nfs4proc.c 89.8 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
/*
 *  Server-side procedures for NFSv4.
 *
 *  Copyright (c) 2002 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Kendrick Smith <kmsmith@umich.edu>
 *  Andy Adamson   <andros@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.
 */
J
J. Bruce Fields 已提交
35
#include <linux/fs_struct.h>
36
#include <linux/file.h>
A
Anna Schumaker 已提交
37
#include <linux/falloc.h>
38
#include <linux/slab.h>
39
#include <linux/kthread.h>
O
Olga Kornievskaia 已提交
40
#include <linux/sunrpc/addr.h>
41
#include <linux/nfs_ssc.h>
L
Linus Torvalds 已提交
42

43
#include "idmap.h"
44 45
#include "cache.h"
#include "xdr4.h"
46
#include "vfs.h"
47
#include "current_stateid.h"
48
#include "netns.h"
49
#include "acl.h"
50
#include "pnfs.h"
C
Christoph Hellwig 已提交
51
#include "trace.h"
L
Linus Torvalds 已提交
52

C
Chuck Lever 已提交
53 54 55 56 57
static bool inter_copy_offload_enable;
module_param(inter_copy_offload_enable, bool, 0644);
MODULE_PARM_DESC(inter_copy_offload_enable,
		 "Enable inter server to server copy offload. Default: false");

58 59 60 61 62 63
#ifdef CONFIG_NFSD_V4_SECURITY_LABEL
#include <linux/security.h>

static inline void
nfsd4_security_inode_setsecctx(struct svc_fh *resfh, struct xdr_netobj *label, u32 *bmval)
{
64
	struct inode *inode = d_inode(resfh->fh_dentry);
65 66
	int status;

A
Al Viro 已提交
67
	inode_lock(inode);
68 69
	status = security_inode_setsecctx(resfh->fh_dentry,
		label->data, label->len);
A
Al Viro 已提交
70
	inode_unlock(inode);
71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87

	if (status)
		/*
		 * XXX: We should really fail the whole open, but we may
		 * already have created a new file, so it may be too
		 * late.  For now this seems the least of evils:
		 */
		bmval[2] &= ~FATTR4_WORD2_SECURITY_LABEL;

	return;
}
#else
static inline void
nfsd4_security_inode_setsecctx(struct svc_fh *resfh, struct xdr_netobj *label, u32 *bmval)
{ }
#endif

L
Linus Torvalds 已提交
88 89
#define NFSDDBG_FACILITY		NFSDDBG_PROC

90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
static u32 nfsd_attrmask[] = {
	NFSD_WRITEABLE_ATTRS_WORD0,
	NFSD_WRITEABLE_ATTRS_WORD1,
	NFSD_WRITEABLE_ATTRS_WORD2
};

static u32 nfsd41_ex_attrmask[] = {
	NFSD_SUPPATTR_EXCLCREAT_WORD0,
	NFSD_SUPPATTR_EXCLCREAT_WORD1,
	NFSD_SUPPATTR_EXCLCREAT_WORD2
};

static __be32
check_attr_support(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
		   u32 *bmval, u32 *writable)
{
	struct dentry *dentry = cstate->current_fh.fh_dentry;
107
	struct svc_export *exp = cstate->current_fh.fh_export;
108

109
	if (!nfsd_attrs_supported(cstate->minorversion, bmval))
110
		return nfserr_attrnotsupp;
111 112
	if ((bmval[0] & FATTR4_WORD0_ACL) && !IS_POSIXACL(d_inode(dentry)))
		return nfserr_attrnotsupp;
113 114 115
	if ((bmval[2] & FATTR4_WORD2_SECURITY_LABEL) &&
			!(exp->ex_flags & NFSEXP_SECURITY_LABEL))
		return nfserr_attrnotsupp;
116 117
	if (writable && !bmval_is_subset(bmval, writable))
		return nfserr_inval;
118 119 120
	if (writable && (bmval[2] & FATTR4_WORD2_MODE_UMASK) &&
			(bmval[1] & FATTR4_WORD1_MODE))
		return nfserr_inval;
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
	return nfs_ok;
}

static __be32
nfsd4_check_open_attributes(struct svc_rqst *rqstp,
	struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
{
	__be32 status = nfs_ok;

	if (open->op_create == NFS4_OPEN_CREATE) {
		if (open->op_createmode == NFS4_CREATE_UNCHECKED
		    || open->op_createmode == NFS4_CREATE_GUARDED)
			status = check_attr_support(rqstp, cstate,
					open->op_bmval, nfsd_attrmask);
		else if (open->op_createmode == NFS4_CREATE_EXCLUSIVE4_1)
			status = check_attr_support(rqstp, cstate,
					open->op_bmval, nfsd41_ex_attrmask);
	}

	return status;
}

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
static int
is_create_with_attrs(struct nfsd4_open *open)
{
	return open->op_create == NFS4_OPEN_CREATE
		&& (open->op_createmode == NFS4_CREATE_UNCHECKED
		    || open->op_createmode == NFS4_CREATE_GUARDED
		    || open->op_createmode == NFS4_CREATE_EXCLUSIVE4_1);
}

/*
 * if error occurs when setting the acl, just clear the acl bit
 * in the returned attr bitmap.
 */
static void
do_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp,
		struct nfs4_acl *acl, u32 *bmval)
{
	__be32 status;

	status = nfsd4_set_nfs4_acl(rqstp, fhp, acl);
	if (status)
		/*
		 * We should probably fail the whole open at this point,
		 * but we've already created the file, so it's too late;
		 * So this seems the least of evils:
		 */
		bmval[0] &= ~FATTR4_WORD0_ACL;
}

L
Linus Torvalds 已提交
172 173 174 175 176 177
static inline void
fh_dup2(struct svc_fh *dst, struct svc_fh *src)
{
	fh_put(dst);
	dget(src->fh_dentry);
	if (src->fh_export)
178
		exp_get(src->fh_export);
L
Linus Torvalds 已提交
179 180 181
	*dst = *src;
}

182
static __be32
183
do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode)
L
Linus Torvalds 已提交
184
{
185
	__be32 status;
L
Linus Torvalds 已提交
186 187 188 189 190

	if (open->op_truncate &&
		!(open->op_share_access & NFS4_SHARE_ACCESS_WRITE))
		return nfserr_inval;

191 192
	accmode |= NFSD_MAY_READ_IF_EXEC;

L
Linus Torvalds 已提交
193
	if (open->op_share_access & NFS4_SHARE_ACCESS_READ)
M
Miklos Szeredi 已提交
194
		accmode |= NFSD_MAY_READ;
195
	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
M
Miklos Szeredi 已提交
196
		accmode |= (NFSD_MAY_WRITE | NFSD_MAY_TRUNC);
197
	if (open->op_share_deny & NFS4_SHARE_DENY_READ)
M
Miklos Szeredi 已提交
198
		accmode |= NFSD_MAY_WRITE;
L
Linus Torvalds 已提交
199 200 201 202 203 204

	status = fh_verify(rqstp, current_fh, S_IFREG, accmode);

	return status;
}

205 206
static __be32 nfsd_check_obj_isreg(struct svc_fh *fh)
{
207
	umode_t mode = d_inode(fh->fh_dentry)->i_mode;
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222

	if (S_ISREG(mode))
		return nfs_ok;
	if (S_ISDIR(mode))
		return nfserr_isdir;
	/*
	 * Using err_symlink as our catch-all case may look odd; but
	 * there's no other obvious error for this case in 4.0, and we
	 * happen to know that it will cause the linux v4 client to do
	 * the right thing on attempts to open something other than a
	 * regular file.
	 */
	return nfserr_symlink;
}

223 224 225 226 227 228 229 230
static void nfsd4_set_open_owner_reply_cache(struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh *resfh)
{
	if (nfsd4_has_session(cstate))
		return;
	fh_copy_shallow(&open->op_openowner->oo_owner.so_replay.rp_openfh,
			&resfh->fh_handle);
}

231
static __be32
232
do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh **resfh)
L
Linus Torvalds 已提交
233
{
234
	struct svc_fh *current_fh = &cstate->current_fh;
235
	int accmode;
236
	__be32 status;
L
Linus Torvalds 已提交
237

238 239
	*resfh = kmalloc(sizeof(struct svc_fh), GFP_KERNEL);
	if (!*resfh)
240
		return nfserr_jukebox;
241
	fh_init(*resfh, NFS4_FHSIZE);
242
	open->op_truncate = false;
L
Linus Torvalds 已提交
243 244

	if (open->op_create) {
B
Benny Halevy 已提交
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
		/* FIXME: check session persistence and pnfs flags.
		 * The nfsv4.1 spec requires the following semantics:
		 *
		 * Persistent   | pNFS   | Server REQUIRED | Client Allowed
		 * Reply Cache  | server |                 |
		 * -------------+--------+-----------------+--------------------
		 * no           | no     | EXCLUSIVE4_1    | EXCLUSIVE4_1
		 *              |        |                 | (SHOULD)
		 *              |        | and EXCLUSIVE4  | or EXCLUSIVE4
		 *              |        |                 | (SHOULD NOT)
		 * no           | yes    | EXCLUSIVE4_1    | EXCLUSIVE4_1
		 * yes          | no     | GUARDED4        | GUARDED4
		 * yes          | yes    | GUARDED4        | GUARDED4
		 */

L
Linus Torvalds 已提交
260 261
		/*
		 * Note: create modes (UNCHECKED,GUARDED...) are the same
262
		 * in NFSv4 as in v3 except EXCLUSIVE4_1.
L
Linus Torvalds 已提交
263
		 */
J
J. Bruce Fields 已提交
264
		current->fs->umask = open->op_umask;
265 266
		status = do_nfsd_create(rqstp, current_fh, open->op_fname,
					open->op_fnamelen, &open->op_iattr,
267
					*resfh, open->op_createmode,
268
					(u32 *)open->op_verf.data,
269
					&open->op_truncate, &open->op_created);
J
J. Bruce Fields 已提交
270
		current->fs->umask = 0;
271

272
		if (!status && open->op_label.len)
273
			nfsd4_security_inode_setsecctx(*resfh, &open->op_label, open->op_bmval);
274

275
		/*
276 277 278
		 * Following rfc 3530 14.2.16, and rfc 5661 18.16.4
		 * use the returned bitmask to indicate which attributes
		 * we used to store the verifier:
279
		 */
280 281 282
		if (nfsd_create_is_exclusive(open->op_createmode) && status == 0)
			open->op_bmval[1] |= (FATTR4_WORD1_TIME_ACCESS |
						FATTR4_WORD1_TIME_MODIFY);
283 284 285 286 287 288 289
	} else
		/*
		 * Note this may exit with the parent still locked.
		 * We will hold the lock until nfsd4_open's final
		 * lookup, to prevent renames or unlinks until we've had
		 * a chance to an acquire a delegation if appropriate.
		 */
L
Linus Torvalds 已提交
290
		status = nfsd_lookup(rqstp, current_fh,
291
				     open->op_fname, open->op_fnamelen, *resfh);
292 293
	if (status)
		goto out;
294
	status = nfsd_check_obj_isreg(*resfh);
295 296
	if (status)
		goto out;
L
Linus Torvalds 已提交
297

298
	if (is_create_with_attrs(open) && open->op_acl != NULL)
299
		do_set_nfs4_acl(rqstp, *resfh, open->op_acl, open->op_bmval);
300

301
	nfsd4_set_open_owner_reply_cache(cstate, open, *resfh);
302
	accmode = NFSD_MAY_NOP;
303 304
	if (open->op_created ||
			open->op_claim_type == NFS4_OPEN_CLAIM_DELEGATE_CUR)
305
		accmode |= NFSD_MAY_OWNER_OVERRIDE;
306
	status = do_open_permission(rqstp, *resfh, open, accmode);
307
	set_change_info(&open->op_cinfo, current_fh);
308
out:
L
Linus Torvalds 已提交
309 310 311
	return status;
}

312
static __be32
313
do_open_fhandle(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
L
Linus Torvalds 已提交
314
{
315
	struct svc_fh *current_fh = &cstate->current_fh;
316
	__be32 status;
317
	int accmode = 0;
L
Linus Torvalds 已提交
318 319 320 321 322 323 324

	/* We don't know the target directory, and therefore can not
	* set the change info
	*/

	memset(&open->op_cinfo, 0, sizeof(struct nfsd4_change_info));

325
	nfsd4_set_open_owner_reply_cache(cstate, open, current_fh);
L
Linus Torvalds 已提交
326 327 328

	open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) &&
		(open->op_iattr.ia_size == 0);
329 330 331 332 333 334 335 336 337 338 339
	/*
	 * In the delegation case, the client is telling us about an
	 * open that it *already* performed locally, some time ago.  We
	 * should let it succeed now if possible.
	 *
	 * In the case of a CLAIM_FH open, on the other hand, the client
	 * may be counting on us to enforce permissions (the Linux 4.1
	 * client uses this for normal opens, for example).
	 */
	if (open->op_claim_type == NFS4_OPEN_CLAIM_DELEG_CUR_FH)
		accmode = NFSD_MAY_OWNER_OVERRIDE;
L
Linus Torvalds 已提交
340

341
	status = do_open_permission(rqstp, current_fh, open, accmode);
L
Linus Torvalds 已提交
342 343 344 345

	return status;
}

A
Andy Adamson 已提交
346 347 348 349 350 351 352 353 354
static void
copy_clientid(clientid_t *clid, struct nfsd4_session *session)
{
	struct nfsd4_sessionid *sid =
			(struct nfsd4_sessionid *)session->se_sessionid.data;

	clid->cl_boot = sid->clientid.cl_boot;
	clid->cl_id = sid->clientid.cl_id;
}
L
Linus Torvalds 已提交
355

356
static __be32
357
nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
358
	   union nfsd4_op_u *u)
L
Linus Torvalds 已提交
359
{
360
	struct nfsd4_open *open = &u->open;
361
	__be32 status;
362
	struct svc_fh *resfh = NULL;
363 364
	struct net *net = SVC_NET(rqstp);
	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
365
	bool reclaim = false;
A
Andy Adamson 已提交
366

367
	dprintk("NFSD: nfsd4_open filename %.*s op_openowner %p\n",
368
		(int)open->op_fnamelen, open->op_fname,
369
		open->op_openowner);
L
Linus Torvalds 已提交
370 371 372 373 374

	/* This check required by spec. */
	if (open->op_create && open->op_claim_type != NFS4_OPEN_CLAIM_NULL)
		return nfserr_inval;

375
	open->op_created = false;
376 377 378 379 380
	/*
	 * RFC5661 18.51.3
	 * Before RECLAIM_COMPLETE done, server should deny new lock
	 */
	if (nfsd4_has_session(cstate) &&
381 382
	    !test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE,
		      &cstate->session->se_client->cl_flags) &&
383 384 385
	    open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
		return nfserr_grace;

A
Andy Adamson 已提交
386 387 388
	if (nfsd4_has_session(cstate))
		copy_clientid(&open->op_clientid, cstate->session);

L
Linus Torvalds 已提交
389
	/* check seqid for replay. set nfs4_owner */
390
	status = nfsd4_process_open1(cstate, open, nn);
A
Al Viro 已提交
391
	if (status == nfserr_replay_me) {
392
		struct nfs4_replay *rp = &open->op_openowner->oo_owner.so_replay;
393
		fh_put(&cstate->current_fh);
394 395
		fh_copy_shallow(&cstate->current_fh.fh_handle,
				&rp->rp_openfh);
M
Miklos Szeredi 已提交
396
		status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
397 398 399 400
		if (status)
			dprintk("nfsd4_open: replay failed"
				" restoring previous filehandle\n");
		else
A
Al Viro 已提交
401
			status = nfserr_replay_me;
L
Linus Torvalds 已提交
402 403 404
	}
	if (status)
		goto out;
405 406 407 408
	if (open->op_xdr_error) {
		status = open->op_xdr_error;
		goto out;
	}
409

410 411 412 413
	status = nfsd4_check_open_attributes(rqstp, cstate, open);
	if (status)
		goto out;

414 415
	/* Openowner is now set, so sequence id will get bumped.  Now we need
	 * these checks before we do any creates: */
416
	status = nfserr_grace;
417
	if (opens_in_grace(net) && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
418 419
		goto out;
	status = nfserr_no_grace;
420
	if (!opens_in_grace(net) && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS)
421
		goto out;
422

