Merge tag 'nfsd-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux

Pull nfsd updates from Chuck Lever:
 "NFSD is finally able to offer write delegations to clients that open
  files with O_WRONLY, thanks to patches from Dai Ngo. We're expecting
  this to accelerate a few interesting corner cases.

  The cap on the number of operations per NFSv4 COMPOUND has been
  lifted. Now, clients that send COMPOUNDs containing dozens of
  operations (for example, a long stream of LOOKUP operations to walk a
  pathname in a single round trip) will no longer be rejected.

  This release re-enables the ability for NFSD to perform NFSv4.2 COPY
  operations asynchronously. This feature has been disabled to mitigate
  the risk of denial-of-service when too many such requests arrive.

  Many thanks to the contributors, reviewers, testers, and bug reporters
  who participated during the v6.17 development cycle"

* tag 'nfsd-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (32 commits)
  nfsd: Drop dprintk in blocklayout xdr functions
  sunrpc: make svc_tcp_sendmsg() take a signed sentp pointer
  sunrpc: rearrange struct svc_rqst for fewer cachelines
  sunrpc: return better error in svcauth_gss_accept() on alloc failure
  sunrpc: reset rq_accept_statp when starting a new RPC
  sunrpc: remove SVC_SYSERR
  sunrpc: fix handling of unknown auth status codes
  NFSD: Simplify struct knfsd_fh
  NFSD: Access a knfsd_fh's fsid by pointer
  Revert "NFSD: Force all NFSv4.2 COPY requests to be synchronous"
  NFSD: Avoid multiple -Wflex-array-member-not-at-end warnings
  NFSD: Use vfs_iocb_iter_write()
  NFSD: Use vfs_iocb_iter_read()
  NFSD: Clean up kdoc for nfsd_open_local_fh()
  NFSD: Clean up kdoc for nfsd_file_put_local()
  NFSD: Remove definition for trace_nfsd_ctl_maxconn
  NFSD: Remove definition for trace_nfsd_file_gc_recent
  NFSD: Remove definitions for unused trace_nfsd_file_lru trace points
  NFSD: Remove definition for trace_nfsd_file_unhash_and_queue
  nfsd: Use correct error code when decoding extents
  ...

fs/nfsd/blocklayout.c

@@ -178,11 +178,13 @@ nfsd4_block_proc_layoutcommit(struct inode *inode,
 {
 	struct iomap *iomaps;
 	int nr_iomaps;
+	__be32 nfserr;
 
-	nr_iomaps = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
-			lcp->lc_up_len, &iomaps, i_blocksize(inode));
-	if (nr_iomaps < 0)
-		return nfserrno(nr_iomaps);
+	nfserr = nfsd4_block_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, &nr_iomaps,
+			i_blocksize(inode));
+	if (nfserr != nfs_ok)
+		return nfserr;
 
 	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
 }
@@ -316,11 +318,13 @@ nfsd4_scsi_proc_layoutcommit(struct inode *inode,
 {
 	struct iomap *iomaps;
 	int nr_iomaps;
+	__be32 nfserr;
 
-	nr_iomaps = nfsd4_scsi_decode_layoutupdate(lcp->lc_up_layout,
-			lcp->lc_up_len, &iomaps, i_blocksize(inode));
-	if (nr_iomaps < 0)
-		return nfserrno(nr_iomaps);
+	nfserr = nfsd4_scsi_decode_layoutupdate(lcp->lc_up_layout,
+			lcp->lc_up_len, &iomaps, &nr_iomaps,
+			i_blocksize(inode));
+	if (nfserr != nfs_ok)
+		return nfserr;
 
 	return nfsd4_block_commit_blocks(inode, lcp, iomaps, nr_iomaps);
 }

fs/nfsd/blocklayoutxdr.c

@@ -112,35 +112,46 @@ nfsd4_block_encode_getdeviceinfo(struct xdr_stream *xdr,
 	return 0;
 }
 
-int
+/**
+ * nfsd4_block_decode_layoutupdate - decode the block layout extent array
+ * @p: pointer to the xdr data
+ * @len: number of bytes to decode
+ * @iomapp: pointer to store the decoded extent array
+ * @nr_iomapsp: pointer to store the number of extents
+ * @block_size: alignment of extent offset and length
+ *
+ * This function decodes the opaque field of the layoutupdate4 structure
+ * in a layoutcommit request for the block layout driver. The field is
+ * actually an array of extents sent by the client. It also checks that
+ * the file offset, storage offset and length of each extent are aligned
+ * by @block_size.
+ *
+ * Return values:
+ *   %nfs_ok: Successful decoding, @iomapp and @nr_iomapsp are valid
+ *   %nfserr_bad_xdr: The encoded array in @p is invalid
+ *   %nfserr_inval: An unaligned extent found
+ *   %nfserr_delay: Failed to allocate memory for @iomapp
+ */
+__be32
 nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size)
+		int *nr_iomapsp, u32 block_size)
 {
 	struct iomap *iomaps;
 	u32 nr_iomaps, i;
 
