_dup() call.  All transactions will contain
	 * the inode, and the inode will always be marked with trans_ihold().
	 * Since the inode will be locked in all transactions, we must log
	 * the inode in every transaction to let it float upward through
	 * the log.
	 */
	args.trans = xfs_trans_alloc(mp, XFS_TRANS_ATTR_RM);

	/*
	 * Root fork attributes can use reserved data blocks for this
	 * operation if necessary
	 */

	if (flags & ATTR_ROOT)
		args.trans->t_flags |= XFS_TRANS_RESERVE;

	if ((error = xfs_trans_reserve(args.trans,
				      XFS_ATTRRM_SPACE_RES(mp),
				      XFS_ATTRRM_LOG_RES(mp),
				      0, XFS_TRANS_PERM_LOG_RES,
				      XFS_ATTRRM_LOG_COUNT))) {
		xfs_trans_cancel(args.trans, 0);
		return(error);
	}

	xfs_ilock(dp, XFS_ILOCK_EXCL);
	/*
	 * No need to make quota reservations here. We expect to release some
	 * blocks not allocate in the common case.
	 */
	xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
	xfs_trans_ihold(args.trans, dp);

	/*
	 * Decide on what work routines to call based on the inode size.
	 */
547
	if (!xfs_inode_hasattr(dp)) {
L
Linus Torvalds 已提交
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
		error = XFS_ERROR(ENOATTR);
		goto out;
	}
	if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
		ASSERT(dp->i_afp->if_flags & XFS_IFINLINE);
		error = xfs_attr_shortform_remove(&args);
		if (error) {
			goto out;
		}
	} else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
		error = xfs_attr_leaf_removename(&args);
	} else {
		error = xfs_attr_node_removename(&args);
	}
	if (error) {
		goto out;
	}

	/*
	 * If this is a synchronous mount, make sure that the
	 * transaction goes to disk before returning to the user.
	 */
	if (mp->m_flags & XFS_MOUNT_WSYNC) {
		xfs_trans_set_sync(args.trans);
	}

	/*
	 * Commit the last in the sequence of transactions.
	 */
	xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
578
	error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
L
Linus Torvalds 已提交
579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
	xfs_iunlock(dp, XFS_ILOCK_EXCL);

	/*
	 * Hit the inode change time.
	 */
	if (!error && (flags & ATTR_KERNOTIME) == 0) {
		xfs_ichgtime(dp, XFS_ICHGTIME_CHG);
	}

	return(error);

out:
	if (args.trans)
		xfs_trans_cancel(args.trans,
			XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
	xfs_iunlock(dp, XFS_ILOCK_EXCL);
	return(error);
}

598
int
599 600 601 602
xfs_attr_remove(
	xfs_inode_t	*dp,
	const char	*name,
	int		flags)
603
{
604 605
	int		error;
	struct xfs_name	xname;
606 607 608 609 610 611

	XFS_STATS_INC(xs_attr_remove);

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

612 613 614 615
	error = xfs_attr_name_to_xname(&xname, name);
	if (error)
		return error;

616
	xfs_ilock(dp, XFS_ILOCK_SHARED);
617
	if (!xfs_inode_hasattr(dp)) {
618
		xfs_iunlock(dp, XFS_ILOCK_SHARED);
619
		return XFS_ERROR(ENOATTR);
620 621 622
	}
	xfs_iunlock(dp, XFS_ILOCK_SHARED);

623
	return xfs_attr_remove_int(dp, &xname, flags);
624 625
}

626
int
627 628 629 630 631
xfs_attr_list_int(xfs_attr_list_context_t *context)
{
	int error;
	xfs_inode_t *dp = context->dp;

632 633 634 635 636 637 638 639
	XFS_STATS_INC(xs_attr_list);

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

	xfs_ilock(dp, XFS_ILOCK_SHARED);
	xfs_attr_trace_l_c("syscall start", context);

640 641 642
	/*
	 * Decide on what work routines to call based on the inode size.
	 */
643
	if (!xfs_inode_hasattr(dp)) {
644 645 646 647 648 649 650 651
		error = 0;
	} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
		error = xfs_attr_shortform_list(context);
	} else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
		error = xfs_attr_leaf_list(context);
	} else {
		error = xfs_attr_node_list(context);
	}
652 653 654 655

	xfs_iunlock(dp, XFS_ILOCK_SHARED);
	xfs_attr_trace_l_c("syscall end", context);

656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671
	return error;
}

#define	ATTR_ENTBASESIZE		/* minimum bytes used by an attr */ \
	(((struct attrlist_ent *) 0)->a_name - (char *) 0)
#define	ATTR_ENTSIZE(namelen)		/* actual bytes used by an attr */ \
	((ATTR_ENTBASESIZE + (namelen) + 1 + sizeof(u_int32_t)-1) \
	 & ~(sizeof(u_int32_t)-1))

/*
 * Format an attribute and copy it out to the user's buffer.
 * Take care to check values and protect against them changing later,
 * we may be reading them directly out of a user buffer.
 */
/*ARGSUSED*/
STATIC int
672
xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags,
673 674 675
		     char *name, int namelen,
		     int valuelen, char *value)
{
676
	struct attrlist *alist = (struct attrlist *)context->alist;
677 678 679 680 681 682
	attrlist_ent_t *aep;
	int arraytop;

	ASSERT(!(context->flags & ATTR_KERNOVAL));
	ASSERT(context->count >= 0);
	ASSERT(context->count < (ATTR_MAX_VALUELEN/8));
683
	ASSERT(context->firstu >= sizeof(*alist));
684 685
	ASSERT(context->firstu <= context->bufsize);

686 687 688 689 690 691 692 693 694 695 696 697
	/*
	 * Only list entries in the right namespace.
	 */
	if (((context->flags & ATTR_SECURE) == 0) !=
	    ((flags & XFS_ATTR_SECURE) == 0))
		return 0;
	if (((context->flags & ATTR_ROOT) == 0) !=
	    ((flags & XFS_ATTR_ROOT) == 0))
		return 0;

	arraytop = sizeof(*alist) +
			context->count * sizeof(alist->al_offset[0]);
698 699 700
	context->firstu -= ATTR_ENTSIZE(namelen);
	if (context->firstu < arraytop) {
		xfs_attr_trace_l_c("buffer full", context);
701
		alist->al_more = 1;
702 703 704 705
		context->seen_enough = 1;
		return 1;
	}

706
	aep = (attrlist_ent_t *)&context->alist[context->firstu];
707 708
	aep->a_valuelen = valuelen;
	memcpy(aep->a_name, name, namelen);
709 710 711
	aep->a_name[namelen] = 0;
	alist->al_offset[context->count++] = context->firstu;
	alist->al_count = context->count;
712 713 714 715
	xfs_attr_trace_l_c("add", context);
	return 0;
}

L
Linus Torvalds 已提交
716 717 718 719 720 721 722 723
/*
 * Generate a list of extended attribute names and optionally
 * also value lengths.  Positive return value follows the XFS
 * convention of being an error, zero or negative return code
 * is the length of the buffer returned (negated), indicating
 * success.
 */
int
724 725 726 727 728 729
xfs_attr_list(
	xfs_inode_t	*dp,
	char		*buffer,
	int		bufsize,
	int		flags,
	attrlist_cursor_kern_t *cursor)
L
Linus Torvalds 已提交
730 731
{
	xfs_attr_list_context_t context;
732
	struct attrlist *alist;
L
Linus Torvalds 已提交
733 734 735 736 737 738 739 740 741
	int error;

