Merge tag 'nfsd-4.19-1' of git://linux-nfs.org/~bfields/linux

Pull nfsd updates from Bruce Fields:
"Chuck Lever fixed a problem with NFSv4.0 callbacks over GSS from
multi-homed servers.

The only new feature is a minor bit of protocol (change_attr_type)
which the client doesn't even use yet.

Other than that, various bugfixes and cleanup"

* tag 'nfsd-4.19-1' of git://linux-nfs.org/~bfields/linux: (27 commits)
sunrpc: Add comment defining gssd upcall API keywords
nfsd: Remove callback_cred
nfsd: Use correct credential for NFSv4.0 callback with GSS
sunrpc: Extract target name into svc_cred
sunrpc: Enable the kernel to specify the hostname part of service principals
sunrpc: Don't use stack buffer with scatterlist
rpc: remove unneeded variable 'ret' in rdma_listen_handler
nfsd: use true and false for boolean values
nfsd: constify write_op[]
fs/nfsd: Delete invalid assignment statements in nfsd4_decode_exchange_id
NFSD: Handle full-length symlinks
NFSD: Refactor the generic write vector fill helper
svcrdma: Clean up Read chunk path
svcrdma: Avoid releasing a page in svc_xprt_release()
nfsd: Mark expected switch fall-through
sunrpc: remove redundant variables 'checksumlen','blocksize' and 'data'
nfsd: fix leaked file lock with nfs exported overlayfs
nfsd: don't advertise a SCSI layout for an unsupported request_queue
nfsd: fix corrupted reply to badly ordered compound
nfsd: clarify check_op_ordering
...

fs/lockd/clntlock.c

···
 			continue;
 		if (!rpc_cmp_addr(nlm_addr(block->b_host), addr))
 			continue;
-		if (nfs_compare_fh(NFS_FH(file_inode(fl_blocked->fl_file)) ,fh) != 0)
+		if (nfs_compare_fh(NFS_FH(locks_inode(fl_blocked->fl_file)), fh) != 0)
 			continue;
 		/* Alright, we found a lock. Set the return status
 		 * and wake up the caller

fs/lockd/clntproc.c

···
 	char *nodename = req->a_host->h_rpcclnt->cl_nodename;
 
 	nlmclnt_next_cookie(&argp->cookie);
-	memcpy(&lock->fh, NFS_FH(file_inode(fl->fl_file)), sizeof(struct nfs_fh));
+	memcpy(&lock->fh, NFS_FH(locks_inode(fl->fl_file)), sizeof(struct nfs_fh));
 	lock->caller  = nodename;
 	lock->oh.data = req->a_owner;
 	lock->oh.len  = snprintf(req->a_owner, sizeof(req->a_owner), "%u@%s",

fs/lockd/svclock.c

···
 	__be32			ret;
 
 	dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n",
-				file_inode(file->f_file)->i_sb->s_id,
-				file_inode(file->f_file)->i_ino,
+				locks_inode(file->f_file)->i_sb->s_id,
+				locks_inode(file->f_file)->i_ino,
 				lock->fl.fl_type, lock->fl.fl_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end,
···
 	__be32			ret;
 
 	dprintk("lockd: nlmsvc_testlock(%s/%ld, ty=%d, %Ld-%Ld)\n",
-				file_inode(file->f_file)->i_sb->s_id,
-				file_inode(file->f_file)->i_ino,
+				locks_inode(file->f_file)->i_sb->s_id,
+				locks_inode(file->f_file)->i_ino,
 				lock->fl.fl_type,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);
···
 	int	error;
 
 	dprintk("lockd: nlmsvc_unlock(%s/%ld, pi=%d, %Ld-%Ld)\n",
-				file_inode(file->f_file)->i_sb->s_id,
-				file_inode(file->f_file)->i_ino,
+				locks_inode(file->f_file)->i_sb->s_id,
+				locks_inode(file->f_file)->i_ino,
 				lock->fl.fl_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);
···
 	int status = 0;
 
 	dprintk("lockd: nlmsvc_cancel(%s/%ld, pi=%d, %Ld-%Ld)\n",
-				file_inode(file->f_file)->i_sb->s_id,
-				file_inode(file->f_file)->i_ino,
+				locks_inode(file->f_file)->i_sb->s_id,
+				locks_inode(file->f_file)->i_ino,
 				lock->fl.fl_pid,
 				(long long)lock->fl.fl_start,
 				(long long)lock->fl.fl_end);

fs/lockd/svcsubs.c

···
 
 static inline void nlm_debug_print_file(char *msg, struct nlm_file *file)
 {
-	struct inode *inode = file_inode(file->f_file);
+	struct inode *inode = locks_inode(file->f_file);
 
 	dprintk("lockd: %s %s/%ld\n",
 		msg, inode->i_sb->s_id, inode->i_ino);
···
 {
 	struct super_block *sb = datap;
 
-	return sb == file_inode(file->f_file)->i_sb;
+	return sb == locks_inode(file->f_file)->i_sb;
 }
 