L
Linus Torvalds 已提交
423
	switch (open->op_claim_type) {
424
		case NFS4_OPEN_CLAIM_DELEGATE_CUR:
L
Linus Torvalds 已提交
425
		case NFS4_OPEN_CLAIM_NULL:
426
			status = do_open_lookup(rqstp, cstate, open, &resfh);
L
Linus Torvalds 已提交
427 428 429 430
			if (status)
				goto out;
			break;
		case NFS4_OPEN_CLAIM_PREVIOUS:
431
			status = nfs4_check_open_reclaim(&open->op_clientid,
432
							 cstate, nn);
433 434
			if (status)
				goto out;
435
			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
436
			reclaim = true;
437
			fallthrough;
438 439
		case NFS4_OPEN_CLAIM_FH:
		case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
440
			status = do_open_fhandle(rqstp, cstate, open);
L
Linus Torvalds 已提交
441 442
			if (status)
				goto out;
443
			resfh = &cstate->current_fh;
L
Linus Torvalds 已提交
444
			break;
445
		case NFS4_OPEN_CLAIM_DELEG_PREV_FH:
L
Linus Torvalds 已提交
446
             	case NFS4_OPEN_CLAIM_DELEGATE_PREV:
447
			dprintk("NFSD: unsupported OPEN claim type %d\n",
L
Linus Torvalds 已提交
448 449 450 451
				open->op_claim_type);
			status = nfserr_notsupp;
			goto out;
		default:
452
			dprintk("NFSD: Invalid OPEN claim type %d\n",
L
Linus Torvalds 已提交
453 454 455 456 457 458 459 460 461
				open->op_claim_type);
			status = nfserr_inval;
			goto out;
	}
	/*
	 * nfsd4_process_open2() does the actual opening of the file.  If
	 * successful, it (1) truncates the file if open->op_truncate was
	 * set, (2) sets open->op_stateid, (3) sets open->op_delegation.
	 */
462
	status = nfsd4_process_open2(rqstp, resfh, open);
463 464 465
	WARN(status && open->op_created,
	     "nfsd4_process_open2 failed to open newly-created file! status=%u\n",
	     be32_to_cpu(status));
466 467
	if (reclaim && !status)
		nn->somebody_reclaimed = true;
L
Linus Torvalds 已提交
468
out:
469 470 471 472 473
	if (resfh && resfh != &cstate->current_fh) {
		fh_dup2(&cstate->current_fh, resfh);
		fh_put(resfh);
		kfree(resfh);
	}
474
	nfsd4_cleanup_open_state(cstate, open);
475
	nfsd4_bump_seqid(cstate, status);
L
Linus Torvalds 已提交
476 477 478
	return status;
}

479 480 481 482 483 484 485 486
/*
 * OPEN is the only seqid-mutating operation whose decoding can fail
 * with a seqid-mutating error (specifically, decoding of user names in
 * the attributes).  Therefore we have to do some processing to look up
 * the stateowner so that we can bump the seqid.
 */
static __be32 nfsd4_open_omfg(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_op *op)
{
487
	struct nfsd4_open *open = &op->u.open;
488 489 490 491 492 493

	if (!seqid_mutating_err(ntohl(op->status)))
		return op->status;
	if (nfsd4_has_session(cstate))
		return op->status;
	open->op_xdr_error = op->status;
494
	return nfsd4_open(rqstp, cstate, &op->u);
495 496
}

L
Linus Torvalds 已提交
497 498 499
/*
 * filehandle-manipulating ops.
 */
500
static __be32
501
nfsd4_getfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
502
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
503
{
504
	u->getfh = &cstate->current_fh;
L
Linus Torvalds 已提交
505 506 507
	return nfs_ok;
}

508
static __be32
509
nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
510
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
511
{
512
	struct nfsd4_putfh *putfh = &u->putfh;
513
	__be32 ret;
514

515 516 517 518
	fh_put(&cstate->current_fh);
	cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen;
	memcpy(&cstate->current_fh.fh_handle.fh_base, putfh->pf_fhval,
	       putfh->pf_fhlen);
519 520 521 522 523 524 525 526
	ret = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS);
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
	if (ret == nfserr_stale && putfh->no_verify) {
		SET_FH_FLAG(&cstate->current_fh, NFSD4_FH_FOREIGN);
		ret = 0;
	}
#endif
	return ret;
L
Linus Torvalds 已提交
527 528
}

529
static __be32
530
nfsd4_putrootfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
531
		union nfsd4_op_u *u)
L
Linus Torvalds 已提交
532
{
533
	__be32 status;
L
Linus Torvalds 已提交
534

535
	fh_put(&cstate->current_fh);
536
	status = exp_pseudoroot(rqstp, &cstate->current_fh);
L
Linus Torvalds 已提交
537 538 539
	return status;
}

540
static __be32
541
nfsd4_restorefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
542
		union nfsd4_op_u *u)
L
Linus Torvalds 已提交
543
{
544
	if (!cstate->save_fh.fh_dentry)
L
Linus Torvalds 已提交
545 546
		return nfserr_restorefh;

547
	fh_dup2(&cstate->current_fh, &cstate->save_fh);
548
	if (HAS_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG)) {
549
		memcpy(&cstate->current_stateid, &cstate->save_stateid, sizeof(stateid_t));
550
		SET_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
551
	}
L
Linus Torvalds 已提交
552 553 554
	return nfs_ok;
}

555
static __be32
556
nfsd4_savefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
557
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
558
{
559
	fh_dup2(&cstate->save_fh, &cstate->current_fh);
560
	if (HAS_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG)) {
561
		memcpy(&cstate->save_stateid, &cstate->current_stateid, sizeof(stateid_t));
562
		SET_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG);
563
	}
L
Linus Torvalds 已提交
564 565 566 567 568 569
	return nfs_ok;
}

/*
 * misc nfsv4 ops
 */
570
static __be32
571
nfsd4_access(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
572
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
573
{
574
	struct nfsd4_access *access = &u->access;
575
	u32 access_full;
576

577 578 579 580 581 582
	access_full = NFS3_ACCESS_FULL;
	if (cstate->minorversion >= 2)
		access_full |= NFS4_ACCESS_XALIST | NFS4_ACCESS_XAREAD |
			       NFS4_ACCESS_XAWRITE;

	if (access->ac_req_access & ~access_full)
L
Linus Torvalds 已提交
583 584 585
		return nfserr_inval;

	access->ac_resp_access = access->ac_req_access;
586 587
	return nfsd_access(rqstp, &cstate->current_fh, &access->ac_resp_access,
			   &access->ac_supported);
L
Linus Torvalds 已提交
588 589
}

590
static void gen_boot_verifier(nfs4_verifier *verifier, struct net *net)
591
{
592
	__be32 *verf = (__be32 *)verifier->data;
593

594 595 596
	BUILD_BUG_ON(2*sizeof(*verf) != sizeof(verifier->data));

	nfsd_copy_boot_verifier(verf, net_generic(net, nfsd_net_id));
597 598
}

599
static __be32
600
nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
601
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
602
{
603 604
	struct nfsd4_commit *commit = &u->commit;

605
	return nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
606 607
			     commit->co_count,
			     (__be32 *)commit->co_verf.data);
L
Linus Torvalds 已提交
608 609
}

610
static __be32
611
nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
612
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
613
{
614
	struct nfsd4_create *create = &u->create;
L
Linus Torvalds 已提交
615
	struct svc_fh resfh;
616
	__be32 status;
L
Linus Torvalds 已提交
617 618 619 620
	dev_t rdev;

	fh_init(&resfh, NFS4_FHSIZE);

621
	status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
622 623 624
	if (status)
		return status;

625 626 627 628 629
	status = check_attr_support(rqstp, cstate, create->cr_bmval,
				    nfsd_attrmask);
	if (status)
		return status;

J
J. Bruce Fields 已提交
630
	current->fs->umask = create->cr_umask;
L
Linus Torvalds 已提交
631 632
	switch (create->cr_type) {
	case NF4LNK:
633 634
		status = nfsd_symlink(rqstp, &cstate->current_fh,
				      create->cr_name, create->cr_namelen,
635
				      create->cr_data, &resfh);
L
Linus Torvalds 已提交
636 637 638
		break;

	case NF4BLK:
J
J. Bruce Fields 已提交
639
		status = nfserr_inval;
L
Linus Torvalds 已提交
640 641 642
		rdev = MKDEV(create->cr_specdata1, create->cr_specdata2);
		if (MAJOR(rdev) != create->cr_specdata1 ||
		    MINOR(rdev) != create->cr_specdata2)
J
J. Bruce Fields 已提交
643
			goto out_umask;
644 645 646
		status = nfsd_create(rqstp, &cstate->current_fh,
				     create->cr_name, create->cr_namelen,
				     &create->cr_iattr, S_IFBLK, rdev, &resfh);
L
Linus Torvalds 已提交
647 648 649
		break;

	case NF4CHR:
J
J. Bruce Fields 已提交
650
		status = nfserr_inval;
L
Linus Torvalds 已提交
651 652 653
		rdev = MKDEV(create->cr_specdata1, create->cr_specdata2);
		if (MAJOR(rdev) != create->cr_specdata1 ||
		    MINOR(rdev) != create->cr_specdata2)
J
J. Bruce Fields 已提交
654
			goto out_umask;
655 656 657
		status = nfsd_create(rqstp, &cstate->current_fh,
				     create->cr_name, create->cr_namelen,
				     &create->cr_iattr,S_IFCHR, rdev, &resfh);
L
Linus Torvalds 已提交
658 659 660
		break;

	case NF4SOCK:
661 662 663
		status = nfsd_create(rqstp, &cstate->current_fh,
				     create->cr_name, create->cr_namelen,
				     &create->cr_iattr, S_IFSOCK, 0, &resfh);
L
Linus Torvalds 已提交
664 665 666
		break;

	case NF4FIFO:
667 668 669
		status = nfsd_create(rqstp, &cstate->current_fh,
				     create->cr_name, create->cr_namelen,
				     &create->cr_iattr, S_IFIFO, 0, &resfh);
L
Linus Torvalds 已提交
670 671 672 673
		break;

	case NF4DIR:
		create->cr_iattr.ia_valid &= ~ATTR_SIZE;
674 675 676
		status = nfsd_create(rqstp, &cstate->current_fh,
				     create->cr_name, create->cr_namelen,
				     &create->cr_iattr, S_IFDIR, 0, &resfh);
L
Linus Torvalds 已提交
677 678 679 680 681 682
		break;

	default:
		status = nfserr_badtype;
	}

683 684
	if (status)
		goto out;
L
Linus Torvalds 已提交
685

686 687 688
	if (create->cr_label.len)
		nfsd4_security_inode_setsecctx(&resfh, &create->cr_label, create->cr_bmval);

689 690 691 692 693 694 695 696
	if (create->cr_acl != NULL)
		do_set_nfs4_acl(rqstp, &resfh, create->cr_acl,
				create->cr_bmval);

	fh_unlock(&cstate->current_fh);
	set_change_info(&create->cr_cinfo, &cstate->current_fh);
	fh_dup2(&cstate->current_fh, &resfh);
out:
L
Linus Torvalds 已提交
697
	fh_put(&resfh);
J
J. Bruce Fields 已提交
698 699
out_umask:
	current->fs->umask = 0;
L
Linus Torvalds 已提交
700 701 702
	return status;
}

703
static __be32
704
nfsd4_getattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
705
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
706
{
707
	struct nfsd4_getattr *getattr = &u->getattr;
708
	__be32 status;
L
Linus Torvalds 已提交
709

M
Miklos Szeredi 已提交
710
	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
711 712 713 714 715 716
	if (status)
		return status;

	if (getattr->ga_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
		return nfserr_inval;

717 718 719
	getattr->ga_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
	getattr->ga_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
	getattr->ga_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
L
Linus Torvalds 已提交
720

721
	getattr->ga_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
722 723 724
	return nfs_ok;
}

725
static __be32
726
nfsd4_link(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
727
	   union nfsd4_op_u *u)
L
Linus Torvalds 已提交
728
{
729
	struct nfsd4_link *link = &u->link;
730
	__be32 status;
L
Linus Torvalds 已提交
731

732 733
	status = nfsd_link(rqstp, &cstate->current_fh,
			   link->li_name, link->li_namelen, &cstate->save_fh);
L
Linus Torvalds 已提交
734
	if (!status)
735
		set_change_info(&link->li_cinfo, &cstate->current_fh);
L
Linus Torvalds 已提交
736 737 738
	return status;
}

739
static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
L
Linus Torvalds 已提交
740 741
{
	struct svc_fh tmp_fh;
742
	__be32 ret;
L
Linus Torvalds 已提交
743 744

	fh_init(&tmp_fh, NFS4_FHSIZE);
745 746
	ret = exp_pseudoroot(rqstp, &tmp_fh);
	if (ret)
L
Linus Torvalds 已提交
747
		return ret;
748
	if (tmp_fh.fh_dentry == fh->fh_dentry) {
L
Linus Torvalds 已提交
749 750 751 752
		fh_put(&tmp_fh);
		return nfserr_noent;
	}
	fh_put(&tmp_fh);
753 754 755 756 757
	return nfsd_lookup(rqstp, fh, "..", 2, fh);
}

static __be32
nfsd4_lookupp(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
758
	      union nfsd4_op_u *u)
759 760
{
	return nfsd4_do_lookupp(rqstp, &cstate->current_fh);
L
Linus Torvalds 已提交
761 762
}

763
static __be32
764
nfsd4_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
765
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
766
{
767
	return nfsd_lookup(rqstp, &cstate->current_fh,
768
			   u->lookup.lo_name, u->lookup.lo_len,
769
			   &cstate->current_fh);
L
Linus Torvalds 已提交
770 771
}

772
static __be32
773
nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
774
	   union nfsd4_op_u *u)
L
Linus Torvalds 已提交
775
{
776
	struct nfsd4_read *read = &u->read;
777
	__be32 status;
L
Linus Torvalds 已提交
778

779
	read->rd_nf = NULL;
L
Linus Torvalds 已提交
780 781 782
	if (read->rd_offset >= OFFSET_MAX)
		return nfserr_inval;

783 784 785
	trace_nfsd_read_start(rqstp, &cstate->current_fh,
			      read->rd_offset, read->rd_length);

786 787 788 789 790 791 792 793 794
	/*
	 * If we do a zero copy read, then a client will see read data
	 * that reflects the state of the file *after* performing the
	 * following compound.
	 *
	 * To ensure proper ordering, we therefore turn off zero copy if
	 * the client wants us to do more in this compound:
	 */
	if (!nfsd4_last_compound_op(rqstp))
795
		clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
796

L
Linus Torvalds 已提交
797
	/* check stateid */
798 799
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
					&read->rd_stateid, RD_STATE,
O
Olga Kornievskaia 已提交
800
					&read->rd_nf, NULL);
801
	if (status) {
L
Linus Torvalds 已提交
802 803 804 805 806 807
		dprintk("NFSD: nfsd4_read: couldn't process stateid!\n");
		goto out;
	}
	status = nfs_ok;
out:
	read->rd_rqstp = rqstp;
808
	read->rd_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
809 810 811
	return status;
}

812 813 814 815

static void
nfsd4_read_release(union nfsd4_op_u *u)
{
816 817
	if (u->read.rd_nf)
		nfsd_file_put(u->read.rd_nf);
818 819
	trace_nfsd_read_done(u->read.rd_rqstp, u->read.rd_fhp,
			     u->read.rd_offset, u->read.rd_length);
820 821
}

822
static __be32
823
nfsd4_readdir(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
824
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
825
{
826
	struct nfsd4_readdir *readdir = &u->readdir;
L
Linus Torvalds 已提交
827 828 829 830 831 832 833 834
	u64 cookie = readdir->rd_cookie;
	static const nfs4_verifier zeroverf;

	/* no need to check permission - this will be done in nfsd_readdir() */

	if (readdir->rd_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1)
		return nfserr_inval;

835 836 837
	readdir->rd_bmval[0] &= nfsd_suppattrs[cstate->minorversion][0];
	readdir->rd_bmval[1] &= nfsd_suppattrs[cstate->minorversion][1];
	readdir->rd_bmval[2] &= nfsd_suppattrs[cstate->minorversion][2];
L
Linus Torvalds 已提交
838

839
	if ((cookie == 1) || (cookie == 2) ||
L
Linus Torvalds 已提交
840 841 842 843
	    (cookie == 0 && memcmp(readdir->rd_verf.data, zeroverf.data, NFS4_VERIFIER_SIZE)))
		return nfserr_bad_cookie;

	readdir->rd_rqstp = rqstp;
844
	readdir->rd_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
845 846 847
	return nfs_ok;
}

848
static __be32
849
nfsd4_readlink(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
850
	       union nfsd4_op_u *u)
L
Linus Torvalds 已提交
851
{
852 853
	u->readlink.rl_rqstp = rqstp;
	u->readlink.rl_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
854 855 856
	return nfs_ok;
}

857
static __be32
858
nfsd4_remove(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
859
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
860
{
861
	struct nfsd4_remove *remove = &u->remove;
862
	__be32 status;
L
Linus Torvalds 已提交
863

