Merge tag 'nfs-for-5.13-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client updates from Trond Myklebust:
 "Highlights include:

  Stable fixes:

   - Add validation of the UDP retrans parameter to prevent shift
     out-of-bounds

   - Don't discard pNFS layout segments that are marked for return

  Bugfixes:

   - Fix a NULL dereference crash in xprt_complete_bc_request() when the
     NFSv4.1 server misbehaves.

   - Fix the handling of NFS READDIR cookie verifiers

   - Sundry fixes to ensure attribute revalidation works correctly when
     the server does not return post-op attributes.

   - nfs4_bitmask_adjust() must not change the server global bitmasks

   - Fix major timeout handling in the RPC code.

   - NFSv4.2 fallocate() fixes.

   - Fix the NFSv4.2 SEEK_HOLE/SEEK_DATA end-of-file handling

   - Copy offload attribute revalidation fixes

   - Fix an incorrect filehandle size check in the pNFS flexfiles driver

   - Fix several RDMA transport setup/teardown races

   - Fix several RDMA queue wrapping issues

   - Fix a misplaced memory read barrier in sunrpc's call_decode()

  Features:

   - Micro optimisation of the TCP transmission queue using TCP_CORK

   - statx() performance improvements by further splitting up the
     tracking of invalid cached file metadata.

   - Support the NFSv4.2 'change_attr_type' attribute and use it to
     optimise handling of change attribute updates"

* tag 'nfs-for-5.13-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (85 commits)
  xprtrdma: Fix a NULL dereference in frwr_unmap_sync()
  sunrpc: Fix misplaced barrier in call_decode
  NFSv4.2: Remove ifdef CONFIG_NFSD from NFSv4.2 client SSC code.
  xprtrdma: Move fr_mr field to struct rpcrdma_mr
  xprtrdma: Move the Work Request union to struct rpcrdma_mr
  xprtrdma: Move fr_linv_done field to struct rpcrdma_mr
  xprtrdma: Move cqe to struct rpcrdma_mr
  xprtrdma: Move fr_cid to struct rpcrdma_mr
  xprtrdma: Remove the RPC/RDMA QP event handler
  xprtrdma: Don't display r_xprt memory addresses in tracepoints
  xprtrdma: Add an rpcrdma_mr_completion_class
  xprtrdma: Add tracepoints showing FastReg WRs and remote invalidation
  xprtrdma: Avoid Send Queue wrapping
  xprtrdma: Do not wake RPC consumer on a failed LocalInv
  xprtrdma: Do not recycle MR after FastReg/LocalInv flushes
  xprtrdma: Clarify use of barrier in frwr_wc_localinv_done()
  xprtrdma: Rename frwr_release_mr()
  xprtrdma: rpcrdma_mr_pop() already does list_del_init()
  xprtrdma: Delete rpcrdma_recv_buffer_put()
  xprtrdma: Fix cwnd update ordering
  ...

Author: torvalds

fs/Kconfig

@@ -338,8 +338,8 @@ config NFS_COMMON
 	default y
 
 config NFS_V4_2_SSC_HELPER
-	tristate
-	default y if NFS_V4=y || NFS_FS=y
+	bool
+	default y if NFS_V4_2
 
 source "net/sunrpc/Kconfig"
 source "fs/ceph/Kconfig"

fs/nfs/callback_proc.c

@@ -137,12 +137,12 @@ static struct inode *nfs_layout_find_inode_by_stateid(struct nfs_client *clp,
 		list_for_each_entry_rcu(lo, &server->layouts, plh_layouts) {
 			if (!pnfs_layout_is_valid(lo))
 				continue;
-			if (stateid != NULL &&
-			    !nfs4_stateid_match_other(stateid, &lo->plh_stateid))
+			if (!nfs4_stateid_match_other(stateid, &lo->plh_stateid))
 				continue;
-			if (!nfs_sb_active(server->super))
-				continue;
-			inode = igrab(lo->plh_inode);
+			if (nfs_sb_active(server->super))
+				inode = igrab(lo->plh_inode);
+			else
+				inode = ERR_PTR(-EAGAIN);
 			rcu_read_unlock();
 			if (inode)
 				return inode;
@@ -176,9 +176,10 @@ static struct inode *nfs_layout_find_inode_by_fh(struct nfs_client *clp,
 				continue;
 			if (nfsi->layout != lo)
 				continue;
-			if (!nfs_sb_active(server->super))
-				continue;
-			inode = igrab(lo->plh_inode);
+			if (nfs_sb_active(server->super))
+				inode = igrab(lo->plh_inode);
+			else
+				inode = ERR_PTR(-EAGAIN);
 			rcu_read_unlock();
 			if (inode)
 				return inode;