	/*
	 * Validate the cursor.
	 */
	if (cursor->pad1 || cursor->pad2)
		return(XFS_ERROR(EINVAL));
	if ((cursor->initted == 0) &&
	    (cursor->hashval || cursor->blkno || cursor->offset))
742
		return XFS_ERROR(EINVAL);
L
Linus Torvalds 已提交
743 744 745 746 747

	/*
	 * Check for a properly aligned buffer.
	 */
	if (((long)buffer) & (sizeof(int)-1))
748
		return XFS_ERROR(EFAULT);
L
Linus Torvalds 已提交
749 750 751 752 753 754
	if (flags & ATTR_KERNOVAL)
		bufsize = 0;

	/*
	 * Initialize the output buffer.
	 */
755
	memset(&context, 0, sizeof(context));
756
	context.dp = dp;
L
Linus Torvalds 已提交
757 758 759
	context.cursor = cursor;
	context.resynch = 1;
	context.flags = flags;
760 761 762 763
	context.alist = buffer;
	context.bufsize = (bufsize & ~(sizeof(int)-1));  /* align */
	context.firstu = context.bufsize;
	context.put_listent = xfs_attr_put_listent;
L
Linus Torvalds 已提交
764

765 766 767 768
	alist = (struct attrlist *)context.alist;
	alist->al_count = 0;
	alist->al_more = 0;
	alist->al_offset[0] = context.bufsize;
769 770

	error = xfs_attr_list_int(&context);
771
	ASSERT(error >= 0);
772
	return error;
L
Linus Torvalds 已提交
773 774 775 776 777 778 779 780 781 782 783 784 785
}

int								/* error */
xfs_attr_inactive(xfs_inode_t *dp)
{
	xfs_trans_t *trans;
	xfs_mount_t *mp;
	int error;

	mp = dp->i_mount;
	ASSERT(! XFS_NOT_DQATTACHED(mp, dp));

	xfs_ilock(dp, XFS_ILOCK_SHARED);
786 787
	if (!xfs_inode_hasattr(dp) ||
	    dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
L
Linus Torvalds 已提交
788
		xfs_iunlock(dp, XFS_ILOCK_SHARED);
789
		return 0;
L
Linus Torvalds 已提交
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
	}
	xfs_iunlock(dp, XFS_ILOCK_SHARED);

	/*
	 * Start our first transaction of the day.
	 *
	 * All future transactions during this code must be "chained" off
	 * this one via the trans_dup() call.  All transactions will contain
	 * the inode, and the inode will always be marked with trans_ihold().
	 * Since the inode will be locked in all transactions, we must log
	 * the inode in every transaction to let it float upward through
	 * the log.
	 */
	trans = xfs_trans_alloc(mp, XFS_TRANS_ATTRINVAL);
	if ((error = xfs_trans_reserve(trans, 0, XFS_ATTRINVAL_LOG_RES(mp), 0,
				      XFS_TRANS_PERM_LOG_RES,
				      XFS_ATTRINVAL_LOG_COUNT))) {
		xfs_trans_cancel(trans, 0);
		return(error);
	}
	xfs_ilock(dp, XFS_ILOCK_EXCL);

	/*
	 * No need to make quota reservations here. We expect to release some
	 * blocks, not allocate, in the common case.
	 */
	xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
	xfs_trans_ihold(trans, dp);

	/*
	 * Decide on what work routines to call based on the inode size.
	 */
822 823
	if (!xfs_inode_hasattr(dp) ||
	    dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
L
Linus Torvalds 已提交
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846
		error = 0;
		goto out;
	}
	error = xfs_attr_root_inactive(&trans, dp);
	if (error)
		goto out;
	/*
	 * signal synchronous inactive transactions unless this
	 * is a synchronous mount filesystem in which case we
	 * know that we're here because we've been called out of
	 * xfs_inactive which means that the last reference is gone
	 * and the unlink transaction has already hit the disk so
	 * async inactive transactions are safe.
	 */
	if ((error = xfs_itruncate_finish(&trans, dp, 0LL, XFS_ATTR_FORK,
				(!(mp->m_flags & XFS_MOUNT_WSYNC)
				 ? 1 : 0))))
		goto out;

	/*
	 * Commit the last in the sequence of transactions.
	 */
	xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE);
847
	error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES);
L
Linus Torvalds 已提交
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
	xfs_iunlock(dp, XFS_ILOCK_EXCL);

	return(error);

out:
	xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
	xfs_iunlock(dp, XFS_ILOCK_EXCL);
	return(error);
}



/*========================================================================
 * External routines when attribute list is inside the inode
 *========================================================================*/

/*
 * Add a name to the shortform attribute list structure
 * This is the external routine.
 */
STATIC int
xfs_attr_shortform_addname(xfs_da_args_t *args)
{
871
	int newsize, forkoff, retval;
L
Linus Torvalds 已提交
872 873 874 875 876 877 878 879 880 881 882

	retval = xfs_attr_shortform_lookup(args);
	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
		return(retval);
	} else if (retval == EEXIST) {
		if (args->flags & ATTR_CREATE)
			return(retval);
		retval = xfs_attr_shortform_remove(args);
		ASSERT(retval == 0);
	}

883 884 885 886
	if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX ||
	    args->valuelen >= XFS_ATTR_SF_ENTSIZE_MAX)
		return(XFS_ERROR(ENOSPC));

L
Linus Torvalds 已提交
887 888
	newsize = XFS_ATTR_SF_TOTSIZE(args->dp);
	newsize += XFS_ATTR_SF_ENTSIZE_BYNAME(args->namelen, args->valuelen);
889 890 891

	forkoff = xfs_attr_shortform_bytesfit(args->dp, newsize);
	if (!forkoff)
L
Linus Torvalds 已提交
892
		return(XFS_ERROR(ENOSPC));
893 894

	xfs_attr_shortform_add(args, forkoff);
L
Linus Torvalds 已提交
895 896 897 898 899 900 901 902 903 904 905 906 907 908
	return(0);
}


/*========================================================================
 * External routines when attribute list is one block
 *========================================================================*/

/*
 * Add a name to the leaf attribute list structure
 *
 * This leaf block cannot have a "remote" value, we only call this routine
 * if bmap_one_block() says there is only one block (ie: no remote blks).
 */
D
David Chinner 已提交
909
STATIC int
L
Linus Torvalds 已提交
910 911 912 913
xfs_attr_leaf_addname(xfs_da_args_t *args)
{
	xfs_inode_t *dp;
	xfs_dabuf_t *bp;
914
	int retval, error, committed, forkoff;
L
Linus Torvalds 已提交
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

	/*
	 * Read the (only) block in the attribute list in.
	 */
	dp = args->dp;
	args->blkno = 0;
	error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
					     XFS_ATTR_FORK);
	if (error)
		return(error);
	ASSERT(bp != NULL);

