inode.c 54.2 KB
Newer Older
1
#include <linux/ceph/ceph_debug.h>
S
Sage Weil 已提交
2 3 4 5 6 7 8 9 10 11

#include <linux/module.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/uaccess.h>
#include <linux/kernel.h>
#include <linux/namei.h>
#include <linux/writeback.h>
#include <linux/vmalloc.h>
12
#include <linux/posix_acl.h>
13
#include <linux/random.h>
S
Sage Weil 已提交
14 15

#include "super.h"
16
#include "mds_client.h"
17
#include "cache.h"
18
#include <linux/ceph/decode.h>
S
Sage Weil 已提交
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

/*
 * Ceph inode operations
 *
 * Implement basic inode helpers (get, alloc) and inode ops (getattr,
 * setattr, etc.), xattr helpers, and helpers for assimilating
 * metadata returned by the MDS into our cache.
 *
 * Also define helpers for doing asynchronous writeback, invalidation,
 * and truncation for the benefit of those who can't afford to block
 * (typically because they are in the message handler path).
 */

static const struct inode_operations ceph_symlink_iops;

34 35 36
static void ceph_invalidate_work(struct work_struct *work);
static void ceph_writeback_work(struct work_struct *work);
static void ceph_vmtruncate_work(struct work_struct *work);
S
Sage Weil 已提交
37 38 39 40

/*
 * find or create an inode, given the ceph ino number
 */
Y
Yehuda Sadeh 已提交
41 42 43 44 45 46 47
static int ceph_set_ino_cb(struct inode *inode, void *data)
{
	ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
	inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data);
	return 0;
}

S
Sage Weil 已提交
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
struct inode *ceph_get_inode(struct super_block *sb, struct ceph_vino vino)
{
	struct inode *inode;
	ino_t t = ceph_vino_to_ino(vino);

	inode = iget5_locked(sb, t, ceph_ino_compare, ceph_set_ino_cb, &vino);
	if (inode == NULL)
		return ERR_PTR(-ENOMEM);
	if (inode->i_state & I_NEW) {
		dout("get_inode created new inode %p %llx.%llx ino %llx\n",
		     inode, ceph_vinop(inode), (u64)inode->i_ino);
		unlock_new_inode(inode);
	}

	dout("get_inode on %lu=%llx.%llx got %p\n", inode->i_ino, vino.ino,
	     vino.snap, inode);
	return inode;
}

/*
 * get/constuct snapdir inode for a given directory
 */
struct inode *ceph_get_snapdir(struct inode *parent)
{
	struct ceph_vino vino = {
		.ino = ceph_ino(parent),
		.snap = CEPH_SNAPDIR,
	};
	struct inode *inode = ceph_get_inode(parent->i_sb, vino);
77
	struct ceph_inode_info *ci = ceph_inode(inode);
S
Sage Weil 已提交
78 79 80

	BUG_ON(!S_ISDIR(parent->i_mode));
	if (IS_ERR(inode))
J
Julia Lawall 已提交
81
		return inode;
S
Sage Weil 已提交
82 83 84
	inode->i_mode = parent->i_mode;
	inode->i_uid = parent->i_uid;
	inode->i_gid = parent->i_gid;
85 86
	inode->i_op = &ceph_snapdir_iops;
	inode->i_fop = &ceph_snapdir_fops;
87 88
	ci->i_snap_caps = CEPH_CAP_PIN; /* so we can open */
	ci->i_rbytes = 0;
S
Sage Weil 已提交
89 90 91 92 93 94 95 96 97 98 99
	return inode;
}

const struct inode_operations ceph_file_iops = {
	.permission = ceph_permission,
	.setattr = ceph_setattr,
	.getattr = ceph_getattr,
	.setxattr = ceph_setxattr,
	.getxattr = ceph_getxattr,
	.listxattr = ceph_listxattr,
	.removexattr = ceph_removexattr,
G
Guangliang Zhao 已提交
100
	.get_acl = ceph_get_acl,
S
Sage Weil 已提交
101
	.set_acl = ceph_set_acl,
S
Sage Weil 已提交
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
};


/*
 * We use a 'frag tree' to keep track of the MDS's directory fragments
 * for a given inode (usually there is just a single fragment).  We
 * need to know when a child frag is delegated to a new MDS, or when
 * it is flagged as replicated, so we can direct our requests
 * accordingly.
 */

/*
 * find/create a frag in the tree
 */
static struct ceph_inode_frag *__get_or_create_frag(struct ceph_inode_info *ci,
						    u32 f)
{
	struct rb_node **p;
	struct rb_node *parent = NULL;
	struct ceph_inode_frag *frag;
	int c;

	p = &ci->i_fragtree.rb_node;
	while (*p) {
		parent = *p;
		frag = rb_entry(parent, struct ceph_inode_frag, node);
		c = ceph_frag_compare(f, frag->frag);
		if (c < 0)
			p = &(*p)->rb_left;
		else if (c > 0)
			p = &(*p)->rb_right;
		else
			return frag;
	}

	frag = kmalloc(sizeof(*frag), GFP_NOFS);
	if (!frag) {
		pr_err("__get_or_create_frag ENOMEM on %p %llx.%llx "
		       "frag %x\n", &ci->vfs_inode,
		       ceph_vinop(&ci->vfs_inode), f);
		return ERR_PTR(-ENOMEM);
	}
	frag->frag = f;
	frag->split_by = 0;
	frag->mds = -1;
	frag->ndist = 0;

	rb_link_node(&frag->node, parent, p);
	rb_insert_color(&frag->node, &ci->i_fragtree);

	dout("get_or_create_frag added %llx.%llx frag %x\n",
	     ceph_vinop(&ci->vfs_inode), f);
	return frag;
}

/*
 * find a specific frag @f
 */
struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci, u32 f)
{
	struct rb_node *n = ci->i_fragtree.rb_node;

	while (n) {
		struct ceph_inode_frag *frag =
			rb_entry(n, struct ceph_inode_frag, node);
		int c = ceph_frag_compare(f, frag->frag);
		if (c < 0)
			n = n->rb_left;
		else if (c > 0)
			n = n->rb_right;
		else
			return frag;
	}
	return NULL;
}

/*
 * Choose frag containing the given value @v.  If @pfrag is
 * specified, copy the frag delegation info to the caller if
 * it is present.
 */
183 184
static u32 __ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
			      struct ceph_inode_frag *pfrag, int *found)
S
Sage Weil 已提交
185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
{
	u32 t = ceph_frag_make(0, 0);
	struct ceph_inode_frag *frag;
	unsigned nway, i;
	u32 n;

	if (found)
		*found = 0;

	while (1) {
		WARN_ON(!ceph_frag_contains_value(t, v));
		frag = __ceph_find_frag(ci, t);
		if (!frag)
			break; /* t is a leaf */
		if (frag->split_by == 0) {
			if (pfrag)
				memcpy(pfrag, frag, sizeof(*pfrag));
			if (found)
				*found = 1;
			break;
		}

		/* choose child */
		nway = 1 << frag->split_by;
		dout("choose_frag(%x) %x splits by %d (%d ways)\n", v, t,
		     frag->split_by, nway);
		for (i = 0; i < nway; i++) {
			n = ceph_frag_make_child(t, frag->split_by, i);
			if (ceph_frag_contains_value(n, v)) {
				t = n;
				break;
			}
		}
		BUG_ON(i == nway);
	}
	dout("choose_frag(%x) = %x\n", v, t);

	return t;
}

225 226 227 228 229 230 231 232 233 234
u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
		     struct ceph_inode_frag *pfrag, int *found)
{
	u32 ret;
	mutex_lock(&ci->i_fragtree_mutex);
	ret = __ceph_choose_frag(ci, v, pfrag, found);
	mutex_unlock(&ci->i_fragtree_mutex);
	return ret;
}

S
Sage Weil 已提交
235 236 237 238 239 240 241 242 243 244 245 246 247
/*
 * Process dirfrag (delegation) info from the mds.  Include leaf
 * fragment in tree ONLY if ndist > 0.  Otherwise, only
 * branches/splits are included in i_fragtree)
 */
static int ceph_fill_dirfrag(struct inode *inode,
			     struct ceph_mds_reply_dirfrag *dirinfo)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_inode_frag *frag;
	u32 id = le32_to_cpu(dirinfo->frag);
	int mds = le32_to_cpu(dirinfo->auth);
	int ndist = le32_to_cpu(dirinfo->ndist);
248
	int diri_auth = -1;
S
Sage Weil 已提交
249 250 251
	int i;
	int err = 0;

252 253 254 255 256
	spin_lock(&ci->i_ceph_lock);
	if (ci->i_auth_cap)
		diri_auth = ci->i_auth_cap->mds;
	spin_unlock(&ci->i_ceph_lock);

S
Sage Weil 已提交
257
	mutex_lock(&ci->i_fragtree_mutex);
258
	if (ndist == 0 && mds == diri_auth) {
S
Sage Weil 已提交
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302
		/* no delegation info needed. */
		frag = __ceph_find_frag(ci, id);
		if (!frag)
			goto out;
		if (frag->split_by == 0) {
			/* tree leaf, remove */
			dout("fill_dirfrag removed %llx.%llx frag %x"
			     " (no ref)\n", ceph_vinop(inode), id);
			rb_erase(&frag->node, &ci->i_fragtree);
			kfree(frag);
		} else {
			/* tree branch, keep and clear */
			dout("fill_dirfrag cleared %llx.%llx frag %x"
			     " referral\n", ceph_vinop(inode), id);
			frag->mds = -1;
			frag->ndist = 0;
		}
		goto out;
	}