fs/nfs/client.c

@@ -476,7 +476,6 @@ void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 			to->to_maxval = to->to_initval;
 		to->to_exponential = 0;
 		break;
-#ifndef CONFIG_NFS_DISABLE_UDP_SUPPORT
 	case XPRT_TRANSPORT_UDP:
 		if (retrans == NFS_UNSPEC_RETRANS)
 			to->to_retries = NFS_DEF_UDP_RETRANS;
@@ -487,7 +486,6 @@ void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
 		to->to_exponential = 1;
 		break;
-#endif
 	default:
 		BUG();
 	}
@@ -698,9 +696,18 @@ static int nfs_init_server(struct nfs_server *server,
 	/* Initialise the client representation from the mount data */
 	server->flags = ctx->flags;
 	server->options = ctx->options;
-	server->caps |= NFS_CAP_HARDLINKS|NFS_CAP_SYMLINKS|NFS_CAP_FILEID|
-		NFS_CAP_MODE|NFS_CAP_NLINK|NFS_CAP_OWNER|NFS_CAP_OWNER_GROUP|
-		NFS_CAP_ATIME|NFS_CAP_CTIME|NFS_CAP_MTIME;
+	server->caps |= NFS_CAP_HARDLINKS | NFS_CAP_SYMLINKS;
+
+	switch (clp->rpc_ops->version) {
+	case 2:
+		server->fattr_valid = NFS_ATTR_FATTR_V2;
+		break;
+	case 3:
+		server->fattr_valid = NFS_ATTR_FATTR_V3;
+		break;
+	default:
+		server->fattr_valid = NFS_ATTR_FATTR_V4;
+	}
 
 	if (ctx->rsize)
 		server->rsize = nfs_block_size(ctx->rsize, NULL);
@@ -794,6 +801,7 @@ static void nfs_server_set_fsinfo(struct nfs_server *server,
 	server->maxfilesize = fsinfo->maxfilesize;
 
 	server->time_delta = fsinfo->time_delta;
+	server->change_attr_type = fsinfo->change_attr_type;
 
 	server->clone_blksize = fsinfo->clone_blksize;
 	/* We're airborne Set socket buffersize */
@@ -935,6 +943,8 @@ struct nfs_server *nfs_alloc_server(void)
 		return NULL;
 	}
 
+	server->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED;
+
 	ida_init(&server->openowner_id);
 	ida_init(&server->lockowner_id);
 	pnfs_init_server(server);

fs/nfs/delegation.c

@@ -114,7 +114,7 @@ nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark)
 	return ret;
 }
 /**
- * nfs_have_delegation - check if inode has a delegation, mark it
+ * nfs4_have_delegation - check if inode has a delegation, mark it
  * NFS_DELEGATION_REFERENCED if there is one.
  * @inode: inode to check
  * @flags: delegation types to check for
@@ -481,18 +481,29 @@ int nfs_inode_set_delegation(struct inode *inode, const struct cred *cred,
 	if (freeme == NULL)
 		goto out;
 add_new:
+	/*
+	 * If we didn't revalidate the change attribute before setting
+	 * the delegation, then pre-emptively ask for a full attribute
+	 * cache revalidation.
+	 */
+	spin_lock(&inode->i_lock);
+	if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_CHANGE)
+		nfs_set_cache_invalid(inode,
+			NFS_INO_INVALID_ATIME | NFS_INO_INVALID_CTIME |
+			NFS_INO_INVALID_MTIME | NFS_INO_INVALID_SIZE |
+			NFS_INO_INVALID_BLOCKS | NFS_INO_INVALID_NLINK |
+			NFS_INO_INVALID_OTHER | NFS_INO_INVALID_DATA |
+			NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL |
+			NFS_INO_INVALID_XATTR);
+	spin_unlock(&inode->i_lock);
+
 	list_add_tail_rcu(&delegation->super_list, &server->delegations);
 	rcu_assign_pointer(nfsi->delegation, delegation);
 	delegation = NULL;
 
