inode.c 58.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
/*
 *  linux/fs/nfs/inode.c
 *
 *  Copyright (C) 1992  Rick Sladkey
 *
 *  nfs inode and superblock handling functions
 *
 *  Modularised by Alan Cox <Alan.Cox@linux.org>, while hacking some
 *  experimental NFS changes. Modularisation taken straight from SYS5 fs.
 *
 *  Change to nfs_read_super() to permit NFS mounts to multi-homed hosts.
 *  J.S.Peatfield@damtp.cam.ac.uk
 *
 */

#include <linux/config.h>
#include <linux/module.h>
#include <linux/init.h>

#include <linux/time.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/stat.h>
#include <linux/errno.h>
#include <linux/unistd.h>
#include <linux/sunrpc/clnt.h>
#include <linux/sunrpc/stats.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_mount.h>
#include <linux/nfs4_mount.h>
#include <linux/lockd/bind.h>
#include <linux/smp_lock.h>
#include <linux/seq_file.h>
#include <linux/mount.h>
#include <linux/nfs_idmap.h>
#include <linux/vfs.h>

#include <asm/system.h>
#include <asm/uaccess.h>

42
#include "nfs4_fs.h"
L
Linus Torvalds 已提交
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
#include "delegation.h"

#define NFSDBG_FACILITY		NFSDBG_VFS
#define NFS_PARANOIA 1

/* Maximum number of readahead requests
 * FIXME: this should really be a sysctl so that users may tune it to suit
 *        their needs. People that do NFS over a slow network, might for
 *        instance want to reduce it to something closer to 1 for improved
 *        interactive response.
 */
#define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)

static void nfs_invalidate_inode(struct inode *);
static int nfs_update_inode(struct inode *, struct nfs_fattr *, unsigned long);

static struct inode *nfs_alloc_inode(struct super_block *sb);
static void nfs_destroy_inode(struct inode *);
static int nfs_write_inode(struct inode *,int);
static void nfs_delete_inode(struct inode *);
static void nfs_clear_inode(struct inode *);
static void nfs_umount_begin(struct super_block *);
static int  nfs_statfs(struct super_block *, struct kstatfs *);
static int  nfs_show_options(struct seq_file *, struct vfsmount *);
67
static void nfs_zap_acl_cache(struct inode *);
L
Linus Torvalds 已提交
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

static struct rpc_program	nfs_program;

static struct super_operations nfs_sops = { 
	.alloc_inode	= nfs_alloc_inode,
	.destroy_inode	= nfs_destroy_inode,
	.write_inode	= nfs_write_inode,
	.delete_inode	= nfs_delete_inode,
	.statfs		= nfs_statfs,
	.clear_inode	= nfs_clear_inode,
	.umount_begin	= nfs_umount_begin,
	.show_options	= nfs_show_options,
};

/*
 * RPC cruft for NFS
 */
static struct rpc_stat		nfs_rpcstat = {
	.program		= &nfs_program
};
static struct rpc_version *	nfs_version[] = {
	NULL,
	NULL,
	&nfs_version2,
#if defined(CONFIG_NFS_V3)
	&nfs_version3,
#elif defined(CONFIG_NFS_V4)
	NULL,
#endif
#if defined(CONFIG_NFS_V4)
	&nfs_version4,
#endif
};

static struct rpc_program	nfs_program = {
	.name			= "nfs",
	.number			= NFS_PROGRAM,
	.nrvers			= sizeof(nfs_version) / sizeof(nfs_version[0]),
	.version		= nfs_version,
	.stats			= &nfs_rpcstat,
	.pipe_dir_name		= "/nfs",
};

111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
#ifdef CONFIG_NFS_V3_ACL
static struct rpc_stat		nfsacl_rpcstat = { &nfsacl_program };
static struct rpc_version *	nfsacl_version[] = {
	[3]			= &nfsacl_version3,
};

struct rpc_program		nfsacl_program = {
	.name =			"nfsacl",
	.number =		NFS_ACL_PROGRAM,
	.nrvers =		sizeof(nfsacl_version) / sizeof(nfsacl_version[0]),
	.version =		nfsacl_version,
	.stats =		&nfsacl_rpcstat,
};
#endif  /* CONFIG_NFS_V3_ACL */

L
Linus Torvalds 已提交
126 127 128 129 130 131 132 133 134 135 136 137
static inline unsigned long
nfs_fattr_to_ino_t(struct nfs_fattr *fattr)
{
	return nfs_fileid_to_ino_t(fattr->fileid);
}

static int
nfs_write_inode(struct inode *inode, int sync)
{
	int flags = sync ? FLUSH_WAIT : 0;
	int ret;

138
	ret = nfs_commit_inode(inode, flags);
L
Linus Torvalds 已提交
139 140 141 142 143 144 145 146 147 148
	if (ret < 0)
		return ret;
	return 0;
}

static void
nfs_delete_inode(struct inode * inode)
{
	dprintk("NFS: delete_inode(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino);

149 150
	truncate_inode_pages(&inode->i_data, 0);

L
Linus Torvalds 已提交
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
	nfs_wb_all(inode);
	/*
	 * The following should never happen...
	 */
	if (nfs_have_writebacks(inode)) {
		printk(KERN_ERR "nfs_delete_inode: inode %ld has pending RPC requests\n", inode->i_ino);
	}

	clear_inode(inode);
}

static void
nfs_clear_inode(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct rpc_cred *cred;

	nfs_wb_all(inode);
	BUG_ON (!list_empty(&nfsi->open_files));
170
	nfs_zap_acl_cache(inode);
L
Linus Torvalds 已提交
171 172 173 174 175 176 177 178 179
	cred = nfsi->cache_access.cred;
	if (cred)
		put_rpccred(cred);
	BUG_ON(atomic_read(&nfsi->data_updates) != 0);
}

void
nfs_umount_begin(struct super_block *sb)
{
180
	struct rpc_clnt	*rpc = NFS_SB(sb)->client;
L
Linus Torvalds 已提交
181 182

	/* -EIO all pending I/O */
183
	if (!IS_ERR(rpc))
L
Linus Torvalds 已提交
184
		rpc_killall_tasks(rpc);
185 186 187
	rpc = NFS_SB(sb)->client_acl;
	if (!IS_ERR(rpc))
		rpc_killall_tasks(rpc);
L
Linus Torvalds 已提交
188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360
}


static inline unsigned long
nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp)
{
	/* make sure blocksize is a power of two */
	if ((bsize & (bsize - 1)) || nrbitsp) {
		unsigned char	nrbits;

		for (nrbits = 31; nrbits && !(bsize & (1 << nrbits)); nrbits--)
			;
		bsize = 1 << nrbits;
		if (nrbitsp)
			*nrbitsp = nrbits;
	}

	return bsize;
}

/*
 * Calculate the number of 512byte blocks used.
 */
static inline unsigned long
nfs_calc_block_size(u64 tsize)
{
	loff_t used = (tsize + 511) >> 9;
	return (used > ULONG_MAX) ? ULONG_MAX : used;
}

/*
 * Compute and set NFS server blocksize
 */
static inline unsigned long
nfs_block_size(unsigned long bsize, unsigned char *nrbitsp)
{
	if (bsize < 1024)
		bsize = NFS_DEF_FILE_IO_BUFFER_SIZE;
	else if (bsize >= NFS_MAX_FILE_IO_BUFFER_SIZE)
		bsize = NFS_MAX_FILE_IO_BUFFER_SIZE;

	return nfs_block_bits(bsize, nrbitsp);
}

/*
 * Obtain the root inode of the file system.
 */
static struct inode *
nfs_get_root(struct super_block *sb, struct nfs_fh *rootfh, struct nfs_fsinfo *fsinfo)
{
	struct nfs_server	*server = NFS_SB(sb);
	struct inode *rooti;
	int			error;

	error = server->rpc_ops->getroot(server, rootfh, fsinfo);
	if (error < 0) {
		dprintk("nfs_get_root: getattr error = %d\n", -error);
		return ERR_PTR(error);
	}

	rooti = nfs_fhget(sb, rootfh, fsinfo->fattr);
	if (!rooti)
		return ERR_PTR(-ENOMEM);
	return rooti;
}

/*
 * Do NFS version-independent mount processing, and sanity checking
 */
static int
nfs_sb_init(struct super_block *sb, rpc_authflavor_t authflavor)
{
	struct nfs_server	*server;
	struct inode		*root_inode;
	struct nfs_fattr	fattr;
	struct nfs_fsinfo	fsinfo = {
					.fattr = &fattr,
				};
	struct nfs_pathconf pathinfo = {
			.fattr = &fattr,
	};
	int no_root_error = 0;
	unsigned long max_rpc_payload;

	/* We probably want something more informative here */
	snprintf(sb->s_id, sizeof(sb->s_id), "%x:%x", MAJOR(sb->s_dev), MINOR(sb->s_dev));

	server = NFS_SB(sb);

	sb->s_magic      = NFS_SUPER_MAGIC;

	root_inode = nfs_get_root(sb, &server->fh, &fsinfo);
	/* Did getting the root inode fail? */
	if (IS_ERR(root_inode)) {
		no_root_error = PTR_ERR(root_inode);
		goto out_no_root;
	}
	sb->s_root = d_alloc_root(root_inode);
	if (!sb->s_root) {
		no_root_error = -ENOMEM;
		goto out_no_root;
	}
	sb->s_root->d_op = server->rpc_ops->dentry_ops;