	/* find/add this frag to store mds delegation info */
	frag = __get_or_create_frag(ci, id);
	if (IS_ERR(frag)) {
		/* this is not the end of the world; we can continue
		   with bad/inaccurate delegation info */
		pr_err("fill_dirfrag ENOMEM on mds ref %llx.%llx fg %x\n",
		       ceph_vinop(inode), le32_to_cpu(dirinfo->frag));
		err = -ENOMEM;
		goto out;
	}

	frag->mds = mds;
	frag->ndist = min_t(u32, ndist, CEPH_MAX_DIRFRAG_REP);
	for (i = 0; i < frag->ndist; i++)
		frag->dist[i] = le32_to_cpu(dirinfo->dist[i]);
	dout("fill_dirfrag %llx.%llx frag %x ndist=%d\n",
	     ceph_vinop(inode), frag->frag, frag->ndist);

out:
	mutex_unlock(&ci->i_fragtree_mutex);
	return err;
}

303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
static int ceph_fill_fragtree(struct inode *inode,
			      struct ceph_frag_tree_head *fragtree,
			      struct ceph_mds_reply_dirfrag *dirinfo)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_inode_frag *frag;
	struct rb_node *rb_node;
	int i;
	u32 id, nsplits;
	bool update = false;

	mutex_lock(&ci->i_fragtree_mutex);
	nsplits = le32_to_cpu(fragtree->nsplits);
	if (nsplits) {
		i = prandom_u32() % nsplits;
		id = le32_to_cpu(fragtree->splits[i].frag);
		if (!__ceph_find_frag(ci, id))
			update = true;
	} else if (!RB_EMPTY_ROOT(&ci->i_fragtree)) {
		rb_node = rb_first(&ci->i_fragtree);
		frag = rb_entry(rb_node, struct ceph_inode_frag, node);
		if (frag->frag != ceph_frag_make(0, 0) || rb_next(rb_node))
			update = true;
	}
	if (!update && dirinfo) {
		id = le32_to_cpu(dirinfo->frag);
		if (id != __ceph_choose_frag(ci, id, NULL, NULL))
			update = true;
	}
	if (!update)
		goto out_unlock;

	dout("fill_fragtree %llx.%llx\n", ceph_vinop(inode));
	rb_node = rb_first(&ci->i_fragtree);
	for (i = 0; i < nsplits; i++) {
		id = le32_to_cpu(fragtree->splits[i].frag);
		frag = NULL;
		while (rb_node) {
			frag = rb_entry(rb_node, struct ceph_inode_frag, node);
			if (ceph_frag_compare(frag->frag, id) >= 0) {
				if (frag->frag != id)
					frag = NULL;
				else
					rb_node = rb_next(rb_node);
				break;
			}
			rb_node = rb_next(rb_node);
			rb_erase(&frag->node, &ci->i_fragtree);
			kfree(frag);
			frag = NULL;
		}
		if (!frag) {
			frag = __get_or_create_frag(ci, id);
			if (IS_ERR(frag))
				continue;
		}
		frag->split_by = le32_to_cpu(fragtree->splits[i].by);
		dout(" frag %x split by %d\n", frag->frag, frag->split_by);
	}
	while (rb_node) {
		frag = rb_entry(rb_node, struct ceph_inode_frag, node);
		rb_node = rb_next(rb_node);
		rb_erase(&frag->node, &ci->i_fragtree);
		kfree(frag);
	}
out_unlock:
	mutex_unlock(&ci->i_fragtree_mutex);
	return 0;
}
S
Sage Weil 已提交
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386

/*
 * initialize a newly allocated inode.
 */
struct inode *ceph_alloc_inode(struct super_block *sb)
{
	struct ceph_inode_info *ci;
	int i;

	ci = kmem_cache_alloc(ceph_inode_cachep, GFP_NOFS);
	if (!ci)
		return NULL;

	dout("alloc_inode %p\n", &ci->vfs_inode);

387 388
	spin_lock_init(&ci->i_ceph_lock);

S
Sage Weil 已提交
389
	ci->i_version = 0;
Y
Yan, Zheng 已提交
390
	ci->i_inline_version = 0;
S
Sage Weil 已提交
391 392
	ci->i_time_warp_seq = 0;
	ci->i_ceph_flags = 0;
393
	ci->i_ordered_count = 0;
394 395
	atomic_set(&ci->i_release_count, 1);
	atomic_set(&ci->i_complete_count, 0);
S
Sage Weil 已提交
396 397
	ci->i_symlink = NULL;

S
Sage Weil 已提交
398 399
	memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout));

S
Sage Weil 已提交
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
	ci->i_fragtree = RB_ROOT;
	mutex_init(&ci->i_fragtree_mutex);

	ci->i_xattrs.blob = NULL;
	ci->i_xattrs.prealloc_blob = NULL;
	ci->i_xattrs.dirty = false;
	ci->i_xattrs.index = RB_ROOT;
	ci->i_xattrs.count = 0;
	ci->i_xattrs.names_size = 0;
	ci->i_xattrs.vals_size = 0;
	ci->i_xattrs.version = 0;
	ci->i_xattrs.index_version = 0;

	ci->i_caps = RB_ROOT;
	ci->i_auth_cap = NULL;
	ci->i_dirty_caps = 0;
	ci->i_flushing_caps = 0;
	INIT_LIST_HEAD(&ci->i_dirty_item);
	INIT_LIST_HEAD(&ci->i_flushing_item);
	ci->i_cap_flush_seq = 0;
	ci->i_cap_flush_last_tid = 0;
	memset(&ci->i_cap_flush_tid, 0, sizeof(ci->i_cap_flush_tid));
	init_waitqueue_head(&ci->i_cap_wq);
	ci->i_hold_caps_min = 0;
	ci->i_hold_caps_max = 0;
	INIT_LIST_HEAD(&ci->i_cap_delay_list);
	INIT_LIST_HEAD(&ci->i_cap_snaps);
	ci->i_head_snapc = NULL;
	ci->i_snap_caps = 0;

	for (i = 0; i < CEPH_FILE_MODE_NUM; i++)
		ci->i_nr_by_mode[i] = 0;

Y
Yan, Zheng 已提交
433
	mutex_init(&ci->i_truncate_mutex);
S
Sage Weil 已提交
434 435 436 437 438 439 440 441 442 443 444 445 446
	ci->i_truncate_seq = 0;
	ci->i_truncate_size = 0;
	ci->i_truncate_pending = 0;

	ci->i_max_size = 0;
	ci->i_reported_size = 0;
	ci->i_wanted_max_size = 0;
	ci->i_requested_max_size = 0;

	ci->i_pin_ref = 0;
	ci->i_rd_ref = 0;
	ci->i_rdcache_ref = 0;
	ci->i_wr_ref = 0;
447
	ci->i_wb_ref = 0;
S
Sage Weil 已提交
448 449 450 451 452 453 454 455 456 457 458 459 460 461
	ci->i_wrbuffer_ref = 0;
	ci->i_wrbuffer_ref_head = 0;
	ci->i_shared_gen = 0;
	ci->i_rdcache_gen = 0;
	ci->i_rdcache_revoking = 0;

	INIT_LIST_HEAD(&ci->i_unsafe_writes);
	INIT_LIST_HEAD(&ci->i_unsafe_dirops);
	spin_lock_init(&ci->i_unsafe_lock);

	ci->i_snap_realm = NULL;
	INIT_LIST_HEAD(&ci->i_snap_realm_item);
	INIT_LIST_HEAD(&ci->i_snap_flush_item);

462 463
	INIT_WORK(&ci->i_wb_work, ceph_writeback_work);
	INIT_WORK(&ci->i_pg_inv_work, ceph_invalidate_work);
S
Sage Weil 已提交
464 465 466

	INIT_WORK(&ci->i_vmtruncate_work, ceph_vmtruncate_work);

467 468
	ceph_fscache_inode_init(ci);

S
Sage Weil 已提交
469 470 471
	return &ci->vfs_inode;
}

N
Nick Piggin 已提交
472 473 474 475 476 477 478 479
static void ceph_i_callback(struct rcu_head *head)
{
	struct inode *inode = container_of(head, struct inode, i_rcu);
	struct ceph_inode_info *ci = ceph_inode(inode);

	kmem_cache_free(ceph_inode_cachep, ci);
}

S
Sage Weil 已提交
480 481 482 483 484 485 486 487
void ceph_destroy_inode(struct inode *inode)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	struct ceph_inode_frag *frag;
	struct rb_node *n;

	dout("destroy_inode %p ino %llx.%llx\n", inode, ceph_vinop(inode));

488 489
	ceph_fscache_unregister_inode_cookie(ci);

S
Sage Weil 已提交
490 491
	ceph_queue_caps_release(inode);

492 493
	/*
	 * we may still have a snap_realm reference if there are stray
494
	 * caps in i_snap_caps.
495 496 497
	 */
	if (ci->i_snap_realm) {
		struct ceph_mds_client *mdsc =
498
			ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
499 500 501 502 503 504 505 506 507
		struct ceph_snap_realm *realm = ci->i_snap_realm;

		dout(" dropping residual ref to snap realm %p\n", realm);
		spin_lock(&realm->inodes_with_caps_lock);
		list_del_init(&ci->i_snap_realm_item);
		spin_unlock(&realm->inodes_with_caps_lock);
		ceph_put_snap_realm(mdsc, realm);
	}

S
Sage Weil 已提交
508 509 510 511 512 513 514 515
	kfree(ci->i_symlink);
	while ((n = rb_first(&ci->i_fragtree)) != NULL) {
		frag = rb_entry(n, struct ceph_inode_frag, node);
		rb_erase(n, &ci->i_fragtree);
		kfree(frag);
	}

	__ceph_destroy_xattrs(ci);
S
Sage Weil 已提交
516 517 518 519
	if (ci->i_xattrs.blob)
		ceph_buffer_put(ci->i_xattrs.blob);
	if (ci->i_xattrs.prealloc_blob)
		ceph_buffer_put(ci->i_xattrs.prealloc_blob);
S
Sage Weil 已提交
520

N
Nick Piggin 已提交
521
	call_rcu(&inode->i_rcu, ceph_i_callback);
S
Sage Weil 已提交
522 523
}

Y
Yan, Zheng 已提交
524 525 526 527 528 529 530 531 532 533
int ceph_drop_inode(struct inode *inode)
{
	/*
	 * Positve dentry and corresponding inode are always accompanied
	 * in MDS reply. So no need to keep inode in the cache after
	 * dropping all its aliases.
	 */
	return 1;
}

S
Sage Weil 已提交
534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557
/*
 * Helpers to fill in size, ctime, mtime, and atime.  We have to be
 * careful because either the client or MDS may have more up to date
 * info, depending on which capabilities are held, and whether
 * time_warp_seq or truncate_seq have increased.  (Ordinarily, mtime
 * and size are monotonically increasing, except when utimes() or
 * truncate() increments the corresponding _seq values.)
 */
int ceph_fill_file_size(struct inode *inode, int issued,
			u32 truncate_seq, u64 truncate_size, u64 size)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int queue_trunc = 0;

	if (ceph_seq_cmp(truncate_seq, ci->i_truncate_seq) > 0 ||
	    (truncate_seq == ci->i_truncate_seq && size > inode->i_size)) {
		dout("size %lld -> %llu\n", inode->i_size, size);
		inode->i_size = size;
		inode->i_blocks = (size + (1<<9) - 1) >> 9;
		ci->i_reported_size = size;
		if (truncate_seq != ci->i_truncate_seq) {
			dout("truncate_seq %u -> %u\n",
			     ci->i_truncate_seq, truncate_seq);
			ci->i_truncate_seq = truncate_seq;
Y
Yan, Zheng 已提交
558 559 560 561 562 563