864
	if (opens_in_grace(SVC_NET(rqstp)))
865
		return nfserr_grace;
866 867
	status = nfsd_unlink(rqstp, &cstate->current_fh, 0,
			     remove->rm_name, remove->rm_namelen);
L
Linus Torvalds 已提交
868
	if (!status) {
869 870
		fh_unlock(&cstate->current_fh);
		set_change_info(&remove->rm_cinfo, &cstate->current_fh);
L
Linus Torvalds 已提交
871 872 873 874
	}
	return status;
}

875
static __be32
876
nfsd4_rename(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
877
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
878
{
879
	struct nfsd4_rename *rename = &u->rename;
880
	__be32 status;
L
Linus Torvalds 已提交
881

882
	if (opens_in_grace(SVC_NET(rqstp)))
883
		return nfserr_grace;
884 885
	status = nfsd_rename(rqstp, &cstate->save_fh, rename->rn_sname,
			     rename->rn_snamelen, &cstate->current_fh,
L
Linus Torvalds 已提交
886
			     rename->rn_tname, rename->rn_tnamelen);
887 888 889 890 891
	if (status)
		return status;
	set_change_info(&rename->rn_sinfo, &cstate->current_fh);
	set_change_info(&rename->rn_tinfo, &cstate->save_fh);
	return nfs_ok;
L
Linus Torvalds 已提交
892 893
}

A
Andy Adamson 已提交
894 895
static __be32
nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
896
	      union nfsd4_op_u *u)
A
Andy Adamson 已提交
897
{
898
	struct nfsd4_secinfo *secinfo = &u->secinfo;
A
Andy Adamson 已提交
899 900 901 902
	struct svc_export *exp;
	struct dentry *dentry;
	__be32 err;

903 904 905
	err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC);
	if (err)
		return err;
A
Andy Adamson 已提交
906 907 908 909 910
	err = nfsd_lookup_dentry(rqstp, &cstate->current_fh,
				    secinfo->si_name, secinfo->si_namelen,
				    &exp, &dentry);
	if (err)
		return err;
911
	fh_unlock(&cstate->current_fh);
912
	if (d_really_is_negative(dentry)) {
A
Andy Adamson 已提交
913 914 915 916 917
		exp_put(exp);
		err = nfserr_noent;
	} else
		secinfo->si_exp = exp;
	dput(dentry);
918 919 920
	if (cstate->minorversion)
		/* See rfc 5661 section 2.6.3.1.1.8 */
		fh_put(&cstate->current_fh);
A
Andy Adamson 已提交
921 922 923
	return err;
}

J
J. Bruce Fields 已提交
924 925
static __be32
nfsd4_secinfo_no_name(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
926
		union nfsd4_op_u *u)
J
J. Bruce Fields 已提交
927 928 929
{
	__be32 err;

930
	switch (u->secinfo_no_name.sin_style) {
J
J. Bruce Fields 已提交
931 932 933 934 935 936 937 938 939 940
	case NFS4_SECINFO_STYLE4_CURRENT_FH:
		break;
	case NFS4_SECINFO_STYLE4_PARENT:
		err = nfsd4_do_lookupp(rqstp, &cstate->current_fh);
		if (err)
			return err;
		break;
	default:
		return nfserr_inval;
	}
941

942
	u->secinfo_no_name.sin_exp = exp_get(cstate->current_fh.fh_export);
J
J. Bruce Fields 已提交
943 944 945 946
	fh_put(&cstate->current_fh);
	return nfs_ok;
}

947 948 949 950 951 952 953
static void
nfsd4_secinfo_release(union nfsd4_op_u *u)
{
	if (u->secinfo.si_exp)
		exp_put(u->secinfo.si_exp);
}

954 955 956 957 958 959 960
static void
nfsd4_secinfo_no_name_release(union nfsd4_op_u *u)
{
	if (u->secinfo_no_name.sin_exp)
		exp_put(u->secinfo_no_name.sin_exp);
}

961
static __be32
962
nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
963
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
964
{
965
	struct nfsd4_setattr *setattr = &u->setattr;
966
	__be32 status = nfs_ok;
967
	int err;
L
Linus Torvalds 已提交
968 969

	if (setattr->sa_iattr.ia_valid & ATTR_SIZE) {
970
		status = nfs4_preprocess_stateid_op(rqstp, cstate,
971
				&cstate->current_fh, &setattr->sa_stateid,
O
Olga Kornievskaia 已提交
972
				WR_STATE, NULL, NULL);
973
		if (status) {
974
			dprintk("NFSD: nfsd4_setattr: couldn't process stateid!\n");
975 976
			return status;
		}
L
Linus Torvalds 已提交
977
	}
978 979 980
	err = fh_want_write(&cstate->current_fh);
	if (err)
		return nfserrno(err);
L
Linus Torvalds 已提交
981
	status = nfs_ok;
982 983 984 985 986 987

	status = check_attr_support(rqstp, cstate, setattr->sa_bmval,
				    nfsd_attrmask);
	if (status)
		goto out;

L
Linus Torvalds 已提交
988
	if (setattr->sa_acl != NULL)
989 990
		status = nfsd4_set_nfs4_acl(rqstp, &cstate->current_fh,
					    setattr->sa_acl);
L
Linus Torvalds 已提交
991
	if (status)
992
		goto out;
993 994 995 996 997
	if (setattr->sa_label.len)
		status = nfsd4_set_nfs4_label(rqstp, &cstate->current_fh,
				&setattr->sa_label);
	if (status)
		goto out;
998
	status = nfsd_setattr(rqstp, &cstate->current_fh, &setattr->sa_iattr,
999
				0, (time64_t)0);
1000
out:
A
Al Viro 已提交
1001
	fh_drop_write(&cstate->current_fh);
L
Linus Torvalds 已提交
1002 1003 1004
	return status;
}

1005
static __be32
1006
nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1007
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
1008
{
1009
	struct nfsd4_write *write = &u->write;
L
Linus Torvalds 已提交
1010
	stateid_t *stateid = &write->wr_stateid;
1011
	struct nfsd_file *nf = NULL;
1012
	__be32 status = nfs_ok;
1013
	unsigned long cnt;
1014
	int nvecs;
L
Linus Torvalds 已提交
1015 1016 1017 1018

	if (write->wr_offset >= OFFSET_MAX)
		return nfserr_inval;

1019 1020 1021
	cnt = write->wr_buflen;
	trace_nfsd_write_start(rqstp, &cstate->current_fh,
			       write->wr_offset, cnt);
1022
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
O
Olga Kornievskaia 已提交
1023
						stateid, WR_STATE, &nf, NULL);
1024 1025 1026 1027 1028
	if (status) {
		dprintk("NFSD: nfsd4_write: couldn't process stateid!\n");
		return status;
	}

L
Linus Torvalds 已提交
1029 1030
	write->wr_how_written = write->wr_stable_how;

1031 1032
	nvecs = svc_fill_write_vector(rqstp, write->wr_payload.pages,
				      write->wr_payload.head, write->wr_buflen);
1033 1034
	WARN_ON_ONCE(nvecs > ARRAY_SIZE(rqstp->rq_vec));

1035
	status = nfsd_vfs_write(rqstp, &cstate->current_fh, nf,
1036
				write->wr_offset, rqstp->rq_vec, nvecs, &cnt,
1037 1038
				write->wr_how_written,
				(__be32 *)write->wr_verifier.data);
1039
	nfsd_file_put(nf);
L
Linus Torvalds 已提交
1040

1041
	write->wr_bytes_written = cnt;
1042 1043
	trace_nfsd_write_done(rqstp, &cstate->current_fh,
			      write->wr_offset, cnt);
L
Linus Torvalds 已提交
1044 1045 1046
	return status;
}

1047
static __be32
A
Anna Schumaker 已提交
1048
nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1049 1050
		  stateid_t *src_stateid, struct nfsd_file **src,
		  stateid_t *dst_stateid, struct nfsd_file **dst)
1051 1052 1053
{
	__be32 status;

1054 1055 1056
	if (!cstate->save_fh.fh_dentry)
		return nfserr_nofilehandle;

1057
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
O
Olga Kornievskaia 已提交
1058
					    src_stateid, RD_STATE, src, NULL);
1059 1060 1061 1062 1063 1064
	if (status) {
		dprintk("NFSD: %s: couldn't process src stateid!\n", __func__);
		goto out;
	}

	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
O
Olga Kornievskaia 已提交
1065
					    dst_stateid, WR_STATE, dst, NULL);
1066 1067 1068 1069 1070 1071
	if (status) {
		dprintk("NFSD: %s: couldn't process dst stateid!\n", __func__);
		goto out_put_src;
	}

	/* fix up for NFS-specific error code */
1072 1073
	if (!S_ISREG(file_inode((*src)->nf_file)->i_mode) ||
	    !S_ISREG(file_inode((*dst)->nf_file)->i_mode)) {
1074 1075 1076 1077
		status = nfserr_wrong_type;
		goto out_put_dst;
	}

A
Anna Schumaker 已提交
1078 1079 1080
out:
	return status;
out_put_dst:
1081
	nfsd_file_put(*dst);
A
Anna Schumaker 已提交
1082
out_put_src:
1083
	nfsd_file_put(*src);
A
Anna Schumaker 已提交
1084 1085 1086 1087 1088
	goto out;
}

static __be32
nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1089
		union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1090
{
1091
	struct nfsd4_clone *clone = &u->clone;
1092
	struct nfsd_file *src, *dst;
A
Anna Schumaker 已提交
1093 1094 1095 1096 1097 1098 1099
	__be32 status;

	status = nfsd4_verify_copy(rqstp, cstate, &clone->cl_src_stateid, &src,
				   &clone->cl_dst_stateid, &dst);
	if (status)
		goto out;

1100 1101
	status = nfsd4_clone_file_range(src, clone->cl_src_pos,
			dst, clone->cl_dst_pos, clone->cl_count,
1102
			EX_ISSYNC(cstate->current_fh.fh_export));
1103

1104 1105
	nfsd_file_put(dst);
	nfsd_file_put(src);
1106 1107 1108 1109
out:
	return status;
}

1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
void nfs4_put_copy(struct nfsd4_copy *copy)
{
	if (!refcount_dec_and_test(&copy->refcount))
		return;
	kfree(copy);
}

static bool
check_and_set_stop_copy(struct nfsd4_copy *copy)
{
	bool value;

	spin_lock(&copy->cp_clp->async_lock);
	value = copy->stopped;
	if (!copy->stopped)
		copy->stopped = true;
	spin_unlock(&copy->cp_clp->async_lock);
	return value;
}

static void nfsd4_stop_copy(struct nfsd4_copy *copy)
{
	/* only 1 thread should stop the copy */
	if (!check_and_set_stop_copy(copy))
		kthread_stop(copy->copy_task);
	nfs4_put_copy(copy);
}

static struct nfsd4_copy *nfsd4_get_copy(struct nfs4_client *clp)
{
	struct nfsd4_copy *copy = NULL;

	spin_lock(&clp->async_lock);
	if (!list_empty(&clp->async_copies)) {
		copy = list_first_entry(&clp->async_copies, struct nfsd4_copy,
					copies);
		refcount_inc(&copy->refcount);
	}
	spin_unlock(&clp->async_lock);
	return copy;
}

void nfsd4_shutdown_copy(struct nfs4_client *clp)
{
	struct nfsd4_copy *copy;

	while ((copy = nfsd4_get_copy(clp)) != NULL)
		nfsd4_stop_copy(copy);
}
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
#ifdef CONFIG_NFSD_V4_2_INTER_SSC

extern struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
				   struct nfs_fh *src_fh,
				   nfs4_stateid *stateid);
extern void nfs42_ssc_close(struct file *filep);

extern void nfs_sb_deactive(struct super_block *sb);

#define NFSD42_INTERSSC_MOUNTOPS "vers=4.2,addr=%s,sec=sys"

1170
/*
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
 * Support one copy source server for now.
 */
static __be32
nfsd4_interssc_connect(struct nl4_server *nss, struct svc_rqst *rqstp,
		       struct vfsmount **mount)
{
	struct file_system_type *type;
	struct vfsmount *ss_mnt;
	struct nfs42_netaddr *naddr;
	struct sockaddr_storage tmp_addr;
	size_t tmp_addrlen, match_netid_len = 3;
	char *startsep = "", *endsep = "", *match_netid = "tcp";
	char *ipaddr, *dev_name, *raw_data;
1184 1185
	int len, raw_len;
	__be32 status = nfserr_inval;
1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222

	naddr = &nss->u.nl4_addr;
	tmp_addrlen = rpc_uaddr2sockaddr(SVC_NET(rqstp), naddr->addr,
					 naddr->addr_len,
					 (struct sockaddr *)&tmp_addr,
					 sizeof(tmp_addr));
	if (tmp_addrlen == 0)
		goto out_err;

	if (tmp_addr.ss_family == AF_INET6) {
		startsep = "[";
		endsep = "]";
		match_netid = "tcp6";
		match_netid_len = 4;
	}

	if (naddr->netid_len != match_netid_len ||
		strncmp(naddr->netid, match_netid, naddr->netid_len))
		goto out_err;

	/* Construct the raw data for the vfs_kern_mount call */
	len = RPC_MAX_ADDRBUFLEN + 1;
	ipaddr = kzalloc(len, GFP_KERNEL);
	if (!ipaddr)
		goto out_err;

	rpc_ntop((struct sockaddr *)&tmp_addr, ipaddr, len);

	/* 2 for ipv6 endsep and startsep. 3 for ":/" and trailing '/0'*/

	raw_len = strlen(NFSD42_INTERSSC_MOUNTOPS) + strlen(ipaddr);
	raw_data = kzalloc(raw_len, GFP_KERNEL);
	if (!raw_data)
		goto out_free_ipaddr;

	snprintf(raw_data, raw_len, NFSD42_INTERSSC_MOUNTOPS, ipaddr);

1223
	status = nfserr_nodev;
1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255
	type = get_fs_type("nfs");
	if (!type)
		goto out_free_rawdata;

	/* Set the server:<export> for the vfs_kern_mount call */
	dev_name = kzalloc(len + 5, GFP_KERNEL);
	if (!dev_name)
		goto out_free_rawdata;
	snprintf(dev_name, len + 5, "%s%s%s:/", startsep, ipaddr, endsep);

	/* Use an 'internal' mount: SB_KERNMOUNT -> MNT_INTERNAL */
	ss_mnt = vfs_kern_mount(type, SB_KERNMOUNT, dev_name, raw_data);
	module_put(type->owner);
	if (IS_ERR(ss_mnt))
		goto out_free_devname;

	status = 0;
	*mount = ss_mnt;

out_free_devname:
	kfree(dev_name);
out_free_rawdata:
	kfree(raw_data);
out_free_ipaddr:
	kfree(ipaddr);
out_err:
	return status;
}

static void
nfsd4_interssc_disconnect(struct vfsmount *ss_mnt)
{
1256
	nfs_do_sb_deactive(ss_mnt->mnt_sb);
1257 1258 1259
	mntput(ss_mnt);
}

1260
/*
1261
 * Verify COPY destination stateid.
1262
 *
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275
 * Connect to the source server with NFSv4.1.
 * Create the source struct file for nfsd_copy_range.
 * Called with COPY cstate:
 *    SAVED_FH: source filehandle
 *    CURRENT_FH: destination filehandle
 */
static __be32
nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
		      struct nfsd4_compound_state *cstate,
		      struct nfsd4_copy *copy, struct vfsmount **mount)
{
	struct svc_fh *s_fh = NULL;
	stateid_t *s_stid = &copy->cp_src_stateid;
1276
	__be32 status = nfserr_inval;
1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292

	/* Verify the destination stateid and set dst struct file*/
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
					    &copy->cp_dst_stateid,
					    WR_STATE, &copy->nf_dst, NULL);
	if (status)
		goto out;

	status = nfsd4_interssc_connect(&copy->cp_src, rqstp, mount);
	if (status)
		goto out;

	s_fh = &cstate->save_fh;

	copy->c_fh.size = s_fh->fh_handle.fh_size;
	memcpy(copy->c_fh.data, &s_fh->fh_handle.fh_base, copy->c_fh.size);
1293
	copy->stateid.seqid = cpu_to_be32(s_stid->si_generation);
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	memcpy(copy->stateid.other, (void *)&s_stid->si_opaque,
	       sizeof(stateid_opaque_t));

	status = 0;
out:
	return status;
}

static void
nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src,
			struct nfsd_file *dst)
{
	nfs42_ssc_close(src->nf_file);
1307
	/* 'src' is freed by nfsd4_do_async_copy */
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
	nfsd_file_put(dst);
	mntput(ss_mnt);
}