	/* Get some general file system info */
	if (server->namelen == 0 &&
	    server->rpc_ops->pathconf(server, &server->fh, &pathinfo) >= 0)
		server->namelen = pathinfo.max_namelen;
	/* Work out a lot of parameters */
	if (server->rsize == 0)
		server->rsize = nfs_block_size(fsinfo.rtpref, NULL);
	if (server->wsize == 0)
		server->wsize = nfs_block_size(fsinfo.wtpref, NULL);

	if (fsinfo.rtmax >= 512 && server->rsize > fsinfo.rtmax)
		server->rsize = nfs_block_size(fsinfo.rtmax, NULL);
	if (fsinfo.wtmax >= 512 && server->wsize > fsinfo.wtmax)
		server->wsize = nfs_block_size(fsinfo.wtmax, NULL);

	max_rpc_payload = nfs_block_size(rpc_max_payload(server->client), NULL);
	if (server->rsize > max_rpc_payload)
		server->rsize = max_rpc_payload;
	if (server->wsize > max_rpc_payload)
		server->wsize = max_rpc_payload;

	server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
	if (server->rpages > NFS_READ_MAXIOV) {
		server->rpages = NFS_READ_MAXIOV;
		server->rsize = server->rpages << PAGE_CACHE_SHIFT;
	}

	server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
        if (server->wpages > NFS_WRITE_MAXIOV) {
		server->wpages = NFS_WRITE_MAXIOV;
                server->wsize = server->wpages << PAGE_CACHE_SHIFT;
	}

	if (sb->s_blocksize == 0)
		sb->s_blocksize = nfs_block_bits(server->wsize,
							 &sb->s_blocksize_bits);
	server->wtmult = nfs_block_bits(fsinfo.wtmult, NULL);

	server->dtsize = nfs_block_size(fsinfo.dtpref, NULL);
	if (server->dtsize > PAGE_CACHE_SIZE)
		server->dtsize = PAGE_CACHE_SIZE;
	if (server->dtsize > server->rsize)
		server->dtsize = server->rsize;

	if (server->flags & NFS_MOUNT_NOAC) {
		server->acregmin = server->acregmax = 0;
		server->acdirmin = server->acdirmax = 0;
		sb->s_flags |= MS_SYNCHRONOUS;
	}
	server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD;

	sb->s_maxbytes = fsinfo.maxfilesize;
	if (sb->s_maxbytes > MAX_LFS_FILESIZE) 
		sb->s_maxbytes = MAX_LFS_FILESIZE; 

	server->client->cl_intr = (server->flags & NFS_MOUNT_INTR) ? 1 : 0;
	server->client->cl_softrtry = (server->flags & NFS_MOUNT_SOFT) ? 1 : 0;

	/* We're airborne Set socket buffersize */
	rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100);
	return 0;
	/* Yargs. It didn't work out. */
out_no_root:
	dprintk("nfs_sb_init: get root inode failed: errno %d\n", -no_root_error);
	if (!IS_ERR(root_inode))
		iput(root_inode);
	return no_root_error;
}

361 362 363 364 365 366 367 368 369 370 371
static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned int timeo, unsigned int retrans)
{
	to->to_initval = timeo * HZ / 10;
	to->to_retries = retrans;
	if (!to->to_retries)
		to->to_retries = 2;

	switch (proto) {
	case IPPROTO_TCP:
		if (!to->to_initval)
			to->to_initval = 60 * HZ;
372 373
		if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
			to->to_initval = NFS_MAX_TCP_TIMEOUT;
374 375 376 377 378 379 380 381
		to->to_increment = to->to_initval;
		to->to_maxval = to->to_initval + (to->to_increment * to->to_retries);
		to->to_exponential = 0;
		break;
	case IPPROTO_UDP:
	default:
		if (!to->to_initval)
			to->to_initval = 11 * HZ / 10;
382 383 384
		if (to->to_initval > NFS_MAX_UDP_TIMEOUT)
			to->to_initval = NFS_MAX_UDP_TIMEOUT;
		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
385 386 387 388 389
		to->to_exponential = 1;
		break;
	}
}

L
Linus Torvalds 已提交
390 391 392 393 394 395 396 397 398
/*
 * Create an RPC client handle.
 */
static struct rpc_clnt *
nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data)
{
	struct rpc_timeout	timeparms;
	struct rpc_xprt		*xprt = NULL;
	struct rpc_clnt		*clnt = NULL;
399
	int			proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP;
L
Linus Torvalds 已提交
400

401
	nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans);
L
Linus Torvalds 已提交
402 403

	/* create transport and client */
404
	xprt = xprt_create_proto(proto, &server->addr, &timeparms);
L
Linus Torvalds 已提交
405
	if (IS_ERR(xprt)) {
406 407
		dprintk("%s: cannot create RPC transport. Error = %ld\n",
				__FUNCTION__, PTR_ERR(xprt));
L
Linus Torvalds 已提交
408 409 410 411 412
		return (struct rpc_clnt *)xprt;
	}
	clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
				 server->rpc_ops->version, data->pseudoflavor);
	if (IS_ERR(clnt)) {
413 414
		dprintk("%s: cannot create RPC client. Error = %ld\n",
				__FUNCTION__, PTR_ERR(xprt));
L
Linus Torvalds 已提交
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
		goto out_fail;
	}

	clnt->cl_intr     = 1;
	clnt->cl_softrtry = 1;
	clnt->cl_chatty   = 1;

	return clnt;

out_fail:
	return clnt;
}

/*
 * The way this works is that the mount process passes a structure
 * in the data argument which contains the server's IP address
 * and the root file handle obtained from the server's mount
 * daemon. We stash these away in the private superblock fields.
 */
static int
nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent)
{
	struct nfs_server	*server;
	rpc_authflavor_t	authflavor;

	server           = NFS_SB(sb);
	sb->s_blocksize_bits = 0;
	sb->s_blocksize = 0;
	if (data->bsize)
		sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits);
	if (data->rsize)
		server->rsize = nfs_block_size(data->rsize, NULL);
	if (data->wsize)
		server->wsize = nfs_block_size(data->wsize, NULL);
	server->flags    = data->flags & NFS_MOUNT_FLAGMASK;

	server->acregmin = data->acregmin*HZ;
	server->acregmax = data->acregmax*HZ;
	server->acdirmin = data->acdirmin*HZ;
	server->acdirmax = data->acdirmax*HZ;

	/* Start lockd here, before we might error out */
	if (!(server->flags & NFS_MOUNT_NONLM))
		lockd_up();

	server->namelen  = data->namlen;
	server->hostname = kmalloc(strlen(data->hostname) + 1, GFP_KERNEL);
	if (!server->hostname)
		return -ENOMEM;
	strcpy(server->hostname, data->hostname);

	/* Check NFS protocol revision and initialize RPC op vector
	 * and file handle pool. */
#ifdef CONFIG_NFS_V3
469
	if (server->flags & NFS_MOUNT_VER3) {
L
Linus Torvalds 已提交
470 471 472 473 474
		server->rpc_ops = &nfs_v3_clientops;
		server->caps |= NFS_CAP_READDIRPLUS;
	} else {
		server->rpc_ops = &nfs_v2_clientops;
	}
475 476 477
#else
	server->rpc_ops = &nfs_v2_clientops;
#endif
L
Linus Torvalds 已提交
478 479 480 481 482 483 484 485 486 487 488 489 490

	/* Fill in pseudoflavor for mount version < 5 */
	if (!(data->flags & NFS_MOUNT_SECFLAVOUR))
		data->pseudoflavor = RPC_AUTH_UNIX;
	authflavor = data->pseudoflavor;	/* save for sb_init() */
	/* XXX maybe we want to add a server->pseudoflavor field */

	/* Create RPC client handles */
	server->client = nfs_create_client(server, data);
	if (IS_ERR(server->client))
		return PTR_ERR(server->client);
	/* RFC 2623, sec 2.3.2 */
	if (authflavor != RPC_AUTH_UNIX) {
491 492
		struct rpc_auth *auth;

L
Linus Torvalds 已提交
493 494 495
		server->client_sys = rpc_clone_client(server->client);
		if (IS_ERR(server->client_sys))
			return PTR_ERR(server->client_sys);
496 497 498
		auth = rpcauth_create(RPC_AUTH_UNIX, server->client_sys);
		if (IS_ERR(auth))
			return PTR_ERR(auth);
L
Linus Torvalds 已提交
499 500 501 502 503
	} else {
		atomic_inc(&server->client->cl_count);
		server->client_sys = server->client;
	}
	if (server->flags & NFS_MOUNT_VER3) {
504 505 506 507 508 509 510 511 512 513
#ifdef CONFIG_NFS_V3_ACL
		if (!(server->flags & NFS_MOUNT_NOACL)) {
			server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3);
			/* No errors! Assume that Sun nfsacls are supported */
			if (!IS_ERR(server->client_acl))
				server->caps |= NFS_CAP_ACLS;
		}
#else
		server->flags &= ~NFS_MOUNT_NOACL;
#endif /* CONFIG_NFS_V3_ACL */
514 515 516 517 518
		/*
		 * The VFS shouldn't apply the umask to mode bits. We will
		 * do so ourselves when necessary.
		 */
		sb->s_flags |= MS_POSIXACL;
L
Linus Torvalds 已提交
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600
		if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN)
			server->namelen = NFS3_MAXNAMLEN;
		sb->s_time_gran = 1;
	} else {
		if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN)
			server->namelen = NFS2_MAXNAMLEN;
	}

	sb->s_op = &nfs_sops;
	return nfs_sb_init(sb, authflavor);
}

static int
nfs_statfs(struct super_block *sb, struct kstatfs *buf)
{
	struct nfs_server *server = NFS_SB(sb);
	unsigned char blockbits;
	unsigned long blockres;
	struct nfs_fh *rootfh = NFS_FH(sb->s_root->d_inode);
	struct nfs_fattr fattr;
	struct nfs_fsstat res = {
			.fattr = &fattr,
	};
	int error;

	lock_kernel();

	error = server->rpc_ops->statfs(server, rootfh, &res);
	buf->f_type = NFS_SUPER_MAGIC;
	if (error < 0)
		goto out_err;

