nfs4proc.c 89.3 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
	    !test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &cstate->clp->cl_flags) &&
382 383 384
	    open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
		return nfserr_grace;

A
Andy Adamson 已提交
385 386 387
	if (nfsd4_has_session(cstate))
		copy_clientid(&open->op_clientid, cstate->session);

L
Linus Torvalds 已提交
388
	/* check seqid for replay. set nfs4_owner */
389
	status = nfsd4_process_open1(cstate, open, nn);
A
Al Viro 已提交
390
	if (status == nfserr_replay_me) {
391
		struct nfs4_replay *rp = &open->op_openowner->oo_owner.so_replay;
392
		fh_put(&cstate->current_fh);
393 394
		fh_copy_shallow(&cstate->current_fh.fh_handle,
				&rp->rp_openfh);
M
Miklos Szeredi 已提交
395
		status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
396 397 398 399
		if (status)
			dprintk("nfsd4_open: replay failed"
				" restoring previous filehandle\n");
		else
A
Al Viro 已提交
400
			status = nfserr_replay_me;
L
Linus Torvalds 已提交
401 402 403
	}
	if (status)
		goto out;
404 405 406 407
	if (open->op_xdr_error) {
		status = open->op_xdr_error;
		goto out;
	}
408

409 410 411 412
	status = nfsd4_check_open_attributes(rqstp, cstate, open);
	if (status)
		goto out;

413 414
	/* Openowner is now set, so sequence id will get bumped.  Now we need
	 * these checks before we do any creates: */
415
	status = nfserr_grace;
416
	if (opens_in_grace(net) && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS)
417 418
		goto out;
	status = nfserr_no_grace;
419
	if (!opens_in_grace(net) && open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS)
420
		goto out;
421

L
Linus Torvalds 已提交
422
	switch (open->op_claim_type) {
423
		case NFS4_OPEN_CLAIM_DELEGATE_CUR:
L
Linus Torvalds 已提交
424
		case NFS4_OPEN_CLAIM_NULL:
425
			status = do_open_lookup(rqstp, cstate, open, &resfh);
L
Linus Torvalds 已提交
426 427 428 429
			if (status)
				goto out;
			break;
		case NFS4_OPEN_CLAIM_PREVIOUS:
430
			status = nfs4_check_open_reclaim(cstate->clp);
431 432
			if (status)
				goto out;
433
			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
434
			reclaim = true;
435
			fallthrough;
436 437
		case NFS4_OPEN_CLAIM_FH:
		case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
438
			status = do_open_fhandle(rqstp, cstate, open);
L
Linus Torvalds 已提交
439 440
			if (status)
				goto out;
441
			resfh = &cstate->current_fh;
L
Linus Torvalds 已提交
442
			break;
443
		case NFS4_OPEN_CLAIM_DELEG_PREV_FH:
L
Linus Torvalds 已提交
444
             	case NFS4_OPEN_CLAIM_DELEGATE_PREV:
445
			dprintk("NFSD: unsupported OPEN claim type %d\n",
L
Linus Torvalds 已提交
446 447 448 449
				open->op_claim_type);
			status = nfserr_notsupp;
			goto out;
		default:
450
			dprintk("NFSD: Invalid OPEN claim type %d\n",
L
Linus Torvalds 已提交
451 452 453 454 455 456 457 458 459
				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.
	 */
460
	status = nfsd4_process_open2(rqstp, resfh, open);
461 462 463
	WARN(status && open->op_created,
	     "nfsd4_process_open2 failed to open newly-created file! status=%u\n",
	     be32_to_cpu(status));
464 465
	if (reclaim && !status)
		nn->somebody_reclaimed = true;
L
Linus Torvalds 已提交
466
out:
467 468 469 470 471
	if (resfh && resfh != &cstate->current_fh) {
		fh_dup2(&cstate->current_fh, resfh);
		fh_put(resfh);
		kfree(resfh);
	}
472
	nfsd4_cleanup_open_state(cstate, open);
473
	nfsd4_bump_seqid(cstate, status);
L
Linus Torvalds 已提交
474 475 476
	return status;
}

477 478 479 480 481 482 483 484
/*
 * 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)
{
485
	struct nfsd4_open *open = &op->u.open;
486 487 488 489 490 491

	if (!seqid_mutating_err(ntohl(op->status)))
		return op->status;
	if (nfsd4_has_session(cstate))
		return op->status;
	open->op_xdr_error = op->status;
492
	return nfsd4_open(rqstp, cstate, &op->u);
493 494
}

L
Linus Torvalds 已提交
495 496 497
/*
 * filehandle-manipulating ops.
 */
498
static __be32
499
nfsd4_getfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
500
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
501
{
502
	u->getfh = &cstate->current_fh;
L
Linus Torvalds 已提交
503 504 505
	return nfs_ok;
}

506
static __be32
507
nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
508
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
509
{
510
	struct nfsd4_putfh *putfh = &u->putfh;
511
	__be32 ret;
512

513 514 515 516
	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);
517 518 519 520 521 522 523 524
	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
525 526
}

527
static __be32
528
nfsd4_putrootfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
529
		union nfsd4_op_u *u)
L
Linus Torvalds 已提交
530
{
531
	__be32 status;
L
Linus Torvalds 已提交
532

533
	fh_put(&cstate->current_fh);
534
	status = exp_pseudoroot(rqstp, &cstate->current_fh);
L
Linus Torvalds 已提交
535 536 537
	return status;
}

538
static __be32
539
nfsd4_restorefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
540
		union nfsd4_op_u *u)
L
Linus Torvalds 已提交
541
{
542
	if (!cstate->save_fh.fh_dentry)
L
Linus Torvalds 已提交
543 544
		return nfserr_restorefh;

545
	fh_dup2(&cstate->current_fh, &cstate->save_fh);
546
	if (HAS_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG)) {
547
		memcpy(&cstate->current_stateid, &cstate->save_stateid, sizeof(stateid_t));
548
		SET_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG);
549
	}
L
Linus Torvalds 已提交
550 551 552
	return nfs_ok;
}

553
static __be32
554
nfsd4_savefh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
555
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
556
{
557
	fh_dup2(&cstate->save_fh, &cstate->current_fh);
558
	if (HAS_CSTATE_FLAG(cstate, CURRENT_STATE_ID_FLAG)) {
559
		memcpy(&cstate->save_stateid, &cstate->current_stateid, sizeof(stateid_t));
560
		SET_CSTATE_FLAG(cstate, SAVED_STATE_ID_FLAG);
561
	}
L
Linus Torvalds 已提交
562 563 564 565 566 567
	return nfs_ok;
}

/*
 * misc nfsv4 ops
 */
568
static __be32
569
nfsd4_access(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
570
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
571
{
572
	struct nfsd4_access *access = &u->access;
573
	u32 access_full;
574

575 576 577 578 579 580
	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
581 582 583
		return nfserr_inval;

	access->ac_resp_access = access->ac_req_access;
584 585
	return nfsd_access(rqstp, &cstate->current_fh, &access->ac_resp_access,
			   &access->ac_supported);
L
Linus Torvalds 已提交
586 587
}

588
static void gen_boot_verifier(nfs4_verifier *verifier, struct net *net)
589
{
590
	__be32 *verf = (__be32 *)verifier->data;
591

592 593 594
	BUILD_BUG_ON(2*sizeof(*verf) != sizeof(verifier->data));

	nfsd_copy_boot_verifier(verf, net_generic(net, nfsd_net_id));
595 596
}

597
static __be32
598
nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
599
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
600
{
601 602
	struct nfsd4_commit *commit = &u->commit;

603
	return nfsd_commit(rqstp, &cstate->current_fh, commit->co_offset,
604 605
			     commit->co_count,
			     (__be32 *)commit->co_verf.data);
L
Linus Torvalds 已提交
606 607
}

608
static __be32
609
nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
610
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
611
{
612
	struct nfsd4_create *create = &u->create;
L
Linus Torvalds 已提交
613
	struct svc_fh resfh;
614
	__be32 status;
L
Linus Torvalds 已提交
615 616 617 618
	dev_t rdev;

	fh_init(&resfh, NFS4_FHSIZE);

619
	status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
620 621 622
	if (status)
		return status;

623 624 625 626 627
	status = check_attr_support(rqstp, cstate, create->cr_bmval,
				    nfsd_attrmask);
	if (status)
		return status;

J
J. Bruce Fields 已提交
628
	current->fs->umask = create->cr_umask;
L
Linus Torvalds 已提交
629 630
	switch (create->cr_type) {
	case NF4LNK:
631 632
		status = nfsd_symlink(rqstp, &cstate->current_fh,
				      create->cr_name, create->cr_namelen,
633
				      create->cr_data, &resfh);
L
Linus Torvalds 已提交
634 635 636
		break;

	case NF4BLK:
J
J. Bruce Fields 已提交
637
		status = nfserr_inval;
L
Linus Torvalds 已提交
638 639 640
		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
641
			goto out_umask;
642 643 644
		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
645 646 647
		break;

	case NF4CHR:
J
J. Bruce Fields 已提交
648
		status = nfserr_inval;
L
Linus Torvalds 已提交
649 650 651
		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
652
			goto out_umask;
653 654 655
		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
656 657 658
		break;

	case NF4SOCK:
659 660 661
		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
662 663 664
		break;

	case NF4FIFO:
665 666 667
		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
668 669 670 671
		break;

	case NF4DIR:
		create->cr_iattr.ia_valid &= ~ATTR_SIZE;
672 673 674
		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
675 676 677 678 679 680
		break;

	default:
		status = nfserr_badtype;
	}

681 682
	if (status)
		goto out;
L
Linus Torvalds 已提交
683

684 685 686
	if (create->cr_label.len)
		nfsd4_security_inode_setsecctx(&resfh, &create->cr_label, create->cr_bmval);