			/* the MDS should have revoked these caps */
			WARN_ON_ONCE(issued & (CEPH_CAP_FILE_EXCL |
					       CEPH_CAP_FILE_RD |
					       CEPH_CAP_FILE_WR |
					       CEPH_CAP_FILE_LAZYIO));
564 565 566 567 568 569
			/*
			 * If we hold relevant caps, or in the case where we're
			 * not the only client referencing this file and we
			 * don't hold those caps, then we need to check whether
			 * the file is either opened or mmaped
			 */
Y
Yan, Zheng 已提交
570 571
			if ((issued & (CEPH_CAP_FILE_CACHE|
				       CEPH_CAP_FILE_BUFFER)) ||
572 573
			    mapping_mapped(inode->i_mapping) ||
			    __ceph_caps_file_wanted(ci)) {
S
Sage Weil 已提交
574 575 576 577 578 579 580 581 582 583 584
				ci->i_truncate_pending++;
				queue_trunc = 1;
			}
		}
	}
	if (ceph_seq_cmp(truncate_seq, ci->i_truncate_seq) >= 0 &&
	    ci->i_truncate_size != truncate_size) {
		dout("truncate_size %lld -> %llu\n", ci->i_truncate_size,
		     truncate_size);
		ci->i_truncate_size = truncate_size;
	}
585 586 587 588

	if (queue_trunc)
		ceph_fscache_invalidate(inode);

S
Sage Weil 已提交
589 590 591 592 593 594 595 596 597 598 599 600
	return queue_trunc;
}

void ceph_fill_file_time(struct inode *inode, int issued,
			 u64 time_warp_seq, struct timespec *ctime,
			 struct timespec *mtime, struct timespec *atime)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int warn = 0;

	if (issued & (CEPH_CAP_FILE_EXCL|
		      CEPH_CAP_FILE_WR|
S
Sage Weil 已提交
601 602 603
		      CEPH_CAP_FILE_BUFFER|
		      CEPH_CAP_AUTH_EXCL|
		      CEPH_CAP_XATTR_EXCL)) {
S
Sage Weil 已提交
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
		if (timespec_compare(ctime, &inode->i_ctime) > 0) {
			dout("ctime %ld.%09ld -> %ld.%09ld inc w/ cap\n",
			     inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
			     ctime->tv_sec, ctime->tv_nsec);
			inode->i_ctime = *ctime;
		}
		if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
			/* the MDS did a utimes() */
			dout("mtime %ld.%09ld -> %ld.%09ld "
			     "tw %d -> %d\n",
			     inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
			     mtime->tv_sec, mtime->tv_nsec,
			     ci->i_time_warp_seq, (int)time_warp_seq);

			inode->i_mtime = *mtime;
			inode->i_atime = *atime;
			ci->i_time_warp_seq = time_warp_seq;
		} else if (time_warp_seq == ci->i_time_warp_seq) {
			/* nobody did utimes(); take the max */
			if (timespec_compare(mtime, &inode->i_mtime) > 0) {
				dout("mtime %ld.%09ld -> %ld.%09ld inc\n",
				     inode->i_mtime.tv_sec,
				     inode->i_mtime.tv_nsec,
				     mtime->tv_sec, mtime->tv_nsec);
				inode->i_mtime = *mtime;
			}
			if (timespec_compare(atime, &inode->i_atime) > 0) {
				dout("atime %ld.%09ld -> %ld.%09ld inc\n",
				     inode->i_atime.tv_sec,
				     inode->i_atime.tv_nsec,
				     atime->tv_sec, atime->tv_nsec);
				inode->i_atime = *atime;
			}
		} else if (issued & CEPH_CAP_FILE_EXCL) {
			/* we did a utimes(); ignore mds values */
		} else {
			warn = 1;
		}
	} else {
S
Sage Weil 已提交
643
		/* we have no write|excl caps; whatever the MDS says is true */
S
Sage Weil 已提交
644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
		if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
			inode->i_ctime = *ctime;
			inode->i_mtime = *mtime;
			inode->i_atime = *atime;
			ci->i_time_warp_seq = time_warp_seq;
		} else {
			warn = 1;
		}
	}
	if (warn) /* time_warp_seq shouldn't go backwards */
		dout("%p mds time_warp_seq %llu < %u\n",
		     inode, time_warp_seq, ci->i_time_warp_seq);
}

/*
 * Populate an inode based on info from mds.  May be called on new or
 * existing inodes.
 */
662
static int fill_inode(struct inode *inode, struct page *locked_page,
S
Sage Weil 已提交
663 664 665 666 667 668
		      struct ceph_mds_reply_info_in *iinfo,
		      struct ceph_mds_reply_dirfrag *dirinfo,
		      struct ceph_mds_session *session,
		      unsigned long ttl_from, int cap_fmode,
		      struct ceph_cap_reservation *caps_reservation)
{
669
	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
S
Sage Weil 已提交
670 671
	struct ceph_mds_reply_inode *info = iinfo->in;
	struct ceph_inode_info *ci = ceph_inode(inode);
672
	int issued = 0, implemented, new_issued;
S
Sage Weil 已提交
673 674
	struct timespec mtime, atime, ctime;
	struct ceph_buffer *xattr_blob = NULL;
675
	struct ceph_cap *new_cap = NULL;
S
Sage Weil 已提交
676
	int err = 0;
677
	bool wake = false;
678 679
	bool queue_trunc = false;
	bool new_version = false;
Y
Yan, Zheng 已提交
680
	bool fill_inline = false;
S
Sage Weil 已提交
681 682 683 684 685

	dout("fill_inode %p ino %llx.%llx v %llu had %llu\n",
	     inode, ceph_vinop(inode), le64_to_cpu(info->version),
	     ci->i_version);

686 687 688 689
	/* prealloc new cap struct */
	if (info->cap.caps && ceph_snap(inode) == CEPH_NOSNAP)
		new_cap = ceph_get_cap(mdsc, caps_reservation);

S
Sage Weil 已提交
690 691 692 693 694 695
	/*
	 * prealloc xattr data, if it looks like we'll need it.  only
	 * if len > 4 (meaning there are actually xattrs; the first 4
	 * bytes are the xattr count).
	 */
	if (iinfo->xattr_len > 4) {
S
Sage Weil 已提交
696
		xattr_blob = ceph_buffer_new(iinfo->xattr_len, GFP_NOFS);
S
Sage Weil 已提交
697 698 699 700 701
		if (!xattr_blob)
			pr_err("fill_inode ENOMEM xattr blob %d bytes\n",
			       iinfo->xattr_len);
	}

702
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
703 704 705

	/*
	 * provided version will be odd if inode value is projected,
706 707 708 709 710 711 712 713
	 * even if stable.  skip the update if we have newer stable
	 * info (ours>=theirs, e.g. due to racing mds replies), unless
	 * we are getting projected (unstable) info (in which case the
	 * version is odd, and we want ours>theirs).
	 *   us   them
	 *   2    2     skip
	 *   3    2     skip
	 *   3    3     update
S
Sage Weil 已提交
714
	 */
715 716 717 718 719
	if (ci->i_version == 0 ||
	    ((info->cap.flags & CEPH_CAP_FLAG_AUTH) &&
	     le64_to_cpu(info->version) > (ci->i_version & ~1)))
		new_version = true;

S
Sage Weil 已提交
720 721
	issued = __ceph_caps_issued(ci, &implemented);
	issued |= implemented | __ceph_caps_dirty(ci);
722
	new_issued = ~issued & le32_to_cpu(info->cap.caps);
S
Sage Weil 已提交
723 724 725 726 727

	/* update inode */
	ci->i_version = le64_to_cpu(info->version);
	inode->i_version++;
	inode->i_rdev = le32_to_cpu(info->rdev);
728
	inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
S
Sage Weil 已提交
729

730 731
	if ((new_version || (new_issued & CEPH_CAP_AUTH_SHARED)) &&
	    (issued & CEPH_CAP_AUTH_EXCL) == 0) {
S
Sage Weil 已提交
732
		inode->i_mode = le32_to_cpu(info->mode);
733 734
		inode->i_uid = make_kuid(&init_user_ns, le32_to_cpu(info->uid));
		inode->i_gid = make_kgid(&init_user_ns, le32_to_cpu(info->gid));
S
Sage Weil 已提交
735
		dout("%p mode 0%o uid.gid %d.%d\n", inode, inode->i_mode,
736 737
		     from_kuid(&init_user_ns, inode->i_uid),
		     from_kgid(&init_user_ns, inode->i_gid));
S
Sage Weil 已提交
738 739
	}

740 741
	if ((new_version || (new_issued & CEPH_CAP_LINK_SHARED)) &&
	    (issued & CEPH_CAP_LINK_EXCL) == 0)
M
Miklos Szeredi 已提交
742
		set_nlink(inode, le32_to_cpu(info->nlink));
S
Sage Weil 已提交
743

744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768
	if (new_version || (new_issued & CEPH_CAP_ANY_RD)) {
		/* be careful with mtime, atime, size */
		ceph_decode_timespec(&atime, &info->atime);
		ceph_decode_timespec(&mtime, &info->mtime);
		ceph_decode_timespec(&ctime, &info->ctime);
		ceph_fill_file_time(inode, issued,
				le32_to_cpu(info->time_warp_seq),
				&ctime, &mtime, &atime);
	}

	if (new_version ||
	    (new_issued & (CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR))) {
		ci->i_layout = info->layout;
		queue_trunc = ceph_fill_file_size(inode, issued,
					le32_to_cpu(info->truncate_seq),
					le64_to_cpu(info->truncate_size),
					le64_to_cpu(info->size));
		/* only update max_size on auth cap */
		if ((info->cap.flags & CEPH_CAP_FLAG_AUTH) &&
		    ci->i_max_size != le64_to_cpu(info->max_size)) {
			dout("max_size %lld -> %llu\n", ci->i_max_size,
					le64_to_cpu(info->max_size));
			ci->i_max_size = le64_to_cpu(info->max_size);
		}
	}
S
Sage Weil 已提交
769 770 771