 /**

fs/nfsd/netns.h

···
 
 	time_t nfsd4_lease;
 	time_t nfsd4_grace;
+	bool somebody_reclaimed;
 
 	bool nfsd_net_up;
 	bool lockd_up;

fs/nfsd/nfs3proc.c

···
 
 	fh_copy(&resp->fh, &argp->fh);
 	resp->committed = argp->stable;
-	nvecs = svc_fill_write_vector(rqstp, &argp->first, cnt);
+	nvecs = svc_fill_write_vector(rqstp, rqstp->rq_arg.pages,
+				      &argp->first, cnt);
 	if (!nvecs)
 		RETURN_STATUS(nfserr_io);
 	nfserr = nfsd_write(rqstp, &resp->fh, argp->offset,
···
 		RETURN_STATUS(nfserr_nametoolong);
 
 	argp->tname = svc_fill_symlink_pathname(rqstp, &argp->first,
+						page_address(rqstp->rq_arg.pages[0]),
 						argp->tlen);
 	if (IS_ERR(argp->tname))
 		RETURN_STATUS(nfserrno(PTR_ERR(argp->tname)));
···
 	fh_init(&resp->fh, NFS3_FHSIZE);
 	nfserr = nfsd_symlink(rqstp, &resp->dirfh, argp->fname, argp->flen,
 						   argp->tname, &resp->fh);
+	kfree(argp->tname);
 	RETURN_STATUS(nfserr);
 }
 

fs/nfsd/nfs4callback.c

···
 	return max(nn->nfsd4_lease/10, (time_t)1) * HZ;
 }
 
-static struct rpc_cred *callback_cred;
-
-int set_callback_cred(void)
-{
-	if (callback_cred)
-		return 0;
-	callback_cred = rpc_lookup_machine_cred("nfs");
-	if (!callback_cred)
-		return -ENOMEM;
-	return 0;
-}
-
-void cleanup_callback_cred(void)
-{
-	if (callback_cred) {
-		put_rpccred(callback_cred);
-		callback_cred = NULL;
-	}
-}
-
 static struct rpc_cred *get_backchannel_cred(struct nfs4_client *clp, struct rpc_clnt *client, struct nfsd4_session *ses)
 {
 	if (clp->cl_minorversion == 0) {
-		return get_rpccred(callback_cred);
+		char *principal = clp->cl_cred.cr_targ_princ ?
+					clp->cl_cred.cr_targ_princ : "nfs";
+		struct rpc_cred *cred;
+
+		cred = rpc_lookup_machine_cred(principal);
+		if (!IS_ERR(cred))
+			get_rpccred(cred);
+		return cred;
 	} else {
 		struct rpc_auth *auth = client->cl_auth;
 		struct auth_cred acred = {};
···
 		break;
 	case -ESERVERFAULT:
 		++session->se_cb_seq_nr;
+		/* Fall through */
 	case 1:
 	case -NFS4ERR_BADSESSION:
 		nfsd4_mark_cb_fault(cb->cb_clp, cb->cb_seq_status);

fs/nfsd/nfs4layouts.c

···
 	if (!(exp->ex_flags & NFSEXP_PNFS))
 		return;
 
-	/*
-	 * If flex file is configured, use it by default. Otherwise
-	 * check if the file system supports exporting a block-like layout.
-	 * If the block device supports reservations prefer the SCSI layout,
-	 * otherwise advertise the block layout.
-	 */
 #ifdef CONFIG_NFSD_FLEXFILELAYOUT
 	exp->ex_layout_types |= 1 << LAYOUT_FLEX_FILES;
 #endif
 #ifdef CONFIG_NFSD_BLOCKLAYOUT
-	/* overwrite flex file layout selection if needed */
 	if (sb->s_export_op->get_uuid &&
 	    sb->s_export_op->map_blocks &&
 	    sb->s_export_op->commit_blocks)
 		exp->ex_layout_types |= 1 << LAYOUT_BLOCK_VOLUME;
 #endif
 #ifdef CONFIG_NFSD_SCSILAYOUT
-	/* overwrite block layout selection if needed */
 	if (sb->s_export_op->map_blocks &&
 	    sb->s_export_op->commit_blocks &&
-	    sb->s_bdev && sb->s_bdev->bd_disk->fops->pr_ops)
+	    sb->s_bdev && sb->s_bdev->bd_disk->fops->pr_ops &&
+		blk_queue_scsi_passthrough(sb->s_bdev->bd_disk->queue))
 		exp->ex_layout_types |= 1 << LAYOUT_SCSI;
 #endif
 }

fs/nfsd/nfs4proc.c

···
 	struct svc_fh *resfh = NULL;
 	struct net *net = SVC_NET(rqstp);
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
+	bool reclaim = false;
 
 	dprintk("NFSD: nfsd4_open filename %.*s op_openowner %p\n",
 		(int)open->op_fname.len, open->op_fname.data,
···
 			if (status)
 				goto out;
 			open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED;
+			reclaim = true;
 		case NFS4_OPEN_CLAIM_FH:
 		case NFS4_OPEN_CLAIM_DELEG_CUR_FH:
 			status = do_open_fhandle(rqstp, cstate, open);
···
 	WARN(status && open->op_created,
 	     "nfsd4_process_open2 failed to open newly-created file! status=%u\n",
 	     be32_to_cpu(status));
+	if (reclaim && !status)
+		nn->somebody_reclaimed = true;
 out:
 	if (resfh && resfh != &cstate->current_fh) {
 		fh_dup2(&cstate->current_fh, resfh);
···
 	return status;
 }
 
-static int fill_in_write_vector(struct kvec *vec, struct nfsd4_write *write)
-{
-        int i = 1;
-        int buflen = write->wr_buflen;
-
-        vec[0].iov_base = write->wr_head.iov_base;
-        vec[0].iov_len = min_t(int, buflen, write->wr_head.iov_len);
-        buflen -= vec[0].iov_len;
-
-        while (buflen) {
-                vec[i].iov_base = page_address(write->wr_pagelist[i - 1]);
-                vec[i].iov_len = min_t(int, PAGE_SIZE, buflen);
-                buflen -= vec[i].iov_len;
-                i++;
-        }
-        return i;
-}
-
 static __be32
 nfsd4_write(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	    union nfsd4_op_u *u)
···
 	write->wr_how_written = write->wr_stable_how;
 	gen_boot_verifier(&write->wr_verifier, SVC_NET(rqstp));
 