687 688 689 690 691 692 693 694
	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
695
	fh_put(&resfh);
J
J. Bruce Fields 已提交
696 697
out_umask:
	current->fs->umask = 0;
L
Linus Torvalds 已提交
698 699 700
	return status;
}

701
static __be32
702
nfsd4_getattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
703
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
704
{
705
	struct nfsd4_getattr *getattr = &u->getattr;
706
	__be32 status;
L
Linus Torvalds 已提交
707

M
Miklos Szeredi 已提交
708
	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
709 710 711 712 713 714
	if (status)
		return status;

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

715 716 717
	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
718

719
	getattr->ga_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
720 721 722
	return nfs_ok;
}

723
static __be32
724
nfsd4_link(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
725
	   union nfsd4_op_u *u)
L
Linus Torvalds 已提交
726
{
727
	struct nfsd4_link *link = &u->link;
728
	__be32 status;
L
Linus Torvalds 已提交
729

730 731
	status = nfsd_link(rqstp, &cstate->current_fh,
			   link->li_name, link->li_namelen, &cstate->save_fh);
L
Linus Torvalds 已提交
732
	if (!status)
733
		set_change_info(&link->li_cinfo, &cstate->current_fh);
L
Linus Torvalds 已提交
734 735 736
	return status;
}

737
static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
L
Linus Torvalds 已提交
738 739
{
	struct svc_fh tmp_fh;
740
	__be32 ret;
L
Linus Torvalds 已提交
741 742

	fh_init(&tmp_fh, NFS4_FHSIZE);
743 744
	ret = exp_pseudoroot(rqstp, &tmp_fh);
	if (ret)
L
Linus Torvalds 已提交
745
		return ret;
746
	if (tmp_fh.fh_dentry == fh->fh_dentry) {
L
Linus Torvalds 已提交
747 748 749 750
		fh_put(&tmp_fh);
		return nfserr_noent;
	}
	fh_put(&tmp_fh);
751 752 753 754 755
	return nfsd_lookup(rqstp, fh, "..", 2, fh);
}

static __be32
nfsd4_lookupp(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
756
	      union nfsd4_op_u *u)
757 758
{
	return nfsd4_do_lookupp(rqstp, &cstate->current_fh);
L
Linus Torvalds 已提交
759 760
}

761
static __be32
762
nfsd4_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
763
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
764
{
765
	return nfsd_lookup(rqstp, &cstate->current_fh,
766
			   u->lookup.lo_name, u->lookup.lo_len,
767
			   &cstate->current_fh);
L
Linus Torvalds 已提交
768 769
}

770
static __be32
771
nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
772
	   union nfsd4_op_u *u)
L
Linus Torvalds 已提交
773
{
774
	struct nfsd4_read *read = &u->read;
775
	__be32 status;
L
Linus Torvalds 已提交
776

777
	read->rd_nf = NULL;
L
Linus Torvalds 已提交
778 779 780
	if (read->rd_offset >= OFFSET_MAX)
		return nfserr_inval;

781 782 783
	trace_nfsd_read_start(rqstp, &cstate->current_fh,
			      read->rd_offset, read->rd_length);

784 785 786 787 788 789 790 791 792
	/*
	 * 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))
793
		clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
794

L
Linus Torvalds 已提交
795
	/* check stateid */
796 797
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
					&read->rd_stateid, RD_STATE,
O
Olga Kornievskaia 已提交
798
					&read->rd_nf, NULL);
799
	if (status) {
L
Linus Torvalds 已提交
800 801 802 803 804 805
		dprintk("NFSD: nfsd4_read: couldn't process stateid!\n");
		goto out;
	}
	status = nfs_ok;
out:
	read->rd_rqstp = rqstp;
806
	read->rd_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
807 808 809
	return status;
}

810 811 812 813

static void
nfsd4_read_release(union nfsd4_op_u *u)
{
814 815
	if (u->read.rd_nf)
		nfsd_file_put(u->read.rd_nf);
816 817
	trace_nfsd_read_done(u->read.rd_rqstp, u->read.rd_fhp,
			     u->read.rd_offset, u->read.rd_length);
818 819
}

820
static __be32
821
nfsd4_readdir(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
822
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
823
{
824
	struct nfsd4_readdir *readdir = &u->readdir;
L
Linus Torvalds 已提交
825 826 827 828 829 830 831 832
	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;

833 834 835
	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
836

837
	if ((cookie == 1) || (cookie == 2) ||
L
Linus Torvalds 已提交
838 839 840 841
	    (cookie == 0 && memcmp(readdir->rd_verf.data, zeroverf.data, NFS4_VERIFIER_SIZE)))
		return nfserr_bad_cookie;

	readdir->rd_rqstp = rqstp;
842
	readdir->rd_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
843 844 845
	return nfs_ok;
}

846
static __be32
847
nfsd4_readlink(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
848
	       union nfsd4_op_u *u)
L
Linus Torvalds 已提交
849
{
850 851
	u->readlink.rl_rqstp = rqstp;
	u->readlink.rl_fhp = &cstate->current_fh;
L
Linus Torvalds 已提交
852 853 854
	return nfs_ok;
}

855
static __be32
856
nfsd4_remove(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
857
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
858
{
859
	struct nfsd4_remove *remove = &u->remove;
860
	__be32 status;
L
Linus Torvalds 已提交
861

862
	if (opens_in_grace(SVC_NET(rqstp)))
863
		return nfserr_grace;
864 865
	status = nfsd_unlink(rqstp, &cstate->current_fh, 0,
			     remove->rm_name, remove->rm_namelen);
L
Linus Torvalds 已提交
866
	if (!status) {
867 868
		fh_unlock(&cstate->current_fh);
		set_change_info(&remove->rm_cinfo, &cstate->current_fh);
L
Linus Torvalds 已提交
869 870 871 872
	}
	return status;
}

873
static __be32
874
nfsd4_rename(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
875
	     union nfsd4_op_u *u)
L
Linus Torvalds 已提交
876
{
877
	struct nfsd4_rename *rename = &u->rename;
878
	__be32 status;
L
Linus Torvalds 已提交
879

880
	if (opens_in_grace(SVC_NET(rqstp)))
881
		return nfserr_grace;
882 883
	status = nfsd_rename(rqstp, &cstate->save_fh, rename->rn_sname,
			     rename->rn_snamelen, &cstate->current_fh,
L
Linus Torvalds 已提交
884
			     rename->rn_tname, rename->rn_tnamelen);
885 886 887 888 889
	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
890 891
}

A
Andy Adamson 已提交
892 893
static __be32
nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
894
	      union nfsd4_op_u *u)
A
Andy Adamson 已提交
895
{
896
	struct nfsd4_secinfo *secinfo = &u->secinfo;
A
Andy Adamson 已提交
897 898 899 900
	struct svc_export *exp;
	struct dentry *dentry;
	__be32 err;

901 902 903
	err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC);
	if (err)
		return err;
A
Andy Adamson 已提交
904 905 906 907 908
	err = nfsd_lookup_dentry(rqstp, &cstate->current_fh,
				    secinfo->si_name, secinfo->si_namelen,
				    &exp, &dentry);
	if (err)
		return err;
909
	fh_unlock(&cstate->current_fh);
910
	if (d_really_is_negative(dentry)) {
A
Andy Adamson 已提交
911 912 913 914 915
		exp_put(exp);
		err = nfserr_noent;
	} else
		secinfo->si_exp = exp;
	dput(dentry);
916 917 918
	if (cstate->minorversion)
		/* See rfc 5661 section 2.6.3.1.1.8 */
		fh_put(&cstate->current_fh);
A
Andy Adamson 已提交
919 920 921
	return err;
}

J
J. Bruce Fields 已提交
922 923
static __be32
nfsd4_secinfo_no_name(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
924
		union nfsd4_op_u *u)
J
J. Bruce Fields 已提交
925 926 927
{
	__be32 err;

928
	switch (u->secinfo_no_name.sin_style) {
J
J. Bruce Fields 已提交
929 930 931 932 933 934 935 936 937 938
	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;
	}
939

940
	u->secinfo_no_name.sin_exp = exp_get(cstate->current_fh.fh_export);
J
J. Bruce Fields 已提交
941 942 943 944
	fh_put(&cstate->current_fh);
	return nfs_ok;
}

945 946 947 948 949 950 951
static void
nfsd4_secinfo_release(union nfsd4_op_u *u)
{
	if (u->secinfo.si_exp)
		exp_put(u->secinfo.si_exp);
}

952 953 954 955 956 957 958
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);
}

959
static __be32
960
nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
961
	      union nfsd4_op_u *u)
L
Linus Torvalds 已提交
962
{
963
	struct nfsd4_setattr *setattr = &u->setattr;
964
	__be32 status = nfs_ok;
965
	int err;
L
Linus Torvalds 已提交
966 967

	if (setattr->sa_iattr.ia_valid & ATTR_SIZE) {
968
		status = nfs4_preprocess_stateid_op(rqstp, cstate,
969
				&cstate->current_fh, &setattr->sa_stateid,
O
Olga Kornievskaia 已提交
970
				WR_STATE, NULL, NULL);
971
		if (status) {
972
			dprintk("NFSD: nfsd4_setattr: couldn't process stateid!\n");
973 974
			return status;
		}
L
Linus Torvalds 已提交
975
	}
976 977 978
	err = fh_want_write(&cstate->current_fh);
	if (err)
		return nfserrno(err);
L
Linus Torvalds 已提交
979
	status = nfs_ok;
980 981 982 983 984 985