#else /* CONFIG_NFSD_V4_2_INTER_SSC */

static __be32
nfsd4_setup_inter_ssc(struct svc_rqst *rqstp,
		      struct nfsd4_compound_state *cstate,
		      struct nfsd4_copy *copy,
		      struct vfsmount **mount)
{
	*mount = NULL;
1321
	return nfserr_inval;
1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 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
}

static void
nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src,
			struct nfsd_file *dst)
{
}

static void
nfsd4_interssc_disconnect(struct vfsmount *ss_mnt)
{
}

static struct file *nfs42_ssc_open(struct vfsmount *ss_mnt,
				   struct nfs_fh *src_fh,
				   nfs4_stateid *stateid)
{
	return NULL;
}
#endif /* CONFIG_NFSD_V4_2_INTER_SSC */

static __be32
nfsd4_setup_intra_ssc(struct svc_rqst *rqstp,
		      struct nfsd4_compound_state *cstate,
		      struct nfsd4_copy *copy)
{
	return nfsd4_verify_copy(rqstp, cstate, &copy->cp_src_stateid,
				 &copy->nf_src, &copy->cp_dst_stateid,
				 &copy->nf_dst);
}

static void
nfsd4_cleanup_intra_ssc(struct nfsd_file *src, struct nfsd_file *dst)
{
	nfsd_file_put(src);
	nfsd_file_put(dst);
}
1359 1360 1361 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 1389 1390 1391 1392 1393 1394

static void nfsd4_cb_offload_release(struct nfsd4_callback *cb)
{
	struct nfsd4_copy *copy = container_of(cb, struct nfsd4_copy, cp_cb);

	nfs4_put_copy(copy);
}

static int nfsd4_cb_offload_done(struct nfsd4_callback *cb,
				 struct rpc_task *task)
{
	return 1;
}

static const struct nfsd4_callback_ops nfsd4_cb_offload_ops = {
	.release = nfsd4_cb_offload_release,
	.done = nfsd4_cb_offload_done
};

static void nfsd4_init_copy_res(struct nfsd4_copy *copy, bool sync)
{
	copy->cp_res.wr_stable_how = NFS_UNSTABLE;
	copy->cp_synchronous = sync;
	gen_boot_verifier(&copy->cp_res.wr_verifier, copy->cp_clp->net);
}

static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy)
{
	ssize_t bytes_copied = 0;
	size_t bytes_total = copy->cp_count;
	u64 src_pos = copy->cp_src_pos;
	u64 dst_pos = copy->cp_dst_pos;

	do {
		if (kthread_should_stop())
			break;
1395 1396 1397
		bytes_copied = nfsd_copy_file_range(copy->nf_src->nf_file,
				src_pos, copy->nf_dst->nf_file, dst_pos,
				bytes_total);
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423
		if (bytes_copied <= 0)
			break;
		bytes_total -= bytes_copied;
		copy->cp_res.wr_bytes_written += bytes_copied;
		src_pos += bytes_copied;
		dst_pos += bytes_copied;
	} while (bytes_total > 0 && !copy->cp_synchronous);
	return bytes_copied;
}

static __be32 nfsd4_do_copy(struct nfsd4_copy *copy, bool sync)
{
	__be32 status;
	ssize_t bytes;

	bytes = _nfsd_copy_file_range(copy);
	/* for async copy, we ignore the error, client can always retry
	 * to get the error
	 */
	if (bytes < 0 && !copy->cp_res.wr_bytes_written)
		status = nfserrno(bytes);
	else {
		nfsd4_init_copy_res(copy, sync);
		status = nfs_ok;
	}

1424 1425 1426 1427 1428 1429
	if (!copy->cp_intra) /* Inter server SSC */
		nfsd4_cleanup_inter_ssc(copy->ss_mnt, copy->nf_src,
					copy->nf_dst);
	else
		nfsd4_cleanup_intra_ssc(copy->nf_src, copy->nf_dst);

1430 1431 1432
	return status;
}

1433
static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
1434 1435 1436 1437 1438 1439 1440 1441
{
	dst->cp_src_pos = src->cp_src_pos;
	dst->cp_dst_pos = src->cp_dst_pos;
	dst->cp_count = src->cp_count;
	dst->cp_synchronous = src->cp_synchronous;
	memcpy(&dst->cp_res, &src->cp_res, sizeof(src->cp_res));
	memcpy(&dst->fh, &src->fh, sizeof(src->fh));
	dst->cp_clp = src->cp_clp;
1442
	dst->nf_dst = nfsd_file_get(src->nf_dst);
1443 1444 1445 1446
	dst->cp_intra = src->cp_intra;
	if (src->cp_intra) /* for inter, file_src doesn't exist yet */
		dst->nf_src = nfsd_file_get(src->nf_src);

1447
	memcpy(&dst->cp_stateid, &src->cp_stateid, sizeof(src->cp_stateid));
1448 1449 1450 1451
	memcpy(&dst->cp_src, &src->cp_src, sizeof(struct nl4_server));
	memcpy(&dst->stateid, &src->stateid, sizeof(src->stateid));
	memcpy(&dst->c_fh, &src->c_fh, sizeof(src->c_fh));
	dst->ss_mnt = src->ss_mnt;
1452 1453 1454 1455
}

static void cleanup_async_copy(struct nfsd4_copy *copy)
{
O
Olga Kornievskaia 已提交
1456
	nfs4_free_copy_state(copy);
1457
	nfsd_file_put(copy->nf_dst);
1458 1459
	if (copy->cp_intra)
		nfsd_file_put(copy->nf_src);
1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470
	spin_lock(&copy->cp_clp->async_lock);
	list_del(&copy->copies);
	spin_unlock(&copy->cp_clp->async_lock);
	nfs4_put_copy(copy);
}

static int nfsd4_do_async_copy(void *data)
{
	struct nfsd4_copy *copy = (struct nfsd4_copy *)data;
	struct nfsd4_copy *cb_copy;

1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
	if (!copy->cp_intra) { /* Inter server SSC */
		copy->nf_src = kzalloc(sizeof(struct nfsd_file), GFP_KERNEL);
		if (!copy->nf_src) {
			copy->nfserr = nfserr_serverfault;
			nfsd4_interssc_disconnect(copy->ss_mnt);
			goto do_callback;
		}
		copy->nf_src->nf_file = nfs42_ssc_open(copy->ss_mnt, &copy->c_fh,
					      &copy->stateid);
		if (IS_ERR(copy->nf_src->nf_file)) {
			copy->nfserr = nfserr_offload_denied;
			nfsd4_interssc_disconnect(copy->ss_mnt);
			goto do_callback;
		}
	}

1487
	copy->nfserr = nfsd4_do_copy(copy, 0);
1488
do_callback:
1489 1490 1491
	cb_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
	if (!cb_copy)
		goto out;
1492
	refcount_set(&cb_copy->refcount, 1);
1493 1494 1495 1496 1497 1498 1499 1500
	memcpy(&cb_copy->cp_res, &copy->cp_res, sizeof(copy->cp_res));
	cb_copy->cp_clp = copy->cp_clp;
	cb_copy->nfserr = copy->nfserr;
	memcpy(&cb_copy->fh, &copy->fh, sizeof(copy->fh));
	nfsd4_init_cb(&cb_copy->cp_cb, cb_copy->cp_clp,
			&nfsd4_cb_offload_ops, NFSPROC4_CLNT_CB_OFFLOAD);
	nfsd4_run_cb(&cb_copy->cp_cb);
out:
1501 1502
	if (!copy->cp_intra)
		kfree(copy->nf_src);
1503 1504 1505 1506
	cleanup_async_copy(copy);
	return 0;
}

A
Anna Schumaker 已提交
1507 1508
static __be32
nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1509
		union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1510
{
1511
	struct nfsd4_copy *copy = &u->copy;
A
Anna Schumaker 已提交
1512
	__be32 status;
1513
	struct nfsd4_copy *async_copy = NULL;
A
Anna Schumaker 已提交
1514

1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
	if (!copy->cp_intra) { /* Inter server SSC */
		if (!inter_copy_offload_enable || copy->cp_synchronous) {
			status = nfserr_notsupp;
			goto out;
		}
		status = nfsd4_setup_inter_ssc(rqstp, cstate, copy,
				&copy->ss_mnt);
		if (status)
			return nfserr_offload_denied;
	} else {
		status = nfsd4_setup_intra_ssc(rqstp, cstate, copy);
		if (status)
			return status;
	}
A
Anna Schumaker 已提交
1529

1530 1531 1532 1533 1534
	copy->cp_clp = cstate->clp;
	memcpy(&copy->fh, &cstate->current_fh.fh_handle,
		sizeof(struct knfsd_fh));
	if (!copy->cp_synchronous) {
		struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
A
Anna Schumaker 已提交
1535

1536 1537 1538
		status = nfserrno(-ENOMEM);
		async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
		if (!async_copy)
1539 1540 1541
			goto out_err;
		if (!nfs4_init_copy_state(nn, copy))
			goto out_err;
1542 1543 1544
		refcount_set(&async_copy->refcount, 1);
		memcpy(&copy->cp_res.cb_stateid, &copy->cp_stateid,
			sizeof(copy->cp_stateid));
1545
		dup_copy_fields(copy, async_copy);
1546 1547 1548 1549 1550 1551 1552 1553 1554
		async_copy->copy_task = kthread_create(nfsd4_do_async_copy,
				async_copy, "%s", "copy thread");
		if (IS_ERR(async_copy->copy_task))
			goto out_err;
		spin_lock(&async_copy->cp_clp->async_lock);
		list_add(&async_copy->copies,
				&async_copy->cp_clp->async_copies);
		spin_unlock(&async_copy->cp_clp->async_lock);
		wake_up_process(async_copy->copy_task);
A
Anna Schumaker 已提交
1555
		status = nfs_ok;
1556
	} else {
1557
		status = nfsd4_do_copy(copy, 1);
1558
	}
A
Anna Schumaker 已提交
1559 1560
out:
	return status;
1561
out_err:
1562 1563
	if (async_copy)
		cleanup_async_copy(async_copy);
1564 1565 1566
	status = nfserrno(-ENOMEM);
	if (!copy->cp_intra)
		nfsd4_interssc_disconnect(copy->ss_mnt);
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576
	goto out;
}

struct nfsd4_copy *
find_async_copy(struct nfs4_client *clp, stateid_t *stateid)
{
	struct nfsd4_copy *copy;

	spin_lock(&clp->async_lock);
	list_for_each_entry(copy, &clp->async_copies, copies) {
1577
		if (memcmp(&copy->cp_stateid.stid, stateid, NFS4_STATEID_SIZE))
1578 1579 1580 1581 1582 1583 1584
			continue;
		refcount_inc(&copy->refcount);
		spin_unlock(&clp->async_lock);
		return copy;
	}
	spin_unlock(&clp->async_lock);
	return NULL;
A
Anna Schumaker 已提交
1585 1586
}

O
Olga Kornievskaia 已提交
1587 1588 1589 1590 1591
static __be32
nfsd4_offload_cancel(struct svc_rqst *rqstp,
		     struct nfsd4_compound_state *cstate,
		     union nfsd4_op_u *u)
{
1592 1593 1594 1595 1596
	struct nfsd4_offload_status *os = &u->offload_status;
	struct nfsd4_copy *copy;
	struct nfs4_client *clp = cstate->clp;

	copy = find_async_copy(clp, &os->stateid);
1597 1598 1599 1600 1601
	if (!copy) {
		struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);

		return manage_cpntf_state(nn, &os->stateid, clp, NULL);
	} else
1602 1603
		nfsd4_stop_copy(copy);

1604
	return nfs_ok;
O
Olga Kornievskaia 已提交
1605 1606
}

O
Olga Kornievskaia 已提交
1607 1608 1609 1610
static __be32
nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
		  union nfsd4_op_u *u)
{
O
Olga Kornievskaia 已提交
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 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
	struct nfsd4_copy_notify *cn = &u->copy_notify;
	__be32 status;
	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
	struct nfs4_stid *stid;
	struct nfs4_cpntf_state *cps;
	struct nfs4_client *clp = cstate->clp;

	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
					&cn->cpn_src_stateid, RD_STATE, NULL,
					&stid);
	if (status)
		return status;

	cn->cpn_sec = nn->nfsd4_lease;
	cn->cpn_nsec = 0;

	status = nfserrno(-ENOMEM);
	cps = nfs4_alloc_init_cpntf_state(nn, stid);
	if (!cps)
		goto out;
	memcpy(&cn->cpn_cnr_stateid, &cps->cp_stateid.stid, sizeof(stateid_t));
	memcpy(&cps->cp_p_stateid, &stid->sc_stateid, sizeof(stateid_t));
	memcpy(&cps->cp_p_clid, &clp->cl_clientid, sizeof(clientid_t));

	/* For now, only return one server address in cpn_src, the
	 * address used by the client to connect to this server.
	 */
	cn->cpn_src.nl4_type = NL4_NETADDR;
	status = nfsd4_set_netaddr((struct sockaddr *)&rqstp->rq_daddr,
				 &cn->cpn_src.u.nl4_addr);
	WARN_ON_ONCE(status);
	if (status) {
		nfs4_put_cpntf_state(nn, cps);
		goto out;
	}
out:
	nfs4_put_stid(stid);
	return status;
O
Olga Kornievskaia 已提交
1649 1650
}

A
Anna Schumaker 已提交
1651 1652 1653 1654
static __be32
nfsd4_fallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
		struct nfsd4_fallocate *fallocate, int flags)
{
J
J. Bruce Fields 已提交
1655
	__be32 status;
1656
	struct nfsd_file *nf;
A
Anna Schumaker 已提交
1657

1658
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
A
Anna Schumaker 已提交
1659
					    &fallocate->falloc_stateid,
O
Olga Kornievskaia 已提交
1660
					    WR_STATE, &nf, NULL);
A
Anna Schumaker 已提交
1661 1662 1663 1664 1665
	if (status != nfs_ok) {
		dprintk("NFSD: nfsd4_fallocate: couldn't process stateid!\n");
		return status;
	}

1666
	status = nfsd4_vfs_fallocate(rqstp, &cstate->current_fh, nf->nf_file,
A
Anna Schumaker 已提交
1667 1668 1669
				     fallocate->falloc_offset,
				     fallocate->falloc_length,
				     flags);
1670
	nfsd_file_put(nf);
A
Anna Schumaker 已提交
1671 1672
	return status;
}
O
Olga Kornievskaia 已提交
1673 1674 1675 1676 1677
static __be32
nfsd4_offload_status(struct svc_rqst *rqstp,
		     struct nfsd4_compound_state *cstate,
		     union nfsd4_op_u *u)
{
1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690
	struct nfsd4_offload_status *os = &u->offload_status;
	__be32 status = 0;
	struct nfsd4_copy *copy;
	struct nfs4_client *clp = cstate->clp;

	copy = find_async_copy(clp, &os->stateid);
	if (copy) {
		os->count = copy->cp_res.wr_bytes_written;
		nfs4_put_copy(copy);
	} else
		status = nfserr_bad_stateid;

	return status;
O
Olga Kornievskaia 已提交
1691
}
A
Anna Schumaker 已提交
1692 1693 1694

static __be32
nfsd4_allocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1695
	       union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1696
{
1697
	return nfsd4_fallocate(rqstp, cstate, &u->allocate, 0);
A
Anna Schumaker 已提交
1698 1699
}

A
Anna Schumaker 已提交
1700 1701
static __be32
nfsd4_deallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1702
		 union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1703
{
1704
	return nfsd4_fallocate(rqstp, cstate, &u->deallocate,
A
Anna Schumaker 已提交
1705 1706 1707
			       FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE);
}

A
Anna Schumaker 已提交
1708 1709
static __be32
nfsd4_seek(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1710
	   union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1711
{
1712
	struct nfsd4_seek *seek = &u->seek;
A
Anna Schumaker 已提交
1713 1714
	int whence;
	__be32 status;
1715
	struct nfsd_file *nf;
A
Anna Schumaker 已提交
1716

1717
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
A
Anna Schumaker 已提交
1718
					    &seek->seek_stateid,
O
Olga Kornievskaia 已提交
1719
					    RD_STATE, &nf, NULL);
A
Anna Schumaker 已提交
1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740
	if (status) {
		dprintk("NFSD: nfsd4_seek: couldn't process stateid!\n");
		return status;
	}

	switch (seek->seek_whence) {
	case NFS4_CONTENT_DATA:
		whence = SEEK_DATA;
		break;
	case NFS4_CONTENT_HOLE:
		whence = SEEK_HOLE;
		break;
	default:
		status = nfserr_union_notsupp;
		goto out;
	}

