nfs4state.c 64.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
/*
*  linux/fs/nfsd/nfs4state.c
*
*  Copyright (c) 2001 The Regents of the University of Michigan.
*  All rights reserved.
*
*  Kendrick Smith <kmsmith@umich.edu>
*  Andy Adamson <kandros@umich.edu>
*
*  Redistribution and use in source and binary forms, with or without
*  modification, are permitted provided that the following conditions
*  are met:
*
*  1. Redistributions of source code must retain the above copyright
*     notice, this list of conditions and the following disclaimer.
*  2. Redistributions in binary form must reproduce the above copyright
*     notice, this list of conditions and the following disclaimer in the
*     documentation and/or other materials provided with the distribution.
*  3. Neither the name of the University nor the names of its
*     contributors may be used to endorse or promote products derived
*     from this software without specific prior written permission.
*
*  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
*  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
*  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
*  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
*  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
*  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
*  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
*  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
*  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
*  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
*  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/

#include <linux/param.h>
#include <linux/major.h>
#include <linux/slab.h>

#include <linux/sunrpc/svc.h>
#include <linux/nfsd/nfsd.h>
#include <linux/nfsd/cache.h>
44
#include <linux/mount.h>
45
#include <linux/workqueue.h>
46
#include <linux/smp_lock.h>
47 48
#include <linux/nfs4.h>
#include <linux/nfsd/state.h>
49
#include <linux/nfsd/xdr4.h>
50 51 52 53 54 55

#define NFSDDBG_FACILITY                NFSDDBG_PROC

/* Globals */
time_t boot_time;
static u32 current_clientid = 1;
56 57 58
static u32 current_ownerid;
static u32 current_fileid;
static u32 nfs4_init;
59 60
stateid_t zerostateid;             /* bits all 0 */
stateid_t onestateid;              /* bits all 1 */
61 62 63 64 65 66

/* debug counters */
u32 list_add_perfile = 0; 
u32 list_del_perfile = 0;
u32 add_perclient = 0;
u32 del_perclient = 0;
67 68 69 70 71 72
u32 alloc_file = 0;
u32 free_file = 0;
u32 alloc_sowner = 0;
u32 free_sowner = 0;
u32 vfsopen = 0;
u32 vfsclose = 0;
73 74 75 76
u32 alloc_lsowner= 0;

/* forward declarations */
struct nfs4_stateid * find_stateid(stateid_t *stid, int flags);
77 78 79 80 81 82 83 84 85

/* Locking:
 *
 * client_sema: 
 * 	protects clientid_hashtbl[], clientstr_hashtbl[],
 * 	unconfstr_hashtbl[], uncofid_hashtbl[].
 */
static struct semaphore client_sema;

86
void
87
nfs4_lock_state(void)
88 89 90 91 92
{
	down(&client_sema);
}

void
93
nfs4_unlock_state(void)
94 95 96 97
{
	up(&client_sema);
}

98 99 100 101 102 103 104 105 106 107 108 109 110
static inline u32
opaque_hashval(const void *ptr, int nbytes)
{
	unsigned char *cptr = (unsigned char *) ptr;

	u32 x = 0;
	while (nbytes--) {
		x *= 37;
		x += *cptr++;
	}
	return x;
}

111 112
/* forward declarations */
static void release_stateowner(struct nfs4_stateowner *sop);
113
static void release_stateid(struct nfs4_stateid *stp, int flags);
114 115
static void release_file(struct nfs4_file *fp);

116 117 118 119 120

/* 
 * SETCLIENTID state 
 */

121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
/* Hash tables for nfs4_clientid state */
#define CLIENT_HASH_BITS                 4
#define CLIENT_HASH_SIZE                (1 << CLIENT_HASH_BITS)
#define CLIENT_HASH_MASK                (CLIENT_HASH_SIZE - 1)

#define clientid_hashval(id) \
	((id) & CLIENT_HASH_MASK)
#define clientstr_hashval(name, namelen) \
	(opaque_hashval((name), (namelen)) & CLIENT_HASH_MASK)

/* conf_id_hashtbl[], and conf_str_hashtbl[] hold confirmed
 * setclientid_confirmed info. 
 *
 * unconf_str_hastbl[] and unconf_id_hashtbl[] hold unconfirmed 
 * setclientid info.
136 137 138
 *
 * client_lru holds client queue ordered by nfs4_client.cl_time
 * for lease renewal.
139 140 141
 *
 * close_lru holds (open) stateowner queue ordered by nfs4_stateowner.so_time
 * for last close replay.
142 143 144 145 146
 */
static struct list_head	conf_id_hashtbl[CLIENT_HASH_SIZE];
static struct list_head	conf_str_hashtbl[CLIENT_HASH_SIZE];
static struct list_head	unconf_str_hashtbl[CLIENT_HASH_SIZE];
static struct list_head	unconf_id_hashtbl[CLIENT_HASH_SIZE];
147
static struct list_head client_lru;
148
static struct list_head close_lru;
149 150 151 152 153 154 155 156 157 158 159 160 161

static inline void
renew_client(struct nfs4_client *clp)
{
	/*
	* Move client to the end to the LRU list.
	*/
	dprintk("renewing client (clientid %08x/%08x)\n", 
			clp->cl_clientid.cl_boot, 
			clp->cl_clientid.cl_id);
	list_move_tail(&clp->cl_lru, &client_lru);
	clp->cl_time = get_seconds();
}
162 163 164 165 166 167 168

/* SETCLIENTID and SETCLIENTID_CONFIRM Helper functions */
static int
STALE_CLIENTID(clientid_t *clid)
{
	if (clid->cl_boot == boot_time)
		return 0;
169 170
	dprintk("NFSD stale clientid (%08x/%08x)\n", 
			clid->cl_boot, clid->cl_id);
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
	return 1;
}

/* 
 * XXX Should we use a slab cache ?
 * This type of memory management is somewhat inefficient, but we use it
 * anyway since SETCLIENTID is not a common operation.
 */
static inline struct nfs4_client *
alloc_client(struct xdr_netobj name)
{
	struct nfs4_client *clp;

	if ((clp = kmalloc(sizeof(struct nfs4_client), GFP_KERNEL))!= NULL) {
		memset(clp, 0, sizeof(*clp));
		if ((clp->cl_name.data = kmalloc(name.len, GFP_KERNEL)) != NULL) {
			memcpy(clp->cl_name.data, name.data, name.len);
			clp->cl_name.len = name.len;
		}
		else {
			kfree(clp);
			clp = NULL;
		}
	}
	return clp;
}

static inline void
free_client(struct nfs4_client *clp)
{
Andrew Morton's avatar
Andrew Morton committed
201 202
	if (clp->cl_cred.cr_group_info)
		put_group_info(clp->cl_cred.cr_group_info);
203 204 205 206 207 208 209
	kfree(clp->cl_name.data);
	kfree(clp);
}

static void
expire_client(struct nfs4_client *clp)
{
210 211
	struct nfs4_stateowner *sop;

212 213 214
	dprintk("NFSD: expire_client\n");
	list_del(&clp->cl_idhash);
	list_del(&clp->cl_strhash);
215
	list_del(&clp->cl_lru);
216 217 218 219
	while (!list_empty(&clp->cl_perclient)) {
		sop = list_entry(clp->cl_perclient.next, struct nfs4_stateowner, so_perclient);
		release_stateowner(sop);
	}
220 221 222 223 224 225 226 227 228 229 230
	free_client(clp);
}

static struct nfs4_client *
create_client(struct xdr_netobj name) {
	struct nfs4_client *clp;

	if(!(clp = alloc_client(name)))
		goto out;
	INIT_LIST_HEAD(&clp->cl_idhash);
	INIT_LIST_HEAD(&clp->cl_strhash);
231
	INIT_LIST_HEAD(&clp->cl_perclient);
232
	INIT_LIST_HEAD(&clp->cl_lru);
233 234 235 236 237
out:
	return clp;
}

static void
238 239
copy_verf(struct nfs4_client *target, nfs4_verifier *source) {
	memcpy(target->cl_verifier.data, source->data, sizeof(target->cl_verifier.data));
240 241 242 243 244 245 246 247 248 249 250 251 252
}

static void
copy_clid(struct nfs4_client *target, struct nfs4_client *source) {
	target->cl_clientid.cl_boot = source->cl_clientid.cl_boot; 
	target->cl_clientid.cl_id = source->cl_clientid.cl_id; 
}

static void
copy_cred(struct svc_cred *target, struct svc_cred *source) {

	target->cr_uid = source->cr_uid;
	target->cr_gid = source->cr_gid;
Andrew Morton's avatar
Andrew Morton committed
253 254
	target->cr_group_info = source->cr_group_info;
	get_group_info(target->cr_group_info);
255 256 257 258 259 260 261 262 263 264
}

static int
cmp_name(struct xdr_netobj *n1, struct xdr_netobj *n2) {
	if(!n1 || !n2)
		return 0;
	return((n1->len == n2->len) && !memcmp(n1->data, n2->data, n2->len));
}

static int
265 266
cmp_verf(nfs4_verifier *v1, nfs4_verifier *v2) {
	return(!memcmp(v1->data,v2->data,sizeof(v1->data)));
267 268 269 270 271 272 273 274 275 276 277
}

static int
cmp_clid(clientid_t * cl1, clientid_t * cl2) {
	return((cl1->cl_boot == cl2->cl_boot) &&
	   	(cl1->cl_id == cl2->cl_id));
}

/* XXX what about NGROUP */
static int
cmp_creds(struct svc_cred *cr1, struct svc_cred *cr2){
278
	return(cr1->cr_uid == cr2->cr_uid);
279 280 281 282 283 284 285 286 287 288 289 290 291 292 293

}

static void
gen_clid(struct nfs4_client *clp) {
	clp->cl_clientid.cl_boot = boot_time;
	clp->cl_clientid.cl_id = current_clientid++; 
}

static void
gen_confirm(struct nfs4_client *clp) {
	struct timespec 	tv;
	u32 *			p;

	tv = CURRENT_TIME;
294
	p = (u32 *)clp->cl_confirm.data;
295 296 297 298 299 300 301 302 303
	*p++ = tv.tv_sec;
	*p++ = tv.tv_nsec;
}

static int
check_name(struct xdr_netobj name) {

	if (name.len == 0) 
		return 0;
304
	if (name.len > NFS4_OPAQUE_LIMIT) {
305 306 307 308 309 310 311 312 313 314 315 316 317 318
		printk("NFSD: check_name: name too long(%d)!\n", name.len);
		return 0;
	}
	return 1;
}

void
add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval)
{
	unsigned int idhashval;

	list_add(&clp->cl_strhash, &unconf_str_hashtbl[strhashval]);
	idhashval = clientid_hashval(clp->cl_clientid.cl_id);
	list_add(&clp->cl_idhash, &unconf_id_hashtbl[idhashval]);
319 320
	list_add_tail(&clp->cl_lru, &client_lru);
	clp->cl_time = get_seconds();
321 322 323 324 325 326 327
}

void
move_to_confirmed(struct nfs4_client *clp, unsigned int idhashval)
{
	unsigned int strhashval;