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

L
Linus Torvalds 已提交
986
	if (setattr->sa_acl != NULL)
987 988
		status = nfsd4_set_nfs4_acl(rqstp, &cstate->current_fh,
					    setattr->sa_acl);
L
Linus Torvalds 已提交
989
	if (status)
990
		goto out;
991 992 993 994 995
	if (setattr->sa_label.len)
		status = nfsd4_set_nfs4_label(rqstp, &cstate->current_fh,
				&setattr->sa_label);
	if (status)
		goto out;
996
	status = nfsd_setattr(rqstp, &cstate->current_fh, &setattr->sa_iattr,
997
				0, (time64_t)0);
998
out:
A
Al Viro 已提交
999
	fh_drop_write(&cstate->current_fh);
L
Linus Torvalds 已提交
1000 1001 1002
	return status;
}

1003
static __be32
1004
nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1005
	    union nfsd4_op_u *u)
L
Linus Torvalds 已提交
1006
{
1007
	struct nfsd4_write *write = &u->write;
L
Linus Torvalds 已提交
1008
	stateid_t *stateid = &write->wr_stateid;
1009
	struct nfsd_file *nf = NULL;
1010
	__be32 status = nfs_ok;
1011
	unsigned long cnt;
1012
	int nvecs;
L
Linus Torvalds 已提交
1013 1014 1015 1016

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

1017 1018 1019
	cnt = write->wr_buflen;
	trace_nfsd_write_start(rqstp, &cstate->current_fh,
			       write->wr_offset, cnt);
1020
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
O
Olga Kornievskaia 已提交
1021
						stateid, WR_STATE, &nf, NULL);
1022 1023 1024 1025 1026
	if (status) {
		dprintk("NFSD: nfsd4_write: couldn't process stateid!\n");
		return status;
	}

L
Linus Torvalds 已提交
1027 1028
	write->wr_how_written = write->wr_stable_how;

1029 1030
	nvecs = svc_fill_write_vector(rqstp, write->wr_payload.pages,
				      write->wr_payload.head, write->wr_buflen);
1031 1032
	WARN_ON_ONCE(nvecs > ARRAY_SIZE(rqstp->rq_vec));

1033
	status = nfsd_vfs_write(rqstp, &cstate->current_fh, nf,
1034
				write->wr_offset, rqstp->rq_vec, nvecs, &cnt,
1035 1036
				write->wr_how_written,
				(__be32 *)write->wr_verifier.data);
1037
	nfsd_file_put(nf);
L
Linus Torvalds 已提交
1038

1039
	write->wr_bytes_written = cnt;
1040 1041
	trace_nfsd_write_done(rqstp, &cstate->current_fh,
			      write->wr_offset, cnt);
L
Linus Torvalds 已提交
1042 1043 1044
	return status;
}

1045
static __be32
A
Anna Schumaker 已提交
1046
nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1047 1048
		  stateid_t *src_stateid, struct nfsd_file **src,
		  stateid_t *dst_stateid, struct nfsd_file **dst)
1049 1050 1051
{
	__be32 status;

1052 1053 1054
	if (!cstate->save_fh.fh_dentry)
		return nfserr_nofilehandle;

1055
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
O
Olga Kornievskaia 已提交
1056
					    src_stateid, RD_STATE, src, NULL);
1057 1058 1059 1060 1061 1062
	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
1063
					    dst_stateid, WR_STATE, dst, NULL);
1064 1065 1066 1067 1068 1069
	if (status) {
		dprintk("NFSD: %s: couldn't process dst stateid!\n", __func__);
		goto out_put_src;
	}

	/* fix up for NFS-specific error code */
1070 1071
	if (!S_ISREG(file_inode((*src)->nf_file)->i_mode) ||
	    !S_ISREG(file_inode((*dst)->nf_file)->i_mode)) {
1072 1073 1074 1075
		status = nfserr_wrong_type;
		goto out_put_dst;
	}

A
Anna Schumaker 已提交
1076 1077 1078
out:
	return status;
out_put_dst:
1079
	nfsd_file_put(*dst);
A
Anna Schumaker 已提交
1080
out_put_src:
1081
	nfsd_file_put(*src);
A
Anna Schumaker 已提交
1082 1083 1084 1085 1086
	goto out;
}

static __be32
nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1087
		union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1088
{
1089
	struct nfsd4_clone *clone = &u->clone;
1090
	struct nfsd_file *src, *dst;
A
Anna Schumaker 已提交
1091 1092 1093 1094 1095 1096 1097
	__be32 status;

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

1098 1099
	status = nfsd4_clone_file_range(src, clone->cl_src_pos,
			dst, clone->cl_dst_pos, clone->cl_count,
1100
			EX_ISSYNC(cstate->current_fh.fh_export));
1101

1102 1103
	nfsd_file_put(dst);
	nfsd_file_put(src);
1104 1105 1106 1107
out:
	return status;
}

1108 1109 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
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);
}
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167
#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"

1168
/*
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
 * 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;
1182 1183
	int len, raw_len;
	__be32 status = nfserr_inval;
1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220

	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);

1221
	status = nfserr_nodev;
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
	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)
{
1254
	nfs_do_sb_deactive(ss_mnt->mnt_sb);
1255 1256 1257
	mntput(ss_mnt);
}

1258
/*
1259
 * Verify COPY destination stateid.
1260
 *
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
 * 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;
1274
	__be32 status = nfserr_inval;
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290

	/* 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);
1291
	copy->stateid.seqid = cpu_to_be32(s_stid->si_generation);
1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304
	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);
1305
	fput(src->nf_file);
1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318
	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;
1319
	return nfserr_inval;
1320 1321 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
}

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);
}
1357 1358 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

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;
1386
	u64 bytes_total = copy->cp_count;
1387 1388 1389
	u64 src_pos = copy->cp_src_pos;
	u64 dst_pos = copy->cp_dst_pos;

1390 1391 1392
	/* See RFC 7862 p.67: */
	if (bytes_total == 0)
		bytes_total = ULLONG_MAX;
1393 1394 1395
	do {
		if (kthread_should_stop())
			break;
1396 1397 1398
		bytes_copied = nfsd_copy_file_range(copy->nf_src->nf_file,
				src_pos, copy->nf_dst->nf_file, dst_pos,
				bytes_total);
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424
		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;
	}

1425 1426 1427 1428 1429 1430
	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);

1431 1432 1433
	return status;
}

1434
static void dup_copy_fields(struct nfsd4_copy *src, struct nfsd4_copy *dst)
1435 1436 1437 1438 1439 1440 1441 1442
{
	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;
1443
	dst->nf_dst = nfsd_file_get(src->nf_dst);
1444 1445 1446 1447
	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);

1448
	memcpy(&dst->cp_stateid, &src->cp_stateid, sizeof(src->cp_stateid));
1449 1450 1451 1452
	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;
1453 1454 1455 1456
}

static void cleanup_async_copy(struct nfsd4_copy *copy)
{
O
Olga Kornievskaia 已提交
1457
	nfs4_free_copy_state(copy);
1458
	nfsd_file_put(copy->nf_dst);
1459 1460
	if (copy->cp_intra)
		nfsd_file_put(copy->nf_src);
1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471
	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;

1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487
	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;
		}
	}

1488
	copy->nfserr = nfsd4_do_copy(copy, 0);
1489
do_callback:
1490 1491 1492
	cb_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
	if (!cb_copy)
		goto out;
1493
	refcount_set(&cb_copy->refcount, 1);
1494 1495 1496 1497 1498 1499 1500 1501
	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:
1502 1503
	if (!copy->cp_intra)
		kfree(copy->nf_src);
1504 1505 1506 1507
	cleanup_async_copy(copy);
	return 0;
}

A
Anna Schumaker 已提交
1508 1509
static __be32
nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1510
		union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1511
{
1512
	struct nfsd4_copy *copy = &u->copy;
A
Anna Schumaker 已提交
1513
	__be32 status;
1514
	struct nfsd4_copy *async_copy = NULL;
A
Anna Schumaker 已提交
1515

1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529
	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
1530

1531 1532 1533 1534 1535
	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
1536

1537 1538 1539
		status = nfserrno(-ENOMEM);
		async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL);
		if (!async_copy)
1540 1541 1542
			goto out_err;
		if (!nfs4_init_copy_state(nn, copy))
			goto out_err;
1543 1544 1545
		refcount_set(&async_copy->refcount, 1);
		memcpy(&copy->cp_res.cb_stateid, &copy->cp_stateid,
			sizeof(copy->cp_stateid));
1546
		dup_copy_fields(copy, async_copy);
1547 1548 1549 1550 1551 1552 1553 1554 1555
		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
1556
		status = nfs_ok;
1557
	} else {
1558
		status = nfsd4_do_copy(copy, 1);
1559
	}
A
Anna Schumaker 已提交
1560 1561
out:
	return status;
1562
out_err:
1563 1564
	if (async_copy)
		cleanup_async_copy(async_copy);
1565 1566 1567
	status = nfserrno(-ENOMEM);
	if (!copy->cp_intra)
		nfsd4_interssc_disconnect(copy->ss_mnt);
1568 1569 1570 1571 1572 1573 1574 1575 1576 1577
	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) {
1578
		if (memcmp(&copy->cp_stateid.stid, stateid, NFS4_STATEID_SIZE))
1579 1580 1581 1582 1583 1584 1585
			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
1586 1587
}