-	if (len < sizeof(u32)) {
-		dprintk("%s: extent array too small: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len < sizeof(u32))
+		return nfserr_bad_xdr;
 	len -= sizeof(u32);
-	if (len % PNFS_BLOCK_EXTENT_SIZE) {
-		dprintk("%s: extent array invalid: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len % PNFS_BLOCK_EXTENT_SIZE)
+		return nfserr_bad_xdr;
 
 	nr_iomaps = be32_to_cpup(p++);
-	if (nr_iomaps != len / PNFS_BLOCK_EXTENT_SIZE) {
-		dprintk("%s: extent array size mismatch: %u/%u\n",
-			__func__, len, nr_iomaps);
-		return -EINVAL;
-	}
+	if (nr_iomaps != len / PNFS_BLOCK_EXTENT_SIZE)
+		return nfserr_bad_xdr;
 
 	iomaps = kcalloc(nr_iomaps, sizeof(*iomaps), GFP_KERNEL);
-	if (!iomaps) {
-		dprintk("%s: failed to allocate extent array\n", __func__);
-		return -ENOMEM;
-	}
+	if (!iomaps)
+		return nfserr_delay;
 
 	for (i = 0; i < nr_iomaps; i++) {
 		struct pnfs_block_extent bex;
@@ -150,26 +161,18 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 
 		p = xdr_decode_hyper(p, &bex.foff);
 		if (bex.foff & (block_size - 1)) {
-			dprintk("%s: unaligned offset 0x%llx\n",
-				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.len);
 		if (bex.len & (block_size - 1)) {
-			dprintk("%s: unaligned length 0x%llx\n",
-				__func__, bex.foff);
 			goto fail;
 		}
 		p = xdr_decode_hyper(p, &bex.soff);
 		if (bex.soff & (block_size - 1)) {
-			dprintk("%s: unaligned disk offset 0x%llx\n",
-				__func__, bex.soff);
 			goto fail;
 		}
 		bex.es = be32_to_cpup(p++);
 		if (bex.es != PNFS_BLOCK_READWRITE_DATA) {
-			dprintk("%s: incorrect extent state %d\n",
-				__func__, bex.es);
 			goto fail;
 		}
 
@@ -178,59 +181,71 @@ nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
 	}
 
 	*iomapp = iomaps;
-	return nr_iomaps;
+	*nr_iomapsp = nr_iomaps;
+	return nfs_ok;
 fail:
 	kfree(iomaps);
-	return -EINVAL;
+	return nfserr_inval;
 }
 
-int
+/**
+ * nfsd4_scsi_decode_layoutupdate - decode the scsi layout extent array
+ * @p: pointer to the xdr data
+ * @len: number of bytes to decode
+ * @iomapp: pointer to store the decoded extent array
+ * @nr_iomapsp: pointer to store the number of extents
+ * @block_size: alignment of extent offset and length
+ *
+ * This function decodes the opaque field of the layoutupdate4 structure
+ * in a layoutcommit request for the scsi layout driver. The field is
+ * actually an array of extents sent by the client. It also checks that
+ * the offset and length of each extent are aligned by @block_size.
+ *
+ * Return values:
+ *   %nfs_ok: Successful decoding, @iomapp and @nr_iomapsp are valid
+ *   %nfserr_bad_xdr: The encoded array in @p is invalid
+ *   %nfserr_inval: An unaligned extent found
+ *   %nfserr_delay: Failed to allocate memory for @iomapp
+ */
+__be32
 nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size)
+		int *nr_iomapsp, u32 block_size)
 {
 	struct iomap *iomaps;
 	u32 nr_iomaps, expected, i;
 
-	if (len < sizeof(u32)) {
-		dprintk("%s: extent array too small: %u\n", __func__, len);
-		return -EINVAL;
-	}
+	if (len < sizeof(u32))
+		return nfserr_bad_xdr;
 
 	nr_iomaps = be32_to_cpup(p++);
 	expected = sizeof(__be32) + nr_iomaps * PNFS_SCSI_RANGE_SIZE;
-	if (len != expected) {
-		dprintk("%s: extent array size mismatch: %u/%u\n",
-			__func__, len, expected);
-		return -EINVAL;
-	}
+	if (len != expected)
+		return nfserr_bad_xdr;
 
 	iomaps = kcalloc(nr_iomaps, sizeof(*iomaps), GFP_KERNEL);
-	if (!iomaps) {
-		dprintk("%s: failed to allocate extent array\n", __func__);
-		return -ENOMEM;
-	}
+	if (!iomaps)
+		return nfserr_delay;
 
 	for (i = 0; i < nr_iomaps; i++) {
 		u64 val;
 
 		p = xdr_decode_hyper(p, &val);
 		if (val & (block_size - 1)) {
-			dprintk("%s: unaligned offset 0x%llx\n", __func__, val);
 			goto fail;
 		}
 		iomaps[i].offset = val;
 
 		p = xdr_decode_hyper(p, &val);
 		if (val & (block_size - 1)) {
-			dprintk("%s: unaligned length 0x%llx\n", __func__, val);
 			goto fail;
 		}
 		iomaps[i].length = val;
 	}
 
 	*iomapp = iomaps;
-	return nr_iomaps;
+	*nr_iomapsp = nr_iomaps;
+	return nfs_ok;
 fail:
 	kfree(iomaps);
-	return -EINVAL;
+	return nfserr_inval;
 }

fs/nfsd/blocklayoutxdr.h

@@ -54,9 +54,9 @@ __be32 nfsd4_block_encode_getdeviceinfo(struct xdr_stream *xdr,
 		const struct nfsd4_getdeviceinfo *gdp);
 __be32 nfsd4_block_encode_layoutget(struct xdr_stream *xdr,
 		const struct nfsd4_layoutget *lgp);
-int nfsd4_block_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size);
-int nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len, struct iomap **iomapp,
-		u32 block_size);
+__be32 nfsd4_block_decode_layoutupdate(__be32 *p, u32 len,
+		struct iomap **iomapp, int *nr_iomapsp, u32 block_size);
+__be32 nfsd4_scsi_decode_layoutupdate(__be32 *p, u32 len,
+		struct iomap **iomapp, int *nr_iomapsp, u32 block_size);
 
 #endif /* _NFSD_BLOCKLAYOUTXDR_H */

fs/nfsd/export.c

@@ -82,8 +82,7 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	int len;
 	struct auth_domain *dom = NULL;
 	int err;
-	int fsidtype;
-	char *ep;
+	u8 fsidtype;
 	struct svc_expkey key;
 	struct svc_expkey *ek = NULL;
 