328
	dprintk("NFSD: move_to_confirm nfs4_client %p\n", clp);
329 330 331 332 333 334
	list_del_init(&clp->cl_strhash);
	list_del_init(&clp->cl_idhash);
	list_add(&clp->cl_idhash, &conf_id_hashtbl[idhashval]);
	strhashval = clientstr_hashval(clp->cl_name.data, 
			clp->cl_name.len);
	list_add(&clp->cl_strhash, &conf_str_hashtbl[strhashval]);
335
	renew_client(clp);
336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
}

/*
 * RFC 3010 has a complex implmentation description of processing a 
 * SETCLIENTID request consisting of 5 bullets, labeled as 
 * CASE0 - CASE4 below.
 *
 * NOTES:
 * 	callback information will be processed in a future patch
 *
 *	an unconfirmed record is added when:
 *      NORMAL (part of CASE 4): there is no confirmed nor unconfirmed record.
 *	CASE 1: confirmed record found with matching name, principal,
 *		verifier, and clientid.
 *	CASE 2: confirmed record found with matching name, principal,
 *		and there is no unconfirmed record with matching
 *		name and principal
 *
 *      an unconfirmed record is replaced when:
 *	CASE 3: confirmed record found with matching name, principal,
 *		and an unconfirmed record is found with matching 
 *		name, principal, and with clientid and
 *		confirm that does not match the confirmed record.
 *	CASE 4: there is no confirmed record with matching name and 
 *		principal. there is an unconfirmed record with 
 *		matching name, principal.
 *
 *	an unconfirmed record is deleted when:
 *	CASE 1: an unconfirmed record that matches input name, verifier,
 *		and confirmed clientid.
 *	CASE 4: any unconfirmed records with matching name and principal
 *		that exist after an unconfirmed record has been replaced
 *		as described above.
 *
 */
int
nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid)
{
	u32 			ip_addr = rqstp->rq_addr.sin_addr.s_addr;
	struct xdr_netobj 	clname = { 
		.len = setclid->se_namelen,
		.data = setclid->se_name,
	};
379
	nfs4_verifier		clverifier = setclid->se_verf;
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
	unsigned int 		strhashval;
	struct nfs4_client *	conf, * unconf, * new, * clp;
	int 			status;
	
	status = nfserr_inval;
	if (!check_name(clname))
		goto out;

	/* 
	 * XXX The Duplicate Request Cache (DRC) has been checked (??)
	 * We get here on a DRC miss.
	 */

	strhashval = clientstr_hashval(clname.data, clname.len);

	conf = NULL;
396
	nfs4_lock_state();
397
	list_for_each_entry(clp, &conf_str_hashtbl[strhashval], cl_strhash) {
398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426
		if (!cmp_name(&clp->cl_name, &clname))
			continue;
		/* 
		 * CASE 0:
		 * clname match, confirmed, different principal
		 * or different ip_address
		 */
		status = nfserr_clid_inuse;
		if (!cmp_creds(&clp->cl_cred,&rqstp->rq_cred)) {
			printk("NFSD: setclientid: string in use by client"
			"(clientid %08x/%08x)\n",
			clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
			goto out;
		}
		if (clp->cl_addr != ip_addr) { 
			printk("NFSD: setclientid: string in use by client"
			"(clientid %08x/%08x)\n",
			clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
			goto out;
		}

		/* 
	 	 * cl_name match from a previous SETCLIENTID operation
	 	 * XXX check for additional matches?
		 */
		conf = clp;
		break;
	}
	unconf = NULL;
427
	list_for_each_entry(clp, &unconf_str_hashtbl[strhashval], cl_strhash) {
428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
		if (!cmp_name(&clp->cl_name, &clname))
			continue;
		/* cl_name match from a previous SETCLIENTID operation */
		unconf = clp;
		break;
	}
	status = nfserr_resource;
	if (!conf) {
		/* 
		 * CASE 4:
		 * placed first, because it is the normal case.
		 */
		if (unconf)
			expire_client(unconf);
		if (!(new = create_client(clname)))
			goto out;
444
		copy_verf(new, &clverifier);
445 446 447 448 449
		new->cl_addr = ip_addr;
		copy_cred(&new->cl_cred,&rqstp->rq_cred);
		gen_clid(new);
		gen_confirm(new);
		add_to_unconfirmed(new, strhashval);
450
	} else if (cmp_verf(&conf->cl_verifier, &clverifier)) {
451 452 453 454 455 456 457 458 459 460 461 462 463 464
		/*
		 * CASE 1:
		 * cl_name match, confirmed, principal match
		 * verifier match: probable callback update
		 *
		 * remove any unconfirmed nfs4_client with 
		 * matching cl_name, cl_verifier, and cl_clientid
		 *
		 * create and insert an unconfirmed nfs4_client with same 
		 * cl_name, cl_verifier, and cl_clientid as existing 
		 * nfs4_client,  but with the new callback info and a 
		 * new cl_confirm
		 */
		if ((unconf) && 
465
		    cmp_verf(&unconf->cl_verifier, &conf->cl_verifier) &&
466 467 468 469 470
		     cmp_clid(&unconf->cl_clientid, &conf->cl_clientid)) {
				expire_client(unconf);
		}
		if (!(new = create_client(clname)))
			goto out;
471
		copy_verf(new,&conf->cl_verifier);
472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
		new->cl_addr = ip_addr;
		copy_cred(&new->cl_cred,&rqstp->rq_cred);
		copy_clid(new, conf);
		gen_confirm(new);
		add_to_unconfirmed(new,strhashval);
	} else if (!unconf) {
		/*
		 * CASE 2:
		 * clname match, confirmed, principal match
		 * verfier does not match
		 * no unconfirmed. create a new unconfirmed nfs4_client
		 * using input clverifier, clname, and callback info
		 * and generate a new cl_clientid and cl_confirm.
		 */
		if (!(new = create_client(clname)))
			goto out;
488
		copy_verf(new,&clverifier);
489 490 491 492 493 494
		new->cl_addr = ip_addr;
		copy_cred(&new->cl_cred,&rqstp->rq_cred);
		gen_clid(new);
		gen_confirm(new);
		add_to_unconfirmed(new, strhashval);
	} else if (!cmp_clid(&conf->cl_clientid, &unconf->cl_clientid) &&
495
	      !cmp_verf(&conf->cl_confirm, &unconf->cl_confirm)) {
496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514
		/*	
		 * CASE3:
		 * confirmed found (name, principal match)
		 * confirmed verifier does not match input clverifier
		 *
		 * unconfirmed found (name match)
		 * confirmed->cl_clientid != unconfirmed->cl_clientid and
		 * confirmed->cl_confirm != unconfirmed->cl_confirm
		 *
		 * remove unconfirmed.
		 *
		 * create an unconfirmed nfs4_client 
		 * with same cl_name as existing confirmed nfs4_client, 
		 * but with new callback info, new cl_clientid,
		 * new cl_verifier and a new cl_confirm
		 */
		expire_client(unconf);
		if (!(new = create_client(clname)))
			goto out;
515
		copy_verf(new,&clverifier);
516 517 518 519 520 521 522 523 524 525 526 527 528
		new->cl_addr = ip_addr;
		copy_cred(&new->cl_cred,&rqstp->rq_cred);
		gen_clid(new);
		gen_confirm(new);
		add_to_unconfirmed(new, strhashval);
	} else {
		/* No cases hit !!! */
		status = nfserr_inval;
		goto out;

	}
	setclid->se_clientid.cl_boot = new->cl_clientid.cl_boot;
	setclid->se_clientid.cl_id = new->cl_clientid.cl_id;
529
	memcpy(setclid->se_confirm.data, new->cl_confirm.data, sizeof(setclid->se_confirm.data));
530 531 532
	printk(KERN_INFO "NFSD: this client will not receive delegations\n");
	status = nfs_ok;
out:
533
	nfs4_unlock_state();
534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550
	return status;
}


/*
 * RFC 3010 has a complex implmentation description of processing a 
 * SETCLIENTID_CONFIRM request consisting of 4 bullets describing
 * processing on a DRC miss, labeled as CASE1 - CASE4 below.
 *
 * NOTE: callback information will be processed here in a future patch
 */
int
nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm)
{
	u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr;
	unsigned int idhashval;
	struct nfs4_client *clp, *conf = NULL, *unconf = NULL;
551
	nfs4_verifier confirm = setclientid_confirm->sc_confirm; 
552 553 554 555 556 557 558 559 560 561 562 563
	clientid_t * clid = &setclientid_confirm->sc_clientid;
	int status;

	status = nfserr_stale_clientid;
	if (STALE_CLIENTID(clid))
		goto out;
	/* 
	 * XXX The Duplicate Request Cache (DRC) has been checked (??)
	 * We get here on a DRC miss.
	 */

	idhashval = clientid_hashval(clid->cl_id);
564
	nfs4_lock_state();
565
	list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) {
566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
		if (!cmp_clid(&clp->cl_clientid, clid))
			continue;

		status = nfserr_inval;
		/* 
		 * Found a record for this clientid. If the IP addresses
		 * don't match, return ERR_INVAL just as if the record had
		 * not been found.
		 */
		if (clp->cl_addr != ip_addr) { 
			printk("NFSD: setclientid: string in use by client"
			"(clientid %08x/%08x)\n",
			clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
			goto out;
		}
		conf = clp;
		break;
	}
584
	list_for_each_entry(clp, &unconf_id_hashtbl[idhashval], cl_idhash) {
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
		if (!cmp_clid(&clp->cl_clientid, clid))
			continue;
		status = nfserr_inval;
		if (clp->cl_addr != ip_addr) { 
			printk("NFSD: setclientid: string in use by client"
			"(clientid %08x/%08x)\n",
			clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
			goto out;
		}
		unconf = clp;
		break;
	}
	/* CASE 1: 
	* unconf record that matches input clientid and input confirm.
	* conf record that matches input clientid.
	* conf  and unconf records match names, verifiers 
	*/
	if ((conf && unconf) && 
603 604
	    (cmp_verf(&unconf->cl_confirm, &confirm)) &&
	    (cmp_verf(&conf->cl_verifier, &unconf->cl_verifier)) &&
605
	    (cmp_name(&conf->cl_name,&unconf->cl_name))  &&
606
	    (!cmp_verf(&conf->cl_confirm, &unconf->cl_confirm))) {
607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622
		if (!cmp_creds(&conf->cl_cred, &unconf->cl_cred)) 
			status = nfserr_clid_inuse;
		else {
			expire_client(conf);
			move_to_confirmed(unconf, idhashval);
			status = nfs_ok;
		}
		goto out;
	} 
	/* CASE 2:
	 * conf record that matches input clientid.
	 * if unconf record that matches input clientid, then unconf->cl_name
	 * or unconf->cl_verifier don't match the conf record.
	 */
	if ((conf && !unconf) || 
	    ((conf && unconf) && 
623
	     (!cmp_verf(&conf->cl_verifier, &unconf->cl_verifier) ||
624 625 626 627 628 629 630 631 632 633 634 635 636
	      !cmp_name(&conf->cl_name, &unconf->cl_name)))) {
		if (!cmp_creds(&conf->cl_cred,&rqstp->rq_cred)) {
			status = nfserr_clid_inuse;
		} else {
			status = nfs_ok;
		}
		goto out;
	}
	/* CASE 3:
	 * conf record not found.
	 * unconf record found. 
	 * unconf->cl_confirm matches input confirm
	 */ 
637
	if (!conf && unconf && cmp_verf(&unconf->cl_confirm, &confirm)) {
638 639 640 641 642 643 644 645 646 647 648 649 650 651
		if (!cmp_creds(&unconf->cl_cred, &rqstp->rq_cred)) {
			status = nfserr_clid_inuse;
		} else {
			status = nfs_ok;
			move_to_confirmed(unconf, idhashval);
		}
		goto out;
	}
	/* CASE 4:
	 * conf record not found, or if conf, then conf->cl_confirm does not
	 * match input confirm.
	 * unconf record not found, or if unconf, then unconf->cl_confirm 
	 * does not match input confirm.
	 */
652 653
	if ((!conf || (conf && !cmp_verf(&conf->cl_confirm, &confirm))) &&
	    (!unconf || (unconf && !cmp_verf(&unconf->cl_confirm, &confirm)))) {
654 655 656 657 658 659 660 661
		status = nfserr_stale_clientid;
		goto out;
	}
	/* check that we have hit one of the cases...*/
	status = nfserr_inval;
	goto out;
out:
	/* XXX if status == nfs_ok, probe callback path */
662
	nfs4_unlock_state();
663 664 665
	return status;
}