O
Olga Kornievskaia 已提交
1588 1589 1590 1591 1592
static __be32
nfsd4_offload_cancel(struct svc_rqst *rqstp,
		     struct nfsd4_compound_state *cstate,
		     union nfsd4_op_u *u)
{
1593 1594 1595 1596 1597
	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);
1598 1599 1600 1601 1602
	if (!copy) {
		struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);

		return manage_cpntf_state(nn, &os->stateid, clp, NULL);
	} else
1603 1604
		nfsd4_stop_copy(copy);

1605
	return nfs_ok;
O
Olga Kornievskaia 已提交
1606 1607
}

O
Olga Kornievskaia 已提交
1608 1609 1610 1611
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
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 1649
	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
1650 1651
}

A
Anna Schumaker 已提交
1652 1653 1654 1655
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
1656
	__be32 status;
1657
	struct nfsd_file *nf;
A
Anna Schumaker 已提交
1658

1659
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
A
Anna Schumaker 已提交
1660
					    &fallocate->falloc_stateid,
O
Olga Kornievskaia 已提交
1661
					    WR_STATE, &nf, NULL);
A
Anna Schumaker 已提交
1662 1663 1664 1665 1666
	if (status != nfs_ok) {
		dprintk("NFSD: nfsd4_fallocate: couldn't process stateid!\n");
		return status;
	}

1667
	status = nfsd4_vfs_fallocate(rqstp, &cstate->current_fh, nf->nf_file,
A
Anna Schumaker 已提交
1668 1669 1670
				     fallocate->falloc_offset,
				     fallocate->falloc_length,
				     flags);
1671
	nfsd_file_put(nf);
A
Anna Schumaker 已提交
1672 1673
	return status;
}
O
Olga Kornievskaia 已提交
1674 1675 1676 1677 1678
static __be32
nfsd4_offload_status(struct svc_rqst *rqstp,
		     struct nfsd4_compound_state *cstate,
		     union nfsd4_op_u *u)
{
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691
	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
1692
}
A
Anna Schumaker 已提交
1693 1694 1695

static __be32
nfsd4_allocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1696
	       union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1697
{
1698
	return nfsd4_fallocate(rqstp, cstate, &u->allocate, 0);
A
Anna Schumaker 已提交
1699 1700
}

A
Anna Schumaker 已提交
1701 1702
static __be32
nfsd4_deallocate(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1703
		 union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1704
{
1705
	return nfsd4_fallocate(rqstp, cstate, &u->deallocate,
A
Anna Schumaker 已提交
1706 1707 1708
			       FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE);
}

A
Anna Schumaker 已提交
1709 1710
static __be32
nfsd4_seek(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1711
	   union nfsd4_op_u *u)
A
Anna Schumaker 已提交
1712
{
1713
	struct nfsd4_seek *seek = &u->seek;
A
Anna Schumaker 已提交
1714 1715
	int whence;
	__be32 status;
1716
	struct nfsd_file *nf;
A
Anna Schumaker 已提交
1717

1718
	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
A
Anna Schumaker 已提交
1719
					    &seek->seek_stateid,
O
Olga Kornievskaia 已提交
1720
					    RD_STATE, &nf, NULL);
A
Anna Schumaker 已提交
1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741
	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.
	 */
1742
	seek->seek_pos = vfs_llseek(nf->nf_file, seek->seek_offset, whence);
A
Anna Schumaker 已提交
1743 1744
	if (seek->seek_pos < 0)
		status = nfserrno(seek->seek_pos);
1745
	else if (seek->seek_pos >= i_size_read(file_inode(nf->nf_file)))
A
Anna Schumaker 已提交
1746 1747 1748
		seek->seek_eof = true;

out:
1749
	nfsd_file_put(nf);
A
Anna Schumaker 已提交
1750 1751 1752
	return status;
}

L
Linus Torvalds 已提交
1753 1754 1755 1756 1757
/* 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.
 */
1758
static __be32
1759
_nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1760
	     struct nfsd4_verify *verify)
L
Linus Torvalds 已提交
1761
{
A
Al Viro 已提交
1762
	__be32 *buf, *p;
L
Linus Torvalds 已提交
1763
	int count;
1764
	__be32 status;
L
Linus Torvalds 已提交
1765

M
Miklos Szeredi 已提交
1766
	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
L
Linus Torvalds 已提交
1767 1768 1769
	if (status)
		return status;

1770 1771 1772 1773
	status = check_attr_support(rqstp, cstate, verify->ve_bmval, NULL);
	if (status)
		return status;

L
Linus Torvalds 已提交
1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
	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)
1786
		return nfserr_jukebox;
L
Linus Torvalds 已提交
1787

1788
	p = buf;
1789
	status = nfsd4_encode_fattr_to_buf(&p, count, &cstate->current_fh,
1790
				    cstate->current_fh.fh_export,
1791 1792
				    cstate->current_fh.fh_dentry,
				    verify->ve_bmval,
1793
				    rqstp, 0);
J
J. Bruce Fields 已提交
1794 1795 1796 1797
	/*
	 * If nfsd4_encode_fattr() ran out of space, assume that's because
	 * the attributes are longer (hence different) than those given:
	 */
1798
	if (status == nfserr_resource)
L
Linus Torvalds 已提交
1799 1800 1801 1802
		status = nfserr_not_same;
	if (status)
		goto out_kfree;

1803 1804
	/* skip bitmap */
	p = buf + 1 + ntohl(buf[0]);
L
Linus Torvalds 已提交
1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815
	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;
}

1816 1817
static __be32
nfsd4_nverify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1818
	      union nfsd4_op_u *u)
1819 1820 1821
{
	__be32 status;

1822
	status = _nfsd4_verify(rqstp, cstate, &u->verify);
1823 1824 1825 1826 1827
	return status == nfserr_not_same ? nfs_ok : status;
}

static __be32
nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
1828
	     union nfsd4_op_u *u)
1829 1830 1831
{
	__be32 status;

1832
	status = _nfsd4_verify(rqstp, cstate, &u->nverify);
1833 1834 1835
	return status == nfserr_same ? nfs_ok : status;
}

1836 1837 1838 1839
#ifdef CONFIG_NFSD_PNFS
static const struct nfsd4_layout_ops *
nfsd4_layout_verify(struct svc_export *exp, unsigned int layout_type)
{
1840
	if (!exp->ex_layout_types) {
1841 1842 1843 1844
		dprintk("%s: export does not support pNFS\n", __func__);
		return NULL;
	}

1845 1846
	if (layout_type >= LAYOUT_TYPE_MAX ||
	    !(exp->ex_layout_types & (1 << layout_type))) {
1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
		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,
1857
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1858
{
1859
	struct nfsd4_getdeviceinfo *gdp = &u->getdeviceinfo;
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 1889
	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;
1890 1891
	if (gdp->gd_maxcount != 0) {
		nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb,
1892
				rqstp, cstate->clp, gdp);
1893
	}
1894 1895 1896

	gdp->gd_notify_types &= ops->notify_types;
out:
1897
	exp_put(exp);
1898 1899 1900
	return nfserr;
}

1901 1902 1903 1904 1905 1906
static void
nfsd4_getdeviceinfo_release(union nfsd4_op_u *u)
{
	kfree(u->getdeviceinfo.gd_device);
}

1907 1908
static __be32
nfsd4_layoutget(struct svc_rqst *rqstp,
1909
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1910
{
1911
	struct nfsd4_layoutget *lgp = &u->layoutget;
1912 1913 1914 1915
	struct svc_fh *current_fh = &cstate->current_fh;
	const struct nfsd4_layout_ops *ops;
	struct nfs4_layout_stateid *ls;
	__be32 nfserr;
1916
	int accmode = NFSD_MAY_READ_IF_EXEC;
1917 1918 1919

	switch (lgp->lg_seg.iomode) {
	case IOMODE_READ:
1920
		accmode |= NFSD_MAY_READ;
1921 1922
		break;
	case IOMODE_RW:
1923
		accmode |= NFSD_MAY_READ | NFSD_MAY_WRITE;
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 1963
		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
1964
	if (nfserr) {
1965
		trace_nfsd_layout_get_lookup_fail(&lgp->lg_sid);
1966
		goto out;
C
Christoph Hellwig 已提交
1967
	}
1968

1969 1970 1971 1972
	nfserr = nfserr_recallconflict;
	if (atomic_read(&ls->ls_stid.sc_file->fi_lo_recalls))
		goto out_put_stid;

1973
	nfserr = ops->proc_layoutget(d_inode(current_fh->fh_dentry),
1974 1975 1976 1977 1978 1979 1980
				     current_fh, lgp);
	if (nfserr)
		goto out_put_stid;

	nfserr = nfsd4_insert_layout(lgp, ls);

out_put_stid:
1981
	mutex_unlock(&ls->ls_mutex);
1982 1983 1984 1985 1986
	nfs4_put_stid(&ls->ls_stid);
out:
	return nfserr;
}

1987 1988 1989 1990 1991 1992
static void
nfsd4_layoutget_release(union nfsd4_op_u *u)
{
	kfree(u->layoutget.lg_content);
}