	/*
	 * Look up the given attribute in the leaf block.  Figure out if
	 * the given flags produce an error or call for an atomic rename.
	 */
	retval = xfs_attr_leaf_lookup_int(bp, args);
	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
		xfs_da_brelse(args->trans, bp);
		return(retval);
	} else if (retval == EEXIST) {
		if (args->flags & ATTR_CREATE) {	/* pure create op */
			xfs_da_brelse(args->trans, bp);
			return(retval);
		}
940
		args->op_flags |= XFS_DA_OP_RENAME;	/* an atomic rename */
L
Linus Torvalds 已提交
941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962
		args->blkno2 = args->blkno;		/* set 2nd entry info*/
		args->index2 = args->index;
		args->rmtblkno2 = args->rmtblkno;
		args->rmtblkcnt2 = args->rmtblkcnt;
	}

	/*
	 * Add the attribute to the leaf block, transitioning to a Btree
	 * if required.
	 */
	retval = xfs_attr_leaf_add(bp, args);
	xfs_da_buf_done(bp);
	if (retval == ENOSPC) {
		/*
		 * Promote the attribute list to the Btree format, then
		 * Commit that transaction so that the node_addname() call
		 * can manage its own transactions.
		 */
		XFS_BMAP_INIT(args->flist, args->firstblock);
		error = xfs_attr_leaf_to_node(args);
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
963
						&committed);
L
Linus Torvalds 已提交
964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			return(error);
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, dp);
		}

		/*
		 * Commit the current trans (including the inode) and start
		 * a new one.
		 */
985 986
		error = xfs_trans_roll(&args->trans, dp);
		if (error)
L
Linus Torvalds 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999
			return (error);

		/*
		 * Fob the whole rest of the problem off on the Btree code.
		 */
		error = xfs_attr_node_addname(args);
		return(error);
	}

	/*
	 * Commit the transaction that added the attr name so that
	 * later routines can manage their own transactions.
	 */
1000 1001
	error = xfs_trans_roll(&args->trans, dp);
	if (error)
L
Linus Torvalds 已提交
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
		return (error);

	/*
	 * If there was an out-of-line value, allocate the blocks we
	 * identified for its storage and copy the value.  This is done
	 * after we create the attribute so that we don't overflow the
	 * maximum size of a transaction and/or hit a deadlock.
	 */
	if (args->rmtblkno > 0) {
		error = xfs_attr_rmtval_set(args);
		if (error)
			return(error);
	}

	/*
	 * If this is an atomic rename operation, we must "flip" the
	 * incomplete flags on the "new" and "old" attribute/value pairs
	 * so that one disappears and one appears atomically.  Then we
	 * must remove the "old" attribute/value pair.
	 */
1022
	if (args->op_flags & XFS_DA_OP_RENAME) {
L
Linus Torvalds 已提交
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
		/*
		 * In a separate transaction, set the incomplete flag on the
		 * "old" attr and clear the incomplete flag on the "new" attr.
		 */
		error = xfs_attr_leaf_flipflags(args);
		if (error)
			return(error);

		/*
		 * Dismantle the "old" attribute/value pair by removing
		 * a "remote" value (if it exists).
		 */
		args->index = args->index2;
		args->blkno = args->blkno2;
		args->rmtblkno = args->rmtblkno2;
		args->rmtblkcnt = args->rmtblkcnt2;
		if (args->rmtblkno) {
			error = xfs_attr_rmtval_remove(args);
			if (error)
				return(error);
		}

		/*
		 * Read in the block containing the "old" attr, then
		 * remove the "old" attr from that block (neat, huh!)
		 */
		error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1,
						     &bp, XFS_ATTR_FORK);
		if (error)
			return(error);
		ASSERT(bp != NULL);
		(void)xfs_attr_leaf_remove(bp, args);

		/*
		 * If the result is small enough, shrink it all into the inode.
		 */
1059
		if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
L
Linus Torvalds 已提交
1060
			XFS_BMAP_INIT(args->flist, args->firstblock);
1061
			error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
L
Linus Torvalds 已提交
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
			/* bp is gone due to xfs_da_shrink_inode */
			if (!error) {
				error = xfs_bmap_finish(&args->trans,
							args->flist,
							&committed);
			}
			if (error) {
				ASSERT(committed);
				args->trans = NULL;
				xfs_bmap_cancel(args->flist);
				return(error);
			}

			/*
			 * bmap_finish() may have committed the last trans
			 * and started a new one.  We need the inode to be
			 * in all transactions.
			 */
			if (committed) {
				xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
				xfs_trans_ihold(args->trans, dp);
			}
		} else
			xfs_da_buf_done(bp);

		/*
		 * Commit the remove and start the next trans in series.
		 */
1090
		error = xfs_trans_roll(&args->trans, dp);
L
Linus Torvalds 已提交
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111

	} else if (args->rmtblkno > 0) {
		/*
		 * Added a "remote" value, just clear the incomplete flag.
		 */
		error = xfs_attr_leaf_clearflag(args);
	}
	return(error);
}

/*
 * Remove a name from the leaf attribute list structure
 *
 * This leaf block cannot have a "remote" value, we only call this routine
 * if bmap_one_block() says there is only one block (ie: no remote blks).
 */
STATIC int
xfs_attr_leaf_removename(xfs_da_args_t *args)
{
	xfs_inode_t *dp;
	xfs_dabuf_t *bp;
1112
	int error, committed, forkoff;
L
Linus Torvalds 已提交
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136

	/*
	 * Remove the attribute.
	 */
	dp = args->dp;
	args->blkno = 0;
	error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
					     XFS_ATTR_FORK);
	if (error) {
		return(error);
	}

	ASSERT(bp != NULL);
	error = xfs_attr_leaf_lookup_int(bp, args);
	if (error == ENOATTR) {
		xfs_da_brelse(args->trans, bp);
		return(error);
	}

	(void)xfs_attr_leaf_remove(bp, args);

	/*
	 * If the result is small enough, shrink it all into the inode.
	 */
1137
	if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
L
Linus Torvalds 已提交
1138
		XFS_BMAP_INIT(args->flist, args->firstblock);
1139
		error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
L
Linus Torvalds 已提交
1140 1141 1142
		/* bp is gone due to xfs_da_shrink_inode */
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
1143
						&committed);
L
Linus Torvalds 已提交
1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			return(error);
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, dp);
		}
	} else
		xfs_da_buf_done(bp);
	return(0);
}

/*
 * Look up a name in a leaf attribute list structure.
 *
 * This leaf block cannot have a "remote" value, we only call this routine
 * if bmap_one_block() says there is only one block (ie: no remote blks).
 */
1171
STATIC int
L
Linus Torvalds 已提交
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
xfs_attr_leaf_get(xfs_da_args_t *args)
{
	xfs_dabuf_t *bp;
	int error;

	args->blkno = 0;
	error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
					     XFS_ATTR_FORK);
	if (error)
		return(error);
	ASSERT(bp != NULL);

	error = xfs_attr_leaf_lookup_int(bp, args);
	if (error != EEXIST)  {
		xfs_da_brelse(args->trans, bp);
		return(error);
	}
	error = xfs_attr_leaf_getvalue(bp, args);
	xfs_da_brelse(args->trans, bp);
	if (!error && (args->rmtblkno > 0) && !(args->flags & ATTR_KERNOVAL)) {
		error = xfs_attr_rmtval_get(args);
	}
	return(error);
}