666 667 668 669 670 671 672 673 674
/* 
 * Open owner state (share locks)
 */

/* hash tables for nfs4_stateowner */
#define OWNER_HASH_BITS              8
#define OWNER_HASH_SIZE             (1 << OWNER_HASH_BITS)
#define OWNER_HASH_MASK             (OWNER_HASH_SIZE - 1)

675 676
#define ownerid_hashval(id) \
        ((id) & OWNER_HASH_MASK)
677 678 679
#define ownerstr_hashval(clientid, ownername) \
        (((clientid) + opaque_hashval((ownername.data), (ownername.len))) & OWNER_HASH_MASK)

680
static struct list_head	ownerid_hashtbl[OWNER_HASH_SIZE];
681 682 683 684 685 686
static struct list_head	ownerstr_hashtbl[OWNER_HASH_SIZE];

/* hash table for nfs4_file */
#define FILE_HASH_BITS                   8
#define FILE_HASH_SIZE                  (1 << FILE_HASH_BITS)
#define FILE_HASH_MASK                  (FILE_HASH_SIZE - 1)
687
/* hash table for (open)nfs4_stateid */
688 689 690
#define STATEID_HASH_BITS              10
#define STATEID_HASH_SIZE              (1 << STATEID_HASH_BITS)
#define STATEID_HASH_MASK              (STATEID_HASH_SIZE - 1)
691 692

#define file_hashval(x) \
693
        hash_ptr(x, FILE_HASH_BITS)
694 695
#define stateid_hashval(owner_id, file_id)  \
        (((owner_id) + (file_id)) & STATEID_HASH_MASK)
696 697

static struct list_head file_hashtbl[FILE_HASH_SIZE];
698
static struct list_head stateid_hashtbl[STATEID_HASH_SIZE];
699 700 701

/* OPEN Share state helper functions */
static inline struct nfs4_file *
702
alloc_init_file(unsigned int hashval, struct inode *ino) {
703 704 705 706 707
	struct nfs4_file *fp;
	if ((fp = kmalloc(sizeof(struct nfs4_file),GFP_KERNEL))) {
		INIT_LIST_HEAD(&fp->fi_hash);
		INIT_LIST_HEAD(&fp->fi_perfile);
		list_add(&fp->fi_hash, &file_hashtbl[hashval]);
708
		fp->fi_inode = igrab(ino);
709
		fp->fi_id = current_fileid++;
710
		alloc_file++;
711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
		return fp;
	}
	return (struct nfs4_file *)NULL;
}

static void
release_all_files(void)
{
	int i;
	struct nfs4_file *fp;

	for (i=0;i<FILE_HASH_SIZE;i++) {
		while (!list_empty(&file_hashtbl[i])) {
			fp = list_entry(file_hashtbl[i].next, struct nfs4_file, fi_hash);
			/* this should never be more than once... */
			if(!list_empty(&fp->fi_perfile)) {
				printk("ERROR: release_all_files: file %p is open, creating dangling state !!!\n",fp);
			}
729
			release_file(fp);
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
		}
	}
}

static inline struct nfs4_stateowner *
alloc_stateowner(struct xdr_netobj *owner)
{
	struct nfs4_stateowner *sop;

	if ((sop = kmalloc(sizeof(struct nfs4_stateowner),GFP_KERNEL))) {
		if((sop->so_owner.data = kmalloc(owner->len, GFP_KERNEL))) {
			memcpy(sop->so_owner.data, owner->data, owner->len);
			sop->so_owner.len = owner->len;
			return sop;
		} 
		kfree(sop);
	}
	return (struct nfs4_stateowner *)NULL;
}

/* should use a slab cache */
static void
free_stateowner(struct nfs4_stateowner *sop) {
	if(sop) {
		kfree(sop->so_owner.data);
		kfree(sop);
		sop = NULL;
757
		free_sowner++;
758 759 760 761
	}
}

static struct nfs4_stateowner *
762
alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfsd4_open *open) {
763
	struct nfs4_stateowner *sop;
764
	struct nfs4_replay *rp;
765
	unsigned int idhashval;
766 767 768

	if (!(sop = alloc_stateowner(&open->op_owner)))
		return (struct nfs4_stateowner *)NULL;
769 770
	idhashval = ownerid_hashval(current_ownerid);
	INIT_LIST_HEAD(&sop->so_idhash);
771 772
	INIT_LIST_HEAD(&sop->so_strhash);
	INIT_LIST_HEAD(&sop->so_perclient);
773
	INIT_LIST_HEAD(&sop->so_perfilestate);
774
	INIT_LIST_HEAD(&sop->so_perlockowner);  /* not used */
775 776
	INIT_LIST_HEAD(&sop->so_close_lru);
	sop->so_time = 0;
777
	list_add(&sop->so_idhash, &ownerid_hashtbl[idhashval]);
778 779 780
	list_add(&sop->so_strhash, &ownerstr_hashtbl[strhashval]);
	list_add(&sop->so_perclient, &clp->cl_perclient);
	add_perclient++;
781
	sop->so_is_open_owner = 1;
782 783 784
	sop->so_id = current_ownerid++;
	sop->so_client = clp;
	sop->so_seqid = open->op_seqid;
785
	sop->so_confirmed = 0;
786 787 788 789
	rp = &sop->so_replay;
	rp->rp_status = NFSERR_SERVERFAULT;
	rp->rp_buflen = 0;
	rp->rp_buf = rp->rp_ibuf;
790
	alloc_sowner++;
791 792 793
	return sop;
}

794 795 796 797 798 799 800 801 802 803 804 805 806 807
static void
release_stateid_lockowner(struct nfs4_stateid *open_stp)
{
	struct nfs4_stateowner *lock_sop;

	while (!list_empty(&open_stp->st_perlockowner)) {
		lock_sop = list_entry(open_stp->st_perlockowner.next,
				struct nfs4_stateowner, so_perlockowner);
		/* list_del(&open_stp->st_perlockowner);  */
		BUG_ON(lock_sop->so_is_open_owner);
		release_stateowner(lock_sop);
	}
}

808 809 810 811 812
static void
release_stateowner(struct nfs4_stateowner *sop)
{
	struct nfs4_stateid *stp;

813 814 815 816
	list_del(&sop->so_idhash);
	list_del(&sop->so_strhash);
	list_del(&sop->so_perclient);
	list_del(&sop->so_perlockowner);
817
	list_del(&sop->so_close_lru);
818
	del_perclient++;
819 820 821
	while (!list_empty(&sop->so_perfilestate)) {
		stp = list_entry(sop->so_perfilestate.next, 
			struct nfs4_stateid, st_perfilestate);
822 823 824 825
		if(sop->so_is_open_owner)
			release_stateid(stp, OPEN_STATE);
		else
			release_stateid(stp, LOCK_STATE);
826 827 828 829 830 831
	}
	free_stateowner(sop);
}

static inline void
init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfs4_stateowner *sop, struct nfsd4_open *open) {
832
	unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id);
833

834
	INIT_LIST_HEAD(&stp->st_hash);
835
	INIT_LIST_HEAD(&stp->st_perfilestate);
836
	INIT_LIST_HEAD(&stp->st_perlockowner);
837
	INIT_LIST_HEAD(&stp->st_perfile);
838 839
	list_add(&stp->st_hash, &stateid_hashtbl[hashval]);
	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
840 841 842 843 844 845 846 847
	list_add_perfile++;
	list_add(&stp->st_perfile, &fp->fi_perfile);
	stp->st_stateowner = sop;
	stp->st_file = fp;
	stp->st_stateid.si_boot = boot_time;
	stp->st_stateid.si_stateownerid = sop->so_id;
	stp->st_stateid.si_fileid = fp->fi_id;
	stp->st_stateid.si_generation = 0;
848 849 850 851
	stp->st_access_bmap = 0;
	stp->st_deny_bmap = 0;
	__set_bit(open->op_share_access, &stp->st_access_bmap);
	__set_bit(open->op_share_deny, &stp->st_deny_bmap);
852 853 854
}

static void
855
release_stateid(struct nfs4_stateid *stp, int flags) {
856

857
	list_del(&stp->st_hash);
858
	list_del_perfile++;
859 860
	list_del(&stp->st_perfile);
	list_del(&stp->st_perfilestate);
861
	if((stp->st_vfs_set) && (flags & OPEN_STATE)) {
862
		release_stateid_lockowner(stp);
863
		nfsd_close(&stp->st_vfs_file);
864
		vfsclose++;
865 866
		dput(stp->st_vfs_file.f_dentry);
		mntput(stp->st_vfs_file.f_vfsmnt);
867 868 869 870
	} else if ((stp->st_vfs_set) && (flags & LOCK_STATE)) {
		struct file *filp = &stp->st_vfs_file;

		locks_remove_posix(filp, (fl_owner_t) stp->st_stateowner);
871 872 873 874 875
	}
	kfree(stp);
	stp = NULL;
}

876 877 878 879
static void
release_file(struct nfs4_file *fp)
{
	free_file++;
880
	list_del(&fp->fi_hash);
881
	iput(fp->fi_inode);
882 883 884
	kfree(fp);
}	

885 886 887 888 889 890 891 892 893 894 895 896 897
void
move_to_close_lru(struct nfs4_stateowner *sop)
{
	dprintk("NFSD: move_to_close_lru nfs4_stateowner %p\n", sop);
	/* remove stateowner from all other hash lists except perclient */
	list_del_init(&sop->so_idhash);
	list_del_init(&sop->so_strhash);
	list_del_init(&sop->so_perlockowner);

        list_add_tail(&sop->so_close_lru, &close_lru);
        sop->so_time = get_seconds();
}