1993 1994
static __be32
nfsd4_layoutcommit(struct svc_rqst *rqstp,
1995
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
1996
{
1997
	struct nfsd4_layoutcommit *lcp = &u->layoutcommit;
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
	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;
2014
	inode = d_inode(current_fh->fh_dentry);
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033

	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) {
2034
		trace_nfsd_layout_commit_lookup_fail(&lcp->lc_sid);
2035 2036 2037 2038 2039 2040
		/* fixup error code as per RFC5661 */
		if (nfserr == nfserr_bad_stateid)
			nfserr = nfserr_badlayout;
		goto out;
	}

2041 2042 2043
	/* LAYOUTCOMMIT does not require any serialization */
	mutex_unlock(&ls->ls_mutex);

2044 2045 2046 2047 2048 2049 2050
	if (new_size > i_size_read(inode)) {
		lcp->lc_size_chg = 1;
		lcp->lc_newsize = new_size;
	} else {
		lcp->lc_size_chg = 0;
	}

2051
	nfserr = ops->proc_layoutcommit(inode, lcp);
2052 2053 2054 2055 2056 2057 2058
	nfs4_put_stid(&ls->ls_stid);
out:
	return nfserr;
}

static __be32
nfsd4_layoutreturn(struct svc_rqst *rqstp,
2059
		struct nfsd4_compound_state *cstate, union nfsd4_op_u *u)
2060
{
2061
	struct nfsd4_layoutreturn *lrp = &u->layoutreturn;
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 2103
	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 */

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 2165
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
2166 2167 2168
/*
 * NULL call.
 */
A
Al Viro 已提交
2169
static __be32
2170
nfsd4_proc_null(struct svc_rqst *rqstp)
L
Linus Torvalds 已提交
2171
{
2172
	return rpc_success;
L
Linus Torvalds 已提交
2173 2174
}

2175 2176 2177
static inline void nfsd4_increment_op_stats(u32 opnum)
{
	if (opnum >= FIRST_NFS4_OP && opnum <= LAST_NFS4_OP)
2178
		percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_NFS4_OP(opnum)]);
2179 2180
}

C
Christoph Hellwig 已提交
2181
static const struct nfsd4_operation nfsd4_ops[];
2182

A
Adrian Bunk 已提交
2183
static const char *nfsd4_op_name(unsigned opnum);
B
Benny Halevy 已提交
2184

2185
/*
2186
 * Enforce NFSv4.1 COMPOUND ordering rules:
2187
 *
2188 2189 2190
 * Also note, enforced elsewhere:
 *	- SEQUENCE other than as first op results in
 *	  NFS4ERR_SEQUENCE_POS. (Enforced in nfsd4_sequence().)
2191 2192
 *	- BIND_CONN_TO_SESSION must be the only op in its compound.
 *	  (Enforced in nfsd4_bind_conn_to_session().)
2193 2194 2195
 *	- 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().)
2196
 */
2197
static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args)
2198
{
2199
	struct nfsd4_op *first_op = &args->ops[0];
2200 2201 2202 2203 2204 2205 2206

	/* 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;
2207
	if (first_op->status == nfserr_op_illegal)
2208
		return nfs_ok;
2209
	if (!(nfsd4_ops[first_op->opnum].op_flags & ALLOWED_AS_FIRST_OP))
2210
		return nfserr_op_not_in_session;
2211
	if (first_op->opnum == OP_SEQUENCE)
2212
		return nfs_ok;
2213 2214 2215 2216 2217
	/*
	 * So first_op is something allowed outside a session, like
	 * EXCHANGE_ID; but then it has to be the only op in the
	 * compound:
	 */
2218 2219 2220
	if (args->opcnt != 1)
		return nfserr_not_only_op;
	return nfs_ok;
2221 2222
}

2223
const struct nfsd4_operation *OPDESC(struct nfsd4_op *op)
J
J. Bruce Fields 已提交
2224 2225 2226 2227
{
	return &nfsd4_ops[op->opnum];
}

2228 2229
bool nfsd4_cache_this_op(struct nfsd4_op *op)
{
2230 2231
	if (op->opnum == OP_ILLEGAL)
		return false;
2232
	return OPDESC(op)->op_flags & OP_CACHEME;
2233 2234
}

2235 2236 2237 2238 2239 2240
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
2241 2242
	const struct nfsd4_operation *thisd = OPDESC(this);
	const struct nfsd4_operation *nextd;
2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256

	/*
	 * 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;
2257 2258
	if (next->opnum == OP_ILLEGAL)
		return false;
2259 2260 2261 2262 2263 2264 2265 2266 2267
	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);
}

2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306
#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
2307 2308 2309
/*
 * COMPOUND call.
 */
A
Al Viro 已提交
2310
static __be32
2311
nfsd4_proc_compound(struct svc_rqst *rqstp)
L
Linus Torvalds 已提交
2312
{
2313 2314
	struct nfsd4_compoundargs *args = rqstp->rq_argp;
	struct nfsd4_compoundres *resp = rqstp->rq_resp;
L
Linus Torvalds 已提交
2315
	struct nfsd4_op	*op;
2316
	struct nfsd4_compound_state *cstate = &resp->cstate;
2317 2318
	struct svc_fh *current_fh = &cstate->current_fh;
	struct svc_fh *save_fh = &cstate->save_fh;
2319
	struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
2320
	__be32		status;
L
Linus Torvalds 已提交
2321

2322 2323 2324 2325 2326 2327
	resp->xdr = &rqstp->rq_res_stream;

	/* reserve space for: NFS status code */
	xdr_reserve_space(resp->xdr, XDR_UNIT);

	resp->tagp = resp->xdr->p;
L
Linus Torvalds 已提交
2328
	/* reserve space for: taglen, tag, and opcnt */
2329
	xdr_reserve_space(resp->xdr, XDR_UNIT * 2 + args->taglen);
L
Linus Torvalds 已提交
2330 2331 2332
	resp->taglen = args->taglen;
	resp->tag = args->tag;
	resp->rqstp = rqstp;
2333 2334 2335
	cstate->minorversion = args->minorversion;
	fh_init(current_fh, NFS4_FHSIZE);
	fh_init(save_fh, NFS4_FHSIZE);
N
NeilBrown 已提交
2336 2337 2338 2339
	/*
	 * Don't use the deferral mechanism for NFSv4; compounds make it
	 * too hard to avoid non-idempotency problems.
	 */
2340
	clear_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
L
Linus Torvalds 已提交
2341 2342 2343 2344 2345

	/*
	 * According to RFC3010, this takes precedence over all other errors.
	 */
	status = nfserr_minor_vers_mismatch;
2346
	if (nfsd_minorversion(nn, args->minorversion, NFSD_TEST) <= 0)
L
Linus Torvalds 已提交
2347
		goto out;
2348 2349 2350
	status = nfserr_resource;
	if (args->opcnt > NFSD_MAX_OPS_PER_COMPOUND)
		goto out;
L
Linus Torvalds 已提交
2351

2352 2353
	status = nfs41_check_op_ordering(args);
	if (status) {
2354
		op = &args->ops[0];
2355
		op->status = status;
2356
		resp->opcnt = 1;
2357 2358
		goto encode_op;
	}
2359
	check_if_stalefh_allowed(args);
2360

2361 2362
	rqstp->rq_lease_breaker = (void **)&cstate->clp;

2363
	trace_nfsd_compound(rqstp, args->opcnt);
L
Linus Torvalds 已提交
2364 2365 2366 2367 2368 2369 2370 2371
	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.
		 */
2372 2373 2374
		if (op->status) {
			if (op->opnum == OP_OPEN)
				op->status = nfsd4_open_omfg(rqstp, cstate, op);
L
Linus Torvalds 已提交
2375
			goto encode_op;
2376
		}
2377 2378
		if (!current_fh->fh_dentry &&
				!HAS_FH_FLAG(current_fh, NFSD4_FH_FOREIGN)) {
2379
			if (!(op->opdesc->op_flags & ALLOWED_WITHOUT_FH)) {
2380 2381 2382
				op->status = nfserr_nofilehandle;
				goto encode_op;
			}
2383 2384
		} else if (current_fh->fh_export &&
			   current_fh->fh_export->ex_fslocs.migrated &&
2385
			  !(op->opdesc->op_flags & ALLOWED_ON_ABSENT_FS)) {
2386
			op->status = nfserr_moved;
L
Linus Torvalds 已提交
2387 2388
			goto encode_op;
		}
2389

2390 2391
		fh_clear_wcc(current_fh);

2392
		/* If op is non-idempotent */
2393
		if (op->opdesc->op_flags & OP_MODIFIES_SOMETHING) {
2394
			/*
J
J. Bruce Fields 已提交
2395 2396 2397
			 * Don't execute this op if we couldn't encode a
			 * succesful reply:
			 */
2398
			u32 plen = op->opdesc->op_rsize_bop(rqstp, op);
J
J. Bruce Fields 已提交
2399 2400
			/*
			 * Plus if there's another operation, make sure
2401 2402 2403 2404
			 * we'll have space to at least encode an error:
			 */
			if (resp->opcnt < args->opcnt)
				plen += COMPOUND_ERR_SLACK_SPACE;
2405 2406 2407 2408 2409 2410
			op->status = nfsd4_check_resp_size(resp, plen);
		}

		if (op->status)
			goto encode_op;