@@ -109,10 +108,9 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen)
 	err = -EINVAL;
 	if (qword_get(&mesg, buf, PAGE_SIZE) <= 0)
 		goto out;
-	fsidtype = simple_strtoul(buf, &ep, 10);
-	if (*ep)
+	if (kstrtou8(buf, 10, &fsidtype))
 		goto out;
-	dprintk("found fsidtype %d\n", fsidtype);
+	dprintk("found fsidtype %u\n", fsidtype);
 	if (key_len(fsidtype)==0) /* invalid type */
 		goto out;
 	if ((len=qword_get(&mesg, buf, PAGE_SIZE)) <= 0)

fs/nfsd/export.h

@@ -88,7 +88,7 @@ struct svc_expkey {
 	struct cache_head	h;
 
 	struct auth_domain *	ek_client;
-	int			ek_fsidtype;
+	u8			ek_fsidtype;
 	u32			ek_fsid[6];
 
 	struct path		ek_path;

fs/nfsd/filecache.c

@@ -372,7 +372,7 @@ nfsd_file_put(struct nfsd_file *nf)
 
 /**
  * nfsd_file_put_local - put nfsd_file reference and arm nfsd_net_put in caller
- * @nf: nfsd_file of which to put the reference
+ * @pnf: nfsd_file of which to put the reference
  *
  * First save the associated net to return to caller, then put
  * the reference of the nfsd_file.

fs/nfsd/localio.c

@@ -32,7 +32,7 @@
  * @rpc_clnt: rpc_clnt that the client established
  * @cred: cred that the client established
  * @nfs_fh: filehandle to lookup
- * @nfp: place to find the nfsd_file, or store it if it was non-NULL
+ * @pnf: place to find the nfsd_file, or store it if it was non-NULL
  * @fmode: fmode_t to use for open
  *
  * This function maps a local fh to a path on a local filesystem.

fs/nfsd/nfs3proc.c

@@ -561,7 +561,7 @@ static void nfsd3_init_dirlist_pages(struct svc_rqst *rqstp,
 	buf->pages = rqstp->rq_next_page;
 	rqstp->rq_next_page += (buf->buflen + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
-	xdr_init_encode_pages(xdr, buf, buf->pages,  NULL);
+	xdr_init_encode_pages(xdr, buf);
 }
 
 /*

fs/nfsd/nfs4layouts.c

@@ -65,7 +65,7 @@ nfsd4_alloc_devid_map(const struct svc_fh *fhp)
 		return;
 
 	map->fsid_type = fh->fh_fsid_type;
-	memcpy(&map->fsid, fh->fh_fsid, fsid_len);
+	memcpy(&map->fsid, fh_fsid(fh), fsid_len);
 
 	spin_lock(&nfsd_devid_lock);
 	if (fhp->fh_export->ex_devid_map)
@@ -75,7 +75,7 @@ nfsd4_alloc_devid_map(const struct svc_fh *fhp)
 		list_for_each_entry(old, &nfsd_devid_hash[i], hash) {
 			if (old->fsid_type != fh->fh_fsid_type)
 				continue;
-			if (memcmp(old->fsid, fh->fh_fsid,
+			if (memcmp(old->fsid, fh_fsid(fh),
 					key_len(old->fsid_type)))
 				continue;
 

fs/nfsd/nfs4proc.c

@@ -1917,13 +1917,6 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	struct nfsd42_write_res *result;
 	__be32 status;
 
-	/*
-	 * Currently, async COPY is not reliable. Force all COPY
-	 * requests to be synchronous to avoid client application
-	 * hangs waiting for COPY completion.
-	 */
-	nfsd4_copy_set_sync(copy, true);
-
 	result = &copy->cp_res;
 	nfsd_copy_write_verifier((__be32 *)&result->wr_verifier.data, nn);
 
@@ -2842,20 +2835,10 @@ nfsd4_proc_compound(struct svc_rqst *rqstp)
 
 	rqstp->rq_lease_breaker = (void **)&cstate->clp;
 
-	trace_nfsd_compound(rqstp, args->tag, args->taglen, args->client_opcnt);
+	trace_nfsd_compound(rqstp, args->tag, args->taglen, args->opcnt);
 	while (!status && resp->opcnt < args->opcnt) {
 		op = &args->ops[resp->opcnt++];
 
-		if (unlikely(resp->opcnt == NFSD_MAX_OPS_PER_COMPOUND)) {
-			/* If there are still more operations to process,
-			 * stop here and report NFS4ERR_RESOURCE. */
-			if (cstate->minorversion == 0 &&
-			    args->client_opcnt > resp->opcnt) {
-				op->status = nfserr_resource;
-				goto encode_op;
-			}
-		}
-
 		/*
 		 * The XDR decode routines may have pre-set op->status;
 		 * for example, if there is a miscellaneous XDR error
@@ -2932,7 +2915,7 @@ encode_op:
 			status = op->status;
 		}
 
