Merge tag 'nfs-for-5.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

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

Features:

- Switch NFS to use readahead instead of the obsolete readpages.

- Readdir fixes to improve cacheability of large directories when
there are multiple readers and writers.

- Readdir performance improvements when doing a seekdir() immediately
after opening the directory (common when re-exporting NFS).

- NFS swap improvements from Neil Brown.

- Loosen up memory allocation to permit direct reclaim and write back
in cases where there is no danger of deadlocking the writeback code
or NFS swap.

- Avoid sillyrename when the NFSv4 server claims to support the
necessary features to recover the unlinked but open file after
reboot.

Bugfixes:

- Patch from Olga to add a mount option to control NFSv4.1 session
trunking discovery, and default it to being off.

- Fix a lockup in nfs_do_recoalesce().

- Two fixes for list iterator variables being used when pointing to
the list head.

- Fix a kernel memory scribble when reading from a non-socket
transport in /sys/kernel/sunrpc.

- Fix a race where reconnecting to a server could leave the TCP
socket stuck forever in the connecting state.

- Patch from Neil to fix a shutdown race which can leave the SUNRPC
transport timer primed after we free the struct xprt itself.

- Patch from Xin Xiong to fix reference count leaks in the NFSv4.2
copy offload.

- Sunrpc patch from Olga to avoid resending a task on an offlined
transport.

Cleanups:

- Patches from Dave Wysochanski to clean up the fscache code"

* tag 'nfs-for-5.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (91 commits)
NFSv4/pNFS: Fix another issue with a list iterator pointing to the head
NFS: Don't loop forever in nfs_do_recoalesce()
SUNRPC: Don't return error values in sysfs read of closed files
SUNRPC: Do not dereference non-socket transports in sysfs
NFSv4.1: don't retry BIND_CONN_TO_SESSION on session error
SUNRPC don't resend a task on an offlined transport
NFS: replace usage of found with dedicated list iterator variable
SUNRPC: avoid race between mod_timer() and del_timer_sync()
pNFS/files: Ensure pNFS allocation modes are consistent with nfsiod
pNFS/flexfiles: Ensure pNFS allocation modes are consistent with nfsiod
NFSv4/pnfs: Ensure pNFS allocation modes are consistent with nfsiod
NFS: Avoid writeback threads getting stuck in mempool_alloc()
NFS: nfsiod should not block forever in mempool_alloc()
SUNRPC: Make the rpciod and xprtiod slab allocation modes consistent
SUNRPC: Fix unx_lookup_cred() allocation
NFS: Fix memory allocation in rpc_alloc_task()
NFS: Fix memory allocation in rpc_malloc()
SUNRPC: Improve accuracy of socket ENOBUFS determination
SUNRPC: Replace internal use of SOCKWQ_ASYNC_NOSPACE
SUNRPC: Fix socket waits for write buffer space
...

fs/nfs/Kconfig

···
 	depends on INET && FILE_LOCKING && MULTIUSER
 	select LOCKD
 	select SUNRPC
+	select CRYPTO
+	select CRYPTO_HASH
+	select XXHASH
+	select CRYPTO_XXHASH
 	select NFS_ACL_SUPPORT if NFS_V3_ACL
 	help
 	  Choose Y here if you want to access files residing on other

fs/nfs/callback_proc.c

···
 				  struct cb_process_state *cps)
 {
 	struct cb_devicenotifyargs *args = argp;
+	const struct pnfs_layoutdriver_type *ld = NULL;
 	uint32_t i;
 	__be32 res = 0;
-	struct nfs_client *clp = cps->clp;
-	struct nfs_server *server = NULL;
 
-	if (!clp) {
+	if (!cps->clp) {
 		res = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION);
 		goto out;
 	}
···
 	for (i = 0; i < args->ndevs; i++) {
 		struct cb_devicenotifyitem *dev = &args->devs[i];
 