-	nvecs = fill_in_write_vector(rqstp->rq_vec, write);
+	nvecs = svc_fill_write_vector(rqstp, write->wr_pagelist,
+				      &write->wr_head, write->wr_buflen);
+	if (!nvecs)
+		return nfserr_io;
 	WARN_ON_ONCE(nvecs > ARRAY_SIZE(rqstp->rq_vec));
 
 	status = nfsd_vfs_write(rqstp, &cstate->current_fh, filp,
···
  */
 static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args)
 {
-	struct nfsd4_op *op = &args->ops[0];
+	struct nfsd4_op *first_op = &args->ops[0];
 
 	/* These ordering requirements don't apply to NFSv4.0: */
 	if (args->minorversion == 0)
···
 	/* This is weird, but OK, not our problem: */
 	if (args->opcnt == 0)
 		return nfs_ok;
-	if (op->status == nfserr_op_illegal)
+	if (first_op->status == nfserr_op_illegal)
 		return nfs_ok;
-	if (!(nfsd4_ops[op->opnum].op_flags & ALLOWED_AS_FIRST_OP))
+	if (!(nfsd4_ops[first_op->opnum].op_flags & ALLOWED_AS_FIRST_OP))
 		return nfserr_op_not_in_session;
-	if (op->opnum == OP_SEQUENCE)
+	if (first_op->opnum == OP_SEQUENCE)
 		return nfs_ok;
+	/*
+	 * So first_op is something allowed outside a session, like
+	 * EXCHANGE_ID; but then it has to be the only op in the
+	 * compound:
+	 */
 	if (args->opcnt != 1)
 		return nfserr_not_only_op;
 	return nfs_ok;
···
 	if (status) {
 		op = &args->ops[0];
 		op->status = status;
+		resp->opcnt = 1;
 		goto encode_op;
 	}
 

fs/nfsd/nfs4state.c

···
 	target->cr_principal = kstrdup(source->cr_principal, GFP_KERNEL);
 	target->cr_raw_principal = kstrdup(source->cr_raw_principal,
 								GFP_KERNEL);
-	if ((source->cr_principal && ! target->cr_principal) ||
-	    (source->cr_raw_principal && ! target->cr_raw_principal))
+	target->cr_targ_princ = kstrdup(source->cr_targ_princ, GFP_KERNEL);
+	if ((source->cr_principal && !target->cr_principal) ||
+	    (source->cr_raw_principal && !target->cr_raw_principal) ||
+	    (source->cr_targ_princ && !target->cr_targ_princ))
 		return -ENOMEM;
 
 	target->cr_flavor = source->cr_flavor;
···
 		|| (!gid_eq(cr1->cr_gid, cr2->cr_gid))
 		|| !groups_equal(cr1->cr_group_info, cr2->cr_group_info))
 		return false;
+	/* XXX: check that cr_targ_princ fields match ? */
 	if (cr1->cr_principal == cr2->cr_principal)
 		return true;
 	if (!cr1->cr_principal || !cr2->cr_principal)
···
 	return status;
 }
 
-static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid)
+static bool nfsd4_compound_in_session(struct nfsd4_compound_state *cstate, struct nfs4_sessionid *sid)
 {
-	if (!session)
+	if (!cstate->session)
 		return false;
-	return !memcmp(sid, &session->se_sessionid, sizeof(*sid));
+	return !memcmp(sid, &cstate->session->se_sessionid, sizeof(*sid));
 }
 
 __be32
 nfsd4_destroy_session(struct svc_rqst *r, struct nfsd4_compound_state *cstate,
 		union nfsd4_op_u *u)
 {
-	struct nfsd4_destroy_session *sessionid = &u->destroy_session;
+	struct nfs4_sessionid *sessionid = &u->destroy_session.sessionid;
 	struct nfsd4_session *ses;
 	__be32 status;
 	int ref_held_by_me = 0;
···
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	status = nfserr_not_only_op;
-	if (nfsd4_compound_in_session(cstate->session, &sessionid->sessionid)) {
+	if (nfsd4_compound_in_session(cstate, sessionid)) {
 		if (!nfsd4_last_compound_op(r))
 			goto out;
 		ref_held_by_me++;
 	}
-	dump_sessionid(__func__, &sessionid->sessionid);
+	dump_sessionid(__func__, sessionid);
 	spin_lock(&nn->client_lock);
-	ses = find_in_sessionid_hashtbl(&sessionid->sessionid, net, &status);
+	ses = find_in_sessionid_hashtbl(sessionid, net, &status);
 	if (!ses)
 		goto out_client_lock;
 	status = nfserr_wrong_cred;
···
 	/*
 	 * We're assuming the state code never drops its reference
 	 * without first removing the lease.  Since we're in this lease
-	 * callback (and since the lease code is serialized by the kernel
-	 * lock) we know the server hasn't removed the lease yet, we know
-	 * it's safe to take a reference.
+	 * callback (and since the lease code is serialized by the
+	 * i_lock) we know the server hasn't removed the lease yet, and
+	 * we know it's safe to take a reference.
 	 */
 	refcount_inc(&dp->dl_stid.sc_count);
 	nfsd4_run_cb(&dp->dl_recall);
···
 	 */
 }
 