	/*
	 * Current versions of glibc do not correctly handle the
	 * case where f_frsize != f_bsize.  Eventually we want to
	 * report the value of wtmult in this field.
	 */
	buf->f_frsize = sb->s_blocksize;

	/*
	 * On most *nix systems, f_blocks, f_bfree, and f_bavail
	 * are reported in units of f_frsize.  Linux hasn't had
	 * an f_frsize field in its statfs struct until recently,
	 * thus historically Linux's sys_statfs reports these
	 * fields in units of f_bsize.
	 */
	buf->f_bsize = sb->s_blocksize;
	blockbits = sb->s_blocksize_bits;
	blockres = (1 << blockbits) - 1;
	buf->f_blocks = (res.tbytes + blockres) >> blockbits;
	buf->f_bfree = (res.fbytes + blockres) >> blockbits;
	buf->f_bavail = (res.abytes + blockres) >> blockbits;

	buf->f_files = res.tfiles;
	buf->f_ffree = res.afiles;

	buf->f_namelen = server->namelen;
 out:
	unlock_kernel();

	return 0;

 out_err:
	printk(KERN_WARNING "nfs_statfs: statfs error = %d\n", -error);
	buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1;
	goto out;

}

static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt)
{
	static struct proc_nfs_info {
		int flag;
		char *str;
		char *nostr;
	} nfs_info[] = {
		{ NFS_MOUNT_SOFT, ",soft", ",hard" },
		{ NFS_MOUNT_INTR, ",intr", "" },
		{ NFS_MOUNT_POSIX, ",posix", "" },
		{ NFS_MOUNT_NOCTO, ",nocto", "" },
		{ NFS_MOUNT_NOAC, ",noac", "" },
		{ NFS_MOUNT_NONLM, ",nolock", ",lock" },
601
		{ NFS_MOUNT_NOACL, ",noacl", "" },
L
Linus Torvalds 已提交
602 603 604 605
		{ 0, NULL, NULL }
	};
	struct proc_nfs_info *nfs_infop;
	struct nfs_server *nfss = NFS_SB(mnt->mnt_sb);
606 607
	char buf[12];
	char *proto;
L
Linus Torvalds 已提交
608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625

	seq_printf(m, ",v%d", nfss->rpc_ops->version);
	seq_printf(m, ",rsize=%d", nfss->rsize);
	seq_printf(m, ",wsize=%d", nfss->wsize);
	if (nfss->acregmin != 3*HZ)
		seq_printf(m, ",acregmin=%d", nfss->acregmin/HZ);
	if (nfss->acregmax != 60*HZ)
		seq_printf(m, ",acregmax=%d", nfss->acregmax/HZ);
	if (nfss->acdirmin != 30*HZ)
		seq_printf(m, ",acdirmin=%d", nfss->acdirmin/HZ);
	if (nfss->acdirmax != 60*HZ)
		seq_printf(m, ",acdirmax=%d", nfss->acdirmax/HZ);
	for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) {
		if (nfss->flags & nfs_infop->flag)
			seq_puts(m, nfs_infop->str);
		else
			seq_puts(m, nfs_infop->nostr);
	}
626 627 628 629 630 631 632 633 634 635 636 637
	switch (nfss->client->cl_xprt->prot) {
		case IPPROTO_TCP:
			proto = "tcp";
			break;
		case IPPROTO_UDP:
			proto = "udp";
			break;
		default:
			snprintf(buf, sizeof(buf), "%u", nfss->client->cl_xprt->prot);
			proto = buf;
	}
	seq_printf(m, ",proto=%s", proto);
L
Linus Torvalds 已提交
638 639 640 641 642 643 644 645 646 647 648 649 650 651
	seq_puts(m, ",addr=");
	seq_escape(m, nfss->hostname, " \t\n\\");
	return 0;
}

/*
 * Invalidate the local caches
 */
void
nfs_zap_caches(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	int mode = inode->i_mode;

652 653
	spin_lock(&inode->i_lock);

L
Linus Torvalds 已提交
654 655 656 657 658
	NFS_ATTRTIMEO(inode) = NFS_MINATTRTIMEO(inode);
	NFS_ATTRTIMEO_UPDATE(inode) = jiffies;

	memset(NFS_COOKIEVERF(inode), 0, sizeof(NFS_COOKIEVERF(inode)));
	if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))
659
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
L
Linus Torvalds 已提交
660
	else
661
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
662 663

	spin_unlock(&inode->i_lock);
664 665 666 667 668 669 670 671 672
}

static void nfs_zap_acl_cache(struct inode *inode)
{
	void (*clear_acl_cache)(struct inode *);

	clear_acl_cache = NFS_PROTO(inode)->clear_acl_cache;
	if (clear_acl_cache != NULL)
		clear_acl_cache(inode);
673
	spin_lock(&inode->i_lock);
674
	NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_ACL;
675
	spin_unlock(&inode->i_lock);
L
Linus Torvalds 已提交
676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771
}

/*
 * Invalidate, but do not unhash, the inode
 */
static void
nfs_invalidate_inode(struct inode *inode)
{
	umode_t save_mode = inode->i_mode;

	make_bad_inode(inode);
	inode->i_mode = save_mode;
	nfs_zap_caches(inode);
}

struct nfs_find_desc {
	struct nfs_fh		*fh;
	struct nfs_fattr	*fattr;
};

/*
 * In NFSv3 we can have 64bit inode numbers. In order to support
 * this, and re-exported directories (also seen in NFSv2)
 * we are forced to allow 2 different inodes to have the same
 * i_ino.
 */
static int
nfs_find_actor(struct inode *inode, void *opaque)
{
	struct nfs_find_desc	*desc = (struct nfs_find_desc *)opaque;
	struct nfs_fh		*fh = desc->fh;
	struct nfs_fattr	*fattr = desc->fattr;

	if (NFS_FILEID(inode) != fattr->fileid)
		return 0;
	if (nfs_compare_fh(NFS_FH(inode), fh))
		return 0;
	if (is_bad_inode(inode) || NFS_STALE(inode))
		return 0;
	return 1;
}

static int
nfs_init_locked(struct inode *inode, void *opaque)
{
	struct nfs_find_desc	*desc = (struct nfs_find_desc *)opaque;
	struct nfs_fattr	*fattr = desc->fattr;

	NFS_FILEID(inode) = fattr->fileid;
	nfs_copy_fh(NFS_FH(inode), desc->fh);
	return 0;
}

/* Don't use READDIRPLUS on directories that we believe are too large */
#define NFS_LIMIT_READDIRPLUS (8*PAGE_SIZE)

/*
 * This is our front-end to iget that looks up inodes by file handle
 * instead of inode number.
 */
struct inode *
nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
{
	struct nfs_find_desc desc = {
		.fh	= fh,
		.fattr	= fattr
	};
	struct inode *inode = NULL;
	unsigned long hash;

	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
		goto out_no_inode;

	if (!fattr->nlink) {
		printk("NFS: Buggy server - nlink == 0!\n");
		goto out_no_inode;
	}

	hash = nfs_fattr_to_ino_t(fattr);

	if (!(inode = iget5_locked(sb, hash, nfs_find_actor, nfs_init_locked, &desc)))
		goto out_no_inode;

	if (inode->i_state & I_NEW) {
		struct nfs_inode *nfsi = NFS_I(inode);

		/* We set i_ino for the few things that still rely on it,
		 * such as stat(2) */
		inode->i_ino = hash;

		/* We can't support update_atime(), since the server will reset it */
		inode->i_flags |= S_NOATIME|S_NOCMTIME;
		inode->i_mode = fattr->mode;
		/* Why so? Because we want revalidate for devices/FIFOs, and
		 * that's precisely what we have in nfs_file_inode_operations.
		 */
772
		inode->i_op = NFS_SB(sb)->rpc_ops->file_inode_ops;
L
Linus Torvalds 已提交
773 774 775 776 777 778 779 780 781
		if (S_ISREG(inode->i_mode)) {
			inode->i_fop = &nfs_file_operations;
			inode->i_data.a_ops = &nfs_file_aops;
			inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info;
		} else if (S_ISDIR(inode->i_mode)) {
			inode->i_op = NFS_SB(sb)->rpc_ops->dir_inode_ops;
			inode->i_fop = &nfs_dir_operations;
			if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS)
			    && fattr->size <= NFS_LIMIT_READDIRPLUS)
782
				set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode));
L
Linus Torvalds 已提交
783 784 785 786 787
		} else if (S_ISLNK(inode->i_mode))
			inode->i_op = &nfs_symlink_inode_operations;
		else
			init_special_inode(inode, inode->i_mode, fattr->rdev);

788 789
		nfsi->read_cache_jiffies = fattr->time_start;
		nfsi->last_updated = jiffies;
L
Linus Torvalds 已提交
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856
		inode->i_atime = fattr->atime;
		inode->i_mtime = fattr->mtime;
		inode->i_ctime = fattr->ctime;
		if (fattr->valid & NFS_ATTR_FATTR_V4)
			nfsi->change_attr = fattr->change_attr;
		inode->i_size = nfs_size_to_loff_t(fattr->size);
		inode->i_nlink = fattr->nlink;
		inode->i_uid = fattr->uid;
		inode->i_gid = fattr->gid;
		if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) {
			/*
			 * report the blocks in 512byte units
			 */
			inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
			inode->i_blksize = inode->i_sb->s_blocksize;
		} else {
			inode->i_blocks = fattr->du.nfs2.blocks;
			inode->i_blksize = fattr->du.nfs2.blocksize;
		}
		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
		nfsi->attrtimeo_timestamp = jiffies;
		memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
		nfsi->cache_access.cred = NULL;

		unlock_new_inode(inode);
	} else
		nfs_refresh_inode(inode, fattr);
	dprintk("NFS: nfs_fhget(%s/%Ld ct=%d)\n",
		inode->i_sb->s_id,
		(long long)NFS_FILEID(inode),
		atomic_read(&inode->i_count));

out:
	return inode;

out_no_inode:
	printk("nfs_fhget: iget failed\n");
	goto out;
}

#define NFS_VALID_ATTRS (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_SIZE|ATTR_ATIME|ATTR_ATIME_SET|ATTR_MTIME|ATTR_MTIME_SET)

int
nfs_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	struct nfs_fattr fattr;
	int error;

	if (attr->ia_valid & ATTR_SIZE) {
		if (!S_ISREG(inode->i_mode) || attr->ia_size == i_size_read(inode))
			attr->ia_valid &= ~ATTR_SIZE;
	}