	/*
	 * Note:  This call does change file->f_pos, but nothing in NFSD
	 *        should ever file->f_pos.
	 */
1741
	seek->seek_pos = vfs_llseek(nf->nf_file, seek->seek_offset, whence);
A
Anna Schumaker 已提交
1742 1743
	if (seek->seek_pos < 0)
		status = nfserrno(seek->seek_pos);
1744
	else if (seek->seek_pos >= i_size_read(file_inode(nf->nf_file)))
A
Anna Schumaker 已提交
1745 1746 1747
		seek->seek_eof = true;

out:
1748
	nfsd_file_put(nf);
A
Anna Schumaker 已提交
1749 1750 1751
	return status;
}

L
Linus Torvalds 已提交
1752 1753 1754 1755 1756
/* This routine never returns NFS_OK!  If there are no other errors, it
 * will return NFSERR_SAME or NFSERR_NOT_SAME depending on whether the
 * attributes matched.  VERIFY is implemented by mapping NFSERR_SAME
 * to NFS_OK after the call; NVERIFY by mapping NFSERR_NOT_SAME to NFS_OK.
 */
1757
static __be32
1758
_nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1759
	     struct nfsd4_verify *verify)
L
Linus Torvalds 已提交
1760
{
A
Al Viro 已提交
1761
	__be32 *buf, *p;
L
Linus Torvalds 已提交
1762
	int count;
1763
	__be32 status;
L
Linus Torvalds 已提交
1764

M
Miklos Szeredi 已提交
1765
	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
1766 1767 1768
	if (status)
		return status;

1769 1770 1771 1772
	status = check_attr_support(rqstp, cstate, verify->ve_bmval, NULL);
	if (status)
		return status;

L
Linus Torvalds 已提交
1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784
	if ((verify->ve_bmval[0] & FATTR4_WORD0_RDATTR_ERROR)
	    || (verify->ve_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1))
		return nfserr_inval;
	if (verify->ve_attrlen & 3)
		return nfserr_inval;

	/* count in words:
	 *   bitmap_len(1) + bitmap(2) + attr_len(1) = 4
	 */
	count = 4 + (verify->ve_attrlen >> 2);
	buf = kmalloc(count << 2, GFP_KERNEL);
	if (!buf)
1785
		return nfserr_jukebox;
L
Linus Torvalds 已提交
1786

1787
	p = buf;
1788
	status = nfsd4_encode_fattr_to_buf(&p, count, &cstate->current_fh,
1789
				    cstate->current_fh.fh_export,
1790 1791
				    cstate->current_fh.fh_dentry,
				    verify->ve_bmval,
1792
				    rqstp, 0);
J
J. Bruce Fields 已提交
1793 1794 1795 1796
	/*
	 * If nfsd4_encode_fattr() ran out of space, assume that's because
	 * the attributes are longer (hence different) than those given:
	 */
1797
	if (status == nfserr_resource)
L
Linus Torvalds 已提交
1798 1799 1800 1801
		status = nfserr_not_same;
	if (status)
		goto out_kfree;

1802 1803
	/* skip bitmap */
	p = buf + 1 + ntohl(buf[0]);
L
Linus Torvalds 已提交
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814
	status = nfserr_not_same;
	if (ntohl(*p++) != verify->ve_attrlen)
		goto out_kfree;
	if (!memcmp(p, verify->ve_attrval, verify->ve_attrlen))
		status = nfserr_same;

out_kfree:
	kfree(buf);
	return status;
}

1815 1816
static __be32
nfsd4_nverify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1817
	      union nfsd4_op_u *u)
1818 1819 1820
{
	__be32 status;

1821
	status = _nfsd4_verify(rqstp, cstate, &u->verify);
1822 1823 1824 1825 1826
	return status == nfserr_not_same ? nfs_ok : status;
}

static __be32
nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1827
	     union nfsd4_op_u *u)
1828 1829 1830
{
	__be32 status;

1831
	status = _nfsd4_verify(rqstp, cstate, &u->nverify);
1832 1833 1834
	return status == nfserr_same ? nfs_ok : status;
}

1835 1836 1837 1838
#ifdef CONFIG_NFSD_PNFS
static const struct nfsd4_layout_ops *
nfsd4_layout_verify(struct svc_export *exp, unsigned int layout_type)
{
1839
	if (!exp->ex_layout_types) {
1840 1841 1842 1843
		dprintk("%s: export does not support pNFS\n", __func__);
		return NULL;
	}

1844 1845
	if (layout_type >= LAYOUT_TYPE_MAX ||
	    !(exp->ex_layout_types & (1 << layout_type))) {
1846 1847 1848 1849 1850 1851 1852 1853 1854 1855
		dprintk("%s: layout type %d not supported\n",
			__func__, layout_type);
		return NULL;
	}

	return nfsd4_layout_ops[layout_type];
}

static __be32
nfsd4_getdeviceinfo(struct svc_rqst *rqstp,
1856
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1857
{
1858
	struct nfsd4_getdeviceinfo *gdp = &u->getdeviceinfo;
1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888
	const struct nfsd4_layout_ops *ops;
	struct nfsd4_deviceid_map *map;
	struct svc_export *exp;
	__be32 nfserr;

	dprintk("%s: layout_type %u dev_id [0x%llx:0x%x] maxcnt %u\n",
	       __func__,
	       gdp->gd_layout_type,
	       gdp->gd_devid.fsid_idx, gdp->gd_devid.generation,
	       gdp->gd_maxcount);

	map = nfsd4_find_devid_map(gdp->gd_devid.fsid_idx);
	if (!map) {
		dprintk("%s: couldn't find device ID to export mapping!\n",
			__func__);
		return nfserr_noent;
	}

	exp = rqst_exp_find(rqstp, map->fsid_type, map->fsid);
	if (IS_ERR(exp)) {
		dprintk("%s: could not find device id\n", __func__);
		return nfserr_noent;
	}

	nfserr = nfserr_layoutunavailable;
	ops = nfsd4_layout_verify(exp, gdp->gd_layout_type);
	if (!ops)
		goto out;

	nfserr = nfs_ok;
1889 1890
	if (gdp->gd_maxcount != 0) {
		nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb,
1891
				rqstp, cstate->session->se_client, gdp);
1892
	}
1893 1894 1895

	gdp->gd_notify_types &= ops->notify_types;
out:
1896
	exp_put(exp);
1897 1898 1899
	return nfserr;
}

1900 1901 1902 1903 1904 1905
static void
nfsd4_getdeviceinfo_release(union nfsd4_op_u *u)
{
	kfree(u->getdeviceinfo.gd_device);
}

1906 1907
static __be32
nfsd4_layoutget(struct svc_rqst *rqstp,
1908
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1909
{
1910
	struct nfsd4_layoutget *lgp = &u->layoutget;
1911 1912 1913 1914
	struct svc_fh *current_fh = &cstate->current_fh;
	const struct nfsd4_layout_ops *ops;
	struct nfs4_layout_stateid *ls;
	__be32 nfserr;
1915
	int accmode = NFSD_MAY_READ_IF_EXEC;
1916 1917 1918

	switch (lgp->lg_seg.iomode) {
	case IOMODE_READ:
1919
		accmode |= NFSD_MAY_READ;
1920 1921
		break;
	case IOMODE_RW:
1922
		accmode |= NFSD_MAY_READ | NFSD_MAY_WRITE;
1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
		break;
	default:
		dprintk("%s: invalid iomode %d\n",
			__func__, lgp->lg_seg.iomode);
		nfserr = nfserr_badiomode;
		goto out;
	}

	nfserr = fh_verify(rqstp, current_fh, 0, accmode);
	if (nfserr)
		goto out;

	nfserr = nfserr_layoutunavailable;
	ops = nfsd4_layout_verify(current_fh->fh_export, lgp->lg_layout_type);
	if (!ops)
		goto out;

	/*
	 * Verify minlength and range as per RFC5661:
	 *  o  If loga_length is less than loga_minlength,
	 *     the metadata server MUST return NFS4ERR_INVAL.
	 *  o  If the sum of loga_offset and loga_minlength exceeds
	 *     NFS4_UINT64_MAX, and loga_minlength is not
	 *     NFS4_UINT64_MAX, the error NFS4ERR_INVAL MUST result.
	 *  o  If the sum of loga_offset and loga_length exceeds
	 *     NFS4_UINT64_MAX, and loga_length is not NFS4_UINT64_MAX,
	 *     the error NFS4ERR_INVAL MUST result.
	 */
	nfserr = nfserr_inval;
	if (lgp->lg_seg.length < lgp->lg_minlength ||
	    (lgp->lg_minlength != NFS4_MAX_UINT64 &&
	     lgp->lg_minlength > NFS4_MAX_UINT64 - lgp->lg_seg.offset) ||
	    (lgp->lg_seg.length != NFS4_MAX_UINT64 &&
	     lgp->lg_seg.length > NFS4_MAX_UINT64 - lgp->lg_seg.offset))
		goto out;
	if (lgp->lg_seg.length == 0)
		goto out;

	nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lgp->lg_sid,
						true, lgp->lg_layout_type, &ls);
C
Christoph Hellwig 已提交
1963
	if (nfserr) {
1964
		trace_nfsd_layout_get_lookup_fail(&lgp->lg_sid);
1965
		goto out;
C
Christoph Hellwig 已提交
1966
	}
1967

1968 1969 1970 1971
	nfserr = nfserr_recallconflict;
	if (atomic_read(&ls->ls_stid.sc_file->fi_lo_recalls))
		goto out_put_stid;

1972
	nfserr = ops->proc_layoutget(d_inode(current_fh->fh_dentry),
1973 1974 1975 1976 1977 1978 1979
				     current_fh, lgp);
	if (nfserr)
		goto out_put_stid;

	nfserr = nfsd4_insert_layout(lgp, ls);

out_put_stid:
1980
	mutex_unlock(&ls->ls_mutex);
1981 1982 1983 1984 1985
	nfs4_put_stid(&ls->ls_stid);
out:
	return nfserr;
}

1986 1987 1988 1989 1990 1991
static void
nfsd4_layoutget_release(union nfsd4_op_u *u)
{
	kfree(u->layoutget.lg_content);
}

1992 1993
static __be32
nfsd4_layoutcommit(struct svc_rqst *rqstp,
1994
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1995
{
1996
	struct nfsd4_layoutcommit *lcp = &u->layoutcommit;
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
	const struct nfsd4_layout_seg *seg = &lcp->lc_seg;
	struct svc_fh *current_fh = &cstate->current_fh;
	const struct nfsd4_layout_ops *ops;
	loff_t new_size = lcp->lc_last_wr + 1;
	struct inode *inode;
	struct nfs4_layout_stateid *ls;
	__be32 nfserr;

	nfserr = fh_verify(rqstp, current_fh, 0, NFSD_MAY_WRITE);
	if (nfserr)
		goto out;

	nfserr = nfserr_layoutunavailable;
	ops = nfsd4_layout_verify(current_fh->fh_export, lcp->lc_layout_type);
	if (!ops)
		goto out;
2013
	inode = d_inode(current_fh->fh_dentry);
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032

	nfserr = nfserr_inval;
	if (new_size <= seg->offset) {
		dprintk("pnfsd: last write before layout segment\n");
		goto out;
	}
	if (new_size > seg->offset + seg->length) {
		dprintk("pnfsd: last write beyond layout segment\n");
		goto out;
	}
	if (!lcp->lc_newoffset && new_size > i_size_read(inode)) {
		dprintk("pnfsd: layoutcommit beyond EOF\n");
		goto out;
	}

	nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lcp->lc_sid,
						false, lcp->lc_layout_type,
						&ls);
	if (nfserr) {
2033
		trace_nfsd_layout_commit_lookup_fail(&lcp->lc_sid);
2034 2035 2036 2037 2038 2039
		/* fixup error code as per RFC5661 */
		if (nfserr == nfserr_bad_stateid)
			nfserr = nfserr_badlayout;
		goto out;
	}

2040 2041 2042
	/* LAYOUTCOMMIT does not require any serialization */
	mutex_unlock(&ls->ls_mutex);

2043 2044 2045 2046 2047 2048 2049
	if (new_size > i_size_read(inode)) {
		lcp->lc_size_chg = 1;
		lcp->lc_newsize = new_size;
	} else {
		lcp->lc_size_chg = 0;
	}

2050
	nfserr = ops->proc_layoutcommit(inode, lcp);
2051 2052 2053 2054 2055 2056 2057
	nfs4_put_stid(&ls->ls_stid);
out:
	return nfserr;
}

static __be32
nfsd4_layoutreturn(struct svc_rqst *rqstp,
2058
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
2059
{
2060
	struct nfsd4_layoutreturn *lrp = &u->layoutreturn;
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 2095 2096 2097 2098 2099 2100 2101 2102
	struct svc_fh *current_fh = &cstate->current_fh;
	__be32 nfserr;

	nfserr = fh_verify(rqstp, current_fh, 0, NFSD_MAY_NOP);
	if (nfserr)
		goto out;

	nfserr = nfserr_layoutunavailable;
	if (!nfsd4_layout_verify(current_fh->fh_export, lrp->lr_layout_type))
		goto out;

	switch (lrp->lr_seg.iomode) {
	case IOMODE_READ:
	case IOMODE_RW:
	case IOMODE_ANY:
		break;
	default:
		dprintk("%s: invalid iomode %d\n", __func__,
			lrp->lr_seg.iomode);
		nfserr = nfserr_inval;
		goto out;
	}

	switch (lrp->lr_return_type) {
	case RETURN_FILE:
		nfserr = nfsd4_return_file_layouts(rqstp, cstate, lrp);
		break;
	case RETURN_FSID:
	case RETURN_ALL:
		nfserr = nfsd4_return_client_layouts(rqstp, cstate, lrp);
		break;
	default:
		dprintk("%s: invalid return_type %d\n", __func__,
			lrp->lr_return_type);
		nfserr = nfserr_inval;
		break;
	}
out:
	return nfserr;
}
#endif /* CONFIG_NFSD_PNFS */

2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 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
static __be32
nfsd4_getxattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
	       union nfsd4_op_u *u)
{
	struct nfsd4_getxattr *getxattr = &u->getxattr;

	return nfsd_getxattr(rqstp, &cstate->current_fh,
			     getxattr->getxa_name, &getxattr->getxa_buf,
			     &getxattr->getxa_len);
}

static __be32
nfsd4_setxattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
	   union nfsd4_op_u *u)
{
	struct nfsd4_setxattr *setxattr = &u->setxattr;
	__be32 ret;

	if (opens_in_grace(SVC_NET(rqstp)))
		return nfserr_grace;

	ret = nfsd_setxattr(rqstp, &cstate->current_fh, setxattr->setxa_name,
			    setxattr->setxa_buf, setxattr->setxa_len,
			    setxattr->setxa_flags);

	if (!ret)
		set_change_info(&setxattr->setxa_cinfo, &cstate->current_fh);

	return ret;
}

static __be32
nfsd4_listxattrs(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
	   union nfsd4_op_u *u)
{
	/*
	 * Get the entire list, then copy out only the user attributes
	 * in the encode function.
	 */
	return nfsd_listxattr(rqstp, &cstate->current_fh,
			     &u->listxattrs.lsxa_buf, &u->listxattrs.lsxa_len);
}

static __be32
nfsd4_removexattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
	   union nfsd4_op_u *u)
{
	struct nfsd4_removexattr *removexattr = &u->removexattr;
	__be32 ret;

	if (opens_in_grace(SVC_NET(rqstp)))
		return nfserr_grace;

	ret = nfsd_removexattr(rqstp, &cstate->current_fh,
	    removexattr->rmxa_name);

	if (!ret)
		set_change_info(&removexattr->rmxa_cinfo, &cstate->current_fh);

	return ret;
}

L
Linus Torvalds 已提交
2165 2166 2167
/*
 * NULL call.
 */
A
Al Viro 已提交
2168
static __be32
2169
nfsd4_proc_null(struct svc_rqst *rqstp)
L
Linus Torvalds 已提交
2170
{
2171
	return rpc_success;
L
Linus Torvalds 已提交
2172 2173
}

2174 2175 2176
static inline void nfsd4_increment_op_stats(u32 opnum)
{
	if (opnum >= FIRST_NFS4_OP && opnum <= LAST_NFS4_OP)
2177
		percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_NFS4_OP(opnum)]);
2178 2179
}

C
Christoph Hellwig 已提交
2180
static const struct nfsd4_operation nfsd4_ops[];
2181

A
Adrian Bunk 已提交
2182
static const char *nfsd4_op_name(unsigned opnum);
B
Benny Halevy 已提交
2183