2411 2412 2413
		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
2414

2415 2416 2417 2418 2419 2420
		/* Only from SEQUENCE */
		if (cstate->status == nfserr_replay_cache) {
			dprintk("%s NFS4.1 replay from cache\n", __func__);
			status = op->status;
			goto out;
		}
2421
		if (!op->status) {
2422 2423
			if (op->opdesc->op_set_currentstateid)
				op->opdesc->op_set_currentstateid(cstate, &op->u);
2424

2425
			if (op->opdesc->op_flags & OP_CLEAR_STATEID)
2426
				clear_current_stateid(cstate);
2427

2428 2429
			if (current_fh->fh_export &&
					need_wrongsec_check(rqstp))
2430
				op->status = check_nfsd_access(current_fh->fh_export, rqstp);
2431
		}
L
Linus Torvalds 已提交
2432
encode_op:
A
Al Viro 已提交
2433
		if (op->status == nfserr_replay_me) {
2434
			op->replay = &cstate->replay_owner->so_replay;
2435
			nfsd4_encode_replay(resp->xdr, op);
L
Linus Torvalds 已提交
2436 2437 2438 2439 2440
			status = op->status = op->replay->rp_status;
		} else {
			nfsd4_encode_operation(resp, op);
			status = op->status;
		}
2441

2442 2443
		trace_nfsd_compound_status(args->opcnt, resp->opcnt, status,
					   nfsd4_op_name(op->opnum));
2444

2445
		nfsd4_cstate_clear_replay(cstate);
2446
		nfsd4_increment_op_stats(op->opnum);
L
Linus Torvalds 已提交
2447 2448
	}

2449 2450 2451
	fh_put(current_fh);
	fh_put(save_fh);
	BUG_ON(cstate->replay_owner);
L
Linus Torvalds 已提交
2452
out:
2453
	cstate->status = status;
2454
	/* Reset deferral mechanism for RPC deferrals */
2455
	set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
2456
	return rpc_success;
L
Linus Torvalds 已提交
2457 2458
}

2459 2460 2461 2462 2463 2464
#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)

2465 2466
/* We'll fall back on returning no lockowner if run out of space: */
#define op_encode_lockowner_maxsz	(0)
2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486
#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);
}

2487 2488 2489 2490 2491 2492
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);
}

2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503
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);
}

2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533
/*
 * 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) {
2534
		ret += NFS4_MAXLABELLEN + 12;
2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546
		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;
}

2547 2548 2549 2550 2551
static inline u32 nfsd4_getfh_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 1) * sizeof(__be32) + NFS4_FHSIZE;
}

2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576
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);
2577
	rlen = min(op->u.read.rd_length, maxcount);
2578

2579
	return (op_encode_hdr_size + 2 + XDR_QUADLEN(rlen)) * sizeof(__be32);
2580 2581
}

A
Anna Schumaker 已提交
2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595
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);
}

2596 2597
static inline u32 nfsd4_readdir_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2598
	u32 maxcount = 0, rlen = 0;
2599

2600 2601
	maxcount = svc_max_payload(rqstp);
	rlen = min(op->u.readdir.rd_maxcount, maxcount);
2602

J
J. Bruce Fields 已提交
2603 2604
	return (op_encode_hdr_size + op_encode_verifier_maxsz +
		XDR_QUADLEN(rlen)) * sizeof(__be32);
2605 2606
}

2607 2608 2609 2610 2611
static inline u32 nfsd4_readlink_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
	return (op_encode_hdr_size + 1) * sizeof(__be32) + PAGE_SIZE;
}

2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623
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);
}

2624 2625 2626
static inline u32 nfsd4_sequence_rsize(struct svc_rqst *rqstp,
				       struct nfsd4_op *op)
{
2627 2628
	return (op_encode_hdr_size
		+ XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5) * sizeof(__be32);
2629 2630
}

2631 2632 2633 2634 2635 2636
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);
}

2637 2638 2639 2640 2641
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);
}

2642 2643 2644 2645 2646 2647
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);
}

2648 2649
static inline u32 nfsd4_setclientid_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2650 2651
	return (op_encode_hdr_size + 2 + XDR_QUADLEN(NFS4_VERIFIER_SIZE)) *
								sizeof(__be32);
2652 2653 2654 2655
}

static inline u32 nfsd4_write_rsize(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
2656
	return (op_encode_hdr_size + 2 + op_encode_verifier_maxsz) * sizeof(__be32);
2657 2658 2659 2660 2661
}

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 */\
2662 2663
		1 + 1 + /* eir_flags, spr_how */\
		4 + /* spo_must_enforce & _allow with bitmap */\
2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688
		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
2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700
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
2701 2702 2703 2704 2705 2706 2707 2708
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
2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723
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);
}

2724
#ifdef CONFIG_NFSD_PNFS
2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737
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);
}

2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766
/*
 * 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 */

2767 2768 2769 2770 2771 2772

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

2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808
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
2809
static const struct nfsd4_operation nfsd4_ops[] = {
2810
	[OP_ACCESS] = {
2811
		.op_func = nfsd4_access,
B
Benny Halevy 已提交
2812
		.op_name = "OP_ACCESS",
C
Christoph Hellwig 已提交
2813
		.op_rsize_bop = nfsd4_access_rsize,
2814 2815
	},
	[OP_CLOSE] = {
2816
		.op_func = nfsd4_close,
2817
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2818
		.op_name = "OP_CLOSE",
C
Christoph Hellwig 已提交
2819
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2820
		.op_get_currentstateid = nfsd4_get_closestateid,
2821
		.op_set_currentstateid = nfsd4_set_closestateid,
2822 2823
	},
	[OP_COMMIT] = {
2824
		.op_func = nfsd4_commit,
2825
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2826
		.op_name = "OP_COMMIT",
C
Christoph Hellwig 已提交
2827
		.op_rsize_bop = nfsd4_commit_rsize,
2828 2829
	},
	[OP_CREATE] = {
2830
		.op_func = nfsd4_create,
2831
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2832
		.op_name = "OP_CREATE",
C
Christoph Hellwig 已提交
2833
		.op_rsize_bop = nfsd4_create_rsize,
2834 2835
	},
	[OP_DELEGRETURN] = {
2836
		.op_func = nfsd4_delegreturn,
2837
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2838
		.op_name = "OP_DELEGRETURN",
2839
		.op_rsize_bop = nfsd4_only_status_rsize,
2840
		.op_get_currentstateid = nfsd4_get_delegreturnstateid,
2841 2842
	},
	[OP_GETATTR] = {
2843
		.op_func = nfsd4_getattr,
2844
		.op_flags = ALLOWED_ON_ABSENT_FS,
2845
		.op_rsize_bop = nfsd4_getattr_rsize,
B
Benny Halevy 已提交
2846
		.op_name = "OP_GETATTR",
2847 2848
	},
	[OP_GETFH] = {
2849
		.op_func = nfsd4_getfh,
B
Benny Halevy 已提交
2850
		.op_name = "OP_GETFH",
C
Christoph Hellwig 已提交
2851
		.op_rsize_bop = nfsd4_getfh_rsize,
2852 2853
	},
	[OP_LINK] = {
2854
		.op_func = nfsd4_link,
2855 2856
		.op_flags = ALLOWED_ON_ABSENT_FS | OP_MODIFIES_SOMETHING
				| OP_CACHEME,
B
Benny Halevy 已提交
2857
		.op_name = "OP_LINK",
C
Christoph Hellwig 已提交
2858
		.op_rsize_bop = nfsd4_link_rsize,
2859 2860
	},
	[OP_LOCK] = {
2861
		.op_func = nfsd4_lock,
2862 2863
		.op_flags = OP_MODIFIES_SOMETHING |
				OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
2864
		.op_name = "OP_LOCK",
C
Christoph Hellwig 已提交
2865
		.op_rsize_bop = nfsd4_lock_rsize,
2866
		.op_set_currentstateid = nfsd4_set_lockstateid,
2867 2868
	},
	[OP_LOCKT] = {
2869
		.op_func = nfsd4_lockt,
2870
		.op_flags = OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
2871
		.op_name = "OP_LOCKT",
C
Christoph Hellwig 已提交
2872
		.op_rsize_bop = nfsd4_lock_rsize,
2873 2874
	},
	[OP_LOCKU] = {
2875
		.op_func = nfsd4_locku,
2876
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2877
		.op_name = "OP_LOCKU",
C
Christoph Hellwig 已提交
2878
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2879
		.op_get_currentstateid = nfsd4_get_lockustateid,
2880 2881
	},
	[OP_LOOKUP] = {
2882
		.op_func = nfsd4_lookup,
2883
		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2884
		.op_name = "OP_LOOKUP",
C
Christoph Hellwig 已提交
2885
		.op_rsize_bop = nfsd4_only_status_rsize,
2886 2887
	},
	[OP_LOOKUPP] = {
2888
		.op_func = nfsd4_lookupp,
2889
		.op_flags = OP_HANDLES_WRONGSEC | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2890
		.op_name = "OP_LOOKUPP",
C
Christoph Hellwig 已提交
2891
		.op_rsize_bop = nfsd4_only_status_rsize,
2892 2893
	},
	[OP_NVERIFY] = {
2894
		.op_func = nfsd4_nverify,
B
Benny Halevy 已提交
2895
		.op_name = "OP_NVERIFY",
C
Christoph Hellwig 已提交
2896
		.op_rsize_bop = nfsd4_only_status_rsize,
2897 2898
	},
	[OP_OPEN] = {
2899
		.op_func = nfsd4_open,
2900
		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2901
		.op_name = "OP_OPEN",
C
Christoph Hellwig 已提交
2902
		.op_rsize_bop = nfsd4_open_rsize,
2903
		.op_set_currentstateid = nfsd4_set_openstateid,
2904 2905
	},
	[OP_OPEN_CONFIRM] = {
2906
		.op_func = nfsd4_open_confirm,
2907
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2908
		.op_name = "OP_OPEN_CONFIRM",
C
Christoph Hellwig 已提交
2909
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2910 2911
	},
	[OP_OPEN_DOWNGRADE] = {
2912
		.op_func = nfsd4_open_downgrade,
2913
		.op_flags = OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2914
		.op_name = "OP_OPEN_DOWNGRADE",
C
Christoph Hellwig 已提交
2915
		.op_rsize_bop = nfsd4_status_stateid_rsize,
2916
		.op_get_currentstateid = nfsd4_get_opendowngradestateid,
2917
		.op_set_currentstateid = nfsd4_set_opendowngradestateid,
2918 2919
	},
	[OP_PUTFH] = {
2920
		.op_func = nfsd4_putfh,
2921
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2922
				| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2923
		.op_name = "OP_PUTFH",
C
Christoph Hellwig 已提交
2924
		.op_rsize_bop = nfsd4_only_status_rsize,
2925
	},
2926
	[OP_PUTPUBFH] = {
2927
		.op_func = nfsd4_putrootfh,
2928
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2929
				| OP_IS_PUTFH_LIKE | OP_CLEAR_STATEID,
B
Benny Halevy 已提交
2930
		.op_name = "OP_PUTPUBFH",
C
Christoph Hellwig 已提交
2931
		.op_rsize_bop = nfsd4_only_status_rsize,
2932
	},
2933
	[OP_PUTROOTFH] = {
2934
		.op_func = nfsd4_putrootfh,
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_PUTROOTFH",
C
Christoph Hellwig 已提交
2938
		.op_rsize_bop = nfsd4_only_status_rsize,
2939 2940
	},
	[OP_READ] = {
2941
		.op_func = nfsd4_read,
2942
		.op_release = nfsd4_read_release,
B
Benny Halevy 已提交
2943
		.op_name = "OP_READ",
C
Christoph Hellwig 已提交
2944
		.op_rsize_bop = nfsd4_read_rsize,
2945
		.op_get_currentstateid = nfsd4_get_readstateid,
2946 2947
	},
	[OP_READDIR] = {
2948
		.op_func = nfsd4_readdir,
B
Benny Halevy 已提交
2949
		.op_name = "OP_READDIR",
C
Christoph Hellwig 已提交
2950
		.op_rsize_bop = nfsd4_readdir_rsize,
2951 2952
	},
	[OP_READLINK] = {
2953
		.op_func = nfsd4_readlink,
B
Benny Halevy 已提交
2954
		.op_name = "OP_READLINK",
C
Christoph Hellwig 已提交
2955
		.op_rsize_bop = nfsd4_readlink_rsize,
2956 2957
	},
	[OP_REMOVE] = {
2958
		.op_func = nfsd4_remove,
2959
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
2960
		.op_name = "OP_REMOVE",
C
Christoph Hellwig 已提交
2961
		.op_rsize_bop = nfsd4_remove_rsize,
2962 2963
	},
	[OP_RENAME] = {
2964
		.op_func = nfsd4_rename,
2965
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
2966
		.op_name = "OP_RENAME",
C
Christoph Hellwig 已提交
2967
		.op_rsize_bop = nfsd4_rename_rsize,
2968 2969
	},
	[OP_RENEW] = {
2970
		.op_func = nfsd4_renew,
2971 2972
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
				| OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2973
		.op_name = "OP_RENEW",
C
Christoph Hellwig 已提交
2974
		.op_rsize_bop = nfsd4_only_status_rsize,
2975