	/* xattrs */
	/* note that if i_xattrs.len <= 4, i_xattrs.data will still be NULL. */
772
	if ((ci->i_xattrs.version == 0 || !(issued & CEPH_CAP_XATTR_EXCL))  &&
S
Sage Weil 已提交
773 774 775 776 777 778 779 780
	    le64_to_cpu(info->xattr_version) > ci->i_xattrs.version) {
		if (ci->i_xattrs.blob)
			ceph_buffer_put(ci->i_xattrs.blob);
		ci->i_xattrs.blob = xattr_blob;
		if (xattr_blob)
			memcpy(ci->i_xattrs.blob->vec.iov_base,
			       iinfo->xattr_data, iinfo->xattr_len);
		ci->i_xattrs.version = le64_to_cpu(info->xattr_version);
G
Guangliang Zhao 已提交
781
		ceph_forget_all_cached_acls(inode);
782
		xattr_blob = NULL;
S
Sage Weil 已提交
783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
	}

	inode->i_mapping->a_ops = &ceph_aops;

	switch (inode->i_mode & S_IFMT) {
	case S_IFIFO:
	case S_IFBLK:
	case S_IFCHR:
	case S_IFSOCK:
		init_special_inode(inode, inode->i_mode, inode->i_rdev);
		inode->i_op = &ceph_file_iops;
		break;
	case S_IFREG:
		inode->i_op = &ceph_file_iops;
		inode->i_fop = &ceph_file_fops;
		break;
	case S_IFLNK:
		inode->i_op = &ceph_symlink_iops;
		if (!ci->i_symlink) {
802
			u32 symlen = iinfo->symlink_len;
S
Sage Weil 已提交
803 804
			char *sym;

805
			spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
806

807 808 809 810
			err = -EINVAL;
			if (WARN_ON(symlen != inode->i_size))
				goto out;

S
Sage Weil 已提交
811
			err = -ENOMEM;
812
			sym = kstrndup(iinfo->symlink, symlen, GFP_NOFS);
S
Sage Weil 已提交
813 814 815
			if (!sym)
				goto out;

816
			spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
817 818 819 820 821 822 823 824 825 826
			if (!ci->i_symlink)
				ci->i_symlink = sym;
			else
				kfree(sym); /* lost a race */
		}
		break;
	case S_IFDIR:
		inode->i_op = &ceph_dir_iops;
		inode->i_fop = &ceph_dir_fops;

827 828
		ci->i_dir_layout = iinfo->dir_layout;

S
Sage Weil 已提交
829 830 831 832 833 834 835 836 837 838 839 840 841 842 843
		ci->i_files = le64_to_cpu(info->files);
		ci->i_subdirs = le64_to_cpu(info->subdirs);
		ci->i_rbytes = le64_to_cpu(info->rbytes);
		ci->i_rfiles = le64_to_cpu(info->rfiles);
		ci->i_rsubdirs = le64_to_cpu(info->rsubdirs);
		ceph_decode_timespec(&ci->i_rctime, &info->rctime);
		break;
	default:
		pr_err("fill_inode %llx.%llx BAD mode 0%o\n",
		       ceph_vinop(inode), inode->i_mode);
	}

	/* were we issued a capability? */
	if (info->cap.caps) {
		if (ceph_snap(inode) == CEPH_NOSNAP) {
844
			unsigned caps = le32_to_cpu(info->cap.caps);
S
Sage Weil 已提交
845 846
			ceph_add_cap(inode, session,
				     le64_to_cpu(info->cap.cap_id),
847
				     cap_fmode, caps,
S
Sage Weil 已提交
848 849 850 851
				     le32_to_cpu(info->cap.wanted),
				     le32_to_cpu(info->cap.seq),
				     le32_to_cpu(info->cap.mseq),
				     le64_to_cpu(info->cap.realm),
852
				     info->cap.flags, &new_cap);
853 854 855 856 857 858 859 860 861 862 863 864 865

			/* set dir completion flag? */
			if (S_ISDIR(inode->i_mode) &&
			    ci->i_files == 0 && ci->i_subdirs == 0 &&
			    (caps & CEPH_CAP_FILE_SHARED) &&
			    (issued & CEPH_CAP_FILE_EXCL) == 0 &&
			    !__ceph_dir_is_complete(ci)) {
				dout(" marking %p complete (empty)\n", inode);
				__ceph_dir_set_complete(ci,
					atomic_read(&ci->i_release_count),
					ci->i_ordered_count);
			}

866
			wake = true;
S
Sage Weil 已提交
867 868 869 870 871 872 873
		} else {
			dout(" %p got snap_caps %s\n", inode,
			     ceph_cap_string(le32_to_cpu(info->cap.caps)));
			ci->i_snap_caps |= le32_to_cpu(info->cap.caps);
			if (cap_fmode >= 0)
				__ceph_get_fmode(ci, cap_fmode);
		}
874
	} else if (cap_fmode >= 0) {
875
		pr_warn("mds issued no caps on %llx.%llx\n",
876 877
			   ceph_vinop(inode));
		__ceph_get_fmode(ci, cap_fmode);
S
Sage Weil 已提交
878
	}
Y
Yan, Zheng 已提交
879 880 881 882 883 884

	if (iinfo->inline_version > 0 &&
	    iinfo->inline_version >= ci->i_inline_version) {
		int cache_caps = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO;
		ci->i_inline_version = iinfo->inline_version;
		if (ci->i_inline_version != CEPH_INLINE_NONE &&
885 886
		    (locked_page ||
		     (le32_to_cpu(info->cap.caps) & cache_caps)))
Y
Yan, Zheng 已提交
887 888 889
			fill_inline = true;
	}

890
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
891

Y
Yan, Zheng 已提交
892
	if (fill_inline)
893
		ceph_fill_inline_data(inode, locked_page,
Y
Yan, Zheng 已提交
894 895
				      iinfo->inline_data, iinfo->inline_len);

896 897 898
	if (wake)
		wake_up_all(&ci->i_cap_wq);

S
Sage Weil 已提交
899 900
	/* queue truncate if we saw i_size decrease */
	if (queue_trunc)
901
		ceph_queue_vmtruncate(inode);
S
Sage Weil 已提交
902 903

	/* populate frag tree */
904 905
	if (S_ISDIR(inode->i_mode))
		ceph_fill_fragtree(inode, &info->fragtree, dirinfo);
S
Sage Weil 已提交
906 907 908 909 910 911 912

	/* update delegation info? */
	if (dirinfo)
		ceph_fill_dirfrag(inode, dirinfo);

	err = 0;
out:
913 914
	if (new_cap)
		ceph_put_cap(mdsc, new_cap);
S
Sage Weil 已提交
915 916
	if (xattr_blob)
		ceph_buffer_put(xattr_blob);
S
Sage Weil 已提交
917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938
	return err;
}

/*
 * caller should hold session s_mutex.
 */
static void update_dentry_lease(struct dentry *dentry,
				struct ceph_mds_reply_lease *lease,
				struct ceph_mds_session *session,
				unsigned long from_time)
{
	struct ceph_dentry_info *di = ceph_dentry(dentry);
	long unsigned duration = le32_to_cpu(lease->duration_ms);
	long unsigned ttl = from_time + (duration * HZ) / 1000;
	long unsigned half_ttl = from_time + (duration * HZ / 2) / 1000;
	struct inode *dir;

	/* only track leases on regular dentries */
	if (dentry->d_op != &ceph_dentry_ops)
		return;

	spin_lock(&dentry->d_lock);
S
Sage Weil 已提交
939 940
	dout("update_dentry_lease %p duration %lu ms ttl %lu\n",
	     dentry, duration, ttl);
S
Sage Weil 已提交
941 942

	/* make lease_rdcache_gen match directory */
943
	dir = d_inode(dentry->d_parent);
S
Sage Weil 已提交
944 945
	di->lease_shared_gen = ceph_inode(dir)->i_shared_gen;

S
Sage Weil 已提交
946
	if (duration == 0)
S
Sage Weil 已提交
947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978
		goto out_unlock;

	if (di->lease_gen == session->s_cap_gen &&
	    time_before(ttl, dentry->d_time))
		goto out_unlock;  /* we already have a newer lease. */

	if (di->lease_session && di->lease_session != session)
		goto out_unlock;

	ceph_dentry_lru_touch(dentry);

	if (!di->lease_session)
		di->lease_session = ceph_get_mds_session(session);
	di->lease_gen = session->s_cap_gen;
	di->lease_seq = le32_to_cpu(lease->seq);
	di->lease_renew_after = half_ttl;
	di->lease_renew_from = 0;
	dentry->d_time = ttl;
out_unlock:
	spin_unlock(&dentry->d_lock);
	return;
}

/*
 * splice a dentry to an inode.
 * caller must hold directory i_mutex for this to be safe.
 *
 * we will only rehash the resulting dentry if @prehash is
 * true; @prehash will be set to false (for the benefit of
 * the caller) if we fail.
 */
static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
979
				    bool *prehash)
S
Sage Weil 已提交
980 981 982
{
	struct dentry *realdn;

983
	BUG_ON(d_inode(dn));
S
Sage Weil 已提交
984

S
Sage Weil 已提交
985 986 987
	/* dn must be unhashed */
	if (!d_unhashed(dn))
		d_drop(dn);
988
	realdn = d_splice_alias(in, dn);
S
Sage Weil 已提交
989
	if (IS_ERR(realdn)) {
990 991
		pr_err("splice_dentry error %ld %p inode %p ino %llx.%llx\n",
		       PTR_ERR(realdn), dn, in, ceph_vinop(in));
S
Sage Weil 已提交
992 993 994 995 996 997 998
		if (prehash)
			*prehash = false; /* don't rehash on error */
		dn = realdn; /* note realdn contains the error */
		goto out;
	} else if (realdn) {
		dout("dn %p (%d) spliced with %p (%d) "
		     "inode %p ino %llx.%llx\n",
A
Al Viro 已提交
999 1000
		     dn, d_count(dn),
		     realdn, d_count(realdn),
1001
		     d_inode(realdn), ceph_vinop(d_inode(realdn)));
S
Sage Weil 已提交
1002 1003 1004 1005 1006
		dput(dn);
		dn = realdn;
	} else {
		BUG_ON(!ceph_dentry(dn));
		dout("dn %p attached to %p ino %llx.%llx\n",
1007
		     dn, d_inode(dn), ceph_vinop(d_inode(dn)));
S
Sage Weil 已提交
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031
	}
	if ((!prehash || *prehash) && d_unhashed(dn))
		d_rehash(dn);
out:
	return dn;
}

/*
 * Incorporate results into the local cache.  This is either just
 * one inode, or a directory, dentry, and possibly linked-to inode (e.g.,
 * after a lookup).
 *
 * A reply may contain
 *         a directory inode along with a dentry.
 *  and/or a target inode
 *
 * Called with snap_rwsem (read).
 */
int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
		    struct ceph_mds_session *session)
{
	struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
	struct inode *in = NULL;
	struct ceph_vino vino;
1032
	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
S
Sage Weil 已提交
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073
	int err = 0;

	dout("fill_trace %p is_dentry %d is_target %d\n", req,
	     rinfo->head->is_dentry, rinfo->head->is_target);