	/* Optimization: if the end result is no change, don't RPC */
	attr->ia_valid &= NFS_VALID_ATTRS;
	if (attr->ia_valid == 0)
		return 0;

	lock_kernel();
	nfs_begin_data_update(inode);
	/* Write all dirty data if we're changing file permissions or size */
	if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_SIZE)) != 0) {
		if (filemap_fdatawrite(inode->i_mapping) == 0)
			filemap_fdatawait(inode->i_mapping);
		nfs_wb_all(inode);
	}
857 858 859 860 861
	/*
	 * Return any delegations if we're going to change ACLs
	 */
	if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0)
		nfs_inode_return_delegation(inode);
L
Linus Torvalds 已提交
862
	error = NFS_PROTO(inode)->setattr(dentry, &fattr, attr);
863
	if (error == 0)
L
Linus Torvalds 已提交
864
		nfs_refresh_inode(inode, &fattr);
865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
	nfs_end_data_update(inode);
	unlock_kernel();
	return error;
}

/**
 * nfs_setattr_update_inode - Update inode metadata after a setattr call.
 * @inode: pointer to struct inode
 * @attr: pointer to struct iattr
 *
 * Note: we do this in the *proc.c in order to ensure that
 *       it works for things like exclusive creates too.
 */
void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr)
{
	if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) {
L
Linus Torvalds 已提交
881
		if ((attr->ia_valid & ATTR_MODE) != 0) {
882 883
			int mode = attr->ia_mode & S_IALLUGO;
			mode |= inode->i_mode & ~S_IALLUGO;
L
Linus Torvalds 已提交
884 885 886 887 888 889
			inode->i_mode = mode;
		}
		if ((attr->ia_valid & ATTR_UID) != 0)
			inode->i_uid = attr->ia_uid;
		if ((attr->ia_valid & ATTR_GID) != 0)
			inode->i_gid = attr->ia_gid;
890
		spin_lock(&inode->i_lock);
891
		NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
892
		spin_unlock(&inode->i_lock);
893 894 895 896 897
	}
	if ((attr->ia_valid & ATTR_SIZE) != 0) {
		inode->i_size = attr->ia_size;
		vmtruncate(inode, attr->ia_size);
	}
L
Linus Torvalds 已提交
898 899
}

900 901 902 903 904 905 906 907
static int nfs_wait_schedule(void *word)
{
	if (signal_pending(current))
		return -ERESTARTSYS;
	schedule();
	return 0;
}

L
Linus Torvalds 已提交
908 909 910
/*
 * Wait for the inode to get unlocked.
 */
911
static int nfs_wait_on_inode(struct inode *inode)
L
Linus Torvalds 已提交
912 913 914
{
	struct rpc_clnt	*clnt = NFS_CLIENT(inode);
	struct nfs_inode *nfsi = NFS_I(inode);
915
	sigset_t oldmask;
L
Linus Torvalds 已提交
916
	int error;
917 918 919 920 921 922

	rpc_clnt_sigmask(clnt, &oldmask);
	error = wait_on_bit_lock(&nfsi->flags, NFS_INO_REVALIDATING,
					nfs_wait_schedule, TASK_INTERRUPTIBLE);
	rpc_clnt_sigunmask(clnt, &oldmask);

L
Linus Torvalds 已提交
923 924 925
	return error;
}

926 927 928 929 930 931 932 933 934
static void nfs_wake_up_inode(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	clear_bit(NFS_INO_REVALIDATING, &nfsi->flags);
	smp_mb__after_clear_bit();
	wake_up_bit(&nfsi->flags, NFS_INO_REVALIDATING);
}

L
Linus Torvalds 已提交
935 936 937
int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
938
	int need_atime = NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATIME;
L
Linus Torvalds 已提交
939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966
	int err;

	if (__IS_FLG(inode, MS_NOATIME))
		need_atime = 0;
	else if (__IS_FLG(inode, MS_NODIRATIME) && S_ISDIR(inode->i_mode))
		need_atime = 0;
	/* We may force a getattr if the user cares about atime */
	if (need_atime)
		err = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
	else
		err = nfs_revalidate_inode(NFS_SERVER(inode), inode);
	if (!err)
		generic_fillattr(inode, stat);
	return err;
}

struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred)
{
	struct nfs_open_context *ctx;

	ctx = (struct nfs_open_context *)kmalloc(sizeof(*ctx), GFP_KERNEL);
	if (ctx != NULL) {
		atomic_set(&ctx->count, 1);
		ctx->dentry = dget(dentry);
		ctx->cred = get_rpccred(cred);
		ctx->state = NULL;
		ctx->lockowner = current->files;
		ctx->error = 0;
967
		ctx->dir_cookie = 0;
L
Linus Torvalds 已提交
968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
	}
	return ctx;
}

struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx)
{
	if (ctx != NULL)
		atomic_inc(&ctx->count);
	return ctx;
}

void put_nfs_open_context(struct nfs_open_context *ctx)
{
	if (atomic_dec_and_test(&ctx->count)) {
		if (!list_empty(&ctx->list)) {
			struct inode *inode = ctx->dentry->d_inode;
			spin_lock(&inode->i_lock);
			list_del(&ctx->list);
			spin_unlock(&inode->i_lock);
		}
		if (ctx->state != NULL)
			nfs4_close_state(ctx->state, ctx->mode);
		if (ctx->cred != NULL)
			put_rpccred(ctx->cred);
		dput(ctx->dentry);
		kfree(ctx);
	}
}

/*
 * Ensure that mmap has a recent RPC credential for use when writing out
 * shared pages
 */
void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx)
{
	struct inode *inode = filp->f_dentry->d_inode;
	struct nfs_inode *nfsi = NFS_I(inode);

	filp->private_data = get_nfs_open_context(ctx);
	spin_lock(&inode->i_lock);
	list_add(&ctx->list, &nfsi->open_files);
	spin_unlock(&inode->i_lock);
}

struct nfs_open_context *nfs_find_open_context(struct inode *inode, int mode)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs_open_context *pos, *ctx = NULL;

	spin_lock(&inode->i_lock);
	list_for_each_entry(pos, &nfsi->open_files, list) {
		if ((pos->mode & mode) == mode) {
			ctx = get_nfs_open_context(pos);
			break;
		}
	}
	spin_unlock(&inode->i_lock);
	return ctx;
}

void nfs_file_clear_open_context(struct file *filp)
{
	struct inode *inode = filp->f_dentry->d_inode;
	struct nfs_open_context *ctx = (struct nfs_open_context *)filp->private_data;

	if (ctx) {
		filp->private_data = NULL;
		spin_lock(&inode->i_lock);
		list_move_tail(&ctx->list, &NFS_I(inode)->open_files);
		spin_unlock(&inode->i_lock);
		put_nfs_open_context(ctx);
	}
}

/*
 * These allocate and release file read/write context information.
 */
int nfs_open(struct inode *inode, struct file *filp)
{
	struct nfs_open_context *ctx;
	struct rpc_cred *cred;

	cred = rpcauth_lookupcred(NFS_CLIENT(inode)->cl_auth, 0);
	if (IS_ERR(cred))
		return PTR_ERR(cred);
	ctx = alloc_nfs_open_context(filp->f_dentry, cred);
	put_rpccred(cred);
	if (ctx == NULL)
		return -ENOMEM;
	ctx->mode = filp->f_mode;
	nfs_file_set_open_context(filp, ctx);
	put_nfs_open_context(ctx);
	if ((filp->f_mode & FMODE_WRITE) != 0)
		nfs_begin_data_update(inode);
	return 0;
}

int nfs_release(struct inode *inode, struct file *filp)
{
	if ((filp->f_mode & FMODE_WRITE) != 0)
		nfs_end_data_update(inode);
	nfs_file_clear_open_context(filp);
	return 0;
}

/*
 * This function is called whenever some part of NFS notices that
 * the cached attributes have to be refreshed.
 */
int
__nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
{
	int		 status = -ESTALE;
	struct nfs_fattr fattr;
	struct nfs_inode *nfsi = NFS_I(inode);
	unsigned long verifier;
1084
	unsigned long cache_validity;
L
Linus Torvalds 已提交
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094

	dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n",
		inode->i_sb->s_id, (long long)NFS_FILEID(inode));

	lock_kernel();
	if (!inode || is_bad_inode(inode))
 		goto out_nowait;
	if (NFS_STALE(inode))
 		goto out_nowait;

1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106
	status = nfs_wait_on_inode(inode);
	if (status < 0)
		goto out;
	if (NFS_STALE(inode)) {
		status = -ESTALE;
		/* Do we trust the cached ESTALE? */
		if (NFS_ATTRTIMEO(inode) != 0) {
			if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ATIME)) {
				/* no */
			} else
				goto out;
		}
L
Linus Torvalds 已提交
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118
	}