/*
 * Copy out attribute entries for attr_list(), for leaf attribute lists.
 */
STATIC int
xfs_attr_leaf_list(xfs_attr_list_context_t *context)
{
	xfs_attr_leafblock_t *leaf;
	int error;
	xfs_dabuf_t *bp;

	context->cursor->blkno = 0;
	error = xfs_da_read_buf(NULL, context->dp, 0, -1, &bp, XFS_ATTR_FORK);
	if (error)
1210
		return XFS_ERROR(error);
L
Linus Torvalds 已提交
1211 1212
	ASSERT(bp != NULL);
	leaf = bp->data;
1213
	if (unlikely(be16_to_cpu(leaf->hdr.info.magic) != XFS_ATTR_LEAF_MAGIC)) {
L
Linus Torvalds 已提交
1214 1215 1216
		XFS_CORRUPTION_ERROR("xfs_attr_leaf_list", XFS_ERRLEVEL_LOW,
				     context->dp->i_mount, leaf);
		xfs_da_brelse(NULL, bp);
1217
		return XFS_ERROR(EFSCORRUPTED);
L
Linus Torvalds 已提交
1218 1219
	}

1220
	error = xfs_attr_leaf_list_int(bp, context);
L
Linus Torvalds 已提交
1221
	xfs_da_brelse(NULL, bp);
1222
	return XFS_ERROR(error);
L
Linus Torvalds 已提交
1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274
}


/*========================================================================
 * External routines when attribute list size > XFS_LBSIZE(mp).
 *========================================================================*/

/*
 * Add a name to a Btree-format attribute list.
 *
 * This will involve walking down the Btree, and may involve splitting
 * leaf nodes and even splitting intermediate nodes up to and including
 * the root node (a special case of an intermediate node).
 *
 * "Remote" attribute values confuse the issue and atomic rename operations
 * add a whole extra layer of confusion on top of that.
 */
STATIC int
xfs_attr_node_addname(xfs_da_args_t *args)
{
	xfs_da_state_t *state;
	xfs_da_state_blk_t *blk;
	xfs_inode_t *dp;
	xfs_mount_t *mp;
	int committed, retval, error;

	/*
	 * Fill in bucket of arguments/results/context to carry around.
	 */
	dp = args->dp;
	mp = dp->i_mount;
restart:
	state = xfs_da_state_alloc();
	state->args = args;
	state->mp = mp;
	state->blocksize = state->mp->m_sb.sb_blocksize;
	state->node_ents = state->mp->m_attr_node_ents;

	/*
	 * Search to see if name already exists, and get back a pointer
	 * to where it should go.
	 */
	error = xfs_da_node_lookup_int(state, &retval);
	if (error)
		goto out;
	blk = &state->path.blk[ state->path.active-1 ];
	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
		goto out;
	} else if (retval == EEXIST) {
		if (args->flags & ATTR_CREATE)
			goto out;
1275
		args->op_flags |= XFS_DA_OP_RENAME;	/* atomic rename op */
L
Linus Torvalds 已提交
1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
		args->blkno2 = args->blkno;		/* set 2nd entry info*/
		args->index2 = args->index;
		args->rmtblkno2 = args->rmtblkno;
		args->rmtblkcnt2 = args->rmtblkcnt;
		args->rmtblkno = 0;
		args->rmtblkcnt = 0;
	}

	retval = xfs_attr_leaf_add(blk->bp, state->args);
	if (retval == ENOSPC) {
		if (state->path.active == 1) {
			/*
			 * Its really a single leaf node, but it had
			 * out-of-line values so it looked like it *might*
			 * have been a b-tree.
			 */
			xfs_da_state_free(state);
			XFS_BMAP_INIT(args->flist, args->firstblock);
			error = xfs_attr_leaf_to_node(args);
			if (!error) {
				error = xfs_bmap_finish(&args->trans,
							args->flist,
							&committed);
			}
			if (error) {
				ASSERT(committed);
				args->trans = NULL;
				xfs_bmap_cancel(args->flist);
				goto out;
			}

			/*
			 * bmap_finish() may have committed the last trans
			 * and started a new one.  We need the inode to be
			 * in all transactions.
			 */
			if (committed) {
				xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
				xfs_trans_ihold(args->trans, dp);
			}

			/*
			 * Commit the node conversion and start the next
			 * trans in the chain.
			 */
1321 1322
			error = xfs_trans_roll(&args->trans, dp);
			if (error)
L
Linus Torvalds 已提交
1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337
				goto out;

			goto restart;
		}

		/*
		 * Split as many Btree elements as required.
		 * This code tracks the new and old attr's location
		 * in the index/blkno/rmtblkno/rmtblkcnt fields and
		 * in the index2/blkno2/rmtblkno2/rmtblkcnt2 fields.
		 */
		XFS_BMAP_INIT(args->flist, args->firstblock);
		error = xfs_da_split(state);
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
1338
						&committed);
L
Linus Torvalds 已提交
1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			goto out;
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, dp);
		}
	} else {
		/*
		 * Addition succeeded, update Btree hashvals.
		 */
		xfs_da_fixhashpath(state, &state->path);
	}

	/*
	 * Kill the state structure, we're done with it and need to
	 * allow the buffers to come back later.
	 */
	xfs_da_state_free(state);
	state = NULL;

	/*
	 * Commit the leaf addition or btree split and start the next
	 * trans in the chain.
	 */
1373 1374
	error = xfs_trans_roll(&args->trans, dp);
	if (error)
L
Linus Torvalds 已提交
1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394
		goto out;

	/*
	 * If there was an out-of-line value, allocate the blocks we
	 * identified for its storage and copy the value.  This is done
	 * after we create the attribute so that we don't overflow the
	 * maximum size of a transaction and/or hit a deadlock.
	 */
	if (args->rmtblkno > 0) {
		error = xfs_attr_rmtval_set(args);
		if (error)
			return(error);
	}

	/*
	 * If this is an atomic rename operation, we must "flip" the
	 * incomplete flags on the "new" and "old" attribute/value pairs
	 * so that one disappears and one appears atomically.  Then we
	 * must remove the "old" attribute/value pair.
	 */
1395
	if (args->op_flags & XFS_DA_OP_RENAME) {
L
Linus Torvalds 已提交
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
		/*
		 * In a separate transaction, set the incomplete flag on the
		 * "old" attr and clear the incomplete flag on the "new" attr.
		 */
		error = xfs_attr_leaf_flipflags(args);
		if (error)
			goto out;

		/*
		 * Dismantle the "old" attribute/value pair by removing
		 * a "remote" value (if it exists).
		 */
		args->index = args->index2;
		args->blkno = args->blkno2;
		args->rmtblkno = args->rmtblkno2;
		args->rmtblkcnt = args->rmtblkcnt2;
		if (args->rmtblkno) {
			error = xfs_attr_rmtval_remove(args);
			if (error)
				return(error);
		}

		/*
		 * Re-find the "old" attribute entry after any split ops.
		 * The INCOMPLETE flag means that we will find the "old"
		 * attr, not the "new" one.
		 */
		args->flags |= XFS_ATTR_INCOMPLETE;
		state = xfs_da_state_alloc();
		state->args = args;
		state->mp = mp;
		state->blocksize = state->mp->m_sb.sb_blocksize;
		state->node_ents = state->mp->m_attr_node_ents;
		state->inleaf = 0;
		error = xfs_da_node_lookup_int(state, &retval);
		if (error)
			goto out;