#if 0
	/*
	 * Debugging hook:
	 *
	 * If we resend completed ops to a recovering mds, we get no
	 * trace.  Since that is very rare, pretend this is the case
	 * to ensure the 'no trace' handlers in the callers behave.
	 *
	 * Fill in inodes unconditionally to avoid breaking cap
	 * invariants.
	 */
	if (rinfo->head->op & CEPH_MDS_OP_WRITE) {
		pr_info("fill_trace faking empty trace on %lld %s\n",
			req->r_tid, ceph_mds_op_name(rinfo->head->op));
		if (rinfo->head->is_dentry) {
			rinfo->head->is_dentry = 0;
			err = fill_inode(req->r_locked_dir,
					 &rinfo->diri, rinfo->dirfrag,
					 session, req->r_request_started, -1);
		}
		if (rinfo->head->is_target) {
			rinfo->head->is_target = 0;
			ininfo = rinfo->targeti.in;
			vino.ino = le64_to_cpu(ininfo->ino);
			vino.snap = le64_to_cpu(ininfo->snapid);
			in = ceph_get_inode(sb, vino);
			err = fill_inode(in, &rinfo->targeti, NULL,
					 session, req->r_request_started,
					 req->r_fmode);
			iput(in);
		}
	}
#endif

	if (!rinfo->head->is_target && !rinfo->head->is_dentry) {
		dout("fill_trace reply is empty!\n");
1074 1075
		if (rinfo->head->result == 0 && req->r_locked_dir)
			ceph_invalidate_dir_request(req);
S
Sage Weil 已提交
1076 1077 1078 1079
		return 0;
	}

	if (rinfo->head->is_dentry) {
1080 1081
		struct inode *dir = req->r_locked_dir;

1082
		if (dir) {
1083 1084
			err = fill_inode(dir, NULL,
					 &rinfo->diri, rinfo->dirfrag,
1085 1086 1087
					 session, req->r_request_started, -1,
					 &req->r_caps_reservation);
			if (err < 0)
1088
				goto done;
1089 1090 1091
		} else {
			WARN_ON_ONCE(1);
		}
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127

		if (dir && req->r_op == CEPH_MDS_OP_LOOKUPNAME) {
			struct qstr dname;
			struct dentry *dn, *parent;

			BUG_ON(!rinfo->head->is_target);
			BUG_ON(req->r_dentry);

			parent = d_find_any_alias(dir);
			BUG_ON(!parent);

			dname.name = rinfo->dname;
			dname.len = rinfo->dname_len;
			dname.hash = full_name_hash(dname.name, dname.len);
			vino.ino = le64_to_cpu(rinfo->targeti.in->ino);
			vino.snap = le64_to_cpu(rinfo->targeti.in->snapid);
retry_lookup:
			dn = d_lookup(parent, &dname);
			dout("d_lookup on parent=%p name=%.*s got %p\n",
			     parent, dname.len, dname.name, dn);

			if (!dn) {
				dn = d_alloc(parent, &dname);
				dout("d_alloc %p '%.*s' = %p\n", parent,
				     dname.len, dname.name, dn);
				if (dn == NULL) {
					dput(parent);
					err = -ENOMEM;
					goto done;
				}
				err = ceph_init_dentry(dn);
				if (err < 0) {
					dput(dn);
					dput(parent);
					goto done;
				}
1128 1129 1130
			} else if (d_really_is_positive(dn) &&
				   (ceph_ino(d_inode(dn)) != vino.ino ||
				    ceph_snap(d_inode(dn)) != vino.snap)) {
1131
				dout(" dn %p points to wrong inode %p\n",
1132
				     dn, d_inode(dn));
1133 1134 1135 1136 1137 1138 1139 1140
				d_delete(dn);
				dput(dn);
				goto retry_lookup;
			}

			req->r_dentry = dn;
			dput(parent);
		}
1141 1142
	}

1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153
	if (rinfo->head->is_target) {
		vino.ino = le64_to_cpu(rinfo->targeti.in->ino);
		vino.snap = le64_to_cpu(rinfo->targeti.in->snapid);

		in = ceph_get_inode(sb, vino);
		if (IS_ERR(in)) {
			err = PTR_ERR(in);
			goto done;
		}
		req->r_target_inode = in;

1154
		err = fill_inode(in, req->r_locked_page, &rinfo->targeti, NULL,
1155
				session, req->r_request_started,
1156
				(!req->r_aborted && rinfo->head->result == 0) ?
1157 1158 1159 1160 1161 1162 1163 1164 1165
				req->r_fmode : -1,
				&req->r_caps_reservation);
		if (err < 0) {
			pr_err("fill_inode badness %p %llx.%llx\n",
				in, ceph_vinop(in));
			goto done;
		}
	}

1166 1167 1168 1169 1170
	/*
	 * ignore null lease/binding on snapdir ENOENT, or else we
	 * will have trouble splicing in the virtual snapdir later
	 */
	if (rinfo->head->is_dentry && !req->r_aborted &&
1171
	    req->r_locked_dir &&
1172
	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
1173
					       fsc->mount_options->snapdir_name,
1174
					       req->r_dentry->d_name.len))) {
S
Sage Weil 已提交
1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185
		/*
		 * lookup link rename   : null -> possibly existing inode
		 * mknod symlink mkdir  : null -> new inode
		 * unlink               : linked -> null
		 */
		struct inode *dir = req->r_locked_dir;
		struct dentry *dn = req->r_dentry;
		bool have_dir_cap, have_lease;

		BUG_ON(!dn);
		BUG_ON(!dir);
1186
		BUG_ON(d_inode(dn->d_parent) != dir);
S
Sage Weil 已提交
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198
		BUG_ON(ceph_ino(dir) !=
		       le64_to_cpu(rinfo->diri.in->ino));
		BUG_ON(ceph_snap(dir) !=
		       le64_to_cpu(rinfo->diri.in->snapid));

		/* do we have a lease on the whole dir? */
		have_dir_cap =
			(le32_to_cpu(rinfo->diri.in->cap.caps) &
			 CEPH_CAP_FILE_SHARED);

		/* do we have a dn lease? */
		have_lease = have_dir_cap ||
S
Sage Weil 已提交
1199
			le32_to_cpu(rinfo->dlease->duration_ms);
S
Sage Weil 已提交
1200 1201 1202 1203 1204
		if (!have_lease)
			dout("fill_trace  no dentry lease or dir cap\n");

		/* rename? */
		if (req->r_old_dentry && req->r_op == CEPH_MDS_OP_RENAME) {
1205 1206 1207
			struct inode *olddir = req->r_old_dentry_dir;
			BUG_ON(!olddir);

A
Al Viro 已提交
1208
			dout(" src %p '%pd' dst %p '%pd'\n",
S
Sage Weil 已提交
1209
			     req->r_old_dentry,
A
Al Viro 已提交
1210 1211
			     req->r_old_dentry,
			     dn, dn);
S
Sage Weil 已提交
1212 1213
			dout("fill_trace doing d_move %p -> %p\n",
			     req->r_old_dentry, dn);
S
Sage Weil 已提交
1214

S
Sage Weil 已提交
1215
			d_move(req->r_old_dentry, dn);
A
Al Viro 已提交
1216 1217
			dout(" src %p '%pd' dst %p '%pd'\n",
			     req->r_old_dentry,
S
Sage Weil 已提交
1218
			     req->r_old_dentry,
A
Al Viro 已提交
1219
			     dn, dn);
1220

1221 1222
			/* ensure target dentry is invalidated, despite
			   rehashing bug in vfs_rename_dir */
1223 1224
			ceph_invalidate_dentry_lease(dn);

1225
			/* d_move screws up sibling dentries' offsets */
1226 1227
			ceph_dir_clear_ordered(dir);
			ceph_dir_clear_ordered(olddir);
1228

1229
			dout("dn %p gets new offset %lld\n", req->r_old_dentry,
S
Sage Weil 已提交
1230
			     ceph_dentry(req->r_old_dentry)->offset);
1231

S
Sage Weil 已提交
1232 1233 1234 1235 1236 1237
			dn = req->r_old_dentry;  /* use old_dentry */
		}

		/* null dentry? */
		if (!rinfo->head->is_target) {
			dout("fill_trace null dentry\n");
1238
			if (d_really_is_positive(dn)) {
1239
				ceph_dir_clear_ordered(dir);
S
Sage Weil 已提交
1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
				dout("d_delete %p\n", dn);
				d_delete(dn);
			} else {
				dout("d_instantiate %p NULL\n", dn);
				d_instantiate(dn, NULL);
				if (have_lease && d_unhashed(dn))
					d_rehash(dn);
				update_dentry_lease(dn, rinfo->dlease,
						    session,
						    req->r_request_started);
			}
			goto done;
		}

		/* attach proper inode */
1255
		if (d_really_is_negative(dn)) {
1256
			ceph_dir_clear_ordered(dir);
1257
			ihold(in);
1258
			dn = splice_dentry(dn, in, &have_lease);
S
Sage Weil 已提交
1259 1260 1261 1262 1263
			if (IS_ERR(dn)) {
				err = PTR_ERR(dn);
				goto done;
			}
			req->r_dentry = dn;  /* may have spliced */
1264
		} else if (d_really_is_positive(dn) && d_inode(dn) != in) {
S
Sage Weil 已提交
1265
			dout(" %p links to %p %llx.%llx, not %llx.%llx\n",
1266
			     dn, d_inode(dn), ceph_vinop(d_inode(dn)),
1267
			     ceph_vinop(in));
S
Sage Weil 已提交
1268 1269 1270 1271 1272 1273 1274
			have_lease = false;
		}

		if (have_lease)
			update_dentry_lease(dn, rinfo->dlease, session,
					    req->r_request_started);
		dout(" final dn %p\n", dn);
1275 1276 1277
	} else if (!req->r_aborted &&
		   (req->r_op == CEPH_MDS_OP_LOOKUPSNAP ||
		    req->r_op == CEPH_MDS_OP_MKSNAP)) {
S
Sage Weil 已提交
1278
		struct dentry *dn = req->r_dentry;
1279
		struct inode *dir = req->r_locked_dir;
S
Sage Weil 已提交
1280 1281 1282