2976 2977
	},
	[OP_RESTOREFH] = {
2978
		.op_func = nfsd4_restorefh,
2979
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
2980
				| OP_IS_PUTFH_LIKE | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2981
		.op_name = "OP_RESTOREFH",
C
Christoph Hellwig 已提交
2982
		.op_rsize_bop = nfsd4_only_status_rsize,
2983 2984
	},
	[OP_SAVEFH] = {
2985
		.op_func = nfsd4_savefh,
2986
		.op_flags = OP_HANDLES_WRONGSEC | OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
2987
		.op_name = "OP_SAVEFH",
C
Christoph Hellwig 已提交
2988
		.op_rsize_bop = nfsd4_only_status_rsize,
2989
	},
A
Andy Adamson 已提交
2990
	[OP_SECINFO] = {
2991
		.op_func = nfsd4_secinfo,
2992
		.op_release = nfsd4_secinfo_release,
2993
		.op_flags = OP_HANDLES_WRONGSEC,
B
Benny Halevy 已提交
2994
		.op_name = "OP_SECINFO",
C
Christoph Hellwig 已提交
2995
		.op_rsize_bop = nfsd4_secinfo_rsize,
A
Andy Adamson 已提交
2996
	},
2997
	[OP_SETATTR] = {
2998
		.op_func = nfsd4_setattr,
B
Benny Halevy 已提交
2999
		.op_name = "OP_SETATTR",
3000 3001
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME
				| OP_NONTRIVIAL_ERROR_ENCODE,
C
Christoph Hellwig 已提交
3002
		.op_rsize_bop = nfsd4_setattr_rsize,
3003
		.op_get_currentstateid = nfsd4_get_setattrstateid,
3004 3005
	},
	[OP_SETCLIENTID] = {
3006
		.op_func = nfsd4_setclientid,
3007
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
3008 3009
				| OP_MODIFIES_SOMETHING | OP_CACHEME
				| OP_NONTRIVIAL_ERROR_ENCODE,
B
Benny Halevy 已提交
3010
		.op_name = "OP_SETCLIENTID",
C
Christoph Hellwig 已提交
3011
		.op_rsize_bop = nfsd4_setclientid_rsize,
3012 3013
	},
	[OP_SETCLIENTID_CONFIRM] = {
3014
		.op_func = nfsd4_setclientid_confirm,
3015
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
3016
				| OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
3017
		.op_name = "OP_SETCLIENTID_CONFIRM",
C
Christoph Hellwig 已提交
3018
		.op_rsize_bop = nfsd4_only_status_rsize,
3019 3020
	},
	[OP_VERIFY] = {
3021
		.op_func = nfsd4_verify,
B
Benny Halevy 已提交
3022
		.op_name = "OP_VERIFY",
C
Christoph Hellwig 已提交
3023
		.op_rsize_bop = nfsd4_only_status_rsize,
3024 3025
	},
	[OP_WRITE] = {
3026
		.op_func = nfsd4_write,
3027
		.op_flags = OP_MODIFIES_SOMETHING | OP_CACHEME,
B
Benny Halevy 已提交
3028
		.op_name = "OP_WRITE",
C
Christoph Hellwig 已提交
3029
		.op_rsize_bop = nfsd4_write_rsize,
3030
		.op_get_currentstateid = nfsd4_get_writestateid,
3031 3032
	},
	[OP_RELEASE_LOCKOWNER] = {
3033
		.op_func = nfsd4_release_lockowner,
3034 3035
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS
				| OP_MODIFIES_SOMETHING,
B
Benny Halevy 已提交
3036
		.op_name = "OP_RELEASE_LOCKOWNER",
C
Christoph Hellwig 已提交
3037
		.op_rsize_bop = nfsd4_only_status_rsize,
3038
	},
A
Andy Adamson 已提交
3039 3040 3041