-		trace_nfsd_compound_status(args->client_opcnt, resp->opcnt,
+		trace_nfsd_compound_status(args->opcnt, resp->opcnt,
 					   status, nfsd4_op_name(op->opnum));
 
 		nfsd4_cstate_clear_replay(cstate);

fs/nfsd/nfs4state.c

@@ -633,18 +633,6 @@ find_readable_file(struct nfs4_file *f)
 	return ret;
 }
 
-static struct nfsd_file *
-find_rw_file(struct nfs4_file *f)
-{
-	struct nfsd_file *ret;
-
-	spin_lock(&f->fi_lock);
-	ret = nfsd_file_get(f->fi_fds[O_RDWR]);
-	spin_unlock(&f->fi_lock);
-
-	return ret;
-}
-
 struct nfsd_file *
 find_any_file(struct nfs4_file *f)
 {
@@ -1218,15 +1206,20 @@ nfs4_inc_and_copy_stateid(stateid_t *dst, struct nfs4_stid *stid)
 
 static void put_deleg_file(struct nfs4_file *fp)
 {
+	struct nfsd_file *rnf = NULL;
 	struct nfsd_file *nf = NULL;
 
 	spin_lock(&fp->fi_lock);
-	if (--fp->fi_delegees == 0)
+	if (--fp->fi_delegees == 0) {
 		swap(nf, fp->fi_deleg_file);
+		swap(rnf, fp->fi_rdeleg_file);
+	}
 	spin_unlock(&fp->fi_lock);
 
 	if (nf)
 		nfsd_file_put(nf);
+	if (rnf)
+		nfs4_file_put_access(fp, NFS4_SHARE_ACCESS_READ);
 }
 
 static void nfs4_unlock_deleg_lease(struct nfs4_delegation *dp)
@@ -3872,7 +3865,6 @@ static __be32 check_forechannel_attrs(struct nfsd4_channel_attrs *ca, struct nfs
 	ca->headerpadsz = 0;
 	ca->maxreq_sz = min_t(u32, ca->maxreq_sz, maxrpc);
 	ca->maxresp_sz = min_t(u32, ca->maxresp_sz, maxrpc);
-	ca->maxops = min_t(u32, ca->maxops, NFSD_MAX_OPS_PER_COMPOUND);
 	ca->maxresp_cached = min_t(u32, ca->maxresp_cached,
 			NFSD_SLOT_CACHE_SIZE + NFSD_MIN_HDR_SEQ_SZ);
 	ca->maxreqs = min_t(u32, ca->maxreqs, NFSD_MAX_SLOTS_PER_SESSION);
@@ -4697,10 +4689,16 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 	}
 	status = nfs_ok;
 	if (conf) {
-		old = unconf;
-		unhash_client_locked(old);
-		nfsd4_change_callback(conf, &unconf->cl_cb_conn);
-	} else {
+		if (get_client_locked(conf) == nfs_ok) {
+			old = unconf;
+			unhash_client_locked(old);
+			nfsd4_change_callback(conf, &unconf->cl_cb_conn);
+		} else {
+			conf = NULL;
+		}
+	}
+
+	if (!conf) {
 		old = find_confirmed_client_by_name(&unconf->cl_name, nn);
 		if (old) {
 			status = nfserr_clid_inuse;
@@ -4717,10 +4715,14 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 			}
 			trace_nfsd_clid_replaced(&old->cl_clientid);
 		}
+		status = get_client_locked(unconf);
+		if (status != nfs_ok) {
+			old = NULL;
+			goto out;
+		}
 		move_to_confirmed(unconf);
 		conf = unconf;
 	}
-	get_client_locked(conf);
 	spin_unlock(&nn->client_lock);
 	if (conf == unconf)
 		fsnotify_dentry(conf->cl_nfsd_info_dentry, FS_MODIFY);
@@ -4750,6 +4752,7 @@ static void nfsd4_file_init(const struct svc_fh *fh, struct nfs4_file *fp)
 	INIT_LIST_HEAD(&fp->fi_clnt_odstate);
 	fh_copy_shallow(&fp->fi_fhandle, &fh->fh_handle);
 	fp->fi_deleg_file = NULL;
+	fp->fi_rdeleg_file = NULL;
 	fp->fi_had_conflict = false;
 	fp->fi_share_deny = 0;
 	memset(fp->fi_fds, 0, sizeof(fp->fi_fds));
@@ -6000,14 +6003,19 @@ nfs4_set_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
 	 *  "An OPEN_DELEGATE_WRITE delegation allows the client to handle,
 	 *   on its own, all opens."
 	 *
-	 * Furthermore the client can use a write delegation for most READ
-	 * operations as well, so we require a O_RDWR file here.
+	 * Furthermore, section 9.1.2 says:
 	 *
-	 * Offer a write delegation in the case of a BOTH open, and ensure
-	 * we get the O_RDWR descriptor.
+	 *  "In the case of READ, the server may perform the corresponding
+	 *  check on the access mode, or it may choose to allow READ for
+	 *  OPEN4_SHARE_ACCESS_WRITE, to accommodate clients whose WRITE
+	 *  implementation may unavoidably do reads (e.g., due to buffer
+	 *  cache constraints)."
+	 *
+	 *  We choose to offer a write delegation for OPEN with the
+	 *  OPEN4_SHARE_ACCESS_WRITE access mode to accommodate such clients.
 	 */
-	if ((open->op_share_access & NFS4_SHARE_ACCESS_BOTH) == NFS4_SHARE_ACCESS_BOTH) {
-		nf = find_rw_file(fp);
+	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) {
+		nf = find_writeable_file(fp);
 		dl_type = deleg_ts ? OPEN_DELEGATE_WRITE_ATTRS_DELEG : OPEN_DELEGATE_WRITE;
 	}
 
@@ -6138,7 +6146,7 @@ static bool
 nfs4_delegation_stat(struct nfs4_delegation *dp, struct svc_fh *currentfh,
 		     struct kstat *stat)
 {
-	struct nfsd_file *nf = find_rw_file(dp->dl_stid.sc_file);
+	struct nfsd_file *nf = find_writeable_file(dp->dl_stid.sc_file);
 	struct path path;
 	int rc;
 
@@ -6156,6 +6164,34 @@ nfs4_delegation_stat(struct nfs4_delegation *dp, struct svc_fh *currentfh,
 	return rc == 0;
 }
 