		/* fill out a snapdir LOOKUPSNAP dentry */
		BUG_ON(!dn);
1283 1284
		BUG_ON(!dir);
		BUG_ON(ceph_snap(dir) != CEPH_SNAPDIR);
S
Sage Weil 已提交
1285
		dout(" linking snapped dir %p to dn %p\n", in, dn);
1286
		ceph_dir_clear_ordered(dir);
1287
		ihold(in);
1288
		dn = splice_dentry(dn, in, NULL);
S
Sage Weil 已提交
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302
		if (IS_ERR(dn)) {
			err = PTR_ERR(dn);
			goto done;
		}
		req->r_dentry = dn;  /* may have spliced */
	}
done:
	dout("fill_trace done err=%d\n", err);
	return err;
}

/*
 * Prepopulate our cache with readdir results, leases, etc.
 */
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322
static int readdir_prepopulate_inodes_only(struct ceph_mds_request *req,
					   struct ceph_mds_session *session)
{
	struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
	int i, err = 0;

	for (i = 0; i < rinfo->dir_nr; i++) {
		struct ceph_vino vino;
		struct inode *in;
		int rc;

		vino.ino = le64_to_cpu(rinfo->dir_in[i].in->ino);
		vino.snap = le64_to_cpu(rinfo->dir_in[i].in->snapid);

		in = ceph_get_inode(req->r_dentry->d_sb, vino);
		if (IS_ERR(in)) {
			err = PTR_ERR(in);
			dout("new_inode badness got %d\n", err);
			continue;
		}
1323
		rc = fill_inode(in, NULL, &rinfo->dir_in[i], NULL, session,
1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
				req->r_request_started, -1,
				&req->r_caps_reservation);
		if (rc < 0) {
			pr_err("fill_inode badness on %p got %d\n", in, rc);
			err = rc;
			continue;
		}
	}

	return err;
}

S
Sage Weil 已提交
1336 1337 1338 1339 1340 1341 1342 1343
int ceph_readdir_prepopulate(struct ceph_mds_request *req,
			     struct ceph_mds_session *session)
{
	struct dentry *parent = req->r_dentry;
	struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
	struct qstr dname;
	struct dentry *dn;
	struct inode *in;
1344
	int err = 0, ret, i;
S
Sage Weil 已提交
1345 1346 1347
	struct inode *snapdir = NULL;
	struct ceph_mds_request_head *rhead = req->r_request->front.iov_base;
	struct ceph_dentry_info *di;
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
	u64 r_readdir_offset = req->r_readdir_offset;
	u32 frag = le32_to_cpu(rhead->args.readdir.frag);

	if (rinfo->dir_dir &&
	    le32_to_cpu(rinfo->dir_dir->frag) != frag) {
		dout("readdir_prepopulate got new frag %x -> %x\n",
		     frag, le32_to_cpu(rinfo->dir_dir->frag));
		frag = le32_to_cpu(rinfo->dir_dir->frag);
		if (ceph_frag_is_leftmost(frag))
			r_readdir_offset = 2;
		else
			r_readdir_offset = 0;
	}
S
Sage Weil 已提交
1361

1362 1363
	if (req->r_aborted)
		return readdir_prepopulate_inodes_only(req, session);
S
Sage Weil 已提交
1364 1365

	if (le32_to_cpu(rinfo->head->op) == CEPH_MDS_OP_LSSNAP) {
1366
		snapdir = ceph_get_snapdir(d_inode(parent));
S
Sage Weil 已提交
1367 1368 1369 1370 1371 1372 1373
		parent = d_find_alias(snapdir);
		dout("readdir_prepopulate %d items under SNAPDIR dn %p\n",
		     rinfo->dir_nr, parent);
	} else {
		dout("readdir_prepopulate %d items under dn %p\n",
		     rinfo->dir_nr, parent);
		if (rinfo->dir_dir)
1374
			ceph_fill_dirfrag(d_inode(parent), rinfo->dir_dir);
S
Sage Weil 已提交
1375 1376
	}

1377
	/* FIXME: release caps/leases if error occurs */
S
Sage Weil 已提交
1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401
	for (i = 0; i < rinfo->dir_nr; i++) {
		struct ceph_vino vino;

		dname.name = rinfo->dir_dname[i];
		dname.len = rinfo->dir_dname_len[i];
		dname.hash = full_name_hash(dname.name, dname.len);

		vino.ino = le64_to_cpu(rinfo->dir_in[i].in->ino);
		vino.snap = le64_to_cpu(rinfo->dir_in[i].in->snapid);

retry_lookup:
		dn = d_lookup(parent, &dname);
		dout("d_lookup on parent=%p name=%.*s got %p\n",
		     parent, dname.len, dname.name, dn);

		if (!dn) {
			dn = d_alloc(parent, &dname);
			dout("d_alloc %p '%.*s' = %p\n", parent,
			     dname.len, dname.name, dn);
			if (dn == NULL) {
				dout("d_alloc badness\n");
				err = -ENOMEM;
				goto out;
			}
1402 1403
			ret = ceph_init_dentry(dn);
			if (ret < 0) {
1404
				dput(dn);
1405
				err = ret;
S
Sage Weil 已提交
1406
				goto out;
1407
			}
1408 1409 1410
		} else if (d_really_is_positive(dn) &&
			   (ceph_ino(d_inode(dn)) != vino.ino ||
			    ceph_snap(d_inode(dn)) != vino.snap)) {
S
Sage Weil 已提交
1411
			dout(" dn %p points to wrong inode %p\n",
1412
			     dn, d_inode(dn));
S
Sage Weil 已提交
1413 1414 1415 1416 1417
			d_delete(dn);
			dput(dn);
			goto retry_lookup;
		} else {
			/* reorder parent's d_subdirs */
N
Nick Piggin 已提交
1418 1419
			spin_lock(&parent->d_lock);
			spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED);
1420
			list_move(&dn->d_child, &parent->d_subdirs);
S
Sage Weil 已提交
1421
			spin_unlock(&dn->d_lock);
N
Nick Piggin 已提交
1422
			spin_unlock(&parent->d_lock);
S
Sage Weil 已提交
1423 1424 1425
		}

		/* inode */
1426 1427
		if (d_really_is_positive(dn)) {
			in = d_inode(dn);
S
Sage Weil 已提交
1428 1429
		} else {
			in = ceph_get_inode(parent->d_sb, vino);
1430
			if (IS_ERR(in)) {
S
Sage Weil 已提交
1431
				dout("new_inode badness\n");
1432
				d_drop(dn);
S
Sage Weil 已提交
1433
				dput(dn);
1434
				err = PTR_ERR(in);
S
Sage Weil 已提交
1435 1436 1437 1438
				goto out;
			}
		}

1439
		if (fill_inode(in, NULL, &rinfo->dir_in[i], NULL, session,
S
Sage Weil 已提交
1440 1441 1442
			       req->r_request_started, -1,
			       &req->r_caps_reservation) < 0) {
			pr_err("fill_inode badness on %p\n", in);
1443
			if (d_really_is_negative(dn))
1444 1445
				iput(in);
			d_drop(dn);
1446
			goto next_item;
S
Sage Weil 已提交
1447
		}
1448

1449
		if (d_really_is_negative(dn)) {
Y
Yan, Zheng 已提交
1450 1451 1452 1453
			struct dentry *realdn = splice_dentry(dn, in, NULL);
			if (IS_ERR(realdn)) {
				err = PTR_ERR(realdn);
				d_drop(dn);
1454 1455 1456
				dn = NULL;
				goto next_item;
			}
Y
Yan, Zheng 已提交
1457
			dn = realdn;
1458 1459 1460 1461 1462 1463 1464 1465
		}

		di = dn->d_fsdata;
		di->offset = ceph_make_fpos(frag, i + r_readdir_offset);

		update_dentry_lease(dn, rinfo->dir_dlease[i],
				    req->r_session,
				    req->r_request_started);
1466 1467 1468
next_item:
		if (dn)
			dput(dn);
S
Sage Weil 已提交
1469
	}
1470 1471
	if (err == 0)
		req->r_did_prepopulate = true;
S
Sage Weil 已提交
1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486

out:
	if (snapdir) {
		iput(snapdir);
		dput(parent);
	}
	dout("readdir_prepopulate done\n");
	return err;
}

int ceph_inode_set_size(struct inode *inode, loff_t size)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int ret = 0;

1487
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1488 1489 1490 1491 1492 1493 1494 1495 1496
	dout("set_size %p %llu -> %llu\n", inode, inode->i_size, size);
	inode->i_size = size;
	inode->i_blocks = (size + (1 << 9) - 1) >> 9;

	/* tell the MDS if we are approaching max_size */
	if ((size << 1) >= ci->i_max_size &&
	    (ci->i_reported_size << 1) < ci->i_max_size)
		ret = 1;

1497
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1498 1499 1500 1501 1502 1503 1504
	return ret;
}

/*
 * Write back inode data in a worker thread.  (This can't be done
 * in the message handler context.)
 */
1505 1506
void ceph_queue_writeback(struct inode *inode)
{
1507
	ihold(inode);
1508 1509
	if (queue_work(ceph_inode_to_client(inode)->wb_wq,
		       &ceph_inode(inode)->i_wb_work)) {
1510
		dout("ceph_queue_writeback %p\n", inode);
1511
	} else {
1512
		dout("ceph_queue_writeback %p failed\n", inode);
1513
		iput(inode);
1514 1515 1516 1517
	}
}

static void ceph_writeback_work(struct work_struct *work)
S
Sage Weil 已提交
1518 1519 1520 1521 1522 1523 1524 1525 1526 1527
{
	struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info,
						  i_wb_work);
	struct inode *inode = &ci->vfs_inode;

	dout("writeback %p\n", inode);
	filemap_fdatawrite(&inode->i_data);
	iput(inode);
}

1528 1529 1530 1531 1532
/*
 * queue an async invalidation
 */
void ceph_queue_invalidate(struct inode *inode)
{
1533
	ihold(inode);
1534 1535 1536 1537 1538
	if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq,
		       &ceph_inode(inode)->i_pg_inv_work)) {
		dout("ceph_queue_invalidate %p\n", inode);
	} else {
		dout("ceph_queue_invalidate %p failed\n", inode);
1539
		iput(inode);
1540 1541 1542
	}
}

S
Sage Weil 已提交
1543 1544 1545 1546
/*
 * Invalidate inode pages in a worker thread.  (This can't be done
 * in the message handler context.)
 */
1547
static void ceph_invalidate_work(struct work_struct *work)
S
Sage Weil 已提交
1548 1549 1550 1551 1552 1553 1554
{
	struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info,
						  i_pg_inv_work);
	struct inode *inode = &ci->vfs_inode;
	u32 orig_gen;
	int check = 0;