	/* Protect against RPC races by saving the change attribute */
	verifier = nfs_save_change_attribute(inode);
	status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), &fattr);
	if (status != 0) {
		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) getattr failed, error=%d\n",
			 inode->i_sb->s_id,
			 (long long)NFS_FILEID(inode), status);
		if (status == -ESTALE) {
			nfs_zap_caches(inode);
			if (!S_ISDIR(inode->i_mode))
1119
				set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
L
Linus Torvalds 已提交
1120 1121 1122 1123
		}
		goto out;
	}

1124
	spin_lock(&inode->i_lock);
L
Linus Torvalds 已提交
1125 1126
	status = nfs_update_inode(inode, &fattr, verifier);
	if (status) {
1127
		spin_unlock(&inode->i_lock);
L
Linus Torvalds 已提交
1128 1129 1130 1131 1132
		dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n",
			 inode->i_sb->s_id,
			 (long long)NFS_FILEID(inode), status);
		goto out;
	}
1133 1134 1135
	cache_validity = nfsi->cache_validity;
	nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;

L
Linus Torvalds 已提交
1136 1137 1138 1139
	/*
	 * We may need to keep the attributes marked as invalid if
	 * we raced with nfs_end_attr_update().
	 */
1140
	if (time_after_eq(verifier, nfsi->cache_change_attribute))
1141
		nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
1142
	spin_unlock(&inode->i_lock);
1143

1144
	nfs_revalidate_mapping(inode, inode->i_mapping);
1145 1146

	if (cache_validity & NFS_INO_INVALID_ACL)
1147
		nfs_zap_acl_cache(inode);
1148

L
Linus Torvalds 已提交
1149 1150 1151 1152
	dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n",
		inode->i_sb->s_id,
		(long long)NFS_FILEID(inode));

1153 1154 1155
 out:
	nfs_wake_up_inode(inode);

L
Linus Torvalds 已提交
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
 out_nowait:
	unlock_kernel();
	return status;
}

int nfs_attribute_timeout(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (nfs_have_delegation(inode, FMODE_READ))
		return 0;
	return time_after(jiffies, nfsi->read_cache_jiffies+nfsi->attrtimeo);
}

/**
 * nfs_revalidate_inode - Revalidate the inode attributes
 * @server - pointer to nfs_server struct
 * @inode - pointer to inode struct
 *
 * Updates inode attribute information by retrieving the data from the server.
 */
int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode)
{
1179
	if (!(NFS_I(inode)->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA))
L
Linus Torvalds 已提交
1180 1181 1182 1183 1184
			&& !nfs_attribute_timeout(inode))
		return NFS_STALE(inode) ? -ESTALE : 0;
	return __nfs_revalidate_inode(server, inode);
}

1185 1186 1187 1188 1189 1190 1191 1192 1193
/**
 * nfs_revalidate_mapping - Revalidate the pagecache
 * @inode - pointer to host inode
 * @mapping - pointer to mapping
 */
void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping)
{
	struct nfs_inode *nfsi = NFS_I(inode);

1194
	if (nfsi->cache_validity & NFS_INO_INVALID_DATA) {
1195 1196 1197 1198 1199 1200
		if (S_ISREG(inode->i_mode)) {
			if (filemap_fdatawrite(mapping) == 0)
				filemap_fdatawait(mapping);
			nfs_wb_all(inode);
		}
		invalidate_inode_pages2(mapping);
1201 1202

		spin_lock(&inode->i_lock);
1203
		nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
1204 1205 1206
		if (S_ISDIR(inode->i_mode)) {
			memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
			/* This ensures we revalidate child dentries */
1207
			nfsi->cache_change_attribute = jiffies;
1208
		}
1209 1210
		spin_unlock(&inode->i_lock);

1211 1212 1213 1214 1215 1216
		dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
				inode->i_sb->s_id,
				(long long)NFS_FILEID(inode));
	}
}

L
Linus Torvalds 已提交
1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238
/**
 * nfs_begin_data_update
 * @inode - pointer to inode
 * Declare that a set of operations will update file data on the server
 */
void nfs_begin_data_update(struct inode *inode)
{
	atomic_inc(&NFS_I(inode)->data_updates);
}

/**
 * nfs_end_data_update
 * @inode - pointer to inode
 * Declare end of the operations that will update file data
 * This will mark the inode as immediately needing revalidation
 * of its attribute cache.
 */
void nfs_end_data_update(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	if (!nfs_have_delegation(inode, FMODE_READ)) {
1239 1240 1241
		/* Directories and symlinks: invalidate page cache */
		if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) {
			spin_lock(&inode->i_lock);
1242
			nfsi->cache_validity |= NFS_INO_INVALID_DATA;
1243 1244
			spin_unlock(&inode->i_lock);
		}
L
Linus Torvalds 已提交
1245
	}
1246
	nfsi->cache_change_attribute = jiffies;
L
Linus Torvalds 已提交
1247 1248 1249 1250
	atomic_dec(&nfsi->data_updates);
}

/**
1251
 * nfs_check_inode_attributes - verify consistency of the inode attribute cache
L
Linus Torvalds 已提交
1252 1253 1254 1255 1256 1257 1258
 * @inode - pointer to inode
 * @fattr - updated attributes
 *
 * Verifies the attribute cache. If we have just changed the attributes,
 * so that fattr carries weak cache consistency data, then it may
 * also update the ctime/mtime/change_attribute.
 */
1259
static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fattr)
L
Linus Torvalds 已提交
1260 1261 1262 1263 1264
{
	struct nfs_inode *nfsi = NFS_I(inode);
	loff_t cur_size, new_isize;
	int data_unstable;

1265

L
Linus Torvalds 已提交
1266 1267 1268 1269 1270 1271 1272
	/* Are we in the process of updating data on the server? */
	data_unstable = nfs_caches_unstable(inode);

	if (fattr->valid & NFS_ATTR_FATTR_V4) {
		if ((fattr->valid & NFS_ATTR_PRE_CHANGE) != 0
				&& nfsi->change_attr == fattr->pre_change_attr)
			nfsi->change_attr = fattr->change_attr;
1273
		if (nfsi->change_attr != fattr->change_attr) {
1274
			nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
1275
			if (!data_unstable)
1276
				nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
1277
		}
L
Linus Torvalds 已提交
1278 1279
	}

1280 1281
	if ((fattr->valid & NFS_ATTR_FATTR) == 0) {
		spin_unlock(&inode->i_lock);
L
Linus Torvalds 已提交
1282
		return 0;
1283
	}
L
Linus Torvalds 已提交
1284 1285 1286

	/* Has the inode gone and changed behind our back? */
	if (nfsi->fileid != fattr->fileid
1287 1288
			|| (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) {
		spin_unlock(&inode->i_lock);
L
Linus Torvalds 已提交
1289
		return -EIO;
1290
	}
L
Linus Torvalds 已提交
1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303

	cur_size = i_size_read(inode);
 	new_isize = nfs_size_to_loff_t(fattr->size);

	/* If we have atomic WCC data, we may update some attributes */
	if ((fattr->valid & NFS_ATTR_WCC) != 0) {
		if (timespec_equal(&inode->i_ctime, &fattr->pre_ctime))
			memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
		if (timespec_equal(&inode->i_mtime, &fattr->pre_mtime))
			memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
	}

	/* Verify a few of the more important attributes */
1304
	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
1305
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
1306
		if (!data_unstable)
1307
			nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
1308 1309
	}
	if (cur_size != new_isize) {
1310
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
1311
		if (nfsi->npages == 0)
1312
			nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE;
1313
	}
L
Linus Torvalds 已提交
1314 1315 1316 1317 1318

	/* Have any file permissions changed? */
	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)
			|| inode->i_uid != fattr->uid
			|| inode->i_gid != fattr->gid)
1319
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL;
L
Linus Torvalds 已提交
1320 1321 1322

	/* Has the link count changed? */
	if (inode->i_nlink != fattr->nlink)
1323
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR;
L
Linus Torvalds 已提交
1324 1325

	if (!timespec_equal(&inode->i_atime, &fattr->atime))
1326
		nfsi->cache_validity |= NFS_INO_INVALID_ATIME;
L
Linus Torvalds 已提交
1327

1328
	nfsi->read_cache_jiffies = fattr->time_start;
L
Linus Torvalds 已提交
1329 1330 1331
	return 0;
}

1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361
/**
 * nfs_refresh_inode - try to update the inode attribute cache
 * @inode - pointer to inode
 * @fattr - updated attributes
 *
 * Check that an RPC call that returned attributes has not overlapped with
 * other recent updates of the inode metadata, then decide whether it is
 * safe to do a full update of the inode attributes, or whether just to
 * call nfs_check_inode_attributes.
 */
int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	int status;

	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
		return 0;
	spin_lock(&inode->i_lock);
	nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE;
	if (nfs_verify_change_attribute(inode, fattr->time_start))
		nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME);
	if (time_after(fattr->time_start, nfsi->last_updated))
		status = nfs_update_inode(inode, fattr, fattr->time_start);
	else
		status = nfs_check_inode_attributes(inode, fattr);

	spin_unlock(&inode->i_lock);
	return status;
}

1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388
/**
 * nfs_post_op_update_inode - try to update the inode attribute cache
 * @inode - pointer to inode
 * @fattr - updated attributes
 *
 * After an operation that has changed the inode metadata, mark the
 * attribute cache as being invalid, then try to update it.
 */
int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	int status = 0;

	spin_lock(&inode->i_lock);
	if (unlikely((fattr->valid & NFS_ATTR_FATTR) == 0)) {
		nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS;
		goto out;
	}
	status = nfs_update_inode(inode, fattr, fattr->time_start);
	if (time_after_eq(fattr->time_start, nfsi->cache_change_attribute))
		nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME|NFS_INO_REVAL_PAGECACHE);
	nfsi->cache_change_attribute = jiffies;
out:
	spin_unlock(&inode->i_lock);
	return status;
}