+/*
+ * If we've waited a lease period but there are still clients trying to
+ * reclaim, wait a little longer to give them a chance to finish.
+ */
+static bool clients_still_reclaiming(struct nfsd_net *nn)
+{
+	unsigned long now = get_seconds();
+	unsigned long double_grace_period_end = nn->boot_time +
+						2 * nn->nfsd4_lease;
+
+	if (!nn->somebody_reclaimed)
+		return false;
+	nn->somebody_reclaimed = false;
+	/*
+	 * If we've given them *two* lease times to reclaim, and they're
+	 * still not done, give up:
+	 */
+	if (time_after(now, double_grace_period_end))
+		return false;
+	return true;
+}
+
 static time_t
 nfs4_laundromat(struct nfsd_net *nn)
 {
···
 	time_t t, new_timeo = nn->nfsd4_lease;
 
 	dprintk("NFSD: laundromat service - starting\n");
+
+	if (clients_still_reclaiming(nn)) {
+		new_timeo = 0;
+		goto out;
+	}
 	nfsd4_end_grace(nn);
 	INIT_LIST_HEAD(&reaplist);
 	spin_lock(&nn->client_lock);
···
 		posix_unblock_lock(&nbl->nbl_lock);
 		free_blocked_lock(nbl);
 	}
-
+out:
 	new_timeo = max_t(time_t, new_timeo, NFSD_LAUNDROMAT_MINTIMEOUT);
 	return new_timeo;
 }
···
 	case 0: /* success! */
 		nfs4_inc_and_copy_stateid(&lock->lk_resp_stateid, &lock_stp->st_stid);
 		status = 0;
+		if (lock->lk_reclaim)
+			nn->somebody_reclaimed = true;
 		break;
 	case FILE_LOCK_DEFERRED:
 		nbl = NULL;
···
 		return status;
 	}
 
-	inode = file_inode(filp);
+	inode = locks_inode(filp);
 	flctx = inode->i_flctx;
 
 	if (flctx && !list_empty_careful(&flctx->flc_posix)) {
···
 {
 	int ret;
 
-	ret = set_callback_cred();
-	if (ret)
-		return ret;
-
 	laundry_wq = alloc_workqueue("%s", WQ_UNBOUND, 0, "nfsd4");
 	if (laundry_wq == NULL) {
 		ret = -ENOMEM;
-		goto out_cleanup_cred;
+		goto out;
 	}
 	ret = nfsd4_create_callback_queue();
 	if (ret)
···
 
 out_free_laundry:
 	destroy_workqueue(laundry_wq);
-out_cleanup_cred:
-	cleanup_callback_cred();
+out:
 	return ret;
 }
 
···
 {
 	destroy_workqueue(laundry_wq);
 	nfsd4_destroy_callback_queue();
-	cleanup_callback_cred();
 }
 
 static void

fs/nfsd/nfs4xdr.c

···
 			p += XDR_QUADLEN(dummy);
 		}
 
-		/* ssp_window and ssp_num_gss_handles */
+		/* ignore ssp_window and ssp_num_gss_handles: */
 		READ_BUF(8);
-		dummy = be32_to_cpup(p++);
-		dummy = be32_to_cpup(p++);
 		break;
 	default:
 		goto xdr_error;
···
 		*p++ = cpu_to_be32(stat->ctime.tv_sec);
 		*p++ = cpu_to_be32(stat->ctime.tv_nsec);
 	}
+	return p;
+}
+
+/*
+ * ctime (in NFSv4, time_metadata) is not writeable, and the client
+ * doesn't really care what resolution could theoretically be stored by
+ * the filesystem.
+ *
+ * The client cares how close together changes can be while still
+ * guaranteeing ctime changes.  For most filesystems (which have
+ * timestamps with nanosecond fields) that is limited by the resolution
+ * of the time returned from current_time() (which I'm assuming to be
+ * 1/HZ).
+ */
+static __be32 *encode_time_delta(__be32 *p, struct inode *inode)
+{
+	struct timespec ts;
+	u32 ns;
+
+	ns = max_t(u32, NSEC_PER_SEC/HZ, inode->i_sb->s_time_gran);
+	ts = ns_to_timespec(ns);
+
+	p = xdr_encode_hyper(p, ts.tv_sec);
+	*p++ = cpu_to_be32(ts.tv_nsec);
+
 	return p;
 }
 
···
 		p = xdr_reserve_space(xdr, 12);
 		if (!p)
 			goto out_resource;
-		*p++ = cpu_to_be32(0);
-		*p++ = cpu_to_be32(1);
-		*p++ = cpu_to_be32(0);
+		p = encode_time_delta(p, d_inode(dentry));
 	}
 	if (bmval1 & FATTR4_WORD1_TIME_METADATA) {
 		p = xdr_reserve_space(xdr, 12);
···
 		status = nfsd4_encode_bitmap(xdr, supp[0], supp[1], supp[2]);
 		if (status)
 			goto out;
+	}
+
+	if (bmval2 & FATTR4_WORD2_CHANGE_ATTR_TYPE) {
+		p = xdr_reserve_space(xdr, 4);
+		if (!p)
+			goto out_resource;
+		if (IS_I_VERSION(d_inode(dentry)))
+			*p++ = cpu_to_be32(NFS4_CHANGE_TYPE_IS_MONOTONIC_INCR);
+		else
+			*p++ = cpu_to_be32(NFS4_CHANGE_TYPE_IS_TIME_METADATA);
 	}
 
 	if (bmval2 & FATTR4_WORD2_SECURITY_LABEL) {

fs/nfsd/nfsctl.c

···
 static ssize_t write_v4_end_grace(struct file *file, char *buf, size_t size);
 #endif
 
-static ssize_t (*write_op[])(struct file *, char *, size_t) = {
+static ssize_t (*const write_op[])(struct file *, char *, size_t) = {
 	[NFSD_Fh] = write_filehandle,
 	[NFSD_FO_UnlockIP] = write_unlock_ip,
 	[NFSD_FO_UnlockFS] = write_unlock_fs,
···
 	retval = nfsd_idmap_init(net);
 	if (retval)
 		goto out_idmap_error;
-	nn->nfsd4_lease = 90;	/* default lease time */
-	nn->nfsd4_grace = 90;
+	nn->nfsd4_lease = 45;	/* default lease time */
+	nn->nfsd4_grace = 45;
+	nn->somebody_reclaimed = false;
 	nn->clverifier_counter = prandom_u32();
 	nn->clientid_counter = prandom_u32();
 