Y
Yan, Zheng 已提交
1555
	mutex_lock(&ci->i_truncate_mutex);
1556
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1557 1558
	dout("invalidate_pages %p gen %d revoking %d\n", inode,
	     ci->i_rdcache_gen, ci->i_rdcache_revoking);
1559
	if (ci->i_rdcache_revoking != ci->i_rdcache_gen) {
Y
Yan, Zheng 已提交
1560 1561
		if (__ceph_caps_revoking_other(ci, NULL, CEPH_CAP_FILE_CACHE))
			check = 1;
1562
		spin_unlock(&ci->i_ceph_lock);
Y
Yan, Zheng 已提交
1563
		mutex_unlock(&ci->i_truncate_mutex);
S
Sage Weil 已提交
1564 1565 1566
		goto out;
	}
	orig_gen = ci->i_rdcache_gen;
1567
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1568

1569
	truncate_pagecache(inode, 0);
S
Sage Weil 已提交
1570

1571
	spin_lock(&ci->i_ceph_lock);
1572 1573
	if (orig_gen == ci->i_rdcache_gen &&
	    orig_gen == ci->i_rdcache_revoking) {
S
Sage Weil 已提交
1574 1575
		dout("invalidate_pages %p gen %d successful\n", inode,
		     ci->i_rdcache_gen);
1576
		ci->i_rdcache_revoking--;
S
Sage Weil 已提交
1577 1578
		check = 1;
	} else {
1579 1580 1581
		dout("invalidate_pages %p gen %d raced, now %d revoking %d\n",
		     inode, orig_gen, ci->i_rdcache_gen,
		     ci->i_rdcache_revoking);
Y
Yan, Zheng 已提交
1582 1583
		if (__ceph_caps_revoking_other(ci, NULL, CEPH_CAP_FILE_CACHE))
			check = 1;
S
Sage Weil 已提交
1584
	}
1585
	spin_unlock(&ci->i_ceph_lock);
Y
Yan, Zheng 已提交
1586
	mutex_unlock(&ci->i_truncate_mutex);
Y
Yan, Zheng 已提交
1587
out:
S
Sage Weil 已提交
1588 1589 1590 1591 1592 1593 1594
	if (check)
		ceph_check_caps(ci, 0, NULL);
	iput(inode);
}


/*
1595
 * called by trunc_wq;
S
Sage Weil 已提交
1596 1597 1598
 *
 * We also truncate in a separate thread as well.
 */
1599
static void ceph_vmtruncate_work(struct work_struct *work)
S
Sage Weil 已提交
1600 1601 1602 1603 1604 1605
{
	struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info,
						  i_vmtruncate_work);
	struct inode *inode = &ci->vfs_inode;

	dout("vmtruncate_work %p\n", inode);
Y
Yan, Zheng 已提交
1606
	__ceph_do_pending_vmtruncate(inode);
S
Sage Weil 已提交
1607 1608 1609
	iput(inode);
}

1610 1611 1612 1613 1614 1615 1616 1617
/*
 * Queue an async vmtruncate.  If we fail to queue work, we will handle
 * the truncation the next time we call __ceph_do_pending_vmtruncate.
 */
void ceph_queue_vmtruncate(struct inode *inode)
{
	struct ceph_inode_info *ci = ceph_inode(inode);

1618
	ihold(inode);
1619

1620
	if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq,
1621 1622 1623 1624 1625
		       &ci->i_vmtruncate_work)) {
		dout("ceph_queue_vmtruncate %p\n", inode);
	} else {
		dout("ceph_queue_vmtruncate %p failed, pending=%d\n",
		     inode, ci->i_truncate_pending);
1626
		iput(inode);
1627 1628 1629
	}
}

S
Sage Weil 已提交
1630 1631 1632 1633
/*
 * Make sure any pending truncation is applied before doing anything
 * that may depend on it.
 */
Y
Yan, Zheng 已提交
1634
void __ceph_do_pending_vmtruncate(struct inode *inode)
S
Sage Weil 已提交
1635 1636 1637
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	u64 to;
1638
	int wrbuffer_refs, finish = 0;
S
Sage Weil 已提交
1639

Y
Yan, Zheng 已提交
1640
	mutex_lock(&ci->i_truncate_mutex);
S
Sage Weil 已提交
1641
retry:
1642
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1643 1644
	if (ci->i_truncate_pending == 0) {
		dout("__do_pending_vmtruncate %p none pending\n", inode);
1645
		spin_unlock(&ci->i_ceph_lock);
Y
Yan, Zheng 已提交
1646
		mutex_unlock(&ci->i_truncate_mutex);
S
Sage Weil 已提交
1647 1648 1649 1650 1651 1652 1653 1654 1655 1656
		return;
	}

	/*
	 * make sure any dirty snapped pages are flushed before we
	 * possibly truncate them.. so write AND block!
	 */
	if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) {
		dout("__do_pending_vmtruncate %p flushing snaps first\n",
		     inode);
1657
		spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1658 1659 1660 1661 1662
		filemap_write_and_wait_range(&inode->i_data, 0,
					     inode->i_sb->s_maxbytes);
		goto retry;
	}

Y
Yan, Zheng 已提交
1663 1664 1665
	/* there should be no reader or writer */
	WARN_ON_ONCE(ci->i_rd_ref || ci->i_wr_ref);

S
Sage Weil 已提交
1666 1667 1668 1669
	to = ci->i_truncate_size;
	wrbuffer_refs = ci->i_wrbuffer_ref;
	dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode,
	     ci->i_truncate_pending, to);
1670
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1671

1672
	truncate_pagecache(inode, to);
S
Sage Weil 已提交
1673

1674
	spin_lock(&ci->i_ceph_lock);
1675 1676 1677 1678
	if (to == ci->i_truncate_size) {
		ci->i_truncate_pending = 0;
		finish = 1;
	}
1679
	spin_unlock(&ci->i_ceph_lock);
1680 1681
	if (!finish)
		goto retry;
S
Sage Weil 已提交
1682

Y
Yan, Zheng 已提交
1683 1684
	mutex_unlock(&ci->i_truncate_mutex);

S
Sage Weil 已提交
1685 1686
	if (wrbuffer_refs == 0)
		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
1687 1688

	wake_up_all(&ci->i_cap_wq);
S
Sage Weil 已提交
1689 1690 1691 1692 1693 1694 1695
}

/*
 * symlinks
 */
static void *ceph_sym_follow_link(struct dentry *dentry, struct nameidata *nd)
{
1696
	struct ceph_inode_info *ci = ceph_inode(d_inode(dentry));
S
Sage Weil 已提交
1697 1698 1699 1700 1701 1702 1703
	nd_set_link(nd, ci->i_symlink);
	return NULL;
}

static const struct inode_operations ceph_symlink_iops = {
	.readlink = generic_readlink,
	.follow_link = ceph_sym_follow_link,
Y
Yan, Zheng 已提交
1704 1705 1706 1707 1708 1709
	.setattr = ceph_setattr,
	.getattr = ceph_getattr,
	.setxattr = ceph_setxattr,
	.getxattr = ceph_getxattr,
	.listxattr = ceph_listxattr,
	.removexattr = ceph_removexattr,
S
Sage Weil 已提交
1710 1711 1712 1713 1714 1715 1716
};

/*
 * setattr
 */
int ceph_setattr(struct dentry *dentry, struct iattr *attr)
{
1717
	struct inode *inode = d_inode(dentry);
S
Sage Weil 已提交
1718 1719 1720
	struct ceph_inode_info *ci = ceph_inode(inode);
	const unsigned int ia_valid = attr->ia_valid;
	struct ceph_mds_request *req;
1721
	struct ceph_mds_client *mdsc = ceph_sb_to_client(dentry->d_sb)->mdsc;
S
Sage Weil 已提交
1722 1723 1724 1725
	int issued;
	int release = 0, dirtied = 0;
	int mask = 0;
	int err = 0;
1726
	int inode_dirty_flags = 0;
S
Sage Weil 已提交
1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return -EROFS;

	err = inode_change_ok(inode, attr);
	if (err != 0)
		return err;

	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETATTR,
				       USE_AUTH_MDS);
	if (IS_ERR(req))
		return PTR_ERR(req);

1740
	spin_lock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1741 1742 1743 1744 1745
	issued = __ceph_caps_issued(ci, NULL);
	dout("setattr %p issued %s\n", inode, ceph_cap_string(issued));

	if (ia_valid & ATTR_UID) {
		dout("setattr %p uid %d -> %d\n", inode,
1746 1747
		     from_kuid(&init_user_ns, inode->i_uid),
		     from_kuid(&init_user_ns, attr->ia_uid));
S
Sage Weil 已提交
1748 1749 1750 1751
		if (issued & CEPH_CAP_AUTH_EXCL) {
			inode->i_uid = attr->ia_uid;
			dirtied |= CEPH_CAP_AUTH_EXCL;
		} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
1752 1753 1754
			   !uid_eq(attr->ia_uid, inode->i_uid)) {
			req->r_args.setattr.uid = cpu_to_le32(
				from_kuid(&init_user_ns, attr->ia_uid));
S
Sage Weil 已提交
1755 1756 1757 1758 1759 1760
			mask |= CEPH_SETATTR_UID;
			release |= CEPH_CAP_AUTH_SHARED;
		}
	}
	if (ia_valid & ATTR_GID) {
		dout("setattr %p gid %d -> %d\n", inode,
1761 1762
		     from_kgid(&init_user_ns, inode->i_gid),
		     from_kgid(&init_user_ns, attr->ia_gid));
S
Sage Weil 已提交
1763 1764 1765 1766
		if (issued & CEPH_CAP_AUTH_EXCL) {
			inode->i_gid = attr->ia_gid;
			dirtied |= CEPH_CAP_AUTH_EXCL;
		} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
1767 1768 1769
			   !gid_eq(attr->ia_gid, inode->i_gid)) {
			req->r_args.setattr.gid = cpu_to_le32(
				from_kgid(&init_user_ns, attr->ia_gid));
S
Sage Weil 已提交
1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781
			mask |= CEPH_SETATTR_GID;
			release |= CEPH_CAP_AUTH_SHARED;
		}
	}
	if (ia_valid & ATTR_MODE) {
		dout("setattr %p mode 0%o -> 0%o\n", inode, inode->i_mode,
		     attr->ia_mode);
		if (issued & CEPH_CAP_AUTH_EXCL) {
			inode->i_mode = attr->ia_mode;
			dirtied |= CEPH_CAP_AUTH_EXCL;
		} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
			   attr->ia_mode != inode->i_mode) {
G
Guangliang Zhao 已提交
1782
			inode->i_mode = attr->ia_mode;
S
Sage Weil 已提交
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 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883
			req->r_args.setattr.mode = cpu_to_le32(attr->ia_mode);
			mask |= CEPH_SETATTR_MODE;
			release |= CEPH_CAP_AUTH_SHARED;
		}
	}