898
void
899 900
release_state_owner(struct nfs4_stateid *stp, struct nfs4_stateowner **sopp,
		int flag)
901 902 903 904
{
	struct nfs4_stateowner *sop = stp->st_stateowner;
	struct nfs4_file *fp = stp->st_file;

905 906
	dprintk("NFSD: release_state_owner\n");
	release_stateid(stp, flag);
907 908

	/* place unused nfs4_stateowners on so_close_lru list to be
909 910 911
	 * released by the laundromat service after the lease period
	 * to enable us to handle CLOSE replay
	 */
912 913
	if (sop->so_confirmed && list_empty(&sop->so_perfilestate))
		move_to_close_lru(sop);
914 915 916 917 918 919
	/* unused nfs4_file's are releseed. XXX slab cache? */
	if (list_empty(&fp->fi_perfile)) {
		release_file(fp);
	}
}

920
static int
921 922 923 924
cmp_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner, clientid_t *clid) {
	return ((sop->so_owner.len == owner->len) && 
	 !memcmp(sop->so_owner.data, owner->data, owner->len) && 
	  (sop->so_client->cl_clientid.cl_id == clid->cl_id));
925 926 927 928
}

/* search ownerstr_hashtbl[] for owner */
static int
929
find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open, struct nfs4_stateowner **op) {
930 931
	struct nfs4_stateowner *local = NULL;

932
	list_for_each_entry(local, &ownerstr_hashtbl[hashval], so_strhash) {
933
		if(!cmp_owner_str(local, &open->op_owner, &open->op_clientid)) 
934 935 936 937 938 939 940 941 942 943 944 945 946 947
			continue;
		*op = local;
		return(1);
	}
	return 0;
}

/* see if clientid is in confirmed hash table */
static int
verify_clientid(struct nfs4_client **client, clientid_t *clid) {

	struct nfs4_client *clp;
	unsigned int idhashval = clientid_hashval(clid->cl_id);

948
	list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) {
949 950 951 952 953 954 955 956 957 958 959
		if (!cmp_clid(&clp->cl_clientid, clid))
			continue;
		*client = clp;
		return 1;
	}
	*client = NULL;
	return 0;
}

/* search file_hashtbl[] for file */
static int
960
find_file(unsigned int hashval, struct inode *ino, struct nfs4_file **fp) {
961 962
	struct nfs4_file *local = NULL;

963
	list_for_each_entry(local, &file_hashtbl[hashval], fi_hash) {
964
		if (local->fi_inode == ino) {
965 966 967 968 969 970 971
			*fp = local;
			return(1);
		}
	}
	return 0;
}

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
#define TEST_ACCESS(x) ((x > 0 || x < 4)?1:0)
#define TEST_DENY(x) ((x >= 0 || x < 5)?1:0)

void
set_access(unsigned int *access, unsigned long bmap) {
	int i;

	*access = 0;
	for (i = 1; i < 4; i++) {
		if(test_bit(i, &bmap))
			*access |= i;
	}
}

void
set_deny(unsigned int *deny, unsigned long bmap) {
	int i;

	*deny = 0;
	for (i = 0; i < 4; i++) {
		if(test_bit(i, &bmap))
			*deny |= i ;
	}
}

997 998
static int
test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) {
999 1000 1001 1002 1003
	unsigned int access, deny;

	set_access(&access, stp->st_access_bmap);
	set_deny(&deny, stp->st_deny_bmap);
	if ((access & open->op_share_deny) || (deny & open->op_share_access))
1004 1005 1006 1007
		return 0;
	return 1;
}

1008 1009 1010 1011
/*
 * Called to check deny when READ with all zero stateid or
 * WRITE with all zero or all one stateid
 */
1012 1013 1014
int
nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type)
{
1015
	struct inode *ino = current_fh->fh_dentry->d_inode;
1016 1017 1018 1019 1020 1021
	unsigned int fi_hashval;
	struct nfs4_file *fp;
	struct nfs4_stateid *stp;

	dprintk("NFSD: nfs4_share_conflict\n");

1022 1023
	fi_hashval = file_hashval(ino);
	if (find_file(fi_hashval, ino, &fp)) {
1024
	/* Search for conflicting share reservations */
1025
		list_for_each_entry(stp, &fp->fi_perfile, st_perfile) {
1026 1027
			if (test_bit(deny_type, &stp->st_deny_bmap) ||
			    test_bit(NFS4_SHARE_DENY_BOTH, &stp->st_deny_bmap))
1028 1029 1030 1031 1032 1033
				return nfserr_share_denied;
		}
	}
	return nfs_ok;
}

1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048
static inline int
nfs4_file_upgrade(struct file *filp, unsigned int share_access)
{
int status;

	if (share_access & NFS4_SHARE_ACCESS_WRITE) {
		status = get_write_access(filp->f_dentry->d_inode);
		if (!status)
			filp->f_mode = FMODE_WRITE;
		else
			return nfserrno(status);
	}
	return nfs_ok;
}

1049 1050 1051 1052 1053 1054 1055 1056 1057
static inline void
nfs4_file_downgrade(struct file *filp, unsigned int share_access)
{
	if (share_access & NFS4_SHARE_ACCESS_WRITE) {
		put_write_access(filp->f_dentry->d_inode);
		filp->f_mode = FMODE_READ;
	}
}

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 1084 1085 1086 1087 1088 1089

/*
 * nfsd4_process_open1()
 * 	lookup stateowner.
 * 		found:
 * 			check confirmed 
 * 				confirmed:
 * 					check seqid
 * 				not confirmed:
 * 					delete owner
 * 					create new owner
 * 		notfound:
 * 			verify clientid
 * 			create new owner
 */
int
nfsd4_process_open1(struct nfsd4_open *open)
{
	int status;
	clientid_t *clientid = &open->op_clientid;
	struct nfs4_client *clp = NULL;
	unsigned int strhashval;
	struct nfs4_stateowner *sop = NULL;

	status = nfserr_inval;
	if (!check_name(open->op_owner))
		goto out;

	status = nfserr_stale_clientid;
	if (STALE_CLIENTID(&open->op_clientid))
		goto out;

1090
	nfs4_lock_state();
1091
	strhashval = ownerstr_hashval(clientid->cl_id, open->op_owner);
1092
	if (find_openstateowner_str(strhashval, open, &sop)) {
1093
		open->op_stateowner = sop;
1094
		/* check for replay */
1095
		if (open->op_seqid == sop->so_seqid){
1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109
			if (!sop->so_replay.rp_buflen) {
			/*
			* The original OPEN failed in so spectacularly that we
			* don't even have replay data saved!  Therefore, we
			* have no choice but to continue processing
			* this OPEN; presumably, we'll fail again for the same
			* reason.
			*/
				dprintk("nfsd4_process_open1: replay with no replay cache\n");
				status = NFS_OK;
				goto renew;
			}
			/* replay: indicate to calling function */
			status = NFSERR_REPLAY_ME;
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
			goto out;
		}
		if (sop->so_confirmed) {
			if (open->op_seqid == sop->so_seqid + 1) { 
				status = nfs_ok;
				goto renew;
			} 
			status = nfserr_bad_seqid;
			goto out;
		}
		/* If we get here, we received and OPEN for an unconfirmed
1121 1122
		 * nfs4_stateowner. 
		 * Since the sequid's are different, purge the 
1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
		 * existing nfs4_stateowner, and instantiate a new one.
		 */
		clp = sop->so_client;
		release_stateowner(sop);
		goto instantiate_new_owner;
	} 
	/* nfs4_stateowner not found. 
	* verify clientid and instantiate new nfs4_stateowner
	* if verify fails this is presumably the result of the 
	* client's lease expiring.
	*
	* XXX compare clp->cl_addr with rqstp addr? 
	*/
	status = nfserr_expired;
	if (!verify_clientid(&clp, clientid))
		goto out;
instantiate_new_owner:
	status = nfserr_resource;
1141
	if (!(sop = alloc_init_open_stateowner(strhashval, clp, open))) 
1142 1143 1144 1145
		goto out;
	open->op_stateowner = sop;
	status = nfs_ok;
renew:
1146
	renew_client(sop->so_client);
1147
out:
1148
	nfs4_unlock_state();
1149 1150 1151 1152 1153 1154 1155 1156 1157
	return status;
}

int
nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open)
{
	struct iattr iattr;
	struct nfs4_stateowner *sop = open->op_stateowner;
	struct nfs4_file *fp;
1158
	struct inode *ino;
1159 1160 1161 1162 1163 1164 1165 1166
	unsigned int fi_hashval;
	struct nfs4_stateid *stq, *stp = NULL;
	int status;

	status = nfserr_resource;
	if (!sop)
		goto out;

1167
	ino = current_fh->fh_dentry->d_inode;
1168

1169 1170 1171 1172
	status = nfserr_inval;
	if (!TEST_ACCESS(open->op_share_access) || !TEST_DENY(open->op_share_deny))
		goto out;

1173
	nfs4_lock_state();
1174 1175
	fi_hashval = file_hashval(ino);
	if (find_file(fi_hashval, ino, &fp)) {
1176 1177
		/* Search for conflicting share reservations */
		status = nfserr_share_denied;
1178
		list_for_each_entry(stq, &fp->fi_perfile, st_perfile) {
1179 1180 1181 1182
			if(stq->st_stateowner == sop) {
				stp = stq;
				continue;
			}
1183 1184 1185
			/* ignore lock owners */
			if (stq->st_stateowner->so_is_open_owner == 0)
				continue;
1186 1187 1188 1189 1190 1191
			if (!test_share(stq,open))	
				goto out;
		}
	} else {
	/* No nfs4_file found; allocate and init a new one */
		status = nfserr_resource;
1192
		if ((fp = alloc_init_file(fi_hashval, ino)) == NULL)
1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203
			goto out;
	}

	if (!stp) {
		int flags = 0;

		status = nfserr_resource;
		if ((stp = kmalloc(sizeof(struct nfs4_stateid),
						GFP_KERNEL)) == NULL)
			goto out;

1204
		if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
1205 1206 1207 1208 1209 1210 1211 1212
			flags = MAY_WRITE;
		else
			flags = MAY_READ;
		if ((status = nfsd_open(rqstp, current_fh,  S_IFREG,
			                      flags,
			                      &stp->st_vfs_file)) != 0)
			goto out_free;

1213
		vfsopen++;
1214 1215 1216 1217 1218 1219 1220 1221 1222
		dget(stp->st_vfs_file.f_dentry);
		mntget(stp->st_vfs_file.f_vfsmnt);

		init_stateid(stp, fp, sop, open);
		stp->st_vfs_set = 1;
	} else {
		/* This is an upgrade of an existing OPEN. 
		 * OR the incoming share with the existing 
		 * nfs4_stateid share */
1223
		unsigned int share_access;
1224