-		if (!server ||
-		    server->pnfs_curr_ld->id != dev->cbd_layout_type) {
-			rcu_read_lock();
-			list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link)
-				if (server->pnfs_curr_ld &&
-				    server->pnfs_curr_ld->id == dev->cbd_layout_type) {
-					rcu_read_unlock();
-					goto found;
-				}
-			rcu_read_unlock();
-			continue;
+		if (!ld || ld->id != dev->cbd_layout_type) {
+			pnfs_put_layoutdriver(ld);
+			ld = pnfs_find_layoutdriver(dev->cbd_layout_type);
+			if (!ld)
+				continue;
 		}
-
-	found:
-		nfs4_delete_deviceid(server->pnfs_curr_ld, clp, &dev->cbd_dev_id);
+		nfs4_delete_deviceid(ld, cps->clp, &dev->cbd_dev_id);
 	}
-
+	pnfs_put_layoutdriver(ld);
 out:
 	kfree(args->devs);
 	return res;
···
 	struct nfs4_copy_state *copy, *tmp_copy;
 	bool found = false;
 
-	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_NOFS);
+	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_KERNEL);
 	if (!copy)
 		return htonl(NFS4ERR_SERVERFAULT);
 

fs/nfs/callback_xdr.c

···
 	n = ntohl(*p++);
 	if (n == 0)
 		goto out;
-	if (n > ULONG_MAX / sizeof(*args->devs)) {
-		status = htonl(NFS4ERR_BADXDR);
-		goto out;
-	}
 
 	args->devs = kmalloc_array(n, sizeof(*args->devs), GFP_KERNEL);
 	if (!args->devs) {

fs/nfs/client.c

···
 	}
 
 	if (clp->rpc_ops->discover_trunking != NULL &&
-			(server->caps & NFS_CAP_FS_LOCATIONS)) {
+			(server->caps & NFS_CAP_FS_LOCATIONS &&
+			 (server->flags & NFS_MOUNT_TRUNK_DISCOVERY))) {
 		error = clp->rpc_ops->discover_trunking(server, mntfh);
 		if (error < 0)
 			return error;

fs/nfs/delegation.c

···
 	struct nfs_delegation *freeme = NULL;
 	int status = 0;
 
-	delegation = kmalloc(sizeof(*delegation), GFP_NOFS);
+	delegation = kmalloc(sizeof(*delegation), GFP_KERNEL_ACCOUNT);
 	if (delegation == NULL)
 		return -ENOMEM;
 	nfs4_stateid_copy(&delegation->stateid, stateid);

fs/nfs/dir.c

···
 #include <linux/sched.h>
 #include <linux/kmemleak.h>
 #include <linux/xattr.h>
+#include <linux/xxhash.h>
 
 #include "delegation.h"
 #include "iostat.h"
···
 	.freepage = nfs_readdir_clear_array,
 };
 
-static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir)
+#define NFS_INIT_DTSIZE PAGE_SIZE
+
+static struct nfs_open_dir_context *
+alloc_nfs_open_dir_context(struct inode *dir)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
 	struct nfs_open_dir_context *ctx;
-	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL_ACCOUNT);
 	if (ctx != NULL) {
-		ctx->duped = 0;
 		ctx->attr_gencount = nfsi->attr_gencount;
-		ctx->dir_cookie = 0;
-		ctx->dup_cookie = 0;
-		ctx->page_index = 0;
-		ctx->eof = false;
+		ctx->dtsize = NFS_INIT_DTSIZE;
 		spin_lock(&dir->i_lock);
 		if (list_empty(&nfsi->open_files) &&
 		    (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER))
 			nfs_set_cache_invalid(dir,
 					      NFS_INO_INVALID_DATA |
 						      NFS_INO_REVAL_FORCED);
-		list_add(&ctx->list, &nfsi->open_files);
-		clear_bit(NFS_INO_FORCE_READDIR, &nfsi->flags);
+		list_add_tail_rcu(&ctx->list, &nfsi->open_files);
+		memcpy(ctx->verf, nfsi->cookieverf, sizeof(ctx->verf));
 		spin_unlock(&dir->i_lock);
 		return ctx;
 	}
···
 static void put_nfs_open_dir_context(struct inode *dir, struct nfs_open_dir_context *ctx)
 {
 	spin_lock(&dir->i_lock);
-	list_del(&ctx->list);
+	list_del_rcu(&ctx->list);
 	spin_unlock(&dir->i_lock);
-	kfree(ctx);
+	kfree_rcu(ctx, rcu_head);
 }
 