fs/nfsd/nfsd.h

···
 
 #define NFSD4_2_SUPPORTED_ATTRS_WORD2 \
 	(NFSD4_1_SUPPORTED_ATTRS_WORD2 | \
+	FATTR4_WORD2_CHANGE_ATTR_TYPE | \
 	FATTR4_WORD2_MODE_UMASK | \
 	NFSD4_2_SECURITY_ATTRS)
 

fs/nfsd/nfsfh.c

···
 	switch (fsid_type) {
 	case FSID_DEV:
 		if (!old_valid_dev(exp_sb(exp)->s_dev))
-			return 0;
+			return false;
 		/* FALL THROUGH */
 	case FSID_MAJOR_MINOR:
 	case FSID_ENCODE_DEV:
···
 	case FSID_UUID8:
 	case FSID_UUID16:
 		if (!is_root_export(exp))
-			return 0;
+			return false;
 		/* fall through */
 	case FSID_UUID4_INUM:
 	case FSID_UUID16_INUM:
 		return exp->ex_uuid != NULL;
 	}
-	return 1;
+	return true;
 }
 
 

fs/nfsd/nfsproc.c

···
 		SVCFH_fmt(&argp->fh),
 		argp->len, argp->offset);
 
-	nvecs = svc_fill_write_vector(rqstp, &argp->first, cnt);
+	nvecs = svc_fill_write_vector(rqstp, rqstp->rq_arg.pages,
+				      &argp->first, cnt);
 	if (!nvecs)
 		return nfserr_io;
 	nfserr = nfsd_write(rqstp, fh_copy(&resp->fh, &argp->fh),
···
 		return nfserr_nametoolong;
 
 	argp->tname = svc_fill_symlink_pathname(rqstp, &argp->first,
+						page_address(rqstp->rq_arg.pages[0]),
 						argp->tlen);
 	if (IS_ERR(argp->tname))
 		return nfserrno(PTR_ERR(argp->tname));
···
 	nfserr = nfsd_symlink(rqstp, &argp->ffh, argp->fname, argp->flen,
 						 argp->tname, &newfh);
 
+	kfree(argp->tname);
 	fh_put(&argp->ffh);
 	fh_put(&newfh);
 	return nfserr;

fs/nfsd/state.h

···
 							struct nfsd_net *nn);
 extern __be32 nfs4_check_open_reclaim(clientid_t *clid,
 		struct nfsd4_compound_state *cstate, struct nfsd_net *nn);
-extern int set_callback_cred(void);
-extern void cleanup_callback_cred(void);
 extern void nfsd4_probe_callback(struct nfs4_client *clp);
 extern void nfsd4_probe_callback_sync(struct nfs4_client *clp);
 extern void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *);

include/linux/lockd/lockd.h

···
 
 static inline struct inode *nlmsvc_file_inode(struct nlm_file *file)
 {
-	return file_inode(file->f_file);
+	return locks_inode(file->f_file);
 }
 
 static inline int __nlm_privileged_request4(const struct sockaddr *sap)
···
 static inline int nlm_compare_locks(const struct file_lock *fl1,
 				    const struct file_lock *fl2)
 {
-	return file_inode(fl1->fl_file) == file_inode(fl2->fl_file)
+	return locks_inode(fl1->fl_file) == locks_inode(fl2->fl_file)
 	     && fl1->fl_pid   == fl2->fl_pid
 	     && fl1->fl_owner == fl2->fl_owner
 	     && fl1->fl_start == fl2->fl_start

include/linux/nfs4.h

···
 	NFS4_WRITEW_LT = 4
 };
 
+enum change_attr_type4 {
+	NFS4_CHANGE_TYPE_IS_MONOTONIC_INCR = 0,
+	NFS4_CHANGE_TYPE_IS_VERSION_COUNTER = 1,
+	NFS4_CHANGE_TYPE_IS_VERSION_COUNTER_NOPNFS = 2,
+	NFS4_CHANGE_TYPE_IS_TIME_METADATA = 3,
+	NFS4_CHANGE_TYPE_IS_UNDEFINED = 4
+};
 
 /* Mandatory Attributes */
 #define FATTR4_WORD0_SUPPORTED_ATTRS    (1UL << 0)
···
 #define FATTR4_WORD2_LAYOUT_BLKSIZE     (1UL << 1)
 #define FATTR4_WORD2_MDSTHRESHOLD       (1UL << 4)
 #define FATTR4_WORD2_CLONE_BLKSIZE	(1UL << 13)
+#define FATTR4_WORD2_CHANGE_ATTR_TYPE	(1UL << 15)
 #define FATTR4_WORD2_SECURITY_LABEL     (1UL << 16)
 #define FATTR4_WORD2_MODE_UMASK		(1UL << 17)
 

include/linux/sunrpc/svc.h

···
 struct svc_pool *  svc_pool_for_cpu(struct svc_serv *serv, int cpu);
 char *		   svc_print_addr(struct svc_rqst *, char *, size_t);
 unsigned int	   svc_fill_write_vector(struct svc_rqst *rqstp,
+					 struct page **pages,
 					 struct kvec *first, size_t total);
 char		  *svc_fill_symlink_pathname(struct svc_rqst *rqstp,
-					     struct kvec *first, size_t total);
+					     struct kvec *first, void *p,
+					     size_t total);
 
 #define	RPC_MAX_ADDRBUFLEN	(63U)
 

include/linux/sunrpc/svcauth.h

···
 	/* name of form servicetype@hostname, passed down by
 	 * rpc.svcgssd, or computed from the above: */
 	char			*cr_principal;
+	char			*cr_targ_princ;
 	struct gss_api_mech	*cr_gss_mech;
 };
 
···
 	cred->cr_group_info = NULL;
 	cred->cr_raw_principal = NULL;
 	cred->cr_principal = NULL;
+	cred->cr_targ_princ = NULL;
 	cred->cr_gss_mech = NULL;
 }
 