	/* NFSv4.1 operations */
	[OP_EXCHANGE_ID] = {
3042
		.op_func = nfsd4_exchange_id,
3043 3044
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3045
		.op_name = "OP_EXCHANGE_ID",
C
Christoph Hellwig 已提交
3046
		.op_rsize_bop = nfsd4_exchange_id_rsize,
A
Andy Adamson 已提交
3047
	},
3048
	[OP_BACKCHANNEL_CTL] = {
3049
		.op_func = nfsd4_backchannel_ctl,
3050 3051
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
		.op_name = "OP_BACKCHANNEL_CTL",
C
Christoph Hellwig 已提交
3052
		.op_rsize_bop = nfsd4_only_status_rsize,
3053
	},
3054
	[OP_BIND_CONN_TO_SESSION] = {
3055
		.op_func = nfsd4_bind_conn_to_session,
3056 3057
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
3058
		.op_name = "OP_BIND_CONN_TO_SESSION",
C
Christoph Hellwig 已提交
3059
		.op_rsize_bop = nfsd4_bind_conn_to_session_rsize,
3060
	},
A
Andy Adamson 已提交
3061
	[OP_CREATE_SESSION] = {
3062
		.op_func = nfsd4_create_session,
3063 3064
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3065
		.op_name = "OP_CREATE_SESSION",
C
Christoph Hellwig 已提交
3066
		.op_rsize_bop = nfsd4_create_session_rsize,
A
Andy Adamson 已提交
3067 3068
	},
	[OP_DESTROY_SESSION] = {
3069
		.op_func = nfsd4_destroy_session,
3070 3071
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
A
Andy Adamson 已提交
3072
		.op_name = "OP_DESTROY_SESSION",
C
Christoph Hellwig 已提交
3073
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Andy Adamson 已提交
3074 3075
	},
	[OP_SEQUENCE] = {
3076
		.op_func = nfsd4_sequence,
3077
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
A
Andy Adamson 已提交
3078
		.op_name = "OP_SEQUENCE",
C
Christoph Hellwig 已提交
3079
		.op_rsize_bop = nfsd4_sequence_rsize,
A
Andy Adamson 已提交
3080
	},
3081
	[OP_DESTROY_CLIENTID] = {
3082
		.op_func = nfsd4_destroy_clientid,
3083 3084
		.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP
				| OP_MODIFIES_SOMETHING,
3085
		.op_name = "OP_DESTROY_CLIENTID",
C
Christoph Hellwig 已提交
3086
		.op_rsize_bop = nfsd4_only_status_rsize,
3087
	},
3088
	[OP_RECLAIM_COMPLETE] = {
3089
		.op_func = nfsd4_reclaim_complete,
3090
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
3091
		.op_name = "OP_RECLAIM_COMPLETE",
C
Christoph Hellwig 已提交
3092
		.op_rsize_bop = nfsd4_only_status_rsize,
3093
	},
J
J. Bruce Fields 已提交
3094
	[OP_SECINFO_NO_NAME] = {
3095
		.op_func = nfsd4_secinfo_no_name,
3096
		.op_release = nfsd4_secinfo_no_name_release,
3097
		.op_flags = OP_HANDLES_WRONGSEC,
J
J. Bruce Fields 已提交
3098
		.op_name = "OP_SECINFO_NO_NAME",
C
Christoph Hellwig 已提交
3099
		.op_rsize_bop = nfsd4_secinfo_rsize,
J
J. Bruce Fields 已提交
3100
	},
3101
	[OP_TEST_STATEID] = {
3102
		.op_func = nfsd4_test_stateid,
3103 3104
		.op_flags = ALLOWED_WITHOUT_FH,
		.op_name = "OP_TEST_STATEID",
C
Christoph Hellwig 已提交
3105
		.op_rsize_bop = nfsd4_test_stateid_rsize,
3106
	},
3107
	[OP_FREE_STATEID] = {
3108
		.op_func = nfsd4_free_stateid,
3109
		.op_flags = ALLOWED_WITHOUT_FH | OP_MODIFIES_SOMETHING,
3110
		.op_name = "OP_FREE_STATEID",
3111
		.op_get_currentstateid = nfsd4_get_freestateid,
C
Christoph Hellwig 已提交
3112
		.op_rsize_bop = nfsd4_only_status_rsize,
3113
	},
3114 3115
#ifdef CONFIG_NFSD_PNFS
	[OP_GETDEVICEINFO] = {
3116
		.op_func = nfsd4_getdeviceinfo,
3117
		.op_release = nfsd4_getdeviceinfo_release,
3118 3119
		.op_flags = ALLOWED_WITHOUT_FH,
		.op_name = "OP_GETDEVICEINFO",
C
Christoph Hellwig 已提交
3120
		.op_rsize_bop = nfsd4_getdeviceinfo_rsize,
3121 3122
	},
	[OP_LAYOUTGET] = {
3123
		.op_func = nfsd4_layoutget,
3124
		.op_release = nfsd4_layoutget_release,
3125 3126
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTGET",
C
Christoph Hellwig 已提交
3127
		.op_rsize_bop = nfsd4_layoutget_rsize,
3128 3129
	},
	[OP_LAYOUTCOMMIT] = {
3130
		.op_func = nfsd4_layoutcommit,
3131 3132
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTCOMMIT",
C
Christoph Hellwig 已提交
3133
		.op_rsize_bop = nfsd4_layoutcommit_rsize,
3134 3135
	},
	[OP_LAYOUTRETURN] = {
3136
		.op_func = nfsd4_layoutreturn,
3137 3138
		.op_flags = OP_MODIFIES_SOMETHING,
		.op_name = "OP_LAYOUTRETURN",
C
Christoph Hellwig 已提交
3139
		.op_rsize_bop = nfsd4_layoutreturn_rsize,
3140 3141
	},
#endif /* CONFIG_NFSD_PNFS */
A
Anna Schumaker 已提交
3142 3143

	/* NFSv4.2 operations */
A
Anna Schumaker 已提交
3144
	[OP_ALLOCATE] = {
3145
		.op_func = nfsd4_allocate,
3146
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3147
		.op_name = "OP_ALLOCATE",
C
Christoph Hellwig 已提交
3148
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Anna Schumaker 已提交
3149
	},
A
Anna Schumaker 已提交
3150
	[OP_DEALLOCATE] = {
3151
		.op_func = nfsd4_deallocate,
3152
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3153
		.op_name = "OP_DEALLOCATE",
C
Christoph Hellwig 已提交
3154
		.op_rsize_bop = nfsd4_only_status_rsize,
A
Anna Schumaker 已提交
3155
	},
3156
	[OP_CLONE] = {
3157
		.op_func = nfsd4_clone,
3158
		.op_flags = OP_MODIFIES_SOMETHING,
3159
		.op_name = "OP_CLONE",
C
Christoph Hellwig 已提交
3160
		.op_rsize_bop = nfsd4_only_status_rsize,
3161
	},
A
Anna Schumaker 已提交
3162
	[OP_COPY] = {
3163
		.op_func = nfsd4_copy,
3164
		.op_flags = OP_MODIFIES_SOMETHING,
A
Anna Schumaker 已提交
3165
		.op_name = "OP_COPY",
C
Christoph Hellwig 已提交
3166
		.op_rsize_bop = nfsd4_copy_rsize,
A
Anna Schumaker 已提交
3167
	},
A
Anna Schumaker 已提交
3168 3169 3170 3171 3172 3173 3174
	[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
3175
	[OP_SEEK] = {
3176
		.op_func = nfsd4_seek,
A
Anna Schumaker 已提交
3177
		.op_name = "OP_SEEK",
C
Christoph Hellwig 已提交
3178
		.op_rsize_bop = nfsd4_seek_rsize,
A
Anna Schumaker 已提交
3179
	},
O
Olga Kornievskaia 已提交
3180 3181 3182 3183 3184
	[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
3185 3186 3187 3188 3189 3190
	[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
3191 3192 3193 3194 3195 3196
	[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,
	},
3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218
	[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,
	},
3219 3220
};

3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242
/**
 * 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
3243
	if (cstate->spo_must_allowed)
3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259
		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;
}

3260 3261
int nfsd4_max_reply(struct svc_rqst *rqstp, struct nfsd4_op *op)
{
3262
	if (op->opnum == OP_ILLEGAL || op->status == nfserr_notsupp)
3263
		return op_encode_hdr_size * sizeof(__be32);
3264 3265 3266

	BUG_ON(OPDESC(op)->op_rsize_bop == NULL);
	return OPDESC(op)->op_rsize_bop(rqstp, op);
3267 3268
}

3269 3270 3271
void warn_on_nonidempotent_op(struct nfsd4_op *op)
{
	if (OPDESC(op)->op_flags & OP_MODIFIES_SOMETHING) {
C
Chuck Lever 已提交
3272
		pr_err("unable to encode reply to nonidempotent op %u (%s)\n",
3273 3274 3275 3276 3277
			op->opnum, nfsd4_op_name(op->opnum));
		WARN_ON_ONCE(1);
	}
}

A
Adrian Bunk 已提交
3278
static const char *nfsd4_op_name(unsigned opnum)
B
Benny Halevy 已提交
3279 3280 3281 3282 3283 3284
{
	if (opnum < ARRAY_SIZE(nfsd4_ops))
		return nfsd4_ops[opnum].op_name;
	return "unknown_operation";
}

3285
static const struct svc_procedure nfsd_procedures4[2] = {
3286
	[NFSPROC4_NULL] = {
3287
		.pc_func = nfsd4_proc_null,
3288 3289 3290 3291
		.pc_decode = nfssvc_decode_voidarg,
		.pc_encode = nfssvc_encode_voidres,
		.pc_argsize = sizeof(struct nfsd_voidargs),
		.pc_ressize = sizeof(struct nfsd_voidres),
3292 3293
		.pc_cachetype = RC_NOCACHE,
		.pc_xdrressize = 1,
3294
		.pc_name = "NULL",
3295 3296
	},
	[NFSPROC4_COMPOUND] = {
3297
		.pc_func = nfsd4_proc_compound,
3298
		.pc_decode = nfs4svc_decode_compoundargs,
3299
		.pc_encode = nfs4svc_encode_compoundres,
3300 3301
		.pc_argsize = sizeof(struct nfsd4_compoundargs),
		.pc_ressize = sizeof(struct nfsd4_compoundres),
3302
		.pc_release = nfsd4_release_compoundargs,
3303 3304
		.pc_cachetype = RC_NOCACHE,
		.pc_xdrressize = NFSD_BUFSIZE/4,
3305
		.pc_name = "COMPOUND",
3306
	},
L
Linus Torvalds 已提交
3307 3308
};

3309
static unsigned int nfsd_count3[ARRAY_SIZE(nfsd_procedures4)];
3310
const struct svc_version nfsd_version4 = {
3311 3312 3313
	.vs_vers		= 4,
	.vs_nproc		= 2,
	.vs_proc		= nfsd_procedures4,
3314
	.vs_count		= nfsd_count3,
3315 3316 3317 3318
	.vs_dispatch		= nfsd_dispatch,
	.vs_xdrsize		= NFS4_SVC_XDRSIZE,
	.vs_rpcb_optnl		= true,
	.vs_need_cong_ctrl	= true,
L
Linus Torvalds 已提交
3319 3320 3321 3322 3323 3324 3325
};

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