 /*
···
 };
 
 struct nfs_cache_array {
+	u64 change_attr;
 	u64 last_cookie;
 	unsigned int size;
 	unsigned char page_full : 1,
···
 	struct page	*page;
 	struct dir_context *ctx;
 	pgoff_t		page_index;
+	pgoff_t		page_index_max;
 	u64		dir_cookie;
 	u64		last_cookie;
-	u64		dup_cookie;
 	loff_t		current_index;
-	loff_t		prev_index;
 
 	__be32		verf[NFS_DIR_VERIFIER_SIZE];
 	unsigned long	dir_verifier;
···
 	unsigned long	gencount;
 	unsigned long	attr_gencount;
 	unsigned int	cache_entry_index;
-	signed char duped;
+	unsigned int	buffer_fills;
+	unsigned int	dtsize;
+	bool clear_cache;
 	bool plus;
 	bool eob;
 	bool eof;
 };
 
-static void nfs_readdir_array_init(struct nfs_cache_array *array)
+static void nfs_set_dtsize(struct nfs_readdir_descriptor *desc, unsigned int sz)
 {
-	memset(array, 0, sizeof(struct nfs_cache_array));
+	struct nfs_server *server = NFS_SERVER(file_inode(desc->file));
+	unsigned int maxsize = server->dtsize;
+
+	if (sz > maxsize)
+		sz = maxsize;
+	if (sz < NFS_MIN_FILE_IO_SIZE)
+		sz = NFS_MIN_FILE_IO_SIZE;
+	desc->dtsize = sz;
 }
 
-static void nfs_readdir_page_init_array(struct page *page, u64 last_cookie)
+static void nfs_shrink_dtsize(struct nfs_readdir_descriptor *desc)
+{
+	nfs_set_dtsize(desc, desc->dtsize >> 1);
+}
+
+static void nfs_grow_dtsize(struct nfs_readdir_descriptor *desc)
+{
+	nfs_set_dtsize(desc, desc->dtsize << 1);
+}
+
+static void nfs_readdir_page_init_array(struct page *page, u64 last_cookie,
+					u64 change_attr)
 {
 	struct nfs_cache_array *array;
 
 	array = kmap_atomic(page);
-	nfs_readdir_array_init(array);
+	array->change_attr = change_attr;
 	array->last_cookie = last_cookie;
+	array->size = 0;
+	array->page_full = 0;
+	array->page_is_eof = 0;
 	array->cookies_are_ordered = 1;
 	kunmap_atomic(array);
 }
···
 /*
  * we are freeing strings created by nfs_add_to_readdir_array()
  */
-static
-void nfs_readdir_clear_array(struct page *page)
+static void nfs_readdir_clear_array(struct page *page)
 {
 	struct nfs_cache_array *array;
-	int i;
+	unsigned int i;
 
 	array = kmap_atomic(page);
 	for (i = 0; i < array->size; i++)
 		kfree(array->array[i].name);
-	nfs_readdir_array_init(array);
+	array->size = 0;
 	kunmap_atomic(array);
+}
+
+static void nfs_readdir_page_reinit_array(struct page *page, u64 last_cookie,
+					  u64 change_attr)
+{
+	nfs_readdir_clear_array(page);
+	nfs_readdir_page_init_array(page, last_cookie, change_attr);
 }
 
 static struct page *
···
 {
 	struct page *page = alloc_page(gfp_flags);
 	if (page)
-		nfs_readdir_page_init_array(page, last_cookie);
+		nfs_readdir_page_init_array(page, last_cookie, 0);
 	return page;
 }
 