2184
/*
2185
 * Enforce NFSv4.1 COMPOUND ordering rules:
2186
 *
2187 2188 2189
 * Also note, enforced elsewhere:
 *	- SEQUENCE other than as first op results in
 *	  NFS4ERR_SEQUENCE_POS. (Enforced in nfsd4_sequence().)
2190 2191
 *	- BIND_CONN_TO_SESSION must be the only op in its compound.
 *	  (Enforced in nfsd4_bind_conn_to_session().)
2192 2193 2194
 *	- DESTROY_SESSION must be the final operation in a compound, if
 *	  sessionid's in SEQUENCE and DESTROY_SESSION are the same.
 *	  (Enforced in nfsd4_destroy_session().)
2195
 */
2196
static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args)
2197
{
2198
	struct nfsd4_op *first_op = &args->ops[0];
2199 2200 2201 2202 2203 2204 2205

	/* These ordering requirements don't apply to NFSv4.0: */
	if (args->minorversion == 0)
		return nfs_ok;
	/* This is weird, but OK, not our problem: */
	if (args->opcnt == 0)
		return nfs_ok;
2206
	if (first_op->status == nfserr_op_illegal)
2207
		return nfs_ok;
2208
	if (!(nfsd4_ops[first_op->opnum].op_flags & ALLOWED_AS_FIRST_OP))
2209
		return nfserr_op_not_in_session;
2210
	if (first_op->opnum == OP_SEQUENCE)
2211
		return nfs_ok;
2212 2213 2214 2215 2216
	/*
	 * So first_op is something allowed outside a session, like
	 * EXCHANGE_ID; but then it has to be the only op in the
	 * compound:
	 */
2217 2218 2219
	if (args->opcnt != 1)
		return nfserr_not_only_op;
	return nfs_ok;
2220 2221
}

2222
const struct nfsd4_operation *OPDESC(struct nfsd4_op *op)
J
J. Bruce Fields 已提交
2223 2224 2225 2226
{
	return &nfsd4_ops[op->opnum];
}

2227 2228
bool nfsd4_cache_this_op(struct nfsd4_op *op)
{
2229 2230
	if (op->opnum == OP_ILLEGAL)
		return false;
2231
	return OPDESC(op)->op_flags & OP_CACHEME;
2232 2233
}

2234 2235 2236 2237 2238 2239
static bool need_wrongsec_check(struct svc_rqst *rqstp)
{
	struct nfsd4_compoundres *resp = rqstp->rq_resp;
	struct nfsd4_compoundargs *argp = rqstp->rq_argp;
	struct nfsd4_op *this = &argp->ops[resp->opcnt - 1];
	struct nfsd4_op *next = &argp->ops[resp->opcnt];
C
Christoph Hellwig 已提交
2240 2241
	const struct nfsd4_operation *thisd = OPDESC(this);
	const struct nfsd4_operation *nextd;
2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255

	/*
	 * Most ops check wronsec on our own; only the putfh-like ops
	 * have special rules.
	 */
	if (!(thisd->op_flags & OP_IS_PUTFH_LIKE))
		return false;
	/*
	 * rfc 5661 2.6.3.1.1.6: don't bother erroring out a
	 * put-filehandle operation if we're not going to use the
	 * result:
	 */
	if (argp->opcnt == resp->opcnt)
		return false;
2256 2257
	if (next->opnum == OP_ILLEGAL)
		return false;
2258 2259 2260 2261 2262 2263 2264 2265 2266
	nextd = OPDESC(next);
	/*
	 * Rest of 2.6.3.1.1: certain operations will return WRONGSEC
	 * errors themselves as necessary; others should check for them
	 * now:
	 */
	return !(nextd->op_flags & OP_HANDLES_WRONGSEC);
}

2267 2268 2269 2270 2271 2272 2273 2274
static void svcxdr_init_encode(struct svc_rqst *rqstp,
			       struct nfsd4_compoundres *resp)
{
	struct xdr_stream *xdr = &resp->xdr;
	struct xdr_buf *buf = &rqstp->rq_res;
	struct kvec *head = buf->head;

	xdr->buf = buf;
2275
	xdr->iov = head;
2276
	xdr->p   = head->iov_base + head->iov_len;
2277
	xdr->end = head->iov_base + PAGE_SIZE - rqstp->rq_auth_slack;
2278 2279
	/* Tail and page_len should be zero at this point: */
	buf->len = buf->head[0].iov_len;
2280
	xdr_reset_scratch_buffer(xdr);
2281
	xdr->page_ptr = buf->pages - 1;
2282
	buf->buflen = PAGE_SIZE * (1 + rqstp->rq_page_end - buf->pages)
2283
		- rqstp->rq_auth_slack;
2284 2285
}

2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324
#ifdef CONFIG_NFSD_V4_2_INTER_SSC
static void
check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
{
	struct nfsd4_op	*op, *current_op = NULL, *saved_op = NULL;
	struct nfsd4_copy *copy;
	struct nfsd4_putfh *putfh;
	int i;

	/* traverse all operation and if it's a COPY compound, mark the
	 * source filehandle to skip verification
	 */
	for (i = 0; i < args->opcnt; i++) {
		op = &args->ops[i];
		if (op->opnum == OP_PUTFH)
			current_op = op;
		else if (op->opnum == OP_SAVEFH)
			saved_op = current_op;
		else if (op->opnum == OP_RESTOREFH)
			current_op = saved_op;
		else if (op->opnum == OP_COPY) {
			copy = (struct nfsd4_copy *)&op->u;
			if (!saved_op) {
				op->status = nfserr_nofilehandle;
				return;
			}
			putfh = (struct nfsd4_putfh *)&saved_op->u;
			if (!copy->cp_intra)
				putfh->no_verify = true;
		}
	}
}
#else
static void
check_if_stalefh_allowed(struct nfsd4_compoundargs *args)
{
}
#endif

L
Linus Torvalds 已提交
2325 2326 2327
/*
 * COMPOUND call.
 */
A
Al Viro 已提交
2328
static __be32
2329
nfsd4_proc_compound(struct svc_rqst *rqstp)
L
Linus Torvalds 已提交
2330
{
2331 2332
	struct nfsd4_compoundargs *args = rqstp->rq_argp;
	struct nfsd4_compoundres *resp = rqstp->rq_resp;
L
Linus Torvalds 已提交
2333
	struct nfsd4_op	*op;
2334
	struct nfsd4_compound_state *cstate = &resp->cstate;
2335 2336
	struct svc_fh *current_fh = &cstate->current_fh;
	struct svc_fh *save_fh = &cstate->save_fh;
2337
	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
2338
	__be32		status;
L
Linus Torvalds 已提交
2339

2340
	svcxdr_init_encode(rqstp, resp);
2341
	resp->tagp = resp->xdr.p;
L
Linus Torvalds 已提交
2342
	/* reserve space for: taglen, tag, and opcnt */
2343
	xdr_reserve_space(&resp->xdr, 8 + args->taglen);
L
Linus Torvalds 已提交
2344 2345 2346
	resp->taglen = args->taglen;
	resp->tag = args->tag;
	resp->rqstp = rqstp;
2347 2348 2349
	cstate->minorversion = args->minorversion;
	fh_init(current_fh, NFS4_FHSIZE);
	fh_init(save_fh, NFS4_FHSIZE);
N
NeilBrown 已提交
2350 2351 2352 2353
	/*
	 * Don't use the deferral mechanism for NFSv4; compounds make it
	 * too hard to avoid non-idempotency problems.
	 */
2354
	clear_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
L
Linus Torvalds 已提交
2355 2356 2357 2358 2359

	/*
	 * According to RFC3010, this takes precedence over all other errors.
	 */
	status = nfserr_minor_vers_mismatch;
2360
	if (nfsd_minorversion(nn, args->minorversion, NFSD_TEST) <= 0)
L
Linus Torvalds 已提交
2361
		goto out;
2362 2363 2364
	status = nfserr_resource;
	if (args->opcnt > NFSD_MAX_OPS_PER_COMPOUND)
		goto out;
L
Linus Torvalds 已提交
2365

2366 2367
	status = nfs41_check_op_ordering(args);
	if (status) {
2368
		op = &args->ops[0];
2369
		op->status = status;
2370
		resp->opcnt = 1;
2371 2372
		goto encode_op;
	}
2373
	check_if_stalefh_allowed(args);
2374

2375 2376
	rqstp->rq_lease_breaker = (void **)&cstate->clp;

2377
	trace_nfsd_compound(rqstp, args->opcnt);
L
Linus Torvalds 已提交
2378 2379 2380 2381 2382 2383 2384 2385
	while (!status && resp->opcnt < args->opcnt) {
		op = &args->ops[resp->opcnt++];

		/*
		 * The XDR decode routines may have pre-set op->status;
		 * for example, if there is a miscellaneous XDR error
		 * it will be set to nfserr_bad_xdr.
		 */
2386 2387 2388
		if (op->status) {
			if (op->opnum == OP_OPEN)
				op->status = nfsd4_open_omfg(rqstp, cstate, op);
L
Linus Torvalds 已提交
2389
			goto encode_op;
2390
		}
2391 2392
		if (!current_fh->fh_dentry &&
				!HAS_FH_FLAG(current_fh, NFSD4_FH_FOREIGN)) {
2393
			if (!(op->opdesc->op_flags & ALLOWED_WITHOUT_FH)) {
2394 2395 2396
				op->status = nfserr_nofilehandle;
				goto encode_op;
			}
2397 2398
		} else if (current_fh->fh_export &&
			   current_fh->fh_export->ex_fslocs.migrated &&
2399
			  !(op->opdesc->op_flags & ALLOWED_ON_ABSENT_FS)) {
2400
			op->status = nfserr_moved;
L
Linus Torvalds 已提交
2401 2402
			goto encode_op;
		}
2403

2404 2405
		fh_clear_wcc(current_fh);

2406
		/* If op is non-idempotent */
2407
		if (op->opdesc->op_flags & OP_MODIFIES_SOMETHING) {
2408
			/*
J
J. Bruce Fields 已提交
2409 2410 2411
			 * Don't execute this op if we couldn't encode a
			 * succesful reply:
			 */
2412
			u32 plen = op->opdesc->op_rsize_bop(rqstp, op);
J
J. Bruce Fields 已提交
2413 2414
			/*
			 * Plus if there's another operation, make sure
2415 2416 2417 2418
			 * we'll have space to at least encode an error:
			 */
			if (resp->opcnt < args->opcnt)
				plen += COMPOUND_ERR_SLACK_SPACE;
2419 2420 2421 2422 2423 2424
			op->status = nfsd4_check_resp_size(resp, plen);
		}

		if (op->status)
			goto encode_op;

2425 2426 2427
		if (op->opdesc->op_get_currentstateid)
			op->opdesc->op_get_currentstateid(cstate, &op->u);
		op->status = op->opdesc->op_func(rqstp, cstate, &op->u);
L
Linus Torvalds 已提交
2428

2429 2430 2431 2432 2433 2434
		/* Only from SEQUENCE */
		if (cstate->status == nfserr_replay_cache) {
			dprintk("%s NFS4.1 replay from cache\n", __func__);
			status = op->status;
			goto out;
		}
2435
		if (!op->status) {
2436 2437
			if (op->opdesc->op_set_currentstateid)
				op->opdesc->op_set_currentstateid(cstate, &op->u);
2438

2439
			if (op->opdesc->op_flags & OP_CLEAR_STATEID)
2440
				clear_current_stateid(cstate);
2441

2442 2443
			if (current_fh->fh_export &&
					need_wrongsec_check(rqstp))
2444
				op->status = check_nfsd_access(current_fh->fh_export, rqstp);
2445
		}
L
Linus Torvalds 已提交
2446
encode_op:
A
Al Viro 已提交
2447
		if (op->status == nfserr_replay_me) {
2448
			op->replay = &cstate->replay_owner->so_replay;
2449
			nfsd4_encode_replay(&resp->xdr, op);
L
Linus Torvalds 已提交
2450 2451 2452 2453 2454
			status = op->status = op->replay->rp_status;
		} else {
			nfsd4_encode_operation(resp, op);
			status = op->status;
		}
2455

2456 2457
		trace_nfsd_compound_status(args->opcnt, resp->opcnt, status,
					   nfsd4_op_name(op->opnum));
2458

2459
		nfsd4_cstate_clear_replay(cstate);
2460
		nfsd4_increment_op_stats(op->opnum);
L
Linus Torvalds 已提交
2461 2462
	}

2463 2464 2465
	fh_put(current_fh);
	fh_put(save_fh);
	BUG_ON(cstate->replay_owner);
L
Linus Torvalds 已提交
2466
out:
2467
	cstate->status = status;
2468
	/* Reset deferral mechanism for RPC deferrals */
2469
	set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
2470
	return rpc_success;
L
Linus Torvalds 已提交
2471 2472
}

2473 2474 2475 2476 2477 2478
#define op_encode_hdr_size		(2)
#define op_encode_stateid_maxsz		(XDR_QUADLEN(NFS4_STATEID_SIZE))
#define op_encode_verifier_maxsz	(XDR_QUADLEN(NFS4_VERIFIER_SIZE))
#define op_encode_change_info_maxsz	(5)
#define nfs4_fattr_bitmap_maxsz		(4)

2479 2480
/* We'll fall back on returning no lockowner if run out of space: */
#define op_encode_lockowner_maxsz	(0)
2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500
#define op_encode_lock_denied_maxsz	(8 + op_encode_lockowner_maxsz)

#define nfs4_owner_maxsz		(1 + XDR_QUADLEN(IDMAP_NAMESZ))

#define op_encode_ace_maxsz		(3 + nfs4_owner_maxsz)
#define op_encode_delegation_maxsz	(1 + op_encode_stateid_maxsz + 1 + \
					 op_encode_ace_maxsz)

#define op_encode_channel_attrs_maxsz	(6 + 1 + 1)

static inline u32 nfsd4_only_status_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size) * sizeof(__be32);
}

static inline u32 nfsd4_status_stateid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_stateid_maxsz)* sizeof(__be32);
}

2501 2502 2503 2504 2505 2506
static inline u32 nfsd4_access_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	/* ac_supported, ac_resp_access */
	return (op_encode_hdr_size + 2)* sizeof(__be32);
}

2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517
static inline u32 nfsd4_commit_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_verifier_maxsz) * sizeof(__be32);
}

static inline u32 nfsd4_create_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz
		+ nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
}

2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547
/*
 * Note since this is an idempotent operation we won't insist on failing
 * the op prematurely if the estimate is too large.  We may turn off splice
 * reads unnecessarily.
 */
static inline u32 nfsd4_getattr_rsize(struct svc_rqst *rqstp,
				      struct nfsd4_op *op)
{
	u32 *bmap = op->u.getattr.ga_bmval;
	u32 bmap0 = bmap[0], bmap1 = bmap[1], bmap2 = bmap[2];
	u32 ret = 0;

	if (bmap0 & FATTR4_WORD0_ACL)
		return svc_max_payload(rqstp);
	if (bmap0 & FATTR4_WORD0_FS_LOCATIONS)
		return svc_max_payload(rqstp);

	if (bmap1 & FATTR4_WORD1_OWNER) {
		ret += IDMAP_NAMESZ + 4;
		bmap1 &= ~FATTR4_WORD1_OWNER;
	}
	if (bmap1 & FATTR4_WORD1_OWNER_GROUP) {
		ret += IDMAP_NAMESZ + 4;
		bmap1 &= ~FATTR4_WORD1_OWNER_GROUP;
	}
	if (bmap0 & FATTR4_WORD0_FILEHANDLE) {
		ret += NFS4_FHSIZE + 4;
		bmap0 &= ~FATTR4_WORD0_FILEHANDLE;
	}
	if (bmap2 & FATTR4_WORD2_SECURITY_LABEL) {
2548
		ret += NFS4_MAXLABELLEN + 12;
2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560
		bmap2 &= ~FATTR4_WORD2_SECURITY_LABEL;
	}
	/*
	 * Largest of remaining attributes are 16 bytes (e.g.,
	 * supported_attributes)
	 */
	ret += 16 * (hweight32(bmap0) + hweight32(bmap1) + hweight32(bmap2));
	/* bitmask, length */
	ret += 20;
	return ret;
}

2561 2562 2563 2564 2565
static inline u32 nfsd4_getfh_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 1) * sizeof(__be32) + NFS4_FHSIZE;
}

2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590
static inline u32 nfsd4_link_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz)
		* sizeof(__be32);
}

static inline u32 nfsd4_lock_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_lock_denied_maxsz)
		* sizeof(__be32);
}

static inline u32 nfsd4_open_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_stateid_maxsz
		+ op_encode_change_info_maxsz + 1
		+ nfs4_fattr_bitmap_maxsz
		+ op_encode_delegation_maxsz) * sizeof(__be32);
}

static inline u32 nfsd4_read_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	u32 maxcount = 0, rlen = 0;

	maxcount = svc_max_payload(rqstp);
2591
	rlen = min(op->u.read.rd_length, maxcount);
2592