		/*
		 * Remove the name and update the hashvals in the tree.
		 */
		blk = &state->path.blk[ state->path.active-1 ];
		ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
		error = xfs_attr_leaf_remove(blk->bp, args);
		xfs_da_fixhashpath(state, &state->path);

		/*
		 * Check to see if the tree needs to be collapsed.
		 */
		if (retval && (state->path.active > 1)) {
			XFS_BMAP_INIT(args->flist, args->firstblock);
			error = xfs_da_join(state);
			if (!error) {
				error = xfs_bmap_finish(&args->trans,
							args->flist,
							&committed);
			}
			if (error) {
				ASSERT(committed);
				args->trans = NULL;
				xfs_bmap_cancel(args->flist);
				goto out;
			}

			/*
			 * bmap_finish() may have committed the last trans
			 * and started a new one.  We need the inode to be
			 * in all transactions.
			 */
			if (committed) {
				xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
				xfs_trans_ihold(args->trans, dp);
			}
		}

		/*
		 * Commit and start the next trans in the chain.
		 */
1474 1475
		error = xfs_trans_roll(&args->trans, dp);
		if (error)
L
Linus Torvalds 已提交
1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509
			goto out;

	} else if (args->rmtblkno > 0) {
		/*
		 * Added a "remote" value, just clear the incomplete flag.
		 */
		error = xfs_attr_leaf_clearflag(args);
		if (error)
			goto out;
	}
	retval = error = 0;

out:
	if (state)
		xfs_da_state_free(state);
	if (error)
		return(error);
	return(retval);
}

/*
 * Remove a name from a B-tree attribute list.
 *
 * This will involve walking down the Btree, and may involve joining
 * leaf nodes and even joining intermediate nodes up to and including
 * the root node (a special case of an intermediate node).
 */
STATIC int
xfs_attr_node_removename(xfs_da_args_t *args)
{
	xfs_da_state_t *state;
	xfs_da_state_blk_t *blk;
	xfs_inode_t *dp;
	xfs_dabuf_t *bp;
1510
	int retval, error, committed, forkoff;
L
Linus Torvalds 已提交
1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585

	/*
	 * Tie a string around our finger to remind us where we are.
	 */
	dp = args->dp;
	state = xfs_da_state_alloc();
	state->args = args;
	state->mp = dp->i_mount;
	state->blocksize = state->mp->m_sb.sb_blocksize;
	state->node_ents = state->mp->m_attr_node_ents;

	/*
	 * Search to see if name exists, and get back a pointer to it.
	 */
	error = xfs_da_node_lookup_int(state, &retval);
	if (error || (retval != EEXIST)) {
		if (error == 0)
			error = retval;
		goto out;
	}

	/*
	 * If there is an out-of-line value, de-allocate the blocks.
	 * This is done before we remove the attribute so that we don't
	 * overflow the maximum size of a transaction and/or hit a deadlock.
	 */
	blk = &state->path.blk[ state->path.active-1 ];
	ASSERT(blk->bp != NULL);
	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
	if (args->rmtblkno > 0) {
		/*
		 * Fill in disk block numbers in the state structure
		 * so that we can get the buffers back after we commit
		 * several transactions in the following calls.
		 */
		error = xfs_attr_fillstate(state);
		if (error)
			goto out;

		/*
		 * Mark the attribute as INCOMPLETE, then bunmapi() the
		 * remote value.
		 */
		error = xfs_attr_leaf_setflag(args);
		if (error)
			goto out;
		error = xfs_attr_rmtval_remove(args);
		if (error)
			goto out;

		/*
		 * Refill the state structure with buffers, the prior calls
		 * released our buffers.
		 */
		error = xfs_attr_refillstate(state);
		if (error)
			goto out;
	}

	/*
	 * Remove the name and update the hashvals in the tree.
	 */
	blk = &state->path.blk[ state->path.active-1 ];
	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
	retval = xfs_attr_leaf_remove(blk->bp, args);
	xfs_da_fixhashpath(state, &state->path);

	/*
	 * Check to see if the tree needs to be collapsed.
	 */
	if (retval && (state->path.active > 1)) {
		XFS_BMAP_INIT(args->flist, args->firstblock);
		error = xfs_da_join(state);
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
1586
						&committed);
L
Linus Torvalds 已提交
1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			goto out;
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, dp);
		}

		/*
		 * Commit the Btree join operation and start a new trans.
		 */
1607 1608
		error = xfs_trans_roll(&args->trans, dp);
		if (error)
L
Linus Torvalds 已提交
1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627
			goto out;
	}

	/*
	 * If the result is small enough, push it all into the inode.
	 */
	if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
		/*
		 * Have to get rid of the copy of this dabuf in the state.
		 */
		ASSERT(state->path.active == 1);
		ASSERT(state->path.blk[0].bp);
		xfs_da_buf_done(state->path.blk[0].bp);
		state->path.blk[0].bp = NULL;

		error = xfs_da_read_buf(args->trans, args->dp, 0, -1, &bp,
						     XFS_ATTR_FORK);
		if (error)
			goto out;
1628 1629
		ASSERT(be16_to_cpu(((xfs_attr_leafblock_t *)
				      bp->data)->hdr.info.magic)
L
Linus Torvalds 已提交
1630 1631
						       == XFS_ATTR_LEAF_MAGIC);

1632
		if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
L
Linus Torvalds 已提交
1633
			XFS_BMAP_INIT(args->flist, args->firstblock);
1634
			error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
L
Linus Torvalds 已提交
1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670
			/* bp is gone due to xfs_da_shrink_inode */
			if (!error) {
				error = xfs_bmap_finish(&args->trans,
							args->flist,
							&committed);
			}
			if (error) {
				ASSERT(committed);
				args->trans = NULL;
				xfs_bmap_cancel(args->flist);
				goto out;
			}

			/*
			 * bmap_finish() may have committed the last trans
			 * and started a new one.  We need the inode to be
			 * in all transactions.
			 */
			if (committed) {
				xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
				xfs_trans_ihold(args->trans, dp);
			}
		} else
			xfs_da_brelse(args->trans, bp);
	}
	error = 0;

out:
	xfs_da_state_free(state);
	return(error);
}

/*
 * Fill in the disk block numbers in the state structure for the buffers
 * that are attached to the state structure.
 * This is done so that we can quickly reattach ourselves to those buffers
1671
 * after some set of transaction commits have released these buffers.
L
Linus Torvalds 已提交
1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717
 */
STATIC int
xfs_attr_fillstate(xfs_da_state_t *state)
{
	xfs_da_state_path_t *path;
	xfs_da_state_blk_t *blk;
	int level;

	/*
	 * Roll down the "path" in the state structure, storing the on-disk
	 * block number for those buffers in the "path".
	 */
	path = &state->path;
	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
		if (blk->bp) {
			blk->disk_blkno = xfs_da_blkno(blk->bp);
			xfs_da_buf_done(blk->bp);
			blk->bp = NULL;
		} else {
			blk->disk_blkno = 0;
		}
	}