 	atomic_long_inc(&nfs_active_delegations);
 
 	trace_nfs4_set_delegation(inode, type);
-
-	spin_lock(&inode->i_lock);
-	if (NFS_I(inode)->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME))
-		NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED;
-	spin_unlock(&inode->i_lock);
 out:
 	spin_unlock(&clp->cl_lock);
 	if (delegation != NULL)
@@ -674,7 +685,7 @@ void nfs_inode_evict_delegation(struct inode *inode)
 }
 
 /**
- * nfs_inode_return_delegation - synchronously return a delegation
+ * nfs4_inode_return_delegation - synchronously return a delegation
  * @inode: inode to process
  *
  * This routine will always flush any dirty data to disk on the
@@ -697,7 +708,7 @@ int nfs4_inode_return_delegation(struct inode *inode)
 }
 
 /**
- * nfs_inode_return_delegation_on_close - asynchronously return a delegation
+ * nfs4_inode_return_delegation_on_close - asynchronously return a delegation
  * @inode: inode to process
  *
  * This routine is called on file close in order to determine if the
@@ -811,7 +822,7 @@ void nfs_expire_all_delegations(struct nfs_client *clp)
 }
 
 /**
- * nfs_super_return_all_delegations - return delegations for one superblock
+ * nfs_server_return_all_delegations - return delegations for one superblock
  * @server: pointer to nfs_server to process
  *
  */

fs/nfs/delegation.h

@@ -84,8 +84,7 @@ int nfs4_inode_make_writeable(struct inode *inode);
 
 static inline int nfs_have_delegated_attributes(struct inode *inode)
 {
-	return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ) &&
-		!(NFS_I(inode)->cache_validity & NFS_INO_REVAL_FORCED);
+	return NFS_PROTO(inode)->have_delegation(inode, FMODE_READ);
 }
 
 #endif

fs/nfs/dir.c

@@ -866,6 +866,8 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc,
 			break;
 		}
 
+		verf_arg = verf_res;
+
 		status = nfs_readdir_page_filler(desc, entry, pages, pglen,
 						 arrays, narrays);
 	} while (!status && nfs_readdir_page_needs_filling(page));
@@ -927,7 +929,12 @@ static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc)
 			}
 			return res;
 		}
-		memcpy(nfsi->cookieverf, verf, sizeof(nfsi->cookieverf));
+		/*
+		 * Set the cookie verifier if the page cache was empty
+		 */
+		if (desc->page_index == 0)
+			memcpy(nfsi->cookieverf, verf,
+			       sizeof(nfsi->cookieverf));
 	}
 	res = nfs_readdir_search_array(desc);
 	if (res == 0) {
@@ -974,10 +981,10 @@ static int readdir_search_pagecache(struct nfs_readdir_descriptor *desc)
 /*
  * Once we've found the start of the dirent within a page: fill 'er up...
  */
-static void nfs_do_filldir(struct nfs_readdir_descriptor *desc)
+static void nfs_do_filldir(struct nfs_readdir_descriptor *desc,
+			   const __be32 *verf)
 {
 	struct file	*file = desc->file;
-	struct nfs_inode *nfsi = NFS_I(file_inode(file));
 	struct nfs_cache_array *array;
 	unsigned int i = 0;
 
@@ -991,7 +998,7 @@ static void nfs_do_filldir(struct nfs_readdir_descriptor *desc)
 			desc->eof = true;
 			break;
 		}
-		memcpy(desc->verf, nfsi->cookieverf, sizeof(desc->verf));
+		memcpy(desc->verf, verf, sizeof(desc->verf));
 		if (i < (array->size-1))
 			desc->dir_cookie = array->array[i+1].cookie;
 		else
@@ -1048,7 +1055,7 @@ static int uncached_readdir(struct nfs_readdir_descriptor *desc)
 
 	for (i = 0; !desc->eof && i < sz && arrays[i]; i++) {
 		desc->page = arrays[i];
-		nfs_do_filldir(desc);
+		nfs_do_filldir(desc, verf);
 	}
 	desc->page = NULL;
 