1225 1226 1227
		set_access(&share_access, stp->st_access_bmap);
		share_access = ~share_access;
		share_access &= open->op_share_access;
1228 1229 1230 1231

		/* update the struct file */
		if ((status = nfs4_file_upgrade(&stp->st_vfs_file, share_access)))
			goto out;
1232 1233 1234
		/* remember the open */
		set_bit(open->op_share_access, &stp->st_access_bmap);
		set_bit(open->op_share_deny, &stp->st_deny_bmap);
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
		/* bump the stateid */
		update_stateid(&stp->st_stateid);
	}
	dprintk("nfs4_process_open2: stateid=(%08x/%08x/%08x/%08x)\n\n",
	            stp->st_stateid.si_boot, stp->st_stateid.si_stateownerid,
	            stp->st_stateid.si_fileid, stp->st_stateid.si_generation);

	if (open->op_truncate) {
		iattr.ia_valid = ATTR_SIZE;
		iattr.ia_size = 0;
		status = nfsd_setattr(rqstp, current_fh, &iattr, 0, (time_t)0);
		if (status)
			goto out;
	}
	memcpy(&open->op_stateid, &stp->st_stateid, sizeof(stateid_t));

	open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE;
	status = nfs_ok;
out:
1254 1255 1256
	if (fp && list_empty(&fp->fi_perfile))
		release_file(fp);

1257 1258 1259 1260 1261 1262 1263
	/*
	* To finish the open response, we just need to set the rflags.
	*/
	open->op_rflags = 0;
	if (!open->op_stateowner->so_confirmed)
		open->op_rflags |= NFS4_OPEN_RESULT_CONFIRM;

1264
	nfs4_unlock_state();
1265 1266 1267 1268 1269
	return status;
out_free:
	kfree(stp);
	goto out;
}
1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280
static struct work_struct laundromat_work;
static void laundromat_main(void *);
static DECLARE_WORK(laundromat_work, laundromat_main, NULL);

int 
nfsd4_renew(clientid_t *clid)
{
	struct nfs4_client *clp;
	unsigned int idhashval;
	int status;

1281
	nfs4_lock_state();
1282
	dprintk("process_renew(%08x/%08x): starting\n", 
1283 1284 1285 1286 1287 1288
			clid->cl_boot, clid->cl_id);
	status = nfserr_stale_clientid;
	if (STALE_CLIENTID(clid))
		goto out;
	status = nfs_ok;
	idhashval = clientid_hashval(clid->cl_id);
1289
	list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) {
1290 1291 1292 1293 1294
		if (!cmp_clid(&clp->cl_clientid, clid))
			continue;
		renew_client(clp);
		goto out;
	}
1295
	list_for_each_entry(clp, &unconf_id_hashtbl[idhashval], cl_idhash) {
1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
		if (!cmp_clid(&clp->cl_clientid, clid))
			continue;
		renew_client(clp);
	goto out;
	}
	/*
	* Couldn't find an nfs4_client for this clientid.  
	* Presumably this is because the client took too long to 
	* RENEW, so return NFS4ERR_EXPIRED.
	*/
1306
	dprintk("nfsd4_renew: clientid not found!\n");
1307 1308
	status = nfserr_expired;
out:
1309
	nfs4_unlock_state();
1310 1311 1312 1313 1314 1315 1316
	return status;
}

time_t
nfs4_laundromat(void)
{
	struct nfs4_client *clp;
1317
	struct nfs4_stateowner *sop;
1318 1319
	struct list_head *pos, *next;
	time_t cutoff = get_seconds() - NFSD_LEASE_TIME;
1320 1321
	time_t t, clientid_val = NFSD_LEASE_TIME;
	time_t u, close_val = NFSD_LEASE_TIME;
1322

1323
	nfs4_lock_state();
1324 1325 1326 1327 1328 1329

	dprintk("NFSD: laundromat service - starting, examining clients\n");
	list_for_each_safe(pos, next, &client_lru) {
		clp = list_entry(pos, struct nfs4_client, cl_lru);
		if (time_after((unsigned long)clp->cl_time, (unsigned long)cutoff)) {
			t = clp->cl_time - cutoff;
1330 1331
			if (clientid_val > t)
				clientid_val = t;
1332 1333 1334 1335 1336 1337
			break;
		}
		dprintk("NFSD: purging unused client (clientid %08x)\n",
			clp->cl_clientid.cl_id);
		expire_client(clp);
	}
1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351
	list_for_each_safe(pos, next, &close_lru) {
		sop = list_entry(pos, struct nfs4_stateowner, so_close_lru);
		if (time_after((unsigned long)sop->so_time, (unsigned long)cutoff)) {
			u = sop->so_time - cutoff;
			if (close_val > u)
				close_val = u;
			break;
		}
		dprintk("NFSD: purging unused open stateowner (so_id %d)\n",
			sop->so_id);
		release_stateowner(sop);
	}
	if (clientid_val < NFSD_LAUNDROMAT_MINTIMEOUT)
		clientid_val = NFSD_LAUNDROMAT_MINTIMEOUT;
1352
	nfs4_unlock_state();
1353
	return clientid_val;
1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
}

void
laundromat_main(void *not_used)
{
	time_t t;

	t = nfs4_laundromat();
	dprintk("NFSD: laundromat_main - sleeping for %ld seconds\n", t);
	schedule_delayed_work(&laundromat_work, t*HZ);
}
1365

1366 1367 1368
/* search ownerid_hashtbl[] and close_lru for stateid owner
 * (stateid->si_stateownerid)
 */
1369
struct nfs4_stateowner *
1370
find_openstateowner_id(u32 st_id, int flags) {
1371 1372
	struct nfs4_stateowner *local = NULL;

1373 1374 1375 1376 1377 1378
	dprintk("NFSD: find_openstateowner_id %d\n", st_id);
	if (flags & CLOSE_STATE) {
		list_for_each_entry(local, &close_lru, so_close_lru) {
			if(local->so_id == st_id)
				return local;
		}
1379 1380 1381 1382 1383 1384 1385
	}
	return NULL;
}

static inline int
nfs4_check_fh(struct svc_fh *fhp, struct nfs4_stateid *stp)
{
1386 1387
	return (stp->st_vfs_set == 0 ||
		fhp->fh_dentry->d_inode != stp->st_vfs_file.f_dentry->d_inode);
1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401
}

static int
STALE_STATEID(stateid_t *stateid)
{
	if (stateid->si_boot == boot_time)
		return 0;
	printk("NFSD: stale stateid (%08x/%08x/%08x/%08x)!\n",
		stateid->si_boot, stateid->si_stateownerid, stateid->si_fileid,
		stateid->si_generation);
	return 1;
}


1402 1403 1404 1405 1406 1407 1408 1409 1410
/*
* Checks for stateid operations
*/
int
nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct nfs4_stateid **stpp)
{
	struct nfs4_stateid *stp;
	int status;

1411
	dprintk("NFSD: preprocess_stateid_op: stateid = (%08x/%08x/%08x/%08x)\n",
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423
		stateid->si_boot, stateid->si_stateownerid, 
		stateid->si_fileid, stateid->si_generation); 

	*stpp = NULL;

	/* STALE STATEID */
	status = nfserr_stale_stateid;
	if (STALE_STATEID(stateid)) 
		goto out;

	/* BAD STATEID */
	status = nfserr_bad_stateid;
1424 1425
	if (!(stp = find_stateid(stateid, flags))) {
		dprintk("NFSD: preprocess_stateid_op: no open stateid!\n");
1426 1427 1428
		goto out;
	}
	if ((flags & CHECK_FH) && nfs4_check_fh(current_fh, stp)) {
1429
		dprintk("NFSD: preprocess_stateid_op: fh-stateid mismatch!\n");
1430
		stp->st_vfs_set = 0;
1431 1432 1433
		goto out;
	}
	if (!stp->st_stateowner->so_confirmed) {
1434
		dprintk("preprocess_stateid_op: lockowner not confirmed yet!\n");
1435 1436 1437
		goto out;
	}
	if (stateid->si_generation > stp->st_stateid.si_generation) {
1438
		dprintk("preprocess_stateid_op: future stateid?!\n");
1439 1440 1441 1442 1443 1444
		goto out;
	}

	/* OLD STATEID */
	status = nfserr_old_stateid;
	if (stateid->si_generation < stp->st_stateid.si_generation) {
1445
		dprintk("preprocess_stateid_op: old stateid!\n");
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455
		goto out;
	}
	*stpp = stp;
	status = nfs_ok;
	renew_client(stp->st_stateowner->so_client);
out:
	return status;
}


1456 1457 1458 1459
/* 
 * Checks for sequence id mutating operations. 
 */
int
1460
nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, clientid_t *lockclid)
1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
{
	int status;
	struct nfs4_stateid *stp;
	struct nfs4_stateowner *sop;

	dprintk("NFSD: preprocess_seqid_op: seqid=%d " 
			"stateid = (%08x/%08x/%08x/%08x)\n", seqid,
		stateid->si_boot, stateid->si_stateownerid, stateid->si_fileid,
		stateid->si_generation);
			        
	*stpp = NULL;
	*sopp = NULL;

	status = nfserr_bad_stateid;
	if (ZERO_STATEID(stateid) || ONE_STATEID(stateid)) {
		printk("NFSD: preprocess_seqid_op: magic stateid!\n");
		goto out;
	}

	status = nfserr_stale_stateid;
	if (STALE_STATEID(stateid))
		goto out;
	/*
	* We return BAD_STATEID if filehandle doesn't match stateid, 
	* the confirmed flag is incorrecly set, or the generation 
	* number is incorrect.  
	* If there is no entry in the openfile table for this id, 
	* we can't always return BAD_STATEID;
	* this might be a retransmitted CLOSE which has arrived after 
	* the openfile has been released.
	*/
1492
	if (!(stp = find_stateid(stateid, flags)))
1493 1494 1495 1496
		goto no_nfs4_stateid;

	status = nfserr_bad_stateid;

1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511
	/* for new lock stateowners, check that the lock->v.new.open_stateid
	 * refers to an open stateowner, and that the lockclid
	 * (nfs4_lock->v.new.clientid) is the same as the
	 * open_stateid->st_stateowner->so_client->clientid
	 */
	if (lockclid) {
		struct nfs4_stateowner *sop = stp->st_stateowner;
		struct nfs4_client *clp = sop->so_client;

		if (!sop->so_is_open_owner)
			goto out;
		if (!cmp_clid(&clp->cl_clientid, lockclid))
			goto out;
	}

1512 1513
	if ((flags & CHECK_FH) && nfs4_check_fh(current_fh, stp)) {
		printk("NFSD: preprocess_seqid_op: fh-stateid mismatch!\n");
1514
		stp->st_vfs_set = 0;
1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563
		goto out;
	}

	*stpp = stp;
	*sopp = sop = stp->st_stateowner;