L
Linus Torvalds 已提交
1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403
/*
 * Many nfs protocol calls return the new file attributes after
 * an operation.  Here we update the inode to reflect the state
 * of the server's inode.
 *
 * This is a bit tricky because we have to make sure all dirty pages
 * have been sent off to the server before calling invalidate_inode_pages.
 * To make sure no other process adds more write requests while we try
 * our best to flush them, we make them sleep during the attribute refresh.
 *
 * A very similar scenario holds for the dir cache.
 */
static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsigned long verifier)
{
	struct nfs_inode *nfsi = NFS_I(inode);
1404
	loff_t cur_isize, new_isize;
L
Linus Torvalds 已提交
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426
	unsigned int	invalid = 0;
	int data_unstable;

	dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n",
			__FUNCTION__, inode->i_sb->s_id, inode->i_ino,
			atomic_read(&inode->i_count), fattr->valid);

	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
		return 0;

	if (nfsi->fileid != fattr->fileid) {
		printk(KERN_ERR "%s: inode number mismatch\n"
		       "expected (%s/0x%Lx), got (%s/0x%Lx)\n",
		       __FUNCTION__,
		       inode->i_sb->s_id, (long long)nfsi->fileid,
		       inode->i_sb->s_id, (long long)fattr->fileid);
		goto out_err;
	}

	/*
	 * Make sure the inode's type hasn't changed.
	 */
1427
	if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT))
L
Linus Torvalds 已提交
1428 1429 1430 1431 1432
		goto out_changed;

	/*
	 * Update the read time so we don't revalidate too often.
	 */
1433 1434
	nfsi->read_cache_jiffies = fattr->time_start;
	nfsi->last_updated = jiffies;
L
Linus Torvalds 已提交
1435 1436

	/* Are we racing with known updates of the metadata on the server? */
1437 1438
	data_unstable = ! (nfs_verify_change_attribute(inode, verifier) ||
		(nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE));
L
Linus Torvalds 已提交
1439

1440
	/* Check if our cached file size is stale */
L
Linus Torvalds 已提交
1441 1442
 	new_isize = nfs_size_to_loff_t(fattr->size);
	cur_isize = i_size_read(inode);
1443 1444 1445 1446
	if (new_isize != cur_isize) {
		/* Do we perhaps have any outstanding writes? */
		if (nfsi->npages == 0) {
			/* No, but did we race with nfs_end_data_update()? */
1447
			if (time_after_eq(verifier,  nfsi->cache_change_attribute)) {
L
Linus Torvalds 已提交
1448
				inode->i_size = new_isize;
1449
				invalid |= NFS_INO_INVALID_DATA;
L
Linus Torvalds 已提交
1450
			}
1451 1452
			invalid |= NFS_INO_INVALID_ATTR;
		} else if (new_isize > cur_isize) {
L
Linus Torvalds 已提交
1453 1454 1455
			inode->i_size = new_isize;
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
		}
1456 1457
		dprintk("NFS: isize change on server for file %s/%ld\n",
				inode->i_sb->s_id, inode->i_ino);
L
Linus Torvalds 已提交
1458 1459
	}

1460
	/* Check if the mtime agrees */
L
Linus Torvalds 已提交
1461 1462
	if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) {
		memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime));
1463 1464
		dprintk("NFS: mtime change on server for file %s/%ld\n",
				inode->i_sb->s_id, inode->i_ino);
L
Linus Torvalds 已提交
1465 1466 1467 1468 1469 1470
		if (!data_unstable)
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
	}

	if ((fattr->valid & NFS_ATTR_FATTR_V4)
	    && nfsi->change_attr != fattr->change_attr) {
1471
		dprintk("NFS: change_attr change on server for file %s/%ld\n",
L
Linus Torvalds 已提交
1472 1473 1474
		       inode->i_sb->s_id, inode->i_ino);
		nfsi->change_attr = fattr->change_attr;
		if (!data_unstable)
1475
			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
L
Linus Torvalds 已提交
1476 1477
	}

1478 1479 1480 1481 1482 1483
	/* If ctime has changed we should definitely clear access+acl caches */
	if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) {
		if (!data_unstable)
			invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
		memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime));
	}
L
Linus Torvalds 已提交
1484 1485 1486 1487 1488
	memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime));

	if ((inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO) ||
	    inode->i_uid != fattr->uid ||
	    inode->i_gid != fattr->gid)
1489
		invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
L
Linus Torvalds 已提交
1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520

	inode->i_mode = fattr->mode;
	inode->i_nlink = fattr->nlink;
	inode->i_uid = fattr->uid;
	inode->i_gid = fattr->gid;

	if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) {
		/*
		 * report the blocks in 512byte units
		 */
		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
		inode->i_blksize = inode->i_sb->s_blocksize;
 	} else {
 		inode->i_blocks = fattr->du.nfs2.blocks;
 		inode->i_blksize = fattr->du.nfs2.blocksize;
 	}

	/* Update attrtimeo value if we're out of the unstable period */
	if (invalid & NFS_INO_INVALID_ATTR) {
		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
		nfsi->attrtimeo_timestamp = jiffies;
	} else if (time_after(jiffies, nfsi->attrtimeo_timestamp+nfsi->attrtimeo)) {
		if ((nfsi->attrtimeo <<= 1) > NFS_MAXATTRTIMEO(inode))
			nfsi->attrtimeo = NFS_MAXATTRTIMEO(inode);
		nfsi->attrtimeo_timestamp = jiffies;
	}
	/* Don't invalidate the data if we were to blame */
	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
				|| S_ISLNK(inode->i_mode)))
		invalid &= ~NFS_INO_INVALID_DATA;
	if (!nfs_have_delegation(inode, FMODE_READ))
1521
		nfsi->cache_validity |= invalid;
L
Linus Torvalds 已提交
1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538

	return 0;
 out_changed:
	/*
	 * Big trouble! The inode has become a different object.
	 */
#ifdef NFS_PARANOIA
	printk(KERN_DEBUG "%s: inode %ld mode changed, %07o to %07o\n",
			__FUNCTION__, inode->i_ino, inode->i_mode, fattr->mode);
#endif
	/*
	 * No need to worry about unhashing the dentry, as the
	 * lookup validation will know that the inode is bad.
	 * (But we fall through to invalidate the caches.)
	 */
	nfs_invalidate_inode(inode);
 out_err:
1539
	set_bit(NFS_INO_STALE, &NFS_FLAGS(inode));
L
Linus Torvalds 已提交
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
	return -ESTALE;
}

/*
 * File system information
 */

static int nfs_set_super(struct super_block *s, void *data)
{
	s->s_fs_info = data;
	return set_anon_super(s, data);
}
 
static int nfs_compare_super(struct super_block *sb, void *data)
{
	struct nfs_server *server = data;
	struct nfs_server *old = NFS_SB(sb);

	if (old->addr.sin_addr.s_addr != server->addr.sin_addr.s_addr)
		return 0;
	if (old->addr.sin_port != server->addr.sin_port)
		return 0;
	return !nfs_compare_fh(&old->fh, &server->fh);
}

static struct super_block *nfs_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *raw_data)
{
	int error;
1569
	struct nfs_server *server = NULL;
L
Linus Torvalds 已提交
1570 1571 1572 1573
	struct super_block *s;
	struct nfs_fh *root;
	struct nfs_mount_data *data = raw_data;

1574 1575 1576 1577 1578 1579 1580 1581
	s = ERR_PTR(-EINVAL);
	if (data == NULL) {
		dprintk("%s: missing data argument\n", __FUNCTION__);
		goto out_err;
	}
	if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) {
		dprintk("%s: bad mount version\n", __FUNCTION__);
		goto out_err;
L
Linus Torvalds 已提交
1582
	}
1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614
	switch (data->version) {
		case 1:
			data->namlen = 0;
		case 2:
			data->bsize  = 0;
		case 3:
			if (data->flags & NFS_MOUNT_VER3) {
				dprintk("%s: mount structure version %d does not support NFSv3\n",
						__FUNCTION__,
						data->version);
				goto out_err;
			}
			data->root.size = NFS2_FHSIZE;
			memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE);
		case 4:
			if (data->flags & NFS_MOUNT_SECFLAVOUR) {
				dprintk("%s: mount structure version %d does not support strong security\n",
						__FUNCTION__,
						data->version);
				goto out_err;
			}
		case 5:
			memset(data->context, 0, sizeof(data->context));
	}
#ifndef CONFIG_NFS_V3
	/* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */
	s = ERR_PTR(-EPROTONOSUPPORT);
	if (data->flags & NFS_MOUNT_VER3) {
		dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__);
		goto out_err;
	}
#endif /* CONFIG_NFS_V3 */
L
Linus Torvalds 已提交
1615

1616
	s = ERR_PTR(-ENOMEM);
L
Linus Torvalds 已提交
1617 1618
	server = kmalloc(sizeof(struct nfs_server), GFP_KERNEL);
	if (!server)
1619
		goto out_err;
L
Linus Torvalds 已提交
1620 1621 1622
	memset(server, 0, sizeof(struct nfs_server));
	/* Zero out the NFS state stuff */
	init_nfsv4_state(server);
1623
	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
L
Linus Torvalds 已提交
1624 1625 1626 1627 1628 1629

	root = &server->fh;
	if (data->flags & NFS_MOUNT_VER3)
		root->size = data->root.size;
	else
		root->size = NFS2_FHSIZE;
1630
	s = ERR_PTR(-EINVAL);
L
Linus Torvalds 已提交
1631
	if (root->size > sizeof(root->data)) {
1632 1633
		dprintk("%s: invalid root filehandle\n", __FUNCTION__);
		goto out_err;
L
Linus Torvalds 已提交
1634 1635 1636 1637 1638 1639
	}
	memcpy(root->data, data->root.data, root->size);