@@ -1069,6 +1076,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 {
 	struct dentry	*dentry = file_dentry(file);
 	struct inode	*inode = d_inode(dentry);
+	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_open_dir_context *dir_ctx = file->private_data;
 	struct nfs_readdir_descriptor *desc;
 	int res;
@@ -1122,7 +1130,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 			break;
 		}
 		if (res == -ETOOSMALL && desc->plus) {
-			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
+			clear_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags);
 			nfs_zap_caches(inode);
 			desc->page_index = 0;
 			desc->plus = false;
@@ -1132,7 +1140,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
 		if (res < 0)
 			break;
 
-		nfs_do_filldir(desc);
+		nfs_do_filldir(desc, nfsi->cookieverf);
 		nfs_readdir_page_unlock_and_put_cached(desc);
 	} while (!desc->eof);
 
@@ -1703,7 +1711,7 @@ static void nfs_drop_nlink(struct inode *inode)
 	NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter();
 	nfs_set_cache_invalid(
 		inode, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME |
-			       NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED);
+			       NFS_INO_INVALID_NLINK);
 	spin_unlock(&inode->i_lock);
 }
 
@@ -2940,7 +2948,7 @@ static int nfs_execute_ok(struct inode *inode, int mask)
 
 	if (S_ISDIR(inode->i_mode))
 		return 0;
-	if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_OTHER)) {
+	if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_MODE)) {
 		if (mask & MAY_NOT_BLOCK)
 			return -ECHILD;
 		ret = __nfs_revalidate_inode(server, inode);
@@ -2998,7 +3006,8 @@ int nfs_permission(struct user_namespace *mnt_userns,
 	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
-	res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
+	res = nfs_revalidate_inode(inode, NFS_INO_INVALID_MODE |
+						  NFS_INO_INVALID_OTHER);
 	if (res == 0)
 		res = generic_permission(&init_user_ns, inode, mask);
 	goto out;

fs/nfs/export.c

@@ -169,19 +169,8 @@ nfs_get_parent(struct dentry *dentry)
 
 static u64 nfs_fetch_iversion(struct inode *inode)
 {
-	struct nfs_server *server = NFS_SERVER(inode);
-
-	/* Is this the right call?: */
-	nfs_revalidate_inode(server, inode);
-	/*
-	 * Also, note we're ignoring any returned error.  That seems to be
-	 * the practice for cache consistency information elsewhere in
-	 * the server, but I'm not sure why.
-	 */
-	if (server->nfs_client->rpc_ops->version >= 4)
-		return inode_peek_iversion_raw(inode);
-	else
-		return time_to_chattr(&inode->i_ctime);
+	nfs_revalidate_inode(inode, NFS_INO_INVALID_CHANGE);
+	return inode_peek_iversion_raw(inode);
 }
 
 const struct export_operations nfs_export_ops = {

fs/nfs/file.c

@@ -105,7 +105,7 @@ static int nfs_revalidate_file_size(struct inode *inode, struct file *filp)
 
 	if (filp->f_flags & O_DIRECT)
 		goto force_reval;
-	if (nfs_check_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE))
+	if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_SIZE))
 		goto force_reval;
 	return 0;
 force_reval:

fs/nfs/flexfilelayout/flexfilelayout.c

@@ -106,7 +106,7 @@ static int decode_nfs_fh(struct xdr_stream *xdr, struct nfs_fh *fh)
 	if (unlikely(!p))
 		return -ENOBUFS;
 	fh->size = be32_to_cpup(p++);
-	if (fh->size > sizeof(struct nfs_fh)) {
+	if (fh->size > NFS_MAXFHSIZE) {
 		printk(KERN_ERR "NFS flexfiles: Too big fh received %d\n",
 		       fh->size);
 		return -EOVERFLOW;