	/*
	 * Roll down the "altpath" in the state structure, storing the on-disk
	 * block number for those buffers in the "altpath".
	 */
	path = &state->altpath;
	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
		if (blk->bp) {
			blk->disk_blkno = xfs_da_blkno(blk->bp);
			xfs_da_buf_done(blk->bp);
			blk->bp = NULL;
		} else {
			blk->disk_blkno = 0;
		}
	}

	return(0);
}

/*
 * Reattach the buffers to the state structure based on the disk block
 * numbers stored in the state structure.
1718
 * This is done after some set of transaction commits have released those
L
Linus Torvalds 已提交
1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775
 * buffers from our grip.
 */
STATIC int
xfs_attr_refillstate(xfs_da_state_t *state)
{
	xfs_da_state_path_t *path;
	xfs_da_state_blk_t *blk;
	int level, error;

	/*
	 * Roll down the "path" in the state structure, storing the on-disk
	 * block number for those buffers in the "path".
	 */
	path = &state->path;
	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
		if (blk->disk_blkno) {
			error = xfs_da_read_buf(state->args->trans,
						state->args->dp,
						blk->blkno, blk->disk_blkno,
						&blk->bp, XFS_ATTR_FORK);
			if (error)
				return(error);
		} else {
			blk->bp = NULL;
		}
	}

	/*
	 * Roll down the "altpath" in the state structure, storing the on-disk
	 * block number for those buffers in the "altpath".
	 */
	path = &state->altpath;
	ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
	for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
		if (blk->disk_blkno) {
			error = xfs_da_read_buf(state->args->trans,
						state->args->dp,
						blk->blkno, blk->disk_blkno,
						&blk->bp, XFS_ATTR_FORK);
			if (error)
				return(error);
		} else {
			blk->bp = NULL;
		}
	}

	return(0);
}

/*
 * Look up a filename in a node attribute list.
 *
 * This routine gets called for any attribute fork that has more than one
 * block, ie: both true Btree attr lists and for single-leaf-blocks with
 * "remote" values taking up more blocks.
 */
1776
STATIC int
L
Linus Torvalds 已提交
1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848
xfs_attr_node_get(xfs_da_args_t *args)
{
	xfs_da_state_t *state;
	xfs_da_state_blk_t *blk;
	int error, retval;
	int i;

	state = xfs_da_state_alloc();
	state->args = args;
	state->mp = args->dp->i_mount;
	state->blocksize = state->mp->m_sb.sb_blocksize;
	state->node_ents = state->mp->m_attr_node_ents;

	/*
	 * Search to see if name exists, and get back a pointer to it.
	 */
	error = xfs_da_node_lookup_int(state, &retval);
	if (error) {
		retval = error;
	} else if (retval == EEXIST) {
		blk = &state->path.blk[ state->path.active-1 ];
		ASSERT(blk->bp != NULL);
		ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);

		/*
		 * Get the value, local or "remote"
		 */
		retval = xfs_attr_leaf_getvalue(blk->bp, args);
		if (!retval && (args->rmtblkno > 0)
		    && !(args->flags & ATTR_KERNOVAL)) {
			retval = xfs_attr_rmtval_get(args);
		}
	}

	/*
	 * If not in a transaction, we have to release all the buffers.
	 */
	for (i = 0; i < state->path.active; i++) {
		xfs_da_brelse(args->trans, state->path.blk[i].bp);
		state->path.blk[i].bp = NULL;
	}

	xfs_da_state_free(state);
	return(retval);
}

STATIC int							/* error */
xfs_attr_node_list(xfs_attr_list_context_t *context)
{
	attrlist_cursor_kern_t *cursor;
	xfs_attr_leafblock_t *leaf;
	xfs_da_intnode_t *node;
	xfs_da_node_entry_t *btree;
	int error, i;
	xfs_dabuf_t *bp;

	cursor = context->cursor;
	cursor->initted = 1;

	/*
	 * Do all sorts of validation on the passed-in cursor structure.
	 * If anything is amiss, ignore the cursor and look up the hashval
	 * starting from the btree root.
	 */
	bp = NULL;
	if (cursor->blkno > 0) {
		error = xfs_da_read_buf(NULL, context->dp, cursor->blkno, -1,
					      &bp, XFS_ATTR_FORK);
		if ((error != 0) && (error != EFSCORRUPTED))
			return(error);
		if (bp) {
			node = bp->data;
1849
			switch (be16_to_cpu(node->hdr.info.magic)) {
L
Linus Torvalds 已提交
1850 1851 1852 1853 1854 1855 1856
			case XFS_DA_NODE_MAGIC:
				xfs_attr_trace_l_cn("wrong blk", context, node);
				xfs_da_brelse(NULL, bp);
				bp = NULL;
				break;
			case XFS_ATTR_LEAF_MAGIC:
				leaf = bp->data;
1857 1858
				if (cursor->hashval > be32_to_cpu(leaf->entries[
				    be16_to_cpu(leaf->hdr.count)-1].hashval)) {
L
Linus Torvalds 已提交
1859 1860 1861 1862 1863
					xfs_attr_trace_l_cl("wrong blk",
							   context, leaf);
					xfs_da_brelse(NULL, bp);
					bp = NULL;
				} else if (cursor->hashval <=
1864
					     be32_to_cpu(leaf->entries[0].hashval)) {
L
Linus Torvalds 已提交
1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898
					xfs_attr_trace_l_cl("maybe wrong blk",
							   context, leaf);
					xfs_da_brelse(NULL, bp);
					bp = NULL;
				}
				break;
			default:
				xfs_attr_trace_l_c("wrong blk - ??", context);
				xfs_da_brelse(NULL, bp);
				bp = NULL;
			}
		}
	}

	/*
	 * We did not find what we expected given the cursor's contents,
	 * so we start from the top and work down based on the hash value.
	 * Note that start of node block is same as start of leaf block.
	 */
	if (bp == NULL) {
		cursor->blkno = 0;
		for (;;) {
			error = xfs_da_read_buf(NULL, context->dp,
						      cursor->blkno, -1, &bp,
						      XFS_ATTR_FORK);
			if (error)
				return(error);
			if (unlikely(bp == NULL)) {
				XFS_ERROR_REPORT("xfs_attr_node_list(2)",
						 XFS_ERRLEVEL_LOW,
						 context->dp->i_mount);
				return(XFS_ERROR(EFSCORRUPTED));
			}
			node = bp->data;
1899
			if (be16_to_cpu(node->hdr.info.magic)
L
Linus Torvalds 已提交
1900 1901
							== XFS_ATTR_LEAF_MAGIC)
				break;
1902
			if (unlikely(be16_to_cpu(node->hdr.info.magic)
L
Linus Torvalds 已提交
1903 1904 1905 1906 1907 1908 1909 1910 1911
							!= XFS_DA_NODE_MAGIC)) {
				XFS_CORRUPTION_ERROR("xfs_attr_node_list(3)",
						     XFS_ERRLEVEL_LOW,
						     context->dp->i_mount,
						     node);
				xfs_da_brelse(NULL, bp);
				return(XFS_ERROR(EFSCORRUPTED));
			}
			btree = node->btree;
1912
			for (i = 0; i < be16_to_cpu(node->hdr.count);
L
Linus Torvalds 已提交
1913 1914
								btree++, i++) {
				if (cursor->hashval
1915 1916
						<= be32_to_cpu(btree->hashval)) {
					cursor->blkno = be32_to_cpu(btree->before);
L
Linus Torvalds 已提交
1917 1918 1919 1920 1921
					xfs_attr_trace_l_cb("descending",
							    context, btree);
					break;
				}
			}
1922
			if (i == be16_to_cpu(node->hdr.count)) {
L
Linus Torvalds 已提交
1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937
				xfs_da_brelse(NULL, bp);
				return(0);
			}
			xfs_da_brelse(NULL, bp);
		}
	}
	ASSERT(bp != NULL);