+/*
+ * Add NFS4_SHARE_ACCESS_READ to the write delegation granted on OPEN
+ * with NFS4_SHARE_ACCESS_WRITE by allocating separate nfsd_file and
+ * struct file to be used for read with delegation stateid.
+ *
+ */
+static bool
+nfsd4_add_rdaccess_to_wrdeleg(struct svc_rqst *rqstp, struct nfsd4_open *open,
+			      struct svc_fh *fh, struct nfs4_ol_stateid *stp)
+{
+	struct nfs4_file *fp;
+	struct nfsd_file *nf = NULL;
+
+	if ((open->op_share_access & NFS4_SHARE_ACCESS_BOTH) ==
+			NFS4_SHARE_ACCESS_WRITE) {
+		if (nfsd_file_acquire_opened(rqstp, fh, NFSD_MAY_READ, NULL, &nf))
+			return (false);
+		fp = stp->st_stid.sc_file;
+		spin_lock(&fp->fi_lock);
+		__nfs4_file_get_access(fp, NFS4_SHARE_ACCESS_READ);
+		fp = stp->st_stid.sc_file;
+		fp->fi_fds[O_RDONLY] = nf;
+		fp->fi_rdeleg_file = nf;
+		spin_unlock(&fp->fi_lock);
+	}
+	return true;
+}
+
 /*
  * The Linux NFS server does not offer write delegations to NFSv4.0
  * clients in order to avoid conflicts between write delegations and
@@ -6181,8 +6217,9 @@ nfs4_delegation_stat(struct nfs4_delegation *dp, struct svc_fh *currentfh,
  * open or lock state.
  */
 static void
-nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
-		     struct svc_fh *currentfh)
+nfs4_open_delegation(struct svc_rqst *rqstp, struct nfsd4_open *open,
+		     struct nfs4_ol_stateid *stp, struct svc_fh *currentfh,
+		     struct svc_fh *fh)
 {
 	struct nfs4_openowner *oo = openowner(stp->st_stateowner);
 	bool deleg_ts = nfsd4_want_deleg_timestamps(open);
@@ -6227,7 +6264,8 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
 	memcpy(&open->op_delegate_stateid, &dp->dl_stid.sc_stateid, sizeof(dp->dl_stid.sc_stateid));
 
 	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) {
-		if (!nfs4_delegation_stat(dp, currentfh, &stat)) {
+		if (!nfsd4_add_rdaccess_to_wrdeleg(rqstp, open, fh, stp) ||
+				!nfs4_delegation_stat(dp, currentfh, &stat)) {
 			nfs4_put_stid(&dp->dl_stid);
 			destroy_delegation(dp);
 			goto out_no_deleg;
@@ -6322,6 +6360,20 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 		status = nfs4_check_deleg(cl, open, &dp);
 		if (status)
 			goto out;
+		if (dp && nfsd4_is_deleg_cur(open) &&
+				(dp->dl_stid.sc_file != fp)) {
+			/*
+			 * RFC8881 section 8.2.4 mandates the server to return
+			 * NFS4ERR_BAD_STATEID if the selected table entry does
+			 * not match the current filehandle. However returning
+			 * NFS4ERR_BAD_STATEID in the OPEN can cause the client
+			 * to repeatedly retry the operation with the same
+			 * stateid, since the stateid itself is valid. To avoid
+			 * this situation NFSD returns NFS4ERR_INVAL instead.
+			 */
+			status = nfserr_inval;
+			goto out;
+		}
 		stp = nfsd4_find_and_lock_existing_open(fp, open);
 	} else {
 		open->op_file = NULL;
@@ -6383,7 +6435,8 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 	* Attempt to hand out a delegation. No error return, because the
 	* OPEN succeeds even if we fail.
 	*/
-	nfs4_open_delegation(open, stp, &resp->cstate.current_fh);
+	nfs4_open_delegation(rqstp, open, stp,
+		&resp->cstate.current_fh, current_fh);
 
 	/*
 	 * If there is an existing open stateid, it must be updated and
@@ -7076,7 +7129,7 @@ nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate,
 		return_revoked = true;
 	if (typemask & SC_TYPE_DELEG)
 		/* Always allow REVOKED for DELEG so we can
-		 * retturn the appropriate error.
+		 * return the appropriate error.
 		 */
 		statusmask |= SC_STATUS_REVOKED;
 
@@ -7119,10 +7172,6 @@ nfs4_find_file(struct nfs4_stid *s, int flags)
 
 	switch (s->sc_type) {
 	case SC_TYPE_DELEG:
-		spin_lock(&s->sc_file->fi_lock);
-		ret = nfsd_file_get(s->sc_file->fi_deleg_file);
-		spin_unlock(&s->sc_file->fi_lock);
-		break;
 	case SC_TYPE_OPEN:
 	case SC_TYPE_LOCK:
 		if (flags & RD_STATE)

fs/nfsd/nfs4xdr.c

@@ -2500,10 +2500,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 
 	if (xdr_stream_decode_u32(argp->xdr, &argp->minorversion) < 0)
 		return false;
-	if (xdr_stream_decode_u32(argp->xdr, &argp->client_opcnt) < 0)
+	if (xdr_stream_decode_u32(argp->xdr, &argp->opcnt) < 0)
 		return false;
-	argp->opcnt = min_t(u32, argp->client_opcnt,
-			    NFSD_MAX_OPS_PER_COMPOUND);
 
 	if (argp->opcnt > ARRAY_SIZE(argp->iops)) {
 		argp->ops = vcalloc(argp->opcnt, sizeof(*argp->ops));

fs/nfsd/nfsctl.c

@@ -1436,7 +1436,7 @@ unsigned int nfsd_net_id;
 
 static int nfsd_genl_rpc_status_compose_msg(struct sk_buff *skb,
 					    struct netlink_callback *cb,
-					    struct nfsd_genl_rqstp *rqstp)
+					    struct nfsd_genl_rqstp *genl_rqstp)
 {
 	void *hdr;
 	u32 i;
@@ -1446,22 +1446,22 @@ static int nfsd_genl_rpc_status_compose_msg(struct sk_buff *skb,
 	if (!hdr)
 		return -ENOBUFS;
 