···
 		put_group_info(cred->cr_group_info);
 	kfree(cred->cr_raw_principal);
 	kfree(cred->cr_principal);
+	kfree(cred->cr_targ_princ);
 	gss_mech_put(cred->cr_gss_mech);
 	init_svc_cred(cred);
 }

net/sunrpc/auth_gss/auth_gss.c

···
 	return p;
 }
 
-#define UPCALL_BUF_LEN 128
+/* XXX: Need some documentation about why UPCALL_BUF_LEN is so small.
+ *	Is user space expecting no more than UPCALL_BUF_LEN bytes?
+ *	Note that there are now _two_ NI_MAXHOST sized data items
+ *	being passed in this string.
+ */
+#define UPCALL_BUF_LEN	256
 
 struct gss_upcall_msg {
 	refcount_t count;
···
 	buflen -= len;
 	p += len;
 	gss_msg->msg.len = len;
+
+	/*
+	 * target= is a full service principal that names the remote
+	 * identity that we are authenticating to.
+	 */
 	if (target_name) {
 		len = scnprintf(p, buflen, "target=%s ", target_name);
 		buflen -= len;
 		p += len;
 		gss_msg->msg.len += len;
 	}
-	if (service_name != NULL) {
-		len = scnprintf(p, buflen, "service=%s ", service_name);
+
+	/*
+	 * gssd uses service= and srchost= to select a matching key from
+	 * the system's keytab to use as the source principal.
+	 *
+	 * service= is the service name part of the source principal,
+	 * or "*" (meaning choose any).
+	 *
+	 * srchost= is the hostname part of the source principal. When
+	 * not provided, gssd uses the local hostname.
+	 */
+	if (service_name) {
+		char *c = strchr(service_name, '@');
+
+		if (!c)
+			len = scnprintf(p, buflen, "service=%s ",
+					service_name);
+		else
+			len = scnprintf(p, buflen,
+					"service=%.*s srchost=%s ",
+					(int)(c - service_name),
+					service_name, c + 1);
 		buflen -= len;
 		p += len;
 		gss_msg->msg.len += len;
 	}
+
 	if (mech->gm_upcall_enctypes) {
 		len = scnprintf(p, buflen, "enctypes=%s ",
 				mech->gm_upcall_enctypes);

net/sunrpc/auth_gss/gss_krb5_crypto.c

···
 	struct scatterlist              sg[1];
 	int err = -1;
 	u8 *checksumdata;
-	u8 rc4salt[4];
+	u8 *rc4salt;
 	struct crypto_ahash *md5;
 	struct crypto_ahash *hmac_md5;
 	struct ahash_request *req;
···
 		return GSS_S_FAILURE;
 	}
 
+	rc4salt = kmalloc_array(4, sizeof(*rc4salt), GFP_NOFS);
+	if (!rc4salt)
+		return GSS_S_FAILURE;
+
 	if (arcfour_hmac_md5_usage_to_salt(usage, rc4salt)) {
 		dprintk("%s: invalid usage value %u\n", __func__, usage);
-		return GSS_S_FAILURE;
+		goto out_free_rc4salt;
 	}
 
 	checksumdata = kmalloc(GSS_KRB5_MAX_CKSUM_LEN, GFP_NOFS);
 	if (!checksumdata)
-		return GSS_S_FAILURE;
+		goto out_free_rc4salt;
 
 	md5 = crypto_alloc_ahash("md5", 0, CRYPTO_ALG_ASYNC);
 	if (IS_ERR(md5))
···
 	crypto_free_ahash(md5);
 out_free_cksum:
 	kfree(checksumdata);
+out_free_rc4salt:
+	kfree(rc4salt);
 	return err ? GSS_S_FAILURE : 0;
 }
 
···
 	struct scatterlist sg[1];
 	int err = -1;
 	u8 *checksumdata;
-	unsigned int checksumlen;
 
 	if (kctx->gk5e->keyed_cksum == 0) {
 		dprintk("%s: expected keyed hash for %s\n",
···
 	tfm = crypto_alloc_ahash(kctx->gk5e->cksum_name, 0, CRYPTO_ALG_ASYNC);
 	if (IS_ERR(tfm))
 		goto out_free_cksum;
-	checksumlen = crypto_ahash_digestsize(tfm);
 
 	req = ahash_request_alloc(tfm, GFP_NOFS);
 	if (!req)

net/sunrpc/auth_gss/gss_krb5_wrap.c

···
 gss_wrap_kerberos_v2(struct krb5_ctx *kctx, u32 offset,
 		     struct xdr_buf *buf, struct page **pages)
 {
-	int		blocksize;
 	u8		*ptr, *plainhdr;
 	s32		now;
 	u8		flags = 0x00;
···
 	*ptr++ = 0xff;
 	be16ptr = (__be16 *)ptr;
 
-	blocksize = crypto_skcipher_blocksize(kctx->acceptor_enc);
 	*be16ptr++ = 0;
 	/* "inner" token header always uses 0 for RRC */
 	*be16ptr++ = 0;

net/sunrpc/auth_gss/gss_rpc_upcall.c

···
 	return 0;
 }
 
+static char *gssp_stringify(struct xdr_netobj *netobj)
+{
+	return kstrndup(netobj->data, netobj->len, GFP_KERNEL);
+}
+
+static void gssp_hostbased_service(char **principal)
+{
+	char *c;
+
+	if (!*principal)
+		return;
+
+	/* terminate and remove realm part */
+	c = strchr(*principal, '@');
+	if (c) {
+		*c = '\0';
+
+		/* change service-hostname delimiter */
+		c = strchr(*principal, '/');
+		if (c)
+			*c = '@';
+	}
+	if (!c) {
+		/* not a service principal */
+		kfree(*principal);
+		*principal = NULL;
+	}
+}
+
 /*
  * Public functions
  */
···
 		 */
 		.exported_context_token.len = GSSX_max_output_handle_sz,
 		.mech.len = GSS_OID_MAX_LEN,
+		.targ_name.display_name.len = GSSX_max_princ_sz,
 		.src_name.display_name.len = GSSX_max_princ_sz
 	};
 	struct gssx_res_accept_sec_context res = {
···
 		.rpc_cred = NULL, /* FIXME ? */
 	};
 	struct xdr_netobj client_name = { 0 , NULL };
+	struct xdr_netobj target_name = { 0, NULL };
 	int ret;
 