···
 		nfs_readdir_clear_array(page);
 		put_page(page);
 	}
+}
+
+static u64 nfs_readdir_array_index_cookie(struct nfs_cache_array *array)
+{
+	return array->size == 0 ? array->last_cookie : array->array[0].cookie;
 }
 
 static void nfs_readdir_array_set_eof(struct nfs_cache_array *array)
···
 	return ret;
 }
 
+static size_t nfs_readdir_array_maxentries(void)
+{
+	return (PAGE_SIZE - sizeof(struct nfs_cache_array)) /
+	       sizeof(struct nfs_cache_array_entry);
+}
+
 /*
  * Check that the next array entry lies entirely within the page bounds
  */
 static int nfs_readdir_array_can_expand(struct nfs_cache_array *array)
 {
-	struct nfs_cache_array_entry *cache_entry;
-
 	if (array->page_full)
 		return -ENOSPC;
-	cache_entry = &array->array[array->size + 1];
-	if ((char *)cache_entry - (char *)array > PAGE_SIZE) {
+	if (array->size == nfs_readdir_array_maxentries()) {
 		array->page_full = 1;
 		return -ENOSPC;
 	}
 	return 0;
 }
 
-static
-int nfs_readdir_add_to_array(struct nfs_entry *entry, struct page *page)
+static int nfs_readdir_page_array_append(struct page *page,
+					 const struct nfs_entry *entry,
+					 u64 *cookie)
 {
 	struct nfs_cache_array *array;
 	struct nfs_cache_array_entry *cache_entry;
 	const char *name;
-	int ret;
+	int ret = -ENOMEM;
 
 	name = nfs_readdir_copy_name(entry->name, entry->len);
-	if (!name)
-		return -ENOMEM;
 
 	array = kmap_atomic(page);
+	if (!name)
+		goto out;
 	ret = nfs_readdir_array_can_expand(array);
 	if (ret) {
 		kfree(name);
···
 	}
 
 	cache_entry = &array->array[array->size];
-	cache_entry->cookie = entry->prev_cookie;
+	cache_entry->cookie = array->last_cookie;
 	cache_entry->ino = entry->ino;
 	cache_entry->d_type = entry->d_type;
 	cache_entry->name_len = entry->len;
···
 	if (entry->eof != 0)
 		nfs_readdir_array_set_eof(array);
 out:
+	*cookie = array->last_cookie;
 	kunmap_atomic(array);
 	return ret;
 }
 
-static struct page *nfs_readdir_page_get_locked(struct address_space *mapping,
-						pgoff_t index, u64 last_cookie)
+#define NFS_READDIR_COOKIE_MASK (U32_MAX >> 14)
+/*
+ * Hash algorithm allowing content addressible access to sequences
+ * of directory cookies. Content is addressed by the value of the
+ * cookie index of the first readdir entry in a page.
+ *
+ * The xxhash algorithm is chosen because it is fast, and is supposed
+ * to result in a decent flat distribution of hashes.
+ *
+ * We then select only the first 18 bits to avoid issues with excessive
+ * memory use for the page cache XArray. 18 bits should allow the caching
+ * of 262144 pages of sequences of readdir entries. Since each page holds
+ * 127 readdir entries for a typical 64-bit system, that works out to a
+ * cache of ~ 33 million entries per directory.
+ */
+static pgoff_t nfs_readdir_page_cookie_hash(u64 cookie)
 {
+	if (cookie == 0)
+		return 0;
+	return xxhash(&cookie, sizeof(cookie), 0) & NFS_READDIR_COOKIE_MASK;
+}
+
+static bool nfs_readdir_page_validate(struct page *page, u64 last_cookie,
+				      u64 change_attr)
+{
+	struct nfs_cache_array *array = kmap_atomic(page);
+	int ret = true;
+
+	if (array->change_attr != change_attr)
+		ret = false;
+	if (nfs_readdir_array_index_cookie(array) != last_cookie)
+		ret = false;
+	kunmap_atomic(array);
+	return ret;
+}
+
+static void nfs_readdir_page_unlock_and_put(struct page *page)
+{
+	unlock_page(page);
+	put_page(page);
+}
+
+static void nfs_readdir_page_init_and_validate(struct page *page, u64 cookie,
+					       u64 change_attr)
+{
+	if (PageUptodate(page)) {
+		if (nfs_readdir_page_validate(page, cookie, change_attr))
+			return;
+		nfs_readdir_clear_array(page);
+	}
+	nfs_readdir_page_init_array(page, cookie, change_attr);
+	SetPageUptodate(page);
+}
+
+static struct page *nfs_readdir_page_get_locked(struct address_space *mapping,
+						u64 cookie, u64 change_attr)
+{
+	pgoff_t index = nfs_readdir_page_cookie_hash(cookie);
 	struct page *page;
 