-	if (nla_put_be32(skb, NFSD_A_RPC_STATUS_XID, rqstp->rq_xid) ||
-	    nla_put_u32(skb, NFSD_A_RPC_STATUS_FLAGS, rqstp->rq_flags) ||
-	    nla_put_u32(skb, NFSD_A_RPC_STATUS_PROG, rqstp->rq_prog) ||
-	    nla_put_u32(skb, NFSD_A_RPC_STATUS_PROC, rqstp->rq_proc) ||
-	    nla_put_u8(skb, NFSD_A_RPC_STATUS_VERSION, rqstp->rq_vers) ||
+	if (nla_put_be32(skb, NFSD_A_RPC_STATUS_XID, genl_rqstp->rq_xid) ||
+	    nla_put_u32(skb, NFSD_A_RPC_STATUS_FLAGS, genl_rqstp->rq_flags) ||
+	    nla_put_u32(skb, NFSD_A_RPC_STATUS_PROG, genl_rqstp->rq_prog) ||
+	    nla_put_u32(skb, NFSD_A_RPC_STATUS_PROC, genl_rqstp->rq_proc) ||
+	    nla_put_u8(skb, NFSD_A_RPC_STATUS_VERSION, genl_rqstp->rq_vers) ||
 	    nla_put_s64(skb, NFSD_A_RPC_STATUS_SERVICE_TIME,
-			ktime_to_us(rqstp->rq_stime),
+			ktime_to_us(genl_rqstp->rq_stime),
 			NFSD_A_RPC_STATUS_PAD))
 		return -ENOBUFS;
 
-	switch (rqstp->rq_saddr.sa_family) {
+	switch (genl_rqstp->rq_saddr.sa_family) {
 	case AF_INET: {
 		const struct sockaddr_in *s_in, *d_in;
 
-		s_in = (const struct sockaddr_in *)&rqstp->rq_saddr;
-		d_in = (const struct sockaddr_in *)&rqstp->rq_daddr;
+		s_in = (const struct sockaddr_in *)&genl_rqstp->rq_saddr;
+		d_in = (const struct sockaddr_in *)&genl_rqstp->rq_daddr;
 		if (nla_put_in_addr(skb, NFSD_A_RPC_STATUS_SADDR4,
 				    s_in->sin_addr.s_addr) ||
 		    nla_put_in_addr(skb, NFSD_A_RPC_STATUS_DADDR4,
@@ -1476,8 +1476,8 @@ static int nfsd_genl_rpc_status_compose_msg(struct sk_buff *skb,
 	case AF_INET6: {
 		const struct sockaddr_in6 *s_in, *d_in;
 
-		s_in = (const struct sockaddr_in6 *)&rqstp->rq_saddr;
-		d_in = (const struct sockaddr_in6 *)&rqstp->rq_daddr;
+		s_in = (const struct sockaddr_in6 *)&genl_rqstp->rq_saddr;
+		d_in = (const struct sockaddr_in6 *)&genl_rqstp->rq_daddr;
 		if (nla_put_in6_addr(skb, NFSD_A_RPC_STATUS_SADDR6,
 				     &s_in->sin6_addr) ||
 		    nla_put_in6_addr(skb, NFSD_A_RPC_STATUS_DADDR6,
@@ -1491,9 +1491,9 @@ static int nfsd_genl_rpc_status_compose_msg(struct sk_buff *skb,
 	}
 	}
 
-	for (i = 0; i < rqstp->rq_opcnt; i++)
+	for (i = 0; i < genl_rqstp->rq_opcnt; i++)
 		if (nla_put_u32(skb, NFSD_A_RPC_STATUS_COMPOUND_OPS,
-				rqstp->rq_opnum[i]))
+				genl_rqstp->rq_opnum[i]))
 			return -ENOBUFS;
 
 	genlmsg_end(skb, hdr);
@@ -1569,7 +1569,8 @@ int nfsd_nl_rpc_status_get_dumpit(struct sk_buff *skb,
 				int j;
 
 				args = rqstp->rq_argp;
-				genl_rqstp.rq_opcnt = args->opcnt;
+				genl_rqstp.rq_opcnt = min_t(u32, args->opcnt,
+							    ARRAY_SIZE(genl_rqstp.rq_opnum));
 				for (j = 0; j < genl_rqstp.rq_opcnt; j++)
 					genl_rqstp.rq_opnum[j] =
 						args->ops[j].opnum;

fs/nfsd/nfsd.h

@@ -57,9 +57,6 @@ struct readdir_cd {
 	__be32			err;	/* 0, nfserr, or nfserr_eof */
 };
 
-/* Maximum number of operations per session compound */
-#define NFSD_MAX_OPS_PER_COMPOUND	50
-
 struct nfsd_genl_rqstp {
 	struct sockaddr		rq_daddr;
 	struct sockaddr		rq_saddr;
@@ -72,7 +69,7 @@ struct nfsd_genl_rqstp {
 