	/* We now require that the mount process passes the remote address */
	memcpy(&server->addr, &data->addr, sizeof(server->addr));
	if (server->addr.sin_addr.s_addr == INADDR_ANY) {
1640 1641 1642
		dprintk("%s: mount program didn't pass remote address!\n",
				__FUNCTION__);
		goto out_err;
L
Linus Torvalds 已提交
1643 1644
	}

1645 1646 1647 1648 1649 1650
	/* Fire up rpciod if not yet running */
	s = ERR_PTR(rpciod_up());
	if (IS_ERR(s)) {
		dprintk("%s: couldn't start rpciod! Error = %ld\n",
				__FUNCTION__, PTR_ERR(s));
		goto out_err;
L
Linus Torvalds 已提交
1651 1652
	}

1653 1654 1655
	s = sget(fs_type, nfs_compare_super, nfs_set_super, server);
	if (IS_ERR(s) || s->s_root)
		goto out_rpciod_down;
L
Linus Torvalds 已提交
1656

1657
	s->s_flags = flags;
L
Linus Torvalds 已提交
1658 1659 1660 1661 1662 1663 1664 1665 1666

	error = nfs_fill_super(s, data, flags & MS_VERBOSE ? 1 : 0);
	if (error) {
		up_write(&s->s_umount);
		deactivate_super(s);
		return ERR_PTR(error);
	}
	s->s_flags |= MS_ACTIVE;
	return s;
1667 1668 1669 1670 1671
out_rpciod_down:
	rpciod_down();
out_err:
	kfree(server);
	return s;
L
Linus Torvalds 已提交
1672 1673 1674 1675 1676 1677 1678 1679
}

static void nfs_kill_super(struct super_block *s)
{
	struct nfs_server *server = NFS_SB(s);

	kill_anon_super(s);

1680
	if (!IS_ERR(server->client))
L
Linus Torvalds 已提交
1681
		rpc_shutdown_client(server->client);
1682
	if (!IS_ERR(server->client_sys))
L
Linus Torvalds 已提交
1683
		rpc_shutdown_client(server->client_sys);
1684 1685
	if (!IS_ERR(server->client_acl))
		rpc_shutdown_client(server->client_acl);
L
Linus Torvalds 已提交
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 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730

	if (!(server->flags & NFS_MOUNT_NONLM))
		lockd_down();	/* release rpc.lockd */

	rpciod_down();		/* release rpciod */

	if (server->hostname != NULL)
		kfree(server->hostname);
	kfree(server);
}

static struct file_system_type nfs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "nfs",
	.get_sb		= nfs_get_sb,
	.kill_sb	= nfs_kill_super,
	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
};

#ifdef CONFIG_NFS_V4

static void nfs4_clear_inode(struct inode *);


static struct super_operations nfs4_sops = { 
	.alloc_inode	= nfs_alloc_inode,
	.destroy_inode	= nfs_destroy_inode,
	.write_inode	= nfs_write_inode,
	.delete_inode	= nfs_delete_inode,
	.statfs		= nfs_statfs,
	.clear_inode	= nfs4_clear_inode,
	.umount_begin	= nfs_umount_begin,
	.show_options	= nfs_show_options,
};

/*
 * Clean out any remaining NFSv4 state that might be left over due
 * to open() calls that passed nfs_atomic_lookup, but failed to call
 * nfs_open().
 */
static void nfs4_clear_inode(struct inode *inode)
{
	struct nfs_inode *nfsi = NFS_I(inode);

	/* If we are holding a delegation, return it! */
1731
	nfs_inode_return_delegation(inode);
L
Linus Torvalds 已提交
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
	/* First call standard NFS clear_inode() code */
	nfs_clear_inode(inode);
	/* Now clear out any remaining state */
	while (!list_empty(&nfsi->open_states)) {
		struct nfs4_state *state;
		
		state = list_entry(nfsi->open_states.next,
				struct nfs4_state,
				inode_states);
		dprintk("%s(%s/%Ld): found unclaimed NFSv4 state %p\n",
				__FUNCTION__,
				inode->i_sb->s_id,
				(long long)NFS_FILEID(inode),
				state);
		BUG_ON(atomic_read(&state->count) != 1);
		nfs4_close_state(state, state->state);
	}
}


static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, int silent)
{
	struct nfs_server *server;
	struct nfs4_client *clp = NULL;
	struct rpc_xprt *xprt = NULL;
	struct rpc_clnt *clnt = NULL;
	struct rpc_timeout timeparms;
	rpc_authflavor_t authflavour;
1760
	int err = -EIO;
L
Linus Torvalds 已提交
1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778

	sb->s_blocksize_bits = 0;
	sb->s_blocksize = 0;
	server = NFS_SB(sb);
	if (data->rsize != 0)
		server->rsize = nfs_block_size(data->rsize, NULL);
	if (data->wsize != 0)
		server->wsize = nfs_block_size(data->wsize, NULL);
	server->flags = data->flags & NFS_MOUNT_FLAGMASK;
	server->caps = NFS_CAP_ATOMIC_OPEN;

	server->acregmin = data->acregmin*HZ;
	server->acregmax = data->acregmax*HZ;
	server->acdirmin = data->acdirmin*HZ;
	server->acdirmax = data->acdirmax*HZ;

	server->rpc_ops = &nfs_v4_clientops;

1779
	nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans);
L
Linus Torvalds 已提交
1780 1781 1782

	clp = nfs4_get_client(&server->addr.sin_addr);
	if (!clp) {
1783
		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
L
Linus Torvalds 已提交
1784 1785 1786 1787 1788 1789
		return -EIO;
	}

	/* Now create transport and client */
	authflavour = RPC_AUTH_UNIX;
	if (data->auth_flavourlen != 0) {
1790 1791 1792 1793 1794 1795
		if (data->auth_flavourlen != 1) {
			dprintk("%s: Invalid number of RPC auth flavours %d.\n",
					__FUNCTION__, data->auth_flavourlen);
			err = -EINVAL;
			goto out_fail;
		}
L
Linus Torvalds 已提交
1796 1797 1798 1799 1800 1801 1802
		if (copy_from_user(&authflavour, data->auth_flavours, sizeof(authflavour))) {
			err = -EFAULT;
			goto out_fail;
		}
	}

	down_write(&clp->cl_sem);
1803
	if (IS_ERR(clp->cl_rpcclient)) {
1804
		xprt = xprt_create_proto(data->proto, &server->addr, &timeparms);
L
Linus Torvalds 已提交
1805 1806 1807
		if (IS_ERR(xprt)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(xprt);
1808 1809
			dprintk("%s: cannot create RPC transport. Error = %d\n",
					__FUNCTION__, err);
L
Linus Torvalds 已提交
1810 1811 1812 1813 1814 1815 1816
			goto out_fail;
		}
		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
				server->rpc_ops->version, authflavour);
		if (IS_ERR(clnt)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(clnt);
1817 1818
			dprintk("%s: cannot create RPC client. Error = %d\n",
					__FUNCTION__, err);
L
Linus Torvalds 已提交
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
			goto out_fail;
		}
		clnt->cl_intr     = 1;
		clnt->cl_softrtry = 1;
		clnt->cl_chatty   = 1;
		clp->cl_rpcclient = clnt;
		clp->cl_cred = rpcauth_lookupcred(clnt->cl_auth, 0);
		if (IS_ERR(clp->cl_cred)) {
			up_write(&clp->cl_sem);
			err = PTR_ERR(clp->cl_cred);
			clp->cl_cred = NULL;
			goto out_fail;
		}
		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
		nfs_idmap_new(clp);
	}
	if (list_empty(&clp->cl_superblocks)) {
		err = nfs4_init_client(clp);
		if (err != 0) {
			up_write(&clp->cl_sem);
			goto out_fail;
		}
	}
	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
	clnt = rpc_clone_client(clp->cl_rpcclient);
	if (!IS_ERR(clnt))
			server->nfs4_state = clp;
	up_write(&clp->cl_sem);
	clp = NULL;

	if (IS_ERR(clnt)) {
1850 1851 1852 1853
		err = PTR_ERR(clnt);
		dprintk("%s: cannot create RPC client. Error = %d\n",
				__FUNCTION__, err);
		return err;
L
Linus Torvalds 已提交
1854 1855 1856 1857 1858
	}

	server->client    = clnt;

	if (server->nfs4_state->cl_idmap == NULL) {
1859
		dprintk("%s: failed to create idmapper.\n", __FUNCTION__);
L
Linus Torvalds 已提交
1860 1861 1862 1863
		return -ENOMEM;
	}

	if (clnt->cl_auth->au_flavor != authflavour) {
1864 1865 1866 1867
		struct rpc_auth *auth;

		auth = rpcauth_create(authflavour, clnt);
		if (IS_ERR(auth)) {
1868
			dprintk("%s: couldn't create credcache!\n", __FUNCTION__);
1869
			return PTR_ERR(auth);
L
Linus Torvalds 已提交
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 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928
		}
	}

	sb->s_time_gran = 1;

	sb->s_op = &nfs4_sops;
	err = nfs_sb_init(sb, authflavour);
	if (err == 0)
		return 0;
out_fail:
	if (clp)
		nfs4_put_client(clp);
	return err;
}

static int nfs4_compare_super(struct super_block *sb, void *data)
{
	struct nfs_server *server = data;
	struct nfs_server *old = NFS_SB(sb);

	if (strcmp(server->hostname, old->hostname) != 0)
		return 0;
	if (strcmp(server->mnt_path, old->mnt_path) != 0)
		return 0;
	return 1;
}

static void *
nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen)
{
	void *p = NULL;

	if (!src->len)
		return ERR_PTR(-EINVAL);
	if (src->len < maxlen)
		maxlen = src->len;
	if (dst == NULL) {
		p = dst = kmalloc(maxlen + 1, GFP_KERNEL);
		if (p == NULL)
			return ERR_PTR(-ENOMEM);
	}
	if (copy_from_user(dst, src->data, maxlen)) {
		if (p != NULL)
			kfree(p);
		return ERR_PTR(-EFAULT);
	}
	dst[maxlen] = '\0';
	return dst;
}

static struct super_block *nfs4_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *raw_data)
{
	int error;
	struct nfs_server *server;
	struct super_block *s;
	struct nfs4_mount_data *data = raw_data;
	void *p;