	/*
	*  We now validate the seqid and stateid generation numbers.
	*  For the moment, we ignore the possibility of 
	*  generation number wraparound.
	*/
	if (seqid != sop->so_seqid + 1)
		goto check_replay;

	if (sop->so_confirmed) {
		if (flags & CONFIRM) {
			printk("NFSD: preprocess_seqid_op: expected unconfirmed stateowner!\n");
			goto out;
		}
	}
	else {
		if (!(flags & CONFIRM)) {
			printk("NFSD: preprocess_seqid_op: stateowner not confirmed yet!\n");
			goto out;
		}
	}
	if (stateid->si_generation > stp->st_stateid.si_generation) {
		printk("NFSD: preprocess_seqid_op: future stateid?!\n");
		goto out;
	}

	status = nfserr_old_stateid;
	if (stateid->si_generation < stp->st_stateid.si_generation) {
		printk("NFSD: preprocess_seqid_op: old stateid!\n");
		goto out;
	}
	/* XXX renew the client lease here */
	status = nfs_ok;

out:
	return status;

no_nfs4_stateid:

	/*
	* We determine whether this is a bad stateid or a replay, 
	* starting by trying to look up the stateowner.
	* If stateowner is not found - stateid is bad.
	*/
1564
	if (!(sop = find_openstateowner_id(stateid->si_stateownerid, flags))) {
1565 1566 1567 1568
		printk("NFSD: preprocess_seqid_op: no stateowner or nfs4_stateid!\n");
		status = nfserr_bad_stateid;
		goto out;
	}
1569
	*sopp = sop;
1570 1571 1572 1573

check_replay:
	if (seqid == sop->so_seqid) {
		printk("NFSD: preprocess_seqid_op: retransmission?\n");
1574 1575
		/* indicate replay to calling function */
		status = NFSERR_REPLAY_ME;
1576
	} else  {
1577 1578
		printk("NFSD: preprocess_seqid_op: bad seqid (expected %d, got %d\n", sop->so_seqid +1, seqid);

1579
		status = nfserr_bad_seqid;
1580
	}
1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
	goto out;
}

int
nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc)
{
	int status;
	struct nfs4_stateowner *sop;
	struct nfs4_stateid *stp;

1591
	dprintk("NFSD: nfsd4_open_confirm on file %.*s\n",
1592
			(int)current_fh->fh_dentry->d_name.len,
1593
			current_fh->fh_dentry->d_name.name);
1594 1595 1596 1597

	if ((status = fh_verify(rqstp, current_fh, S_IFREG, 0)))
		goto out;

1598
	oc->oc_stateowner = NULL;
1599
	nfs4_lock_state();
1600 1601 1602

	if ((status = nfs4_preprocess_seqid_op(current_fh, oc->oc_seqid,
					&oc->oc_req_stateid,
1603
					CHECK_FH | CONFIRM | OPEN_STATE,
1604
					&oc->oc_stateowner, &stp, NULL)))
1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618
		goto out; 

	sop = oc->oc_stateowner;
	sop->so_confirmed = 1;
	update_stateid(&stp->st_stateid);
	memcpy(&oc->oc_resp_stateid, &stp->st_stateid, sizeof(stateid_t));
	dprintk("NFSD: nfsd4_open_confirm: success, seqid=%d " 
		"stateid=(%08x/%08x/%08x/%08x)\n", oc->oc_seqid,
		         stp->st_stateid.si_boot,
		         stp->st_stateid.si_stateownerid,
		         stp->st_stateid.si_fileid,
		         stp->st_stateid.si_generation);
	status = nfs_ok;
out:
1619
	nfs4_unlock_state();
1620 1621
	return status;
}
1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648


/*
 * unset all bits in union bitmap (bmap) that
 * do not exist in share (from successful OPEN_DOWNGRADE)
 */
static void
reset_union_bmap_access(unsigned long access, unsigned long *bmap)
{
	int i;
	for (i = 1; i < 4; i++) {
		if ((i & access) != i)
			__clear_bit(i, bmap);
	}
}

static void
reset_union_bmap_deny(unsigned long deny, unsigned long *bmap)
{
	int i;
	for (i = 0; i < 4; i++) {
		if ((i & deny) != i)
			__clear_bit(i, bmap);
	}
}


1649 1650 1651 1652 1653
int
nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od)
{
	int status;
	struct nfs4_stateid *stp;
1654
	unsigned int share_access;
1655 1656

	dprintk("NFSD: nfsd4_open_downgrade on file %.*s\n", 
1657
			(int)current_fh->fh_dentry->d_name.len,
1658 1659
			current_fh->fh_dentry->d_name.name);

1660 1661 1662 1663
	status = nfserr_inval;
	if (!TEST_ACCESS(od->od_share_access) || !TEST_DENY(od->od_share_deny))
		goto out;

1664
	nfs4_lock_state();
1665 1666
	if ((status = nfs4_preprocess_seqid_op(current_fh, od->od_seqid, 
					&od->od_stateid, 
1667
					CHECK_FH | OPEN_STATE, 
1668
					&od->od_stateowner, &stp, NULL)))
1669 1670 1671
		goto out; 

	status = nfserr_inval;
1672 1673 1674
	if (!test_bit(od->od_share_access, &stp->st_access_bmap)) {
		dprintk("NFSD:access not a subset current bitmap: 0x%lx, input access=%08x\n",
			stp->st_access_bmap, od->od_share_access);
1675 1676
		goto out;
	}
1677 1678 1679
	if (!test_bit(od->od_share_deny, &stp->st_deny_bmap)) {
		dprintk("NFSD:deny not a subset current bitmap: 0x%lx, input deny=%08x\n",
			stp->st_deny_bmap, od->od_share_deny);
1680 1681
		goto out;
	}
1682
	set_access(&share_access, stp->st_access_bmap);
1683
	nfs4_file_downgrade(&stp->st_vfs_file, 
1684 1685 1686 1687 1688
	                    share_access & ~od->od_share_access);

	reset_union_bmap_access(od->od_share_access, &stp->st_access_bmap);
	reset_union_bmap_deny(od->od_share_deny, &stp->st_deny_bmap);

1689 1690 1691 1692
	update_stateid(&stp->st_stateid);
	memcpy(&od->od_stateid, &stp->st_stateid, sizeof(stateid_t));
	status = nfs_ok;
out:
1693
	nfs4_unlock_state();
1694 1695 1696
	return status;
}

1697 1698 1699 1700 1701 1702 1703
int
nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close)
{
	int status;
	struct nfs4_stateid *stp;

	dprintk("NFSD: nfsd4_close on file %.*s\n", 
1704
			(int)current_fh->fh_dentry->d_name.len,
1705 1706
			current_fh->fh_dentry->d_name.name);

1707
	nfs4_lock_state();
1708
	/* check close_lru for replay */
1709 1710
	if ((status = nfs4_preprocess_seqid_op(current_fh, close->cl_seqid, 
					&close->cl_stateid, 
1711
					CHECK_FH | OPEN_STATE | CLOSE_STATE,
1712
					&close->cl_stateowner, &stp, NULL)))
1713 1714 1715 1716 1717 1718 1719 1720
		goto out; 
	/*
	*  Return success, but first update the stateid.
	*/
	status = nfs_ok;
	update_stateid(&stp->st_stateid);
	memcpy(&close->cl_stateid, &stp->st_stateid, sizeof(stateid_t));

1721 1722
	/* release_state_owner() calls nfsd_close() if needed */
	release_state_owner(stp, &close->cl_stateowner, OPEN_STATE);
1723
out:
1724
	nfs4_unlock_state();
1725 1726 1727
	return status;
}

1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755
/* 
 * Lock owner state (byte-range locks)
 */
#define LOFF_OVERFLOW(start, len)      ((u64)(len) > ~(u64)(start))
#define LOCK_HASH_BITS              8
#define LOCK_HASH_SIZE             (1 << LOCK_HASH_BITS)
#define LOCK_HASH_MASK             (LOCK_HASH_SIZE - 1)

#define lockownerid_hashval(id) \
        ((id) & LOCK_HASH_MASK)
#define lock_ownerstr_hashval(x, clientid, ownername) \
        ((file_hashval(x) + (clientid) + opaque_hashval((ownername.data), (ownername.len))) & LOCK_HASH_MASK)

static struct list_head lock_ownerid_hashtbl[LOCK_HASH_SIZE];
static struct list_head	lock_ownerstr_hashtbl[LOCK_HASH_SIZE];
static struct list_head lockstateid_hashtbl[STATEID_HASH_SIZE];

struct nfs4_stateid *
find_stateid(stateid_t *stid, int flags)
{
	struct nfs4_stateid *local = NULL;
	u32 st_id = stid->si_stateownerid;
	u32 f_id = stid->si_fileid;
	unsigned int hashval;

	dprintk("NFSD: find_stateid flags 0x%x\n",flags);
	if ((flags & LOCK_STATE) || (flags & RDWR_STATE)) {
		hashval = stateid_hashval(st_id, f_id);
1756
		list_for_each_entry(local, &lockstateid_hashtbl[hashval], st_hash) {
1757 1758 1759 1760 1761 1762 1763
			if((local->st_stateid.si_stateownerid == st_id) &&
			   (local->st_stateid.si_fileid == f_id))
				return local;
		}
	} 
	if ((flags & OPEN_STATE) || (flags & RDWR_STATE)) {
		hashval = stateid_hashval(st_id, f_id);
1764
		list_for_each_entry(local, &stateid_hashtbl[hashval], st_hash) {
1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799
			if((local->st_stateid.si_stateownerid == st_id) &&
			   (local->st_stateid.si_fileid == f_id))
				return local;
		}
	} else
		printk("NFSD: find_stateid: ERROR: no state flag\n");
	return NULL;
}


/*
 * TODO: Linux file offsets are _signed_ 64-bit quantities, which means that
 * we can't properly handle lock requests that go beyond the (2^63 - 1)-th
 * byte, because of sign extension problems.  Since NFSv4 calls for 64-bit
 * locking, this prevents us from being completely protocol-compliant.  The
 * real solution to this problem is to start using unsigned file offsets in
 * the VFS, but this is a very deep change!
 */
static inline void
nfs4_transform_lock_offset(struct file_lock *lock)
{
	if (lock->fl_start < 0)
		lock->fl_start = OFFSET_MAX;
	if (lock->fl_end < 0)
		lock->fl_end = OFFSET_MAX;
}

int
nfs4_verify_lock_stateowner(struct nfs4_stateowner *sop, unsigned int hashval)
{
	struct nfs4_stateowner *local = NULL;
	int status = 0;
			        
	if (hashval >= LOCK_HASH_SIZE)
		goto out;
1800
	list_for_each_entry(local, &lock_ownerid_hashtbl[hashval], so_idhash) {
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
		if (local == sop) {
			status = 1;
			goto out;
		}
	}
out:
	return status;
}


static inline void
nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny)
{
	struct nfs4_stateowner *sop = (struct nfs4_stateowner *) fl->fl_owner;

	deny->ld_sop = NULL;
	if (nfs4_verify_lock_stateowner(sop, fl->fl_pid))
		deny->ld_sop = sop;
	deny->ld_start = fl->fl_start;
	deny->ld_length = ~(u64)0;
	if (fl->fl_end != ~(u64)0)
		deny->ld_length = fl->fl_end - fl->fl_start + 1;        
	deny->ld_type = NFS4_READ_LT;
	if (fl->fl_type != F_RDLCK)
		deny->ld_type = NFS4_WRITE_LT;
}


static int
find_lockstateowner_str(unsigned int hashval, struct xdr_netobj *owner, clientid_t *clid, struct nfs4_stateowner **op) {
	struct nfs4_stateowner *local = NULL;