 	/* NFSv4 compound */
 	u32			rq_opcnt;
-	u32			rq_opnum[NFSD_MAX_OPS_PER_COMPOUND];
+	u32			rq_opnum[16];
 };
 
 extern struct svc_program	nfsd_programs[];
@@ -283,6 +280,7 @@ void		nfsd_lockd_shutdown(void);
 #define	nfserr_cb_path_down	cpu_to_be32(NFSERR_CB_PATH_DOWN)
 #define	nfserr_locked		cpu_to_be32(NFSERR_LOCKED)
 #define	nfserr_wrongsec		cpu_to_be32(NFSERR_WRONGSEC)
+#define nfserr_delay			cpu_to_be32(NFS4ERR_DELAY)
 #define nfserr_badiomode		cpu_to_be32(NFS4ERR_BADIOMODE)
 #define nfserr_badlayout		cpu_to_be32(NFS4ERR_BADLAYOUT)
 #define nfserr_bad_session_digest	cpu_to_be32(NFS4ERR_BAD_SESSION_DIGEST)

fs/nfsd/nfsfh.c

@@ -172,6 +172,8 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct net *net,
 	if (len == 0)
 		return error;
 	if (fh->fh_fsid_type == FSID_MAJOR_MINOR) {
+		u32 *fsid = fh_fsid(fh);
+
 		/* deprecated, convert to type 3 */
 		len = key_len(FSID_ENCODE_DEV)/4;
 		fh->fh_fsid_type = FSID_ENCODE_DEV;
@@ -181,17 +183,17 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct net *net,
 		 * confuses sparse, so we must use __force here to
 		 * keep it from complaining.
 		 */
-		fh->fh_fsid[0] = new_encode_dev(MKDEV(ntohl((__force __be32)fh->fh_fsid[0]),
-						      ntohl((__force __be32)fh->fh_fsid[1])));
-		fh->fh_fsid[1] = fh->fh_fsid[2];
+		fsid[0] = new_encode_dev(MKDEV(ntohl((__force __be32)fsid[0]),
+					       ntohl((__force __be32)fsid[1])));
+		fsid[1] = fsid[2];
 	}
 	data_left -= len;
 	if (data_left < 0)
 		return error;
 	exp = rqst_exp_find(rqstp ? &rqstp->rq_chandle : NULL,
 			    net, client, gssclient,
-			    fh->fh_fsid_type, fh->fh_fsid);
-	fid = (struct fid *)(fh->fh_fsid + len);
+			    fh->fh_fsid_type, fh_fsid(fh));
+	fid = (struct fid *)(fh_fsid(fh) + len);
 
 	error = nfserr_stale;
 	if (IS_ERR(exp)) {
@@ -463,7 +465,7 @@ static void _fh_update(struct svc_fh *fhp, struct svc_export *exp,
 {
 	if (dentry != exp->ex_path.dentry) {
 		struct fid *fid = (struct fid *)
-			(fhp->fh_handle.fh_fsid + fhp->fh_handle.fh_size/4 - 1);
+			(fh_fsid(&fhp->fh_handle) + fhp->fh_handle.fh_size/4 - 1);
 		int maxsize = (fhp->fh_maxsize - fhp->fh_handle.fh_size)/4;
 		int fh_flags = (exp->ex_flags & NFSEXP_NOSUBTREECHECK) ? 0 :
 				EXPORT_FH_CONNECTABLE;
@@ -614,7 +616,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry,
 	fhp->fh_handle.fh_auth_type = 0;
 
 	mk_fsid(fhp->fh_handle.fh_fsid_type,
-		fhp->fh_handle.fh_fsid,
+		fh_fsid(&fhp->fh_handle),
 		ex_dev,
 		d_inode(exp->ex_path.dentry)->i_ino,
 		exp->ex_fsid, exp->ex_uuid);

fs/nfsd/nfsfh.h

@@ -49,18 +49,19 @@ struct knfsd_fh {
 					 * Points to the current size while
 					 * building a new file handle.
 					 */
-	union {
-		char			fh_raw[NFS4_FHSIZE];
-		struct {
-			u8		fh_version;	/* == 1 */
-			u8		fh_auth_type;	/* deprecated */
-			u8		fh_fsid_type;
-			u8		fh_fileid_type;
-			u32		fh_fsid[]; /* flexible-array member */
-		};
-	};
+	u8		fh_raw[NFS4_FHSIZE];
 };
 
+#define fh_version		fh_raw[0]
+#define fh_auth_type		fh_raw[1]
+#define fh_fsid_type		fh_raw[2]
+#define fh_fileid_type		fh_raw[3]
+
+static inline u32 *fh_fsid(const struct knfsd_fh *fh)
+{
+	return (u32 *)&fh->fh_raw[4];
+}
+
 static inline __u32 ino_t_to_u32(ino_t ino)
 {
 	return (__u32) ino;
@@ -260,9 +261,12 @@ static inline bool fh_match(const struct knfsd_fh *fh1,
 static inline bool fh_fsid_match(const struct knfsd_fh *fh1,
 				 const struct knfsd_fh *fh2)
 {
+	u32 *fsid1 = fh_fsid(fh1);
+	u32 *fsid2 = fh_fsid(fh2);
+
 	if (fh1->fh_fsid_type != fh2->fh_fsid_type)
 		return false;
-	if (memcmp(fh1->fh_fsid, fh2->fh_fsid, key_len(fh1->fh_fsid_type)) != 0)
+	if (memcmp(fsid1, fsid2, key_len(fh1->fh_fsid_type)) != 0)
 		return false;
 	return true;
 }

fs/nfsd/nfsproc.c

@@ -575,7 +575,7 @@ static void nfsd_init_dirlist_pages(struct svc_rqst *rqstp,
 	buf->pages = rqstp->rq_next_page;
 	rqstp->rq_next_page++;
 
-	xdr_init_encode_pages(xdr, buf, buf->pages,  NULL);
+	xdr_init_encode_pages(xdr, buf);
 }
 