	/*
	 * Roll upward through the blocks, processing each leaf block in
	 * order.  As long as there is space in the result buffer, keep
	 * adding the information.
	 */
	for (;;) {
		leaf = bp->data;
1938
		if (unlikely(be16_to_cpu(leaf->hdr.info.magic)
L
Linus Torvalds 已提交
1939 1940 1941 1942 1943 1944 1945 1946
						!= XFS_ATTR_LEAF_MAGIC)) {
			XFS_CORRUPTION_ERROR("xfs_attr_node_list(4)",
					     XFS_ERRLEVEL_LOW,
					     context->dp->i_mount, leaf);
			xfs_da_brelse(NULL, bp);
			return(XFS_ERROR(EFSCORRUPTED));
		}
		error = xfs_attr_leaf_list_int(bp, context);
1947 1948 1949 1950 1951 1952
		if (error) {
			xfs_da_brelse(NULL, bp);
			return error;
		}
		if (context->seen_enough || leaf->hdr.info.forw == 0)
			break;
1953
		cursor->blkno = be32_to_cpu(leaf->hdr.info.forw);
L
Linus Torvalds 已提交
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978
		xfs_da_brelse(NULL, bp);
		error = xfs_da_read_buf(NULL, context->dp, cursor->blkno, -1,
					      &bp, XFS_ATTR_FORK);
		if (error)
			return(error);
		if (unlikely((bp == NULL))) {
			XFS_ERROR_REPORT("xfs_attr_node_list(5)",
					 XFS_ERRLEVEL_LOW,
					 context->dp->i_mount);
			return(XFS_ERROR(EFSCORRUPTED));
		}
	}
	xfs_da_brelse(NULL, bp);
	return(0);
}


/*========================================================================
 * External routines for manipulating out-of-line attribute values.
 *========================================================================*/

/*
 * Read the value associated with an attribute from the out-of-line buffer
 * that we stored it in.
 */
1979
int
L
Linus Torvalds 已提交
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
xfs_attr_rmtval_get(xfs_da_args_t *args)
{
	xfs_bmbt_irec_t map[ATTR_RMTVALUE_MAPSIZE];
	xfs_mount_t *mp;
	xfs_daddr_t dblkno;
	xfs_caddr_t dst;
	xfs_buf_t *bp;
	int nmap, error, tmp, valuelen, blkcnt, i;
	xfs_dablk_t lblkno;

	ASSERT(!(args->flags & ATTR_KERNOVAL));

	mp = args->dp->i_mount;
	dst = args->value;
	valuelen = args->valuelen;
	lblkno = args->rmtblkno;
	while (valuelen > 0) {
		nmap = ATTR_RMTVALUE_MAPSIZE;
		error = xfs_bmapi(args->trans, args->dp, (xfs_fileoff_t)lblkno,
				  args->rmtblkcnt,
				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
2001
				  NULL, 0, map, &nmap, NULL, NULL);
L
Linus Torvalds 已提交
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078
		if (error)
			return(error);
		ASSERT(nmap >= 1);

		for (i = 0; (i < nmap) && (valuelen > 0); i++) {
			ASSERT((map[i].br_startblock != DELAYSTARTBLOCK) &&
			       (map[i].br_startblock != HOLESTARTBLOCK));
			dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
			blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
			error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno,
					     blkcnt, XFS_BUF_LOCK, &bp);
			if (error)
				return(error);

			tmp = (valuelen < XFS_BUF_SIZE(bp))
				? valuelen : XFS_BUF_SIZE(bp);
			xfs_biomove(bp, 0, tmp, dst, XFS_B_READ);
			xfs_buf_relse(bp);
			dst += tmp;
			valuelen -= tmp;

			lblkno += map[i].br_blockcount;
		}
	}
	ASSERT(valuelen == 0);
	return(0);
}

/*
 * Write the value associated with an attribute into the out-of-line buffer
 * that we have defined for it.
 */
STATIC int
xfs_attr_rmtval_set(xfs_da_args_t *args)
{
	xfs_mount_t *mp;
	xfs_fileoff_t lfileoff;
	xfs_inode_t *dp;
	xfs_bmbt_irec_t map;
	xfs_daddr_t dblkno;
	xfs_caddr_t src;
	xfs_buf_t *bp;
	xfs_dablk_t lblkno;
	int blkcnt, valuelen, nmap, error, tmp, committed;

	dp = args->dp;
	mp = dp->i_mount;
	src = args->value;

	/*
	 * Find a "hole" in the attribute address space large enough for
	 * us to drop the new attribute's value into.
	 */
	blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
	lfileoff = 0;
	error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff,
						   XFS_ATTR_FORK);
	if (error) {
		return(error);
	}
	args->rmtblkno = lblkno = (xfs_dablk_t)lfileoff;
	args->rmtblkcnt = blkcnt;

	/*
	 * Roll through the "value", allocating blocks on disk as required.
	 */
	while (blkcnt > 0) {
		/*
		 * Allocate a single extent, up to the size of the value.
		 */
		XFS_BMAP_INIT(args->flist, args->firstblock);
		nmap = 1;
		error = xfs_bmapi(args->trans, dp, (xfs_fileoff_t)lblkno,
				  blkcnt,
				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA |
							XFS_BMAPI_WRITE,
				  args->firstblock, args->total, &map, &nmap,
2079
				  args->flist, NULL);
L
Linus Torvalds 已提交
2080 2081
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
2082
						&committed);
L
Linus Torvalds 已提交
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			return(error);
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, dp);
		}

		ASSERT(nmap == 1);
		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
		       (map.br_startblock != HOLESTARTBLOCK));
		lblkno += map.br_blockcount;
		blkcnt -= map.br_blockcount;

		/*
		 * Start the next trans in the chain.
		 */
2109 2110
		error = xfs_trans_roll(&args->trans, dp);
		if (error)
L
Linus Torvalds 已提交
2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130
			return (error);
	}

	/*
	 * Roll through the "value", copying the attribute value to the
	 * already-allocated blocks.  Blocks are written synchronously
	 * so that we can know they are all on disk before we turn off
	 * the INCOMPLETE flag.
	 */
	lblkno = args->rmtblkno;
	valuelen = args->valuelen;
	while (valuelen > 0) {
		/*
		 * Try to remember where we decided to put the value.
		 */
		XFS_BMAP_INIT(args->flist, args->firstblock);
		nmap = 1;
		error = xfs_bmapi(NULL, dp, (xfs_fileoff_t)lblkno,
				  args->rmtblkcnt,
				  XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
2131 2132
				  args->firstblock, 0, &map, &nmap,
				  NULL, NULL);
L
Linus Torvalds 已提交
2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196
		if (error) {
			return(error);
		}
		ASSERT(nmap == 1);
		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
		       (map.br_startblock != HOLESTARTBLOCK));