	if (ia_valid & ATTR_ATIME) {
		dout("setattr %p atime %ld.%ld -> %ld.%ld\n", inode,
		     inode->i_atime.tv_sec, inode->i_atime.tv_nsec,
		     attr->ia_atime.tv_sec, attr->ia_atime.tv_nsec);
		if (issued & CEPH_CAP_FILE_EXCL) {
			ci->i_time_warp_seq++;
			inode->i_atime = attr->ia_atime;
			dirtied |= CEPH_CAP_FILE_EXCL;
		} else if ((issued & CEPH_CAP_FILE_WR) &&
			   timespec_compare(&inode->i_atime,
					    &attr->ia_atime) < 0) {
			inode->i_atime = attr->ia_atime;
			dirtied |= CEPH_CAP_FILE_WR;
		} else if ((issued & CEPH_CAP_FILE_SHARED) == 0 ||
			   !timespec_equal(&inode->i_atime, &attr->ia_atime)) {
			ceph_encode_timespec(&req->r_args.setattr.atime,
					     &attr->ia_atime);
			mask |= CEPH_SETATTR_ATIME;
			release |= CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_RD |
				CEPH_CAP_FILE_WR;
		}
	}
	if (ia_valid & ATTR_MTIME) {
		dout("setattr %p mtime %ld.%ld -> %ld.%ld\n", inode,
		     inode->i_mtime.tv_sec, inode->i_mtime.tv_nsec,
		     attr->ia_mtime.tv_sec, attr->ia_mtime.tv_nsec);
		if (issued & CEPH_CAP_FILE_EXCL) {
			ci->i_time_warp_seq++;
			inode->i_mtime = attr->ia_mtime;
			dirtied |= CEPH_CAP_FILE_EXCL;
		} else if ((issued & CEPH_CAP_FILE_WR) &&
			   timespec_compare(&inode->i_mtime,
					    &attr->ia_mtime) < 0) {
			inode->i_mtime = attr->ia_mtime;
			dirtied |= CEPH_CAP_FILE_WR;
		} else if ((issued & CEPH_CAP_FILE_SHARED) == 0 ||
			   !timespec_equal(&inode->i_mtime, &attr->ia_mtime)) {
			ceph_encode_timespec(&req->r_args.setattr.mtime,
					     &attr->ia_mtime);
			mask |= CEPH_SETATTR_MTIME;
			release |= CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_RD |
				CEPH_CAP_FILE_WR;
		}
	}
	if (ia_valid & ATTR_SIZE) {
		dout("setattr %p size %lld -> %lld\n", inode,
		     inode->i_size, attr->ia_size);
		if ((issued & CEPH_CAP_FILE_EXCL) &&
		    attr->ia_size > inode->i_size) {
			inode->i_size = attr->ia_size;
			inode->i_blocks =
				(attr->ia_size + (1 << 9) - 1) >> 9;
			inode->i_ctime = attr->ia_ctime;
			ci->i_reported_size = attr->ia_size;
			dirtied |= CEPH_CAP_FILE_EXCL;
		} else if ((issued & CEPH_CAP_FILE_SHARED) == 0 ||
			   attr->ia_size != inode->i_size) {
			req->r_args.setattr.size = cpu_to_le64(attr->ia_size);
			req->r_args.setattr.old_size =
				cpu_to_le64(inode->i_size);
			mask |= CEPH_SETATTR_SIZE;
			release |= CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_RD |
				CEPH_CAP_FILE_WR;
		}
	}

	/* these do nothing */
	if (ia_valid & ATTR_CTIME) {
		bool only = (ia_valid & (ATTR_SIZE|ATTR_MTIME|ATTR_ATIME|
					 ATTR_MODE|ATTR_UID|ATTR_GID)) == 0;
		dout("setattr %p ctime %ld.%ld -> %ld.%ld (%s)\n", inode,
		     inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
		     attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec,
		     only ? "ctime only" : "ignored");
		inode->i_ctime = attr->ia_ctime;
		if (only) {
			/*
			 * if kernel wants to dirty ctime but nothing else,
			 * we need to choose a cap to dirty under, or do
			 * a almost-no-op setattr
			 */
			if (issued & CEPH_CAP_AUTH_EXCL)
				dirtied |= CEPH_CAP_AUTH_EXCL;
			else if (issued & CEPH_CAP_FILE_EXCL)
				dirtied |= CEPH_CAP_FILE_EXCL;
			else if (issued & CEPH_CAP_XATTR_EXCL)
				dirtied |= CEPH_CAP_XATTR_EXCL;
			else
				mask |= CEPH_SETATTR_CTIME;
		}
	}
	if (ia_valid & ATTR_FILE)
		dout("setattr %p ATTR_FILE ... hrm!\n", inode);

	if (dirtied) {
1884
		inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied);
S
Sage Weil 已提交
1885 1886 1887 1888
		inode->i_ctime = CURRENT_TIME;
	}

	release &= issued;
1889
	spin_unlock(&ci->i_ceph_lock);
S
Sage Weil 已提交
1890

1891 1892 1893
	if (inode_dirty_flags)
		__mark_inode_dirty(inode, inode_dirty_flags);

G
Guangliang Zhao 已提交
1894
	if (ia_valid & ATTR_MODE) {
1895
		err = posix_acl_chmod(inode, attr->ia_mode);
G
Guangliang Zhao 已提交
1896 1897 1898 1899
		if (err)
			goto out_put;
	}

S
Sage Weil 已提交
1900
	if (mask) {
1901 1902
		req->r_inode = inode;
		ihold(inode);
S
Sage Weil 已提交
1903 1904 1905
		req->r_inode_drop = release;
		req->r_args.setattr.mask = cpu_to_le32(mask);
		req->r_num_caps = 1;
1906
		err = ceph_mdsc_do_request(mdsc, NULL, req);
S
Sage Weil 已提交
1907 1908 1909 1910 1911
	}
	dout("setattr %p result=%d (%s locally, %d remote)\n", inode, err,
	     ceph_cap_string(dirtied), mask);

	ceph_mdsc_put_request(req);
Y
Yan, Zheng 已提交
1912 1913
	if (mask & CEPH_SETATTR_SIZE)
		__ceph_do_pending_vmtruncate(inode);
S
Sage Weil 已提交
1914
	return err;
G
Guangliang Zhao 已提交
1915
out_put:
S
Sage Weil 已提交
1916 1917 1918 1919 1920 1921 1922 1923
	ceph_mdsc_put_request(req);
	return err;
}

/*
 * Verify that we have a lease on the given mask.  If not,
 * do a getattr against an mds.
 */
1924 1925
int __ceph_do_getattr(struct inode *inode, struct page *locked_page,
		      int mask, bool force)
S
Sage Weil 已提交
1926
{
1927 1928
	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
	struct ceph_mds_client *mdsc = fsc->mdsc;
S
Sage Weil 已提交
1929 1930 1931 1932 1933 1934 1935 1936
	struct ceph_mds_request *req;
	int err;

	if (ceph_snap(inode) == CEPH_SNAPDIR) {
		dout("do_getattr inode %p SNAPDIR\n", inode);
		return 0;
	}

1937 1938
	dout("do_getattr inode %p mask %s mode 0%o\n",
	     inode, ceph_cap_string(mask), inode->i_mode);
1939
	if (!force && ceph_caps_issued_mask(ceph_inode(inode), mask, 1))
S
Sage Weil 已提交
1940 1941 1942 1943 1944
		return 0;

	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS);
	if (IS_ERR(req))
		return PTR_ERR(req);
1945 1946
	req->r_inode = inode;
	ihold(inode);
S
Sage Weil 已提交
1947 1948
	req->r_num_caps = 1;
	req->r_args.getattr.mask = cpu_to_le32(mask);
1949
	req->r_locked_page = locked_page;
S
Sage Weil 已提交
1950
	err = ceph_mdsc_do_request(mdsc, NULL, req);
1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961
	if (locked_page && err == 0) {
		u64 inline_version = req->r_reply_info.targeti.inline_version;
		if (inline_version == 0) {
			/* the reply is supposed to contain inline data */
			err = -EINVAL;
		} else if (inline_version == CEPH_INLINE_NONE) {
			err = -ENODATA;
		} else {
			err = req->r_reply_info.targeti.inline_len;
		}
	}
S
Sage Weil 已提交
1962 1963 1964 1965 1966 1967 1968 1969 1970 1971
	ceph_mdsc_put_request(req);
	dout("do_getattr result=%d\n", err);
	return err;
}


/*
 * Check inode permissions.  We verify we have a valid value for
 * the AUTH cap, then call the generic handler.
 */
1972
int ceph_permission(struct inode *inode, int mask)
S
Sage Weil 已提交
1973
{
1974 1975
	int err;

1976
	if (mask & MAY_NOT_BLOCK)
1977 1978
		return -ECHILD;

1979
	err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED, false);
S
Sage Weil 已提交
1980 1981

	if (!err)
1982
		err = generic_permission(inode, mask);
S
Sage Weil 已提交
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
	return err;
}

/*
 * Get all attributes.  Hopefully somedata we'll have a statlite()
 * and can limit the fields we require to be accurate.
 */
int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry,
		 struct kstat *stat)
{
1993
	struct inode *inode = d_inode(dentry);
1994
	struct ceph_inode_info *ci = ceph_inode(inode);
S
Sage Weil 已提交
1995 1996
	int err;

1997
	err = ceph_do_getattr(inode, CEPH_STAT_CAP_INODE_ALL, false);
S
Sage Weil 已提交
1998 1999
	if (!err) {
		generic_fillattr(inode, stat);
Y
Yehuda Sadeh 已提交
2000
		stat->ino = ceph_translate_ino(inode->i_sb, inode->i_ino);
S
Sage Weil 已提交
2001 2002 2003 2004
		if (ceph_snap(inode) != CEPH_NOSNAP)
			stat->dev = ceph_snap(inode);
		else
			stat->dev = 0;
2005
		if (S_ISDIR(inode->i_mode)) {
2006 2007 2008 2009 2010
			if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb),
						RBYTES))
				stat->size = ci->i_rbytes;
			else
				stat->size = ci->i_files + ci->i_subdirs;
2011
			stat->blocks = 0;
S
Sage Weil 已提交
2012
			stat->blksize = 65536;
2013
		}
S
Sage Weil 已提交
2014 2015 2016
	}
	return err;
}