1929 1930 1931 1932 1933 1934
	if (data == NULL) {
		dprintk("%s: missing data argument\n", __FUNCTION__);
		return ERR_PTR(-EINVAL);
	}
	if (data->version <= 0 || data->version > NFS4_MOUNT_VERSION) {
		dprintk("%s: bad mount version\n", __FUNCTION__);
L
Linus Torvalds 已提交
1935 1936 1937 1938 1939 1940 1941 1942 1943
		return ERR_PTR(-EINVAL);
	}

	server = kmalloc(sizeof(struct nfs_server), GFP_KERNEL);
	if (!server)
		return ERR_PTR(-ENOMEM);
	memset(server, 0, sizeof(struct nfs_server));
	/* Zero out the NFS state stuff */
	init_nfsv4_state(server);
1944
	server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL);
L
Linus Torvalds 已提交
1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971

	p = nfs_copy_user_string(NULL, &data->hostname, 256);
	if (IS_ERR(p))
		goto out_err;
	server->hostname = p;

	p = nfs_copy_user_string(NULL, &data->mnt_path, 1024);
	if (IS_ERR(p))
		goto out_err;
	server->mnt_path = p;

	p = nfs_copy_user_string(server->ip_addr, &data->client_addr,
			sizeof(server->ip_addr) - 1);
	if (IS_ERR(p))
		goto out_err;

	/* We now require that the mount process passes the remote address */
	if (data->host_addrlen != sizeof(server->addr)) {
		s = ERR_PTR(-EINVAL);
		goto out_free;
	}
	if (copy_from_user(&server->addr, data->host_addr, sizeof(server->addr))) {
		s = ERR_PTR(-EFAULT);
		goto out_free;
	}
	if (server->addr.sin_family != AF_INET ||
	    server->addr.sin_addr.s_addr == INADDR_ANY) {
1972 1973
		dprintk("%s: mount program didn't pass remote IP address!\n",
				__FUNCTION__);
L
Linus Torvalds 已提交
1974 1975 1976 1977
		s = ERR_PTR(-EINVAL);
		goto out_free;
	}

1978 1979 1980 1981 1982 1983 1984 1985
	/* Fire up rpciod if not yet running */
	s = ERR_PTR(rpciod_up());
	if (IS_ERR(s)) {
		dprintk("%s: couldn't start rpciod! Error = %ld\n",
				__FUNCTION__, PTR_ERR(s));
		goto out_free;
	}

L
Linus Torvalds 已提交
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 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
	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);

	if (IS_ERR(s) || s->s_root)
		goto out_free;

	s->s_flags = flags;

	error = nfs4_fill_super(s, data, flags & MS_VERBOSE ? 1 : 0);
	if (error) {
		up_write(&s->s_umount);
		deactivate_super(s);
		return ERR_PTR(error);
	}
	s->s_flags |= MS_ACTIVE;
	return s;
out_err:
	s = (struct super_block *)p;
out_free:
	if (server->mnt_path)
		kfree(server->mnt_path);
	if (server->hostname)
		kfree(server->hostname);
	kfree(server);
	return s;
}

static void nfs4_kill_super(struct super_block *sb)
{
	struct nfs_server *server = NFS_SB(sb);

	nfs_return_all_delegations(sb);
	kill_anon_super(sb);

	nfs4_renewd_prepare_shutdown(server);

	if (server->client != NULL && !IS_ERR(server->client))
		rpc_shutdown_client(server->client);
	rpciod_down();		/* release rpciod */

	destroy_nfsv4_state(server);

	if (server->hostname != NULL)
		kfree(server->hostname);
	kfree(server);
}

static struct file_system_type nfs4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "nfs4",
	.get_sb		= nfs4_get_sb,
	.kill_sb	= nfs4_kill_super,
	.fs_flags	= FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
};

#define nfs4_init_once(nfsi) \
	do { \
		INIT_LIST_HEAD(&(nfsi)->open_states); \
		nfsi->delegation = NULL; \
		nfsi->delegation_state = 0; \
		init_rwsem(&nfsi->rwsem); \
	} while(0)
#define register_nfs4fs() register_filesystem(&nfs4_fs_type)
#define unregister_nfs4fs() unregister_filesystem(&nfs4_fs_type)
#else
#define nfs4_init_once(nfsi) \
	do { } while (0)
#define register_nfs4fs() (0)
#define unregister_nfs4fs()
#endif

extern int nfs_init_nfspagecache(void);
extern void nfs_destroy_nfspagecache(void);
extern int nfs_init_readpagecache(void);
extern void nfs_destroy_readpagecache(void);
extern int nfs_init_writepagecache(void);
extern void nfs_destroy_writepagecache(void);
#ifdef CONFIG_NFS_DIRECTIO
extern int nfs_init_directcache(void);
extern void nfs_destroy_directcache(void);
#endif

static kmem_cache_t * nfs_inode_cachep;

static struct inode *nfs_alloc_inode(struct super_block *sb)
{
	struct nfs_inode *nfsi;
	nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, SLAB_KERNEL);
	if (!nfsi)
		return NULL;
2075 2076
	nfsi->flags = 0UL;
	nfsi->cache_validity = 0UL;
2077 2078 2079 2080
#ifdef CONFIG_NFS_V3_ACL
	nfsi->acl_access = ERR_PTR(-EAGAIN);
	nfsi->acl_default = ERR_PTR(-EAGAIN);
#endif
2081 2082 2083
#ifdef CONFIG_NFS_V4
	nfsi->nfs4_acl = NULL;
#endif /* CONFIG_NFS_V4 */
L
Linus Torvalds 已提交
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 2109 2110 2111
	return &nfsi->vfs_inode;
}

static void nfs_destroy_inode(struct inode *inode)
{
	kmem_cache_free(nfs_inode_cachep, NFS_I(inode));
}

static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
{
	struct nfs_inode *nfsi = (struct nfs_inode *) foo;

	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
	    SLAB_CTOR_CONSTRUCTOR) {
		inode_init_once(&nfsi->vfs_inode);
		spin_lock_init(&nfsi->req_lock);
		INIT_LIST_HEAD(&nfsi->dirty);
		INIT_LIST_HEAD(&nfsi->commit);
		INIT_LIST_HEAD(&nfsi->open_files);
		INIT_RADIX_TREE(&nfsi->nfs_page_tree, GFP_ATOMIC);
		atomic_set(&nfsi->data_updates, 0);
		nfsi->ndirty = 0;
		nfsi->ncommit = 0;
		nfsi->npages = 0;
		nfs4_init_once(nfsi);
	}
}
 
A
Adrian Bunk 已提交
2112
static int nfs_init_inodecache(void)
L
Linus Torvalds 已提交
2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123
{
	nfs_inode_cachep = kmem_cache_create("nfs_inode_cache",
					     sizeof(struct nfs_inode),
					     0, SLAB_RECLAIM_ACCOUNT,
					     init_once, NULL);
	if (nfs_inode_cachep == NULL)
		return -ENOMEM;

	return 0;
}

A
Adrian Bunk 已提交
2124
static void nfs_destroy_inodecache(void)
L
Linus Torvalds 已提交
2125 2126 2127 2128 2129 2130 2131 2132 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 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208
{
	if (kmem_cache_destroy(nfs_inode_cachep))
		printk(KERN_INFO "nfs_inode_cache: not all structures were freed\n");
}

/*
 * Initialize NFS
 */
static int __init init_nfs_fs(void)
{
	int err;

	err = nfs_init_nfspagecache();
	if (err)
		goto out4;

	err = nfs_init_inodecache();
	if (err)
		goto out3;

	err = nfs_init_readpagecache();
	if (err)
		goto out2;

	err = nfs_init_writepagecache();
	if (err)
		goto out1;

#ifdef CONFIG_NFS_DIRECTIO
	err = nfs_init_directcache();
	if (err)
		goto out0;
#endif

#ifdef CONFIG_PROC_FS
	rpc_proc_register(&nfs_rpcstat);
#endif
        err = register_filesystem(&nfs_fs_type);
	if (err)
		goto out;
	if ((err = register_nfs4fs()) != 0)
		goto out;
	return 0;
out:
#ifdef CONFIG_PROC_FS
	rpc_proc_unregister("nfs");
#endif
	nfs_destroy_writepagecache();
#ifdef CONFIG_NFS_DIRECTIO
out0:
	nfs_destroy_directcache();
#endif
out1:
	nfs_destroy_readpagecache();
out2:
	nfs_destroy_inodecache();
out3:
	nfs_destroy_nfspagecache();
out4:
	return err;
}

static void __exit exit_nfs_fs(void)
{
#ifdef CONFIG_NFS_DIRECTIO
	nfs_destroy_directcache();
#endif
	nfs_destroy_writepagecache();
	nfs_destroy_readpagecache();
	nfs_destroy_inodecache();
	nfs_destroy_nfspagecache();
#ifdef CONFIG_PROC_FS
	rpc_proc_unregister("nfs");
#endif
	unregister_filesystem(&nfs_fs_type);
	unregister_nfs4fs();
}

/* Not quite true; I just maintain it */
MODULE_AUTHOR("Olaf Kirch <okir@monad.swb.de>");
MODULE_LICENSE("GPL");

module_init(init_nfs_fs)
module_exit(exit_nfs_fs)