		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);

		bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno,
							blkcnt, XFS_BUF_LOCK);
		ASSERT(bp);
		ASSERT(!XFS_BUF_GETERROR(bp));

		tmp = (valuelen < XFS_BUF_SIZE(bp)) ? valuelen :
							XFS_BUF_SIZE(bp);
		xfs_biomove(bp, 0, tmp, src, XFS_B_WRITE);
		if (tmp < XFS_BUF_SIZE(bp))
			xfs_biozero(bp, tmp, XFS_BUF_SIZE(bp) - tmp);
		if ((error = xfs_bwrite(mp, bp))) {/* GROT: NOTE: synchronous write */
			return (error);
		}
		src += tmp;
		valuelen -= tmp;

		lblkno += map.br_blockcount;
	}
	ASSERT(valuelen == 0);
	return(0);
}

/*
 * Remove the value associated with an attribute by deleting the
 * out-of-line buffer that it is stored on.
 */
STATIC int
xfs_attr_rmtval_remove(xfs_da_args_t *args)
{
	xfs_mount_t *mp;
	xfs_bmbt_irec_t map;
	xfs_buf_t *bp;
	xfs_daddr_t dblkno;
	xfs_dablk_t lblkno;
	int valuelen, blkcnt, nmap, error, done, committed;

	mp = args->dp->i_mount;

	/*
	 * Roll through the "value", invalidating the attribute value's
	 * blocks.
	 */
	lblkno = args->rmtblkno;
	valuelen = args->rmtblkcnt;
	while (valuelen > 0) {
		/*
		 * Try to remember where we decided to put the value.
		 */
		XFS_BMAP_INIT(args->flist, args->firstblock);
		nmap = 1;
		error = xfs_bmapi(NULL, args->dp, (xfs_fileoff_t)lblkno,
					args->rmtblkcnt,
					XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
					args->firstblock, 0, &map, &nmap,
2197
					args->flist, NULL);
L
Linus Torvalds 已提交
2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234
		if (error) {
			return(error);
		}
		ASSERT(nmap == 1);
		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
		       (map.br_startblock != HOLESTARTBLOCK));

		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);

		/*
		 * If the "remote" value is in the cache, remove it.
		 */
		bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt,
				XFS_INCORE_TRYLOCK);
		if (bp) {
			XFS_BUF_STALE(bp);
			XFS_BUF_UNDELAYWRITE(bp);
			xfs_buf_relse(bp);
			bp = NULL;
		}

		valuelen -= map.br_blockcount;

		lblkno += map.br_blockcount;
	}

	/*
	 * Keep de-allocating extents until the remote-value region is gone.
	 */
	lblkno = args->rmtblkno;
	blkcnt = args->rmtblkcnt;
	done = 0;
	while (!done) {
		XFS_BMAP_INIT(args->flist, args->firstblock);
		error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt,
				    XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
2235 2236
				    1, args->firstblock, args->flist,
				    NULL, &done);
L
Linus Torvalds 已提交
2237 2238
		if (!error) {
			error = xfs_bmap_finish(&args->trans, args->flist,
2239
						&committed);
L
Linus Torvalds 已提交
2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259
		}
		if (error) {
			ASSERT(committed);
			args->trans = NULL;
			xfs_bmap_cancel(args->flist);
			return(error);
		}

		/*
		 * bmap_finish() may have committed the last trans and started
		 * a new one.  We need the inode to be in all transactions.
		 */
		if (committed) {
			xfs_trans_ijoin(args->trans, args->dp, XFS_ILOCK_EXCL);
			xfs_trans_ihold(args->trans, args->dp);
		}

		/*
		 * Close out trans and start the next one in the chain.
		 */
2260 2261
		error = xfs_trans_roll(&args->trans, args->dp);
		if (error)
L
Linus Torvalds 已提交
2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273
			return (error);
	}
	return(0);
}

#if defined(XFS_ATTR_TRACE)
/*
 * Add a trace buffer entry for an attr_list context structure.
 */
void
xfs_attr_trace_l_c(char *where, struct xfs_attr_list_context *context)
{
E
[XFS]  
Eric Sandeen 已提交
2274
	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where, context,
L
Linus Torvalds 已提交
2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286
		(__psunsigned_t)NULL,
		(__psunsigned_t)NULL,
		(__psunsigned_t)NULL);
}

/*
 * Add a trace buffer entry for a context structure and a Btree node.
 */
void
xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context,
			 struct xfs_da_intnode *node)
{
E
[XFS]  
Eric Sandeen 已提交
2287
	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where, context,
2288
		(__psunsigned_t)be16_to_cpu(node->hdr.count),
2289
		(__psunsigned_t)be32_to_cpu(node->btree[0].hashval),
2290 2291
		(__psunsigned_t)be32_to_cpu(node->btree[
				    be16_to_cpu(node->hdr.count)-1].hashval));
L
Linus Torvalds 已提交
2292 2293 2294 2295 2296 2297 2298 2299 2300
}

/*
 * Add a trace buffer entry for a context structure and a Btree element.
 */
void
xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
			  struct xfs_da_node_entry *btree)
{
E
[XFS]  
Eric Sandeen 已提交
2301
	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where, context,
2302 2303
		(__psunsigned_t)be32_to_cpu(btree->hashval),
		(__psunsigned_t)be32_to_cpu(btree->before),
L
Linus Torvalds 已提交
2304 2305 2306 2307 2308 2309 2310 2311 2312 2313
		(__psunsigned_t)NULL);
}

/*
 * Add a trace buffer entry for a context structure and a leaf block.
 */
void
xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
			      struct xfs_attr_leafblock *leaf)
{
E
[XFS]  
Eric Sandeen 已提交
2314
	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where, context,
2315
		(__psunsigned_t)be16_to_cpu(leaf->hdr.count),
2316 2317 2318
		(__psunsigned_t)be32_to_cpu(leaf->entries[0].hashval),
		(__psunsigned_t)be32_to_cpu(leaf->entries[
				be16_to_cpu(leaf->hdr.count)-1].hashval));
L
Linus Torvalds 已提交
2319 2320 2321 2322 2323 2324 2325 2326
}

/*
 * Add a trace buffer entry for the arguments given to the routine,
 * generic form.
 */
void
xfs_attr_trace_enter(int type, char *where,
E
[XFS]  
Eric Sandeen 已提交
2327 2328 2329
			 struct xfs_attr_list_context *context,
			 __psunsigned_t a13, __psunsigned_t a14,
			 __psunsigned_t a15)
L
Linus Torvalds 已提交
2330 2331 2332
{
	ASSERT(xfs_attr_trace_buf);
	ktrace_enter(xfs_attr_trace_buf, (void *)((__psunsigned_t)type),
E
[XFS]  
Eric Sandeen 已提交
2333 2334 2335 2336 2337 2338 2339 2340 2341
		(void *)((__psunsigned_t)where),
		(void *)((__psunsigned_t)context->dp),
		(void *)((__psunsigned_t)context->cursor->hashval),
		(void *)((__psunsigned_t)context->cursor->blkno),
		(void *)((__psunsigned_t)context->cursor->offset),
		(void *)((__psunsigned_t)context->alist),
		(void *)((__psunsigned_t)context->bufsize),
		(void *)((__psunsigned_t)context->count),
		(void *)((__psunsigned_t)context->firstu),
2342
		NULL,
E
[XFS]  
Eric Sandeen 已提交
2343 2344 2345
		(void *)((__psunsigned_t)context->dupcnt),
		(void *)((__psunsigned_t)context->flags),
		(void *)a13, (void *)a14, (void *)a15);
L
Linus Torvalds 已提交
2346 2347
}
#endif	/* XFS_ATTR_TRACE */