 	if (data->in_handle.len != 0)
···
 	ret = gssp_alloc_receive_pages(&arg);
 	if (ret)
 		return ret;
-
-	/* use nfs/ for targ_name ? */
 
 	ret = gssp_call(net, &msg);
 
···
 			kfree(rctxh.mech.data);
 		}
 		client_name = rctxh.src_name.display_name;
+		target_name = rctxh.targ_name.display_name;
 	}
 
 	if (res.options.count == 1) {
···
 	}
 
 	/* convert to GSS_NT_HOSTBASED_SERVICE form and set into creds */
-	if (data->found_creds && client_name.data != NULL) {
-		char *c;
-
-		data->creds.cr_raw_principal = kstrndup(client_name.data,
-						client_name.len, GFP_KERNEL);
-
-		data->creds.cr_principal = kstrndup(client_name.data,
-						client_name.len, GFP_KERNEL);
-		if (data->creds.cr_principal) {
-			/* terminate and remove realm part */
-			c = strchr(data->creds.cr_principal, '@');
-			if (c) {
-				*c = '\0';
-
-				/* change service-hostname delimiter */
-				c = strchr(data->creds.cr_principal, '/');
-				if (c) *c = '@';
-			}
-			if (!c) {
-				/* not a service principal */
-				kfree(data->creds.cr_principal);
-				data->creds.cr_principal = NULL;
-			}
+	if (data->found_creds) {
+		if (client_name.data) {
+			data->creds.cr_raw_principal =
+					gssp_stringify(&client_name);
+			data->creds.cr_principal =
+					gssp_stringify(&client_name);
+			gssp_hostbased_service(&data->creds.cr_principal);
+		}
+		if (target_name.data) {
+			data->creds.cr_targ_princ =
+					gssp_stringify(&target_name);
+			gssp_hostbased_service(&data->creds.cr_targ_princ);
 		}
 	}
 	kfree(client_name.data);
+	kfree(target_name.data);
 
 	return ret;
 }

net/sunrpc/svc.c

···
 /**
  * svc_fill_write_vector - Construct data argument for VFS write call
  * @rqstp: svc_rqst to operate on
+ * @pages: list of pages containing data payload
  * @first: buffer containing first section of write payload
  * @total: total number of bytes of write payload
  *
- * Returns the number of elements populated in the data argument array.
+ * Fills in rqstp::rq_vec, and returns the number of elements.
  */
-unsigned int svc_fill_write_vector(struct svc_rqst *rqstp, struct kvec *first,
-				   size_t total)
+unsigned int svc_fill_write_vector(struct svc_rqst *rqstp, struct page **pages,
+				   struct kvec *first, size_t total)
 {
 	struct kvec *vec = rqstp->rq_vec;
-	struct page **pages;
 	unsigned int i;
 
 	/* Some types of transport can present the write payload
···
 		++i;
 	}
 
-	WARN_ON_ONCE(rqstp->rq_arg.page_base != 0);
-	pages = rqstp->rq_arg.pages;
 	while (total) {
 		vec[i].iov_base = page_address(*pages);
 		vec[i].iov_len = min_t(size_t, total, PAGE_SIZE);
 		total -= vec[i].iov_len;
 		++i;
-
 		++pages;
 	}
 
···
  * svc_fill_symlink_pathname - Construct pathname argument for VFS symlink call
  * @rqstp: svc_rqst to operate on
  * @first: buffer containing first section of pathname
+ * @p: buffer containing remaining section of pathname
  * @total: total length of the pathname argument
  *
- * Returns pointer to a NUL-terminated string, or an ERR_PTR. The buffer is
- * released automatically when @rqstp is recycled.
+ * The VFS symlink API demands a NUL-terminated pathname in mapped memory.
+ * Returns pointer to a NUL-terminated string, or an ERR_PTR. Caller must free
+ * the returned string.
  */
 char *svc_fill_symlink_pathname(struct svc_rqst *rqstp, struct kvec *first,
-				size_t total)
+				void *p, size_t total)
 {
-	struct xdr_buf *arg = &rqstp->rq_arg;
-	struct page **pages;
-	char *result;
+	size_t len, remaining;
+	char *result, *dst;
 
-	/* VFS API demands a NUL-terminated pathname. This function
-	 * uses a page from @rqstp as the pathname buffer, to enable
-	 * direct placement. Thus the total buffer size is PAGE_SIZE.
-	 * Space in this buffer for NUL-termination requires that we
-	 * cap the size of the returned symlink pathname just a
-	 * little early.
-	 */
-	if (total > PAGE_SIZE - 1)
-		return ERR_PTR(-ENAMETOOLONG);
+	result = kmalloc(total + 1, GFP_KERNEL);
+	if (!result)
+		return ERR_PTR(-ESERVERFAULT);
 
-	/* Some types of transport can present the pathname entirely
-	 * in rq_arg.pages. If not, then copy the pathname into one
-	 * page.
-	 */
-	pages = arg->pages;
-	WARN_ON_ONCE(arg->page_base != 0);
-	if (first->iov_base == 0) {
-		result = page_address(*pages);
-		result[total] = '\0';
-	} else {
-		size_t len, remaining;
-		char *dst;
+	dst = result;
+	remaining = total;
 