 /*

fs/nfsd/state.h

@@ -665,6 +665,7 @@ struct nfs4_file {
 	atomic_t		fi_access[2];
 	u32			fi_share_deny;
 	struct nfsd_file	*fi_deleg_file;
+	struct nfsd_file	*fi_rdeleg_file;
 	int			fi_delegees;
 	struct knfsd_fh		fi_fhandle;
 	bool			fi_had_conflict;

fs/nfsd/trace.h

@@ -344,7 +344,7 @@ TRACE_EVENT(nfsd_exp_find_key,
 		 int status),
 	TP_ARGS(key, status),
 	TP_STRUCT__entry(
-		__field(int, fsidtype)
+		__field(u8, fsidtype)
 		__array(u32, fsid, 6)
 		__string(auth_domain, key->ek_client->name)
 		__field(int, status)
@@ -367,7 +367,7 @@ TRACE_EVENT(nfsd_expkey_update,
 	TP_PROTO(const struct svc_expkey *key, const char *exp_path),
 	TP_ARGS(key, exp_path),
 	TP_STRUCT__entry(
-		__field(int, fsidtype)
+		__field(u8, fsidtype)
 		__array(u32, fsid, 6)
 		__string(auth_domain, key->ek_client->name)
 		__string(path, exp_path)
@@ -1108,7 +1108,6 @@ DEFINE_NFSD_FILE_EVENT(nfsd_file_free);
 DEFINE_NFSD_FILE_EVENT(nfsd_file_unhash);
 DEFINE_NFSD_FILE_EVENT(nfsd_file_put);
 DEFINE_NFSD_FILE_EVENT(nfsd_file_closing);
-DEFINE_NFSD_FILE_EVENT(nfsd_file_unhash_and_queue);
 
 TRACE_EVENT(nfsd_file_alloc,
 	TP_PROTO(
@@ -1344,9 +1343,7 @@ DEFINE_EVENT(nfsd_file_gc_class, name,					\
 	TP_ARGS(nf))
 
 DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_lru_add);
-DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_lru_add_disposed);
 DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_lru_del);
-DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_lru_del_disposed);
 DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_gc_in_use);
 DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_gc_writeback);
 DEFINE_NFSD_FILE_GC_EVENT(nfsd_file_gc_referenced);
@@ -1380,7 +1377,6 @@ DEFINE_EVENT(nfsd_file_lruwalk_class, name,				\
 	TP_ARGS(removed, remaining))
 
 DEFINE_NFSD_FILE_LRUWALK_EVENT(nfsd_file_gc_removed);
-DEFINE_NFSD_FILE_LRUWALK_EVENT(nfsd_file_gc_recent);
 DEFINE_NFSD_FILE_LRUWALK_EVENT(nfsd_file_shrinker_removed);
 
 TRACE_EVENT(nfsd_file_close,
@@ -2103,25 +2099,6 @@ TRACE_EVENT(nfsd_ctl_maxblksize,
 	)
 );
 
-TRACE_EVENT(nfsd_ctl_maxconn,
-	TP_PROTO(
-		const struct net *net,
-		int maxconn
-	),
-	TP_ARGS(net, maxconn),
-	TP_STRUCT__entry(
-		__field(unsigned int, netns_ino)
-		__field(int, maxconn)
-	),
-	TP_fast_assign(
-		__entry->netns_ino = net->ns.inum;
-		__entry->maxconn = maxconn;
-	),
-	TP_printk("maxconn=%d",
-		__entry->maxconn
-	)
-);
-
 TRACE_EVENT(nfsd_ctl_time,
 	TP_PROTO(
 		const struct net *net,

fs/nfsd/vfs.c

@@ -1086,10 +1086,13 @@ __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 {
 	unsigned long v, total;
 	struct iov_iter iter;
-	loff_t ppos = offset;
+	struct kiocb kiocb;
 	ssize_t host_err;
 	size_t len;
 
+	init_sync_kiocb(&kiocb, file);
+	kiocb.ki_pos = offset;
+
 	v = 0;
 	total = *count;
 	while (total) {
@@ -1104,7 +1107,7 @@ __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 	trace_nfsd_read_vector(rqstp, fhp, offset, *count);
 	iov_iter_bvec(&iter, ITER_DEST, rqstp->rq_bvec, v, *count);
-	host_err = vfs_iter_read(file, &iter, &ppos, 0);
+	host_err = vfs_iocb_iter_read(file, &kiocb, &iter);
 	return nfsd_finish_read(rqstp, fhp, file, offset, count, eof, host_err);
 }
 
@@ -1170,15 +1173,14 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct nfsd_net		*nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
 	struct file		*file = nf->nf_file;
 	struct super_block	*sb = file_inode(file)->i_sb;
+	struct kiocb		kiocb;
 	struct svc_export	*exp;
 	struct iov_iter		iter;
 	errseq_t		since;
 	__be32			nfserr;
 	int			host_err;
-	loff_t			pos = offset;
 	unsigned long		exp_op_flags = 0;
 	unsigned int		pflags = current->flags;
-	rwf_t			flags = 0;
 	bool			restore_flags = false;
 	unsigned int		nvecs;
 
@@ -1204,16 +1206,17 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
 
 	if (!EX_ISSYNC(exp))
 		stable = NFS_UNSTABLE;
-
+	init_sync_kiocb(&kiocb, file);
+	kiocb.ki_pos = offset;
 	if (stable && !fhp->fh_use_wgather)
-		flags |= RWF_SYNC;
+		kiocb.ki_flags |= IOCB_DSYNC;
 
 	nvecs = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, payload);
 	iov_iter_bvec(&iter, ITER_SOURCE, rqstp->rq_bvec, nvecs, *cnt);
 	since = READ_ONCE(file->f_wb_err);
 	if (verf)
 		nfsd_copy_write_verifier(verf, nn);
-	host_err = vfs_iter_write(file, &iter, &pos, flags);
+	host_err = vfs_iocb_iter_write(file, &kiocb, &iter);
 	if (host_err < 0) {
 		commit_reset_write_verifier(nn, rqstp, host_err);
 		goto out_nfserr;