1833
	list_for_each_entry(local, &lock_ownerstr_hashtbl[hashval], so_strhash) {
1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852
		if(!cmp_owner_str(local, owner, clid)) 
			continue;
		*op = local;
		return(1);
	}
	*op = NULL;
	return 0;
}

/*
 * Alloc a lock owner structure.
 * Called in nfsd4_lock - therefore, OPEN and OPEN_CONFIRM (if needed) has 
 * occured. 
 *
 * strhashval = lock_ownerstr_hashval 
 * so_seqid = lock->lk_new_lock_seqid - 1: it gets bumped in encode 
 */

static struct nfs4_stateowner *
1853
alloc_init_lock_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfs4_stateid *open_stp, struct nfsd4_lock *lock) {
1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864
	struct nfs4_stateowner *sop;
	struct nfs4_replay *rp;
	unsigned int idhashval;

	if (!(sop = alloc_stateowner(&lock->lk_new_owner)))
		return (struct nfs4_stateowner *)NULL;
	idhashval = lockownerid_hashval(current_ownerid);
	INIT_LIST_HEAD(&sop->so_idhash);
	INIT_LIST_HEAD(&sop->so_strhash);
	INIT_LIST_HEAD(&sop->so_perclient);
	INIT_LIST_HEAD(&sop->so_perfilestate);
1865
	INIT_LIST_HEAD(&sop->so_perlockowner);
1866 1867
	INIT_LIST_HEAD(&sop->so_close_lru); /* not used */
	sop->so_time = 0;
1868 1869 1870
	list_add(&sop->so_idhash, &lock_ownerid_hashtbl[idhashval]);
	list_add(&sop->so_strhash, &lock_ownerstr_hashtbl[strhashval]);
	list_add(&sop->so_perclient, &clp->cl_perclient);
1871
	list_add(&sop->so_perlockowner, &open_stp->st_perlockowner);
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
	add_perclient++;
	sop->so_is_open_owner = 0;
	sop->so_id = current_ownerid++;
	sop->so_client = clp;
	sop->so_seqid = lock->lk_new_lock_seqid - 1;
	sop->so_confirmed = 1;
	rp = &sop->so_replay;
	rp->rp_status = NFSERR_SERVERFAULT;
	rp->rp_buflen = 0;
	rp->rp_buf = rp->rp_ibuf;
	alloc_lsowner++;
	return sop;
}

struct nfs4_stateid *
alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struct nfs4_stateid *open_stp)
{
	struct nfs4_stateid *stp;
	unsigned int hashval = stateid_hashval(sop->so_id, fp->fi_id);

	if ((stp = kmalloc(sizeof(struct nfs4_stateid), 
					GFP_KERNEL)) == NULL)
		goto out;
	INIT_LIST_HEAD(&stp->st_hash);
	INIT_LIST_HEAD(&stp->st_perfile);
	INIT_LIST_HEAD(&stp->st_perfilestate);
1898
	INIT_LIST_HEAD(&stp->st_perlockowner); /* not used */
1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910
	list_add(&stp->st_hash, &lockstateid_hashtbl[hashval]);
	list_add(&stp->st_perfile, &fp->fi_perfile);
	list_add_perfile++;
	list_add(&stp->st_perfilestate, &sop->so_perfilestate);
	stp->st_stateowner = sop;
	stp->st_file = fp;
	stp->st_stateid.si_boot = boot_time;
	stp->st_stateid.si_stateownerid = sop->so_id;
	stp->st_stateid.si_fileid = fp->fi_id;
	stp->st_stateid.si_generation = 0;
	stp->st_vfs_file = open_stp->st_vfs_file;
	stp->st_vfs_set = open_stp->st_vfs_set;
1911 1912
	stp->st_access_bmap = open_stp->st_access_bmap;
	stp->st_deny_bmap = open_stp->st_deny_bmap;
1913 1914 1915 1916 1917

out:
	return stp;
}

1918 1919 1920 1921 1922 1923 1924
int
check_lock_length(u64 offset, u64 length)
{
	return ((length == 0)  || ((length != ~(u64)0) &&
 	     LOFF_OVERFLOW(offset, length)));
}

1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939
/*
 *  LOCK operation 
 */
int
nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock)
{
	struct nfs4_stateowner *lock_sop = NULL, *open_sop = NULL;
	struct nfs4_stateid *lock_stp;
	struct file *filp;
	struct file_lock file_lock;
	struct file_lock *conflock;
	int status = 0;
	unsigned int strhashval;

	dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n",
1940 1941
		(long long) lock->lk_offset,
		(long long) lock->lk_length);
1942

1943 1944 1945
	if (check_lock_length(lock->lk_offset, lock->lk_length))
		 return nfserr_inval;

1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
	lock->lk_stateowner = NULL;
	nfs4_lock_state();

	if (lock->lk_is_new) {
	/*
	 * Client indicates that this is a new lockowner.
	 * Use open owner and open stateid to create lock owner and lock 
	 * stateid.
	 */
		struct nfs4_stateid *open_stp = NULL;
		struct nfs4_file *fp;
		
		status = nfserr_stale_clientid;
		if (STALE_CLIENTID(&lock->lk_new_clientid)) {
			printk("NFSD: nfsd4_lock: clientid is stale!\n");
			goto out;
		}
1963 1964
		/* does the clientid in the lock owner own the open stateid? */

1965 1966 1967 1968 1969
		/* validate and update open stateid and open seqid */
		status = nfs4_preprocess_seqid_op(current_fh, 
				        lock->lk_new_open_seqid,
		                        &lock->lk_new_open_stateid,
		                        CHECK_FH | OPEN_STATE,
1970 1971
		                        &open_sop, &open_stp,
					&lock->v.new.clientid);
1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
		if (status)
			goto out;
		/* create lockowner and lock stateid */
		fp = open_stp->st_file;
		strhashval = lock_ownerstr_hashval(fp->fi_inode, 
				open_sop->so_client->cl_clientid.cl_id, 
				lock->v.new.owner);

		/* 
		 * If we already have this lock owner, the client is in 
		 * error (or our bookeeping is wrong!) 
		 * for asking for a 'new lock'.
		 */
		status = nfserr_bad_stateid;
		if (find_lockstateowner_str(strhashval, &lock->v.new.owner,
					&lock->v.new.clientid, &lock_sop))
			goto out;
		status = nfserr_resource;
1990
		if (!(lock->lk_stateowner = alloc_init_lock_stateowner(strhashval, open_sop->so_client, open_stp, lock)))
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
			goto out;
		if ((lock_stp = alloc_init_lock_stateid(lock->lk_stateowner, 
						fp, open_stp)) == NULL)
			goto out;
		/* bump the open seqid used to create the lock */
		open_sop->so_seqid++;
	} else {
		/* lock (lock owner + lock stateid) already exists */
		status = nfs4_preprocess_seqid_op(current_fh,
				       lock->lk_old_lock_seqid, 
				       &lock->lk_old_lock_stateid, 
				       CHECK_FH | LOCK_STATE, 
2003
				       &lock->lk_stateowner, &lock_stp, NULL);
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094
		if (status)
			goto out;
	}
	/* lock->lk_stateowner and lock_stp have been created or found */
	filp = &lock_stp->st_vfs_file;

	if ((status = fh_verify(rqstp, current_fh, S_IFREG, MAY_LOCK))) {
		printk("NFSD: nfsd4_lock: permission denied!\n");
		goto out;
	}

	switch (lock->lk_type) {
		case NFS4_READ_LT:
		case NFS4_READW_LT:
			file_lock.fl_type = F_RDLCK;
		break;
		case NFS4_WRITE_LT:
		case NFS4_WRITEW_LT:
			file_lock.fl_type = F_WRLCK;
		break;
		default:
			status = nfserr_inval;
		goto out;
	}
	file_lock.fl_owner = (fl_owner_t) lock->lk_stateowner;
	file_lock.fl_pid = lockownerid_hashval(lock->lk_stateowner->so_id);
	file_lock.fl_file = filp;
	file_lock.fl_flags = FL_POSIX;
	file_lock.fl_notify = NULL;
	file_lock.fl_insert = NULL;
	file_lock.fl_remove = NULL;

	file_lock.fl_start = lock->lk_offset;
	if ((lock->lk_length == ~(u64)0) || 
			LOFF_OVERFLOW(lock->lk_offset, lock->lk_length))
		file_lock.fl_end = ~(u64)0;
	else
		file_lock.fl_end = lock->lk_offset + lock->lk_length - 1;
	nfs4_transform_lock_offset(&file_lock);

	/*
	* Try to lock the file in the VFS.
	* Note: locks.c uses the BKL to protect the inode's lock list.
	*/

	status = posix_lock_file(filp, &file_lock);
	dprintk("NFSD: nfsd4_lock: posix_test_lock passed. posix_lock_file status %d\n",status);
	switch (-status) {
	case 0: /* success! */
		update_stateid(&lock_stp->st_stateid);
		memcpy(&lock->lk_resp_stateid, &lock_stp->st_stateid, 
				sizeof(stateid_t));
		goto out;
	case (EAGAIN):
		goto conflicting_lock;
	case (EDEADLK):
		status = nfserr_deadlock;
	default:        
		dprintk("NFSD: nfsd4_lock: posix_lock_file() failed! status %d\n",status);
		goto out_destroy_new_stateid;
	}

conflicting_lock:
	dprintk("NFSD: nfsd4_lock: conflicting lock found!\n");
	status = nfserr_denied;
	/* XXX There is a race here. Future patch needed to provide 
	 * an atomic posix_lock_and_test_file
	 */
	if (!(conflock = posix_test_lock(filp, &file_lock))) {
		status = nfserr_serverfault;
		goto out;
	}
	nfs4_set_lock_denied(conflock, &lock->lk_denied);

out_destroy_new_stateid:
	if (lock->lk_is_new) {
		dprintk("NFSD: nfsd4_lock: destroy new stateid!\n");
	/*
	* An error encountered after instantiation of the new
	* stateid has forced us to destroy it.
	*/
		if (!seqid_mutating_err(status))
			open_sop->so_seqid--;

		release_state_owner(lock_stp, &lock->lk_stateowner, LOCK_STATE);
	}
out:
	nfs4_unlock_state();
	return status;
}

2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108
/*
 * LOCKT operation
 */
int
nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt)
{
	struct inode *inode;
	struct nfs4_stateowner *sop;
	struct file file;
	struct file_lock file_lock;
	struct file_lock *conflicting_lock;
	unsigned int strhashval;
	int status;

2109 2110 2111
	if (check_lock_length(lockt->lt_offset, lockt->lt_length))
		 return nfserr_inval;