2593
	return (op_encode_hdr_size + 2 + XDR_QUADLEN(rlen)) * sizeof(__be32);
2594 2595
}

A
Anna Schumaker 已提交
2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609
static inline u32 nfsd4_read_plus_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	u32 maxcount = svc_max_payload(rqstp);
	u32 rlen = min(op->u.read.rd_length, maxcount);
	/*
	 * If we detect that the file changed during hole encoding, then we
	 * recover by encoding the remaining reply as data. This means we need
	 * to set aside enough room to encode two data segments.
	 */
	u32 seg_len = 2 * (1 + 2 + 1);

	return (op_encode_hdr_size + 2 + seg_len + XDR_QUADLEN(rlen)) * sizeof(__be32);
}

2610 2611
static inline u32 nfsd4_readdir_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2612
	u32 maxcount = 0, rlen = 0;
2613

2614 2615
	maxcount = svc_max_payload(rqstp);
	rlen = min(op->u.readdir.rd_maxcount, maxcount);
2616

J
J. Bruce Fields 已提交
2617 2618
	return (op_encode_hdr_size + op_encode_verifier_maxsz +
		XDR_QUADLEN(rlen)) * sizeof(__be32);
2619 2620
}

2621 2622 2623 2624 2625
static inline u32 nfsd4_readlink_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 1) * sizeof(__be32) + PAGE_SIZE;
}

2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637
static inline u32 nfsd4_remove_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz)
		* sizeof(__be32);
}

static inline u32 nfsd4_rename_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz
		+ op_encode_change_info_maxsz) * sizeof(__be32);
}

2638 2639 2640
static inline u32 nfsd4_sequence_rsize(struct svc_rqst *rqstp,
				       struct nfsd4_op *op)
{
2641 2642
	return (op_encode_hdr_size
		+ XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5) * sizeof(__be32);
2643 2644
}

2645 2646 2647 2648 2649 2650
static inline u32 nfsd4_test_stateid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 1 + op->u.test_stateid.ts_num_ids)
		* sizeof(__be32);
}

2651 2652 2653 2654 2655
static inline u32 nfsd4_setattr_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + nfs4_fattr_bitmap_maxsz) * sizeof(__be32);
}

2656 2657 2658 2659 2660 2661
static inline u32 nfsd4_secinfo_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + RPC_AUTH_MAXFLAVOR *
		(4 + XDR_QUADLEN(GSS_OID_MAX_LEN))) * sizeof(__be32);
}

2662 2663
static inline u32 nfsd4_setclientid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2664 2665
	return (op_encode_hdr_size + 2 + XDR_QUADLEN(NFS4_VERIFIER_SIZE)) *
								sizeof(__be32);
2666 2667 2668 2669
}

static inline u32 nfsd4_write_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2670
	return (op_encode_hdr_size + 2 + op_encode_verifier_maxsz) * sizeof(__be32);
2671 2672 2673 2674 2675
}

static inline u32 nfsd4_exchange_id_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 2 + 1 + /* eir_clientid, eir_sequenceid */\
2676 2677
		1 + 1 + /* eir_flags, spr_how */\
		4 + /* spo_must_enforce & _allow with bitmap */\
2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702
		2 + /*eir_server_owner.so_minor_id */\
		/* eir_server_owner.so_major_id<> */\
		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
		/* eir_server_scope<> */\
		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 +\
		1 + /* eir_server_impl_id array length */\
		0 /* ignored eir_server_impl_id contents */) * sizeof(__be32);
}

static inline u32 nfsd4_bind_conn_to_session_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + \
		XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* bctsr_sessid */\
		2 /* bctsr_dir, use_conn_in_rdma_mode */) * sizeof(__be32);
}

static inline u32 nfsd4_create_session_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + \
		XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + /* sessionid */\
		2 + /* csr_sequence, csr_flags */\
		op_encode_channel_attrs_maxsz + \
		op_encode_channel_attrs_maxsz) * sizeof(__be32);
}

A
Anna Schumaker 已提交
2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714
static inline u32 nfsd4_copy_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		1 /* wr_callback */ +
		op_encode_stateid_maxsz /* wr_callback */ +
		2 /* wr_count */ +
		1 /* wr_committed */ +
		op_encode_verifier_maxsz +
		1 /* cr_consecutive */ +
		1 /* cr_synchronous */) * sizeof(__be32);
}

O
Olga Kornievskaia 已提交
2715 2716 2717 2718 2719 2720 2721 2722
static inline u32 nfsd4_offload_status_rsize(struct svc_rqst *rqstp,
					     struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		2 /* osr_count */ +
		1 /* osr_complete<1> optional 0 for now */) * sizeof(__be32);
}

O
Olga Kornievskaia 已提交
2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737
static inline u32 nfsd4_copy_notify_rsize(struct svc_rqst *rqstp,
					struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		3 /* cnr_lease_time */ +
		1 /* We support one cnr_source_server */ +
		1 /* cnr_stateid seq */ +
		op_encode_stateid_maxsz /* cnr_stateid */ +
		1 /* num cnr_source_server*/ +
		1 /* nl4_type */ +
		1 /* nl4 size */ +
		XDR_QUADLEN(NFS4_OPAQUE_LIMIT) /*nl4_loc + nl4_loc_sz */)
		* sizeof(__be32);
}

2738
#ifdef CONFIG_NFSD_PNFS
2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751
static inline u32 nfsd4_getdeviceinfo_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	u32 maxcount = 0, rlen = 0;

	maxcount = svc_max_payload(rqstp);
	rlen = min(op->u.getdeviceinfo.gd_maxcount, maxcount);

	return (op_encode_hdr_size +
		1 /* gd_layout_type*/ +
		XDR_QUADLEN(rlen) +
		2 /* gd_notify_types */) * sizeof(__be32);
}

2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780
/*
 * At this stage we don't really know what layout driver will handle the request,
 * so we need to define an arbitrary upper bound here.
 */
#define MAX_LAYOUT_SIZE		128
static inline u32 nfsd4_layoutget_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		1 /* logr_return_on_close */ +
		op_encode_stateid_maxsz +
		1 /* nr of layouts */ +
		MAX_LAYOUT_SIZE) * sizeof(__be32);
}

static inline u32 nfsd4_layoutcommit_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		1 /* locr_newsize */ +
		2 /* ns_size */) * sizeof(__be32);
}

static inline u32 nfsd4_layoutreturn_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size +
		1 /* lrs_stateid */ +
		op_encode_stateid_maxsz) * sizeof(__be32);
}
#endif /* CONFIG_NFSD_PNFS */

2781 2782 2783 2784 2785 2786

static inline u32 nfsd4_seek_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 3) * sizeof(__be32);
}

2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822
static inline u32 nfsd4_getxattr_rsize(struct svc_rqst *rqstp,
				       struct nfsd4_op *op)
{
	u32 maxcount, rlen;

	maxcount = svc_max_payload(rqstp);
	rlen = min_t(u32, XATTR_SIZE_MAX, maxcount);

	return (op_encode_hdr_size + 1 + XDR_QUADLEN(rlen)) * sizeof(__be32);
}

static inline u32 nfsd4_setxattr_rsize(struct svc_rqst *rqstp,
				       struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz)
		* sizeof(__be32);
}
static inline u32 nfsd4_listxattrs_rsize(struct svc_rqst *rqstp,
					 struct nfsd4_op *op)
{
	u32 maxcount, rlen;

	maxcount = svc_max_payload(rqstp);
	rlen = min(op->u.listxattrs.lsxa_maxcount, maxcount);

	return (op_encode_hdr_size + 4 + XDR_QUADLEN(rlen)) * sizeof(__be32);
}

static inline u32 nfsd4_removexattr_rsize(struct svc_rqst *rqstp,
					  struct nfsd4_op *op)
{
	return (op_encode_hdr_size + op_encode_change_info_maxsz)
		* sizeof(__be32);
}


C
Christoph Hellwig 已提交
2823
static const struct nfsd4_operation nfsd4_ops[] = {
2824
	[OP_ACCESS] = {
2825
		.op_func = nfsd4_access,
B
Benny Halevy 已提交
2826
		.op_name = "OP_ACCESS",
C
Christoph Hellwig 已提交
2827
		.op_rsize_bop = nfsd4_access_rsize,
2828 2829
	},
	[OP_CLOSE] = {
2830
		.op_func = nfsd4_close,
2831
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2832
		.op_name = "OP_CLOSE",
C
Christoph Hellwig 已提交
2833
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2834
		.op_get_currentstateid = nfsd4_get_closestateid,
2835
		.op_set_currentstateid = nfsd4_set_closestateid,
2836 2837
	},
	[OP_COMMIT] = {
2838
		.op_func = nfsd4_commit,
2839
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2840
		.op_name = "OP_COMMIT",
C
Christoph Hellwig 已提交
2841
		.op_rsize_bop = nfsd4_commit_rsize,
2842 2843
	},
	[OP_CREATE] = {
2844
		.op_func = nfsd4_create,
2845
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2846
		.op_name = "OP_CREATE",
C
Christoph Hellwig 已提交
2847
		.op_rsize_bop = nfsd4_create_rsize,
2848 2849
	},
	[OP_DELEGRETURN] = {
2850
		.op_func = nfsd4_delegreturn,
2851
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2852
		.op_name = "OP_DELEGRETURN",
2853
		.op_rsize_bop = nfsd4_only_status_rsize,
2854
		.op_get_currentstateid = nfsd4_get_delegreturnstateid,
2855 2856
	},
	[OP_GETATTR] = {
2857
		.op_func = nfsd4_getattr,
2858
		.op_flags = ALLOWED_ON_ABSENT_FS,
2859
		.op_rsize_bop = nfsd4_getattr_rsize,
B
Benny Halevy 已提交
2860
		.op_name = "OP_GETATTR",
2861 2862
	},
	[OP_GETFH] = {
2863
		.op_func = nfsd4_getfh,
B
Benny Halevy 已提交
2864
		.op_name = "OP_GETFH",
C
Christoph Hellwig 已提交
2865
		.op_rsize_bop = nfsd4_getfh_rsize,
2866 2867
	},
	[OP_LINK] = {
2868
		.op_func = nfsd4_link,
2869 2870
		.op_flags = ALLOWED_ON_ABSENT_FS | OP_MODIFIES_SOMETHING
				| OP_CACHEME,
B
Benny Halevy 已提交
2871
		.op_name = "OP_LINK",
C
Christoph Hellwig 已提交
2872
		.op_rsize_bop = nfsd4_link_rsize,
2873 2874
	},
	[OP_LOCK] = {
2875
		.op_func = nfsd4_lock,
2876 2877
		.op_flags = OP_MODIFIES_SOMETHING |
				OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
2878
		.op_name = "OP_LOCK",
C
Christoph Hellwig 已提交
2879
		.op_rsize_bop = nfsd4_lock_rsize,
2880
		.op_set_currentstateid = nfsd4_set_lockstateid,
2881 2882
	},
	[OP_LOCKT] = {
2883
		.op_func = nfsd4_lockt,
2884
		.op_flags = OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
2885
		.op_name = "OP_LOCKT",
C
Christoph Hellwig 已提交
2886
		.op_rsize_bop = nfsd4_lock_rsize,
2887 2888
	},
	[OP_LOCKU] = {
2889
		.op_func = nfsd4_locku,
2890
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2891
		.op_name = "OP_LOCKU",
C
Christoph Hellwig 已提交
2892
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2893
		.op_get_currentstateid = nfsd4_get_lockustateid,
2894 2895
	},
	[OP_LOOKUP] = {
2896
		.op_func = nfsd4_lookup,
2897
		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2898
		.op_name = "OP_LOOKUP",
C
Christoph Hellwig 已提交
2899
		.op_rsize_bop = nfsd4_only_status_rsize,
2900 2901
	},
	[OP_LOOKUPP] = {
2902
		.op_func = nfsd4_lookupp,
2903
		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2904
		.op_name = "OP_LOOKUPP",
C
Christoph Hellwig 已提交
2905
		.op_rsize_bop = nfsd4_only_status_rsize,
2906 2907
	},
	[OP_NVERIFY] = {
2908
		.op_func = nfsd4_nverify,
B
Benny Halevy 已提交
2909
		.op_name = "OP_NVERIFY",
C
Christoph Hellwig 已提交
2910
		.op_rsize_bop = nfsd4_only_status_rsize,
2911 2912
	},
	[OP_OPEN] = {
2913
		.op_func = nfsd4_open,
2914
		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2915
		.op_name = "OP_OPEN",
C
Christoph Hellwig 已提交
2916
		.op_rsize_bop = nfsd4_open_rsize,
2917
		.op_set_currentstateid = nfsd4_set_openstateid,
2918 2919
	},
	[OP_OPEN_CONFIRM] = {
2920
		.op_func = nfsd4_open_confirm,
2921
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2922
		.op_name = "OP_OPEN_CONFIRM",
C
Christoph Hellwig 已提交
2923
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2924 2925
	},
	[OP_OPEN_DOWNGRADE] = {
2926
		.op_func = nfsd4_open_downgrade,
2927
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2928
		.op_name = "OP_OPEN_DOWNGRADE",
C
Christoph Hellwig 已提交
2929
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2930
		.op_get_currentstateid = nfsd4_get_opendowngradestateid,
2931
		.op_set_currentstateid = nfsd4_set_opendowngradestateid,
2932 2933
	},
	[OP_PUTFH] = {
2934
		.op_func = nfsd4_putfh,
2935
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2936
				| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2937
		.op_name = "OP_PUTFH",
C
Christoph Hellwig 已提交
2938
		.op_rsize_bop = nfsd4_only_status_rsize,
2939
	},
2940
	[OP_PUTPUBFH] = {
2941
		.op_func = nfsd4_putrootfh,
2942
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2943
				| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2944
		.op_name = "OP_PUTPUBFH",
C
Christoph Hellwig 已提交
2945
		.op_rsize_bop = nfsd4_only_status_rsize,
2946
	},
2947
	[OP_PUTROOTFH] = {
2948
		.op_func = nfsd4_putrootfh,
2949
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2950
				| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2951
		.op_name = "OP_PUTROOTFH",
C
Christoph Hellwig 已提交
2952
		.op_rsize_bop = nfsd4_only_status_rsize,
2953 2954
	},
	[OP_READ] = {
2955
		.op_func = nfsd4_read,
2956
		.op_release = nfsd4_read_release,
B
Benny Halevy 已提交
2957
		.op_name = "OP_READ",
C
Christoph Hellwig 已提交
2958
		.op_rsize_bop = nfsd4_read_rsize,
2959
		.op_get_currentstateid = nfsd4_get_readstateid,
2960 2961
	},
	[OP_READDIR] = {
2962
		.op_func = nfsd4_readdir,
B
Benny Halevy 已提交
2963
		.op_name = "OP_READDIR",
C
Christoph Hellwig 已提交
2964
		.op_rsize_bop = nfsd4_readdir_rsize,
2965 2966
	},
	[OP_READLINK] = {
2967
		.op_func = nfsd4_readlink,
B
Benny Halevy 已提交
2968
		.op_name = "OP_READLINK",
C
Christoph Hellwig 已提交
2969
		.op_rsize_bop = nfsd4_readlink_rsize,
2970 2971
	},
	[OP_REMOVE] = {
2972
		.op_func = nfsd4_remove,
2973
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
2974
		.op_name = "OP_REMOVE",
C
Christoph Hellwig 已提交
2975
		.op_rsize_bop = nfsd4_remove_rsize,
2976 2977
	},
	[OP_RENAME] = {
2978
		.op_func = nfsd4_rename,
2979
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
2980
		.op_name = "OP_RENAME",
C
Christoph Hellwig 已提交
2981
		.op_rsize_bop = nfsd4_rename_rsize,
2982 2983
	},
	[OP_RENEW] = {
2984
		.op_func = nfsd4_renew,
2985 2986
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
				| OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2987
		.op_name = "OP_RENEW",
C
Christoph Hellwig 已提交
2988
		.op_rsize_bop = nfsd4_only_status_rsize,
2989