-		result = page_address(*(rqstp->rq_next_page++));
-		dst = result;
-		remaining = total;
-
-		len = min_t(size_t, total, first->iov_len);
+	len = min_t(size_t, total, first->iov_len);
+	if (len) {
 		memcpy(dst, first->iov_base, len);
 		dst += len;
 		remaining -= len;
-
-		/* No more than one page left */
-		if (remaining) {
-			len = min_t(size_t, remaining, PAGE_SIZE);
-			memcpy(dst, page_address(*pages), len);
-			dst += len;
-		}
-
-		*dst = '\0';
 	}
 
-	/* Sanity check: we don't allow the pathname argument to
+	if (remaining) {
+		len = min_t(size_t, remaining, PAGE_SIZE);
+		memcpy(dst, p, len);
+		dst += len;
+	}
+
+	*dst = '\0';
+
+	/* Sanity check: Linux doesn't allow the pathname argument to
 	 * contain a NUL byte.
 	 */
-	if (strlen(result) != total)
+	if (strlen(result) != total) {
+		kfree(result);
 		return ERR_PTR(-EINVAL);
+	}
 	return result;
 }
 EXPORT_SYMBOL_GPL(svc_fill_symlink_pathname);

net/sunrpc/xprtrdma/svc_rdma.c

···
 		atomic_set(stat, 0);
 	else {
 		char str_buf[32];
-		char *data;
 		int len = snprintf(str_buf, 32, "%d\n", atomic_read(stat));
 		if (len >= 32)
 			return -EFAULT;
···
 			*lenp = 0;
 			return 0;
 		}
-		data = &str_buf[*ppos];
 		len -= *ppos;
 		if (len > *lenp)
 			len = *lenp;

net/sunrpc/xprtrdma/svc_rdma_recvfrom.c

···
 	arg->page_base = 0;
 	arg->buflen = ctxt->rc_byte_len;
 	arg->len = ctxt->rc_byte_len;
-
-	rqstp->rq_respages = &rqstp->rq_pages[0];
-	rqstp->rq_next_page = rqstp->rq_respages + 1;
 }
 
 /* This accommodates the largest possible Write chunk,
···
 	atomic_inc(&rdma_stat_recv);
 
 	svc_rdma_build_arg_xdr(rqstp, ctxt);
+
+	/* Prevent svc_xprt_release from releasing pages in rq_pages
+	 * if we return 0 or an error.
+	 */
+	rqstp->rq_respages = rqstp->rq_pages;
+	rqstp->rq_next_page = rqstp->rq_respages;
 
 	p = (__be32 *)rqstp->rq_arg.head[0].iov_base;
 	ret = svc_rdma_xdr_decode_req(&rqstp->rq_arg);

net/sunrpc/xprtrdma/svc_rdma_rw.c

···
 				     struct svc_rdma_read_info *info,
 				     __be32 *p)
 {
+	unsigned int i;
 	int ret;
 
 	ret = -EINVAL;
···
 		trace_svcrdma_encode_rseg(rs_handle, rs_length, rs_offset);
 		info->ri_chunklen += rs_length;
 	}
+
+	/* Pages under I/O have been copied to head->rc_pages.
+	 * Prevent their premature release by svc_xprt_release() .
+	 */
+	for (i = 0; i < info->ri_readctxt->rc_page_count; i++)
+		rqstp->rq_pages[i] = NULL;
 
 	return ret;
 }
···
 			     struct svc_rdma_recv_ctxt *head, __be32 *p)
 {
 	struct svc_rdma_read_info *info;
-	struct page **page;
 	int ret;
 
 	/* The request (with page list) is constructed in
···
 		ret = svc_rdma_build_normal_read_chunk(rqstp, info, p);
 	else
 		ret = svc_rdma_build_pz_read_chunk(rqstp, info, p);
-
-	/* Mark the start of the pages that can be used for the reply */
-	if (info->ri_pageoff > 0)
-		info->ri_pageno++;
-	rqstp->rq_respages = &rqstp->rq_pages[info->ri_pageno];
-	rqstp->rq_next_page = rqstp->rq_respages + 1;
-
 	if (ret < 0)
-		goto out;
+		goto out_err;
 
 	ret = svc_rdma_post_chunk_ctxt(&info->ri_cc);
-
-out:
-	/* Read sink pages have been moved from rqstp->rq_pages to
-	 * head->rc_arg.pages. Force svc_recv to refill those slots
-	 * in rq_pages.
-	 */
-	for (page = rqstp->rq_pages; page < rqstp->rq_respages; page++)
-		*page = NULL;
-
 	if (ret < 0)
-		svc_rdma_read_info_free(info);
+		goto out_err;
+	return 0;
+
+out_err:
+	svc_rdma_read_info_free(info);
 	return ret;
 }

net/sunrpc/xprtrdma/svc_rdma_sendto.c

···
 		ctxt->sc_pages[i] = rqstp->rq_respages[i];
 		rqstp->rq_respages[i] = NULL;
 	}
-	rqstp->rq_next_page = rqstp->rq_respages + 1;
+
+	/* Prevent svc_xprt_release from releasing pages in rq_pages */
+	rqstp->rq_next_page = rqstp->rq_respages;
 }
 
 /* Prepare the portion of the RPC Reply that will be transmitted

net/sunrpc/xprtrdma/svc_rdma_transport.c

···
 			       struct rdma_cm_event *event)
 {
 	struct sockaddr *sap = (struct sockaddr *)&cma_id->route.addr.src_addr;
-	int ret = 0;
 
 	trace_svcrdma_cm_event(event, sap);
 
···
 		break;
 	}
 
-	return ret;
+	return 0;
 }
 
 static int rdma_cma_handler(struct rdma_cm_id *cma_id,