2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122
	lockt->lt_stateowner = NULL;
	nfs4_lock_state();

	status = nfserr_stale_clientid;
	if (STALE_CLIENTID(&lockt->lt_clientid)) {
		printk("NFSD: nfsd4_lockt: clientid is stale!\n");
		goto out;
	}

	if ((status = fh_verify(rqstp, current_fh, S_IFREG, 0))) {
		printk("NFSD: nfsd4_lockt: fh_verify() failed!\n");
2123 2124
		if (status == nfserr_symlink)
			status = nfserr_inval;
2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
		goto out;
	}

	inode = current_fh->fh_dentry->d_inode;
	switch (lockt->lt_type) {
		case NFS4_READ_LT:
		case NFS4_READW_LT:
			file_lock.fl_type = F_RDLCK;
		break;
		case NFS4_WRITE_LT:
		case NFS4_WRITEW_LT:
			file_lock.fl_type = F_WRLCK;
		break;
		default:
			printk("NFSD: nfs4_lockt: bad lock type!\n");
			status = nfserr_inval;
		goto out;
	}

	strhashval = lock_ownerstr_hashval(inode, 
			lockt->lt_clientid.cl_id, lockt->lt_owner);

2147
	find_lockstateowner_str(strhashval, &lockt->lt_owner,
2148
					&lockt->lt_clientid, 
2149
					&lockt->lt_stateowner);
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
	sop = lockt->lt_stateowner;
	if (sop) {
		file_lock.fl_owner = (fl_owner_t) sop;
		file_lock.fl_pid = lockownerid_hashval(sop->so_id);
	} else {
		file_lock.fl_owner = NULL;
		file_lock.fl_pid = 0;
	}
	file_lock.fl_flags = FL_POSIX;

	file_lock.fl_start = lockt->lt_offset;
	if ((lockt->lt_length == ~(u64)0) || LOFF_OVERFLOW(lockt->lt_offset, lockt->lt_length))
		file_lock.fl_end = ~(u64)0;
	else
		file_lock.fl_end = lockt->lt_offset + lockt->lt_length - 1;

	nfs4_transform_lock_offset(&file_lock);

	/* posix_test_lock uses the struct file _only_ to resolve the inode.
	 * since LOCKT doesn't require an OPEN, and therefore a struct
	 * file may not exist, pass posix_test_lock a struct file with
	 * only the dentry:inode set.
	 */
	memset(&file, 0, sizeof (struct file));
	file.f_dentry = current_fh->fh_dentry;

	status = nfs_ok;
	conflicting_lock = posix_test_lock(&file, &file_lock);
	if (conflicting_lock) {
		status = nfserr_denied;
		nfs4_set_lock_denied(conflicting_lock, &lockt->lt_denied);
	}
out:
	nfs4_unlock_state();
	return status;
}

2187 2188 2189 2190 2191 2192 2193 2194 2195
int
nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku)
{
	struct nfs4_stateid *stp;
	struct file *filp = NULL;
	struct file_lock file_lock;
	int status;
						        
	dprintk("NFSD: nfsd4_locku: start=%Ld length=%Ld\n",
2196 2197
		(long long) locku->lu_offset,
		(long long) locku->lu_length);
2198 2199 2200 2201

	if (check_lock_length(locku->lu_offset, locku->lu_length))
		 return nfserr_inval;

2202 2203 2204 2205 2206 2207
	nfs4_lock_state();
									        
	if ((status = nfs4_preprocess_seqid_op(current_fh, 
					locku->lu_seqid, 
					&locku->lu_stateid, 
					CHECK_FH | LOCK_STATE, 
2208
					&locku->lu_stateowner, &stp, NULL)))
2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250
		goto out;

	filp = &stp->st_vfs_file;
	BUG_ON(!filp);
	file_lock.fl_type = F_UNLCK;
	file_lock.fl_owner = (fl_owner_t) locku->lu_stateowner;
	file_lock.fl_pid = lockownerid_hashval(locku->lu_stateowner->so_id);
	file_lock.fl_file = filp;
	file_lock.fl_flags = FL_POSIX; 
	file_lock.fl_notify = NULL;
	file_lock.fl_insert = NULL;
	file_lock.fl_remove = NULL;
	file_lock.fl_start = locku->lu_offset;

	if ((locku->lu_length == ~(u64)0) || LOFF_OVERFLOW(locku->lu_offset, locku->lu_length))
		file_lock.fl_end = ~(u64)0;
	else
		file_lock.fl_end = locku->lu_offset + locku->lu_length - 1;
	nfs4_transform_lock_offset(&file_lock);

	/*
	*  Try to unlock the file in the VFS.
	*/
	status = posix_lock_file(filp, &file_lock); 
	if (status) {
		printk("NFSD: nfs4_locku: posix_lock_file failed!\n");
		goto out_nfserr;
	}
	/*
	* OK, unlock succeeded; the only thing left to do is update the stateid.
	*/
	update_stateid(&stp->st_stateid);
	memcpy(&locku->lu_stateid, &stp->st_stateid, sizeof(stateid_t));

out:
	nfs4_unlock_state();
	return status;

out_nfserr:
	status = nfserrno(status);
	goto out;
}
2251

2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298
/*
 * returns
 * 	1: locks held by lockowner
 * 	0: no locks held by lockowner
 */
static int
check_for_locks(struct file *filp, struct nfs4_stateowner *lowner)
{
	struct file_lock **flpp;
	struct inode *inode = filp->f_dentry->d_inode;
	int status = 0;

	lock_kernel();
	for (flpp = &inode->i_flock; *flpp != NULL; flpp = &(*flpp)->fl_next) {
		if ((*flpp)->fl_owner == (fl_owner_t)lowner)
			status = 1;
			goto out;
	}
out:
	unlock_kernel();
	return status;
}

int
nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner)
{
	clientid_t *clid = &rlockowner->rl_clientid;
	struct nfs4_stateowner *local = NULL;
	struct xdr_netobj *owner = &rlockowner->rl_owner;
	int status, i;

	dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n",
		clid->cl_boot, clid->cl_id);

	/* XXX check for lease expiration */

	status = nfserr_stale_clientid;
	if (STALE_CLIENTID(clid)) {
		printk("NFSD: nfsd4_release_lockowner: clientid is stale!\n");
		return status;
	}

	nfs4_lock_state();

	/* find the lockowner */
        status = nfs_ok;
	for (i=0; i < LOCK_HASH_SIZE; i++) {
2299
		list_for_each_entry(local, &lock_ownerstr_hashtbl[i], so_strhash) {
2300 2301 2302 2303 2304 2305 2306 2307 2308 2309
			if(cmp_owner_str(local, owner, clid))
				break;
		}
	}
	if (local) {
		struct nfs4_stateid *stp;

		/* check for any locks held by any stateid associated with the
		 * (lock) stateowner */
		status = nfserr_locks_held;
2310
		list_for_each_entry(stp, &local->so_perfilestate, st_perfilestate) {
2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324
			if(stp->st_vfs_set) {
				if (check_for_locks(&stp->st_vfs_file, local))
					goto out;
			}
		}
		/* no locks held by (lock) stateowner */
		status = nfs_ok;
		release_stateowner(local);
	}
out:
	nfs4_unlock_state();
	return status;
}

2325 2326 2327 2328
/* 
 * Start and stop routines
 */

2329
void 
2330 2331 2332 2333
nfs4_state_init(void)
{
	int i;

2334 2335
	if (nfs4_init)
		return;
2336 2337 2338 2339 2340 2341
	for (i = 0; i < CLIENT_HASH_SIZE; i++) {
		INIT_LIST_HEAD(&conf_id_hashtbl[i]);
		INIT_LIST_HEAD(&conf_str_hashtbl[i]);
		INIT_LIST_HEAD(&unconf_str_hashtbl[i]);
		INIT_LIST_HEAD(&unconf_id_hashtbl[i]);
	}
2342 2343 2344 2345 2346
	for (i = 0; i < FILE_HASH_SIZE; i++) {
		INIT_LIST_HEAD(&file_hashtbl[i]);
	}
	for (i = 0; i < OWNER_HASH_SIZE; i++) {
		INIT_LIST_HEAD(&ownerstr_hashtbl[i]);
2347 2348
		INIT_LIST_HEAD(&ownerid_hashtbl[i]);
	}
2349 2350
	for (i = 0; i < STATEID_HASH_SIZE; i++) {
		INIT_LIST_HEAD(&stateid_hashtbl[i]);
2351 2352 2353 2354 2355
		INIT_LIST_HEAD(&lockstateid_hashtbl[i]);
	}
	for (i = 0; i < LOCK_HASH_SIZE; i++) {
		INIT_LIST_HEAD(&lock_ownerid_hashtbl[i]);
		INIT_LIST_HEAD(&lock_ownerstr_hashtbl[i]);
2356
	}
2357 2358 2359
	memset(&zerostateid, 0, sizeof(stateid_t));
	memset(&onestateid, ~0, sizeof(stateid_t));

2360
	INIT_LIST_HEAD(&close_lru);
2361
	INIT_LIST_HEAD(&client_lru);
2362
	init_MUTEX(&client_sema);
2363 2364 2365
	boot_time = get_seconds();
	INIT_WORK(&laundromat_work,laundromat_main, NULL);
	schedule_delayed_work(&laundromat_work, NFSD_LEASE_TIME*HZ);
2366
	nfs4_init = 1;
2367

2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385
}

static void
__nfs4_state_shutdown(void)
{
	int i;
	struct nfs4_client *clp = NULL;

	for (i = 0; i < CLIENT_HASH_SIZE; i++) {
		while (!list_empty(&conf_id_hashtbl[i])) {
			clp = list_entry(conf_id_hashtbl[i].next, struct nfs4_client, cl_idhash);
			expire_client(clp);
		}
		while (!list_empty(&unconf_str_hashtbl[i])) {
			clp = list_entry(unconf_str_hashtbl[i].next, struct nfs4_client, cl_strhash);
			expire_client(clp);
		}
	}
2386
	release_all_files();
2387 2388
	cancel_delayed_work(&laundromat_work);
	flush_scheduled_work();
2389
	nfs4_init = 0;
2390 2391 2392 2393
	dprintk("NFSD: list_add_perfile %d list_del_perfile %d\n",
			list_add_perfile, list_del_perfile);
	dprintk("NFSD: add_perclient %d del_perclient %d\n",
			add_perclient, del_perclient);
2394 2395
	dprintk("NFSD: alloc_file %d free_file %d\n",
			alloc_file, free_file);
2396 2397
	dprintk("NFSD: alloc_sowner %d alloc_lsowner %d free_sowner %d\n",
			alloc_sowner, alloc_lsowner, free_sowner);
2398 2399
	dprintk("NFSD: vfsopen %d vfsclose %d\n",
			vfsopen, vfsclose);
2400 2401 2402 2403 2404
}

void
nfs4_state_shutdown(void)
{
2405
	nfs4_lock_state();
2406
	__nfs4_state_shutdown();
2407
	nfs4_unlock_state();
2408
}