2990 2991
	},
	[OP_RESTOREFH] = {
2992
		.op_func = nfsd4_restorefh,
2993
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2994
				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2995
		.op_name = "OP_RESTOREFH",
C
Christoph Hellwig 已提交
2996
		.op_rsize_bop = nfsd4_only_status_rsize,
2997 2998
	},
	[OP_SAVEFH] = {
2999
		.op_func = nfsd4_savefh,
3000
		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
3001
		.op_name = "OP_SAVEFH",
C
Christoph Hellwig 已提交
3002
		.op_rsize_bop = nfsd4_only_status_rsize,
3003
	},
A
Andy Adamson 已提交
3004
	[OP_SECINFO] = {
3005
		.op_func = nfsd4_secinfo,
3006
		.op_release = nfsd4_secinfo_release,
3007
		.op_flags = OP_HANDLES_WRONGSEC,
B
Benny Halevy 已提交
3008
		.op_name = "OP_SECINFO",
C
Christoph Hellwig 已提交
3009
		.op_rsize_bop = nfsd4_secinfo_rsize,
A
Andy Adamson 已提交
3010
	},
3011
	[OP_SETATTR] = {
3012
		.op_func = nfsd4_setattr,
B
Benny Halevy 已提交
3013
		.op_name = "OP_SETATTR",
3014 3015
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME
				| OP_NONTRIVIAL_ERROR_ENCODE,
C
Christoph Hellwig 已提交
3016
		.op_rsize_bop = nfsd4_setattr_rsize,
3017
		.op_get_currentstateid = nfsd4_get_setattrstateid,
3018 3019
	},
	[OP_SETCLIENTID] = {
3020
		.op_func = nfsd4_setclientid,
3021
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
3022 3023
				| OP_MODIFIES_SOMETHING | OP_CACHEME
				| OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
3024
		.op_name = "OP_SETCLIENTID",
C
Christoph Hellwig 已提交
3025
		.op_rsize_bop = nfsd4_setclientid_rsize,
3026 3027
	},
	[OP_SETCLIENTID_CONFIRM] = {
3028
		.op_func = nfsd4_setclientid_confirm,
3029
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
3030
				| OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
3031
		.op_name = "OP_SETCLIENTID_CONFIRM",
C
Christoph Hellwig 已提交
3032
		.op_rsize_bop = nfsd4_only_status_rsize,
3033 3034
	},
	[OP_VERIFY] = {
3035
		.op_func = nfsd4_verify,
B
Benny Halevy 已提交
3036
		.op_name = "OP_VERIFY",
C
Christoph Hellwig 已提交
3037
		.op_rsize_bop = nfsd4_only_status_rsize,
3038 3039
	},
	[OP_WRITE] = {
3040
		.op_func = nfsd4_write,
3041
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
3042
		.op_name = "OP_WRITE",
C
Christoph Hellwig 已提交
3043
		.op_rsize_bop = nfsd4_write_rsize,
3044
		.op_get_currentstateid = nfsd4_get_writestateid,
3045 3046
	},
	[OP_RELEASE_LOCKOWNER] = {
3047
		.op_func = nfsd4_release_lockowner,
3048 3049
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
				| OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
3050
		.op_name = "OP_RELEASE_LOCKOWNER",
C
Christoph Hellwig 已提交
3051
		.op_rsize_bop = nfsd4_only_status_rsize,
3052
	},
A
Andy Adamson 已提交
3053 3054 3055

	/* NFSv4.1 operations */
	[OP_EXCHANGE_ID] = {
3056
		.op_func = nfsd4_exchange_id,
3057 3058
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3059
		.op_name = "OP_EXCHANGE_ID",
C
Christoph Hellwig 已提交
3060
		.op_rsize_bop = nfsd4_exchange_id_rsize,
A
Andy Adamson 已提交
3061
	},
3062
	[OP_BACKCHANNEL_CTL] = {
3063
		.op_func = nfsd4_backchannel_ctl,
3064 3065
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
		.op_name = "OP_BACKCHANNEL_CTL",
C
Christoph Hellwig 已提交
3066
		.op_rsize_bop = nfsd4_only_status_rsize,
3067
	},
3068
	[OP_BIND_CONN_TO_SESSION] = {
3069
		.op_func = nfsd4_bind_conn_to_session,
3070 3071
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
3072
		.op_name = "OP_BIND_CONN_TO_SESSION",
C
Christoph Hellwig 已提交
3073
		.op_rsize_bop = nfsd4_bind_conn_to_session_rsize,
3074
	},
A
Andy Adamson 已提交
3075
	[OP_CREATE_SESSION] = {
3076
		.op_func = nfsd4_create_session,
3077 3078
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3079
		.op_name = "OP_CREATE_SESSION",
C
Christoph Hellwig 已提交
3080
		.op_rsize_bop = nfsd4_create_session_rsize,
A
Andy Adamson 已提交
3081 3082
	},
	[OP_DESTROY_SESSION] = {
3083
		.op_func = nfsd4_destroy_session,
3084 3085
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3086
		.op_name = "OP_DESTROY_SESSION",
C
Christoph Hellwig 已提交
3087
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Andy Adamson 已提交
3088 3089
	},
	[OP_SEQUENCE] = {
3090
		.op_func = nfsd4_sequence,
3091
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
A
Andy Adamson 已提交
3092
		.op_name = "OP_SEQUENCE",
C
Christoph Hellwig 已提交
3093
		.op_rsize_bop = nfsd4_sequence_rsize,
A
Andy Adamson 已提交
3094
	},
3095
	[OP_DESTROY_CLIENTID] = {
3096
		.op_func = nfsd4_destroy_clientid,
3097 3098
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
3099
		.op_name = "OP_DESTROY_CLIENTID",
C
Christoph Hellwig 已提交
3100
		.op_rsize_bop = nfsd4_only_status_rsize,
3101
	},
3102
	[OP_RECLAIM_COMPLETE] = {
3103
		.op_func = nfsd4_reclaim_complete,
3104
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
3105
		.op_name = "OP_RECLAIM_COMPLETE",
C
Christoph Hellwig 已提交
3106
		.op_rsize_bop = nfsd4_only_status_rsize,
3107
	},
J
J. Bruce Fields 已提交
3108
	[OP_SECINFO_NO_NAME] = {
3109
		.op_func = nfsd4_secinfo_no_name,
3110
		.op_release = nfsd4_secinfo_no_name_release,
3111
		.op_flags = OP_HANDLES_WRONGSEC,
J
J. Bruce Fields 已提交
3112
		.op_name = "OP_SECINFO_NO_NAME",
C
Christoph Hellwig 已提交
3113
		.op_rsize_bop = nfsd4_secinfo_rsize,
J
J. Bruce Fields 已提交
3114
	},
3115
	[OP_TEST_STATEID] = {
3116
		.op_func = nfsd4_test_stateid,
3117 3118
		.op_flags = ALLOWED_WITHOUT_FH,
		.op_name = "OP_TEST_STATEID",
C
Christoph Hellwig 已提交
3119
		.op_rsize_bop = nfsd4_test_stateid_rsize,
3120
	},
3121
	[OP_FREE_STATEID] = {
3122
		.op_func = nfsd4_free_stateid,
3123
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
3124
		.op_name = "OP_FREE_STATEID",
3125
		.op_get_currentstateid = nfsd4_get_freestateid,
C
Christoph Hellwig 已提交
3126
		.op_rsize_bop = nfsd4_only_status_rsize,
3127
	},
3128 3129
#ifdef CONFIG_NFSD_PNFS
	[OP_GETDEVICEINFO] = {
3130
		.op_func = nfsd4_getdeviceinfo,
3131
		.op_release = nfsd4_getdeviceinfo_release,
3132 3133
		.op_flags = ALLOWED_WITHOUT_FH,
		.op_name = "OP_GETDEVICEINFO",
C
Christoph Hellwig 已提交
3134
		.op_rsize_bop = nfsd4_getdeviceinfo_rsize,
3135 3136
	},
	[OP_LAYOUTGET] = {
3137
		.op_func = nfsd4_layoutget,
3138
		.op_release = nfsd4_layoutget_release,
3139 3140
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTGET",
C
Christoph Hellwig 已提交
3141
		.op_rsize_bop = nfsd4_layoutget_rsize,
3142 3143
	},
	[OP_LAYOUTCOMMIT] = {
3144
		.op_func = nfsd4_layoutcommit,
3145 3146
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTCOMMIT",
C
Christoph Hellwig 已提交
3147
		.op_rsize_bop = nfsd4_layoutcommit_rsize,
3148 3149
	},
	[OP_LAYOUTRETURN] = {
3150
		.op_func = nfsd4_layoutreturn,
3151 3152
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTRETURN",
C
Christoph Hellwig 已提交
3153
		.op_rsize_bop = nfsd4_layoutreturn_rsize,
3154 3155
	},
#endif /* CONFIG_NFSD_PNFS */
A
Anna Schumaker 已提交
3156 3157

	/* NFSv4.2 operations */
A
Anna Schumaker 已提交
3158
	[OP_ALLOCATE] = {
3159
		.op_func = nfsd4_allocate,
3160
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3161
		.op_name = "OP_ALLOCATE",
C
Christoph Hellwig 已提交
3162
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Anna Schumaker 已提交
3163
	},
A
Anna Schumaker 已提交
3164
	[OP_DEALLOCATE] = {
3165
		.op_func = nfsd4_deallocate,
3166
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3167
		.op_name = "OP_DEALLOCATE",
C
Christoph Hellwig 已提交
3168
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Anna Schumaker 已提交
3169
	},
3170
	[OP_CLONE] = {
3171
		.op_func = nfsd4_clone,
3172
		.op_flags = OP_MODIFIES_SOMETHING,
3173
		.op_name = "OP_CLONE",
C
Christoph Hellwig 已提交
3174
		.op_rsize_bop = nfsd4_only_status_rsize,
3175
	},
A
Anna Schumaker 已提交
3176
	[OP_COPY] = {
3177
		.op_func = nfsd4_copy,
3178
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3179
		.op_name = "OP_COPY",
C
Christoph Hellwig 已提交
3180
		.op_rsize_bop = nfsd4_copy_rsize,
A
Anna Schumaker 已提交
3181
	},
A
Anna Schumaker 已提交
3182 3183 3184 3185 3186 3187 3188
	[OP_READ_PLUS] = {
		.op_func = nfsd4_read,
		.op_release = nfsd4_read_release,
		.op_name = "OP_READ_PLUS",
		.op_rsize_bop = nfsd4_read_plus_rsize,
		.op_get_currentstateid = nfsd4_get_readstateid,
	},
A
Anna Schumaker 已提交
3189
	[OP_SEEK] = {
3190
		.op_func = nfsd4_seek,
A
Anna Schumaker 已提交
3191
		.op_name = "OP_SEEK",
C
Christoph Hellwig 已提交
3192
		.op_rsize_bop = nfsd4_seek_rsize,
A
Anna Schumaker 已提交
3193
	},
O
Olga Kornievskaia 已提交
3194 3195 3196 3197 3198
	[OP_OFFLOAD_STATUS] = {
		.op_func = nfsd4_offload_status,
		.op_name = "OP_OFFLOAD_STATUS",
		.op_rsize_bop = nfsd4_offload_status_rsize,
	},
O
Olga Kornievskaia 已提交
3199 3200 3201 3202 3203 3204
	[OP_OFFLOAD_CANCEL] = {
		.op_func = nfsd4_offload_cancel,
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_OFFLOAD_CANCEL",
		.op_rsize_bop = nfsd4_only_status_rsize,
	},
O
Olga Kornievskaia 已提交
3205 3206 3207 3208 3209 3210
	[OP_COPY_NOTIFY] = {
		.op_func = nfsd4_copy_notify,
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_COPY_NOTIFY",
		.op_rsize_bop = nfsd4_copy_notify_rsize,
	},
3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232
	[OP_GETXATTR] = {
		.op_func = nfsd4_getxattr,
		.op_name = "OP_GETXATTR",
		.op_rsize_bop = nfsd4_getxattr_rsize,
	},
	[OP_SETXATTR] = {
		.op_func = nfsd4_setxattr,
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
		.op_name = "OP_SETXATTR",
		.op_rsize_bop = nfsd4_setxattr_rsize,
	},
	[OP_LISTXATTRS] = {
		.op_func = nfsd4_listxattrs,
		.op_name = "OP_LISTXATTRS",
		.op_rsize_bop = nfsd4_listxattrs_rsize,
	},
	[OP_REMOVEXATTR] = {
		.op_func = nfsd4_removexattr,
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
		.op_name = "OP_REMOVEXATTR",
		.op_rsize_bop = nfsd4_removexattr_rsize,
	},
3233 3234
};

3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256
/**
 * nfsd4_spo_must_allow - Determine if the compound op contains an
 * operation that is allowed to be sent with machine credentials
 *
 * @rqstp: a pointer to the struct svc_rqst
 *
 * Checks to see if the compound contains a spo_must_allow op
 * and confirms that it was sent with the proper machine creds.
 */

bool nfsd4_spo_must_allow(struct svc_rqst *rqstp)
{
	struct nfsd4_compoundres *resp = rqstp->rq_resp;
	struct nfsd4_compoundargs *argp = rqstp->rq_argp;
	struct nfsd4_op *this = &argp->ops[resp->opcnt - 1];
	struct nfsd4_compound_state *cstate = &resp->cstate;
	struct nfs4_op_map *allow = &cstate->clp->cl_spo_must_allow;
	u32 opiter;

	if (!cstate->minorversion)
		return false;

Z
Zheng Bin 已提交
3257
	if (cstate->spo_must_allowed)
3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273
		return true;

	opiter = resp->opcnt;
	while (opiter < argp->opcnt) {
		this = &argp->ops[opiter++];
		if (test_bit(this->opnum, allow->u.longs) &&
			cstate->clp->cl_mach_cred &&
			nfsd4_mach_creds_match(cstate->clp, rqstp)) {
			cstate->spo_must_allowed = true;
			return true;
		}
	}
	cstate->spo_must_allowed = false;
	return false;
}

3274 3275
int nfsd4_max_reply(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
3276
	if (op->opnum == OP_ILLEGAL || op->status == nfserr_notsupp)
3277
		return op_encode_hdr_size * sizeof(__be32);
3278 3279 3280

	BUG_ON(OPDESC(op)->op_rsize_bop == NULL);
	return OPDESC(op)->op_rsize_bop(rqstp, op);
3281 3282
}

3283 3284 3285
void warn_on_nonidempotent_op(struct nfsd4_op *op)
{
	if (OPDESC(op)->op_flags & OP_MODIFIES_SOMETHING) {
C
Chuck Lever 已提交
3286
		pr_err("unable to encode reply to nonidempotent op %u (%s)\n",
3287 3288 3289 3290 3291
			op->opnum, nfsd4_op_name(op->opnum));
		WARN_ON_ONCE(1);
	}
}

A
Adrian Bunk 已提交
3292
static const char *nfsd4_op_name(unsigned opnum)
B
Benny Halevy 已提交
3293 3294 3295 3296 3297 3298
{
	if (opnum < ARRAY_SIZE(nfsd4_ops))
		return nfsd4_ops[opnum].op_name;
	return "unknown_operation";
}

3299
static const struct svc_procedure nfsd_procedures4[2] = {
3300
	[NFSPROC4_NULL] = {
3301
		.pc_func = nfsd4_proc_null,
3302 3303 3304 3305
		.pc_decode = nfssvc_decode_voidarg,
		.pc_encode = nfssvc_encode_voidres,
		.pc_argsize = sizeof(struct nfsd_voidargs),
		.pc_ressize = sizeof(struct nfsd_voidres),
3306 3307
		.pc_cachetype = RC_NOCACHE,
		.pc_xdrressize = 1,
3308
		.pc_name = "NULL",
3309 3310
	},
	[NFSPROC4_COMPOUND] = {
3311
		.pc_func = nfsd4_proc_compound,
3312
		.pc_decode = nfs4svc_decode_compoundargs,
3313
		.pc_encode = nfs4svc_encode_compoundres,
3314 3315
		.pc_argsize = sizeof(struct nfsd4_compoundargs),
		.pc_ressize = sizeof(struct nfsd4_compoundres),
3316
		.pc_release = nfsd4_release_compoundargs,
3317 3318
		.pc_cachetype = RC_NOCACHE,
		.pc_xdrressize = NFSD_BUFSIZE/4,
3319
		.pc_name = "COMPOUND",
3320
	},
L
Linus Torvalds 已提交
3321 3322
};

3323
static unsigned int nfsd_count3[ARRAY_SIZE(nfsd_procedures4)];
3324
const struct svc_version nfsd_version4 = {
3325 3326 3327
	.vs_vers		= 4,
	.vs_nproc		= 2,
	.vs_proc		= nfsd_procedures4,
3328
	.vs_count		= nfsd_count3,
3329 3330 3331 3332
	.vs_dispatch		= nfsd_dispatch,
	.vs_xdrsize		= NFS4_SVC_XDRSIZE,
	.vs_rpcb_optnl		= true,
	.vs_need_cong_ctrl	= true,
L
Linus Torvalds 已提交
3333 3334 3335 3336 3337 3338 3339
};

/*
 * Local variables:
 *  c-basic-offset: 8
 * End:
 */