 	page = grab_cache_page(mapping, index);
-	if (page && !PageUptodate(page)) {
-		nfs_readdir_page_init_array(page, last_cookie);
-		if (invalidate_inode_pages2_range(mapping, index + 1, -1) < 0)
-			nfs_zap_mapping(mapping->host, mapping);
-		SetPageUptodate(page);
-	}
-
+	if (!page)
+		return NULL;
+	nfs_readdir_page_init_and_validate(page, cookie, change_attr);
 	return page;
 }
 
···
 	kunmap_atomic(array);
 }
 
-static void nfs_readdir_page_unlock_and_put(struct page *page)
-{
-	unlock_page(page);
-	put_page(page);
-}
-
 static struct page *nfs_readdir_page_get_next(struct address_space *mapping,
-					      pgoff_t index, u64 cookie)
+					      u64 cookie, u64 change_attr)
 {
+	pgoff_t index = nfs_readdir_page_cookie_hash(cookie);
 	struct page *page;
 
-	page = nfs_readdir_page_get_locked(mapping, index, cookie);
-	if (page) {
-		if (nfs_readdir_page_last_cookie(page) == cookie)
-			return page;
-		nfs_readdir_page_unlock_and_put(page);
-	}
-	return NULL;
+	page = grab_cache_page_nowait(mapping, index);
+	if (!page)
+		return NULL;
+	nfs_readdir_page_init_and_validate(page, cookie, change_attr);
+	if (nfs_readdir_page_last_cookie(page) != cookie)
+		nfs_readdir_page_reinit_array(page, cookie, change_attr);
+	return page;
 }
 
 static inline
···
 	return true;
 }
 
+static void nfs_readdir_seek_next_array(struct nfs_cache_array *array,
+					struct nfs_readdir_descriptor *desc)
+{
+	if (array->page_full) {
+		desc->last_cookie = array->last_cookie;
+		desc->current_index += array->size;
+		desc->cache_entry_index = 0;
+		desc->page_index++;
+	} else
+		desc->last_cookie = nfs_readdir_array_index_cookie(array);
+}
+
+static void nfs_readdir_rewind_search(struct nfs_readdir_descriptor *desc)
+{
+	desc->current_index = 0;
+	desc->last_cookie = 0;
+	desc->page_index = 0;
+}
+
 static int nfs_readdir_search_for_pos(struct nfs_cache_array *array,
 				      struct nfs_readdir_descriptor *desc)
 {
···
 	if (diff >= array->size) {
 		if (array->page_is_eof)
 			goto out_eof;
+		nfs_readdir_seek_next_array(array, desc);
 		return -EAGAIN;
 	}
 
···
 out_eof:
 	desc->eof = true;
 	return -EBADCOOKIE;
-}
-
-static bool
-nfs_readdir_inode_mapping_valid(struct nfs_inode *nfsi)
-{
-	if (nfsi->cache_validity & (NFS_INO_INVALID_CHANGE |
-				    NFS_INO_INVALID_DATA))
-		return false;
-	smp_rmb();
-	return !test_bit(NFS_INO_INVALIDATING, &nfsi->flags);
 }
 
 static bool nfs_readdir_array_cookie_in_range(struct nfs_cache_array *array,
···
 static int nfs_readdir_search_for_cookie(struct nfs_cache_array *array,
 					 struct nfs_readdir_descriptor *desc)
 {
-	int i;
-	loff_t new_pos;
+	unsigned int i;
 	int status = -EAGAIN;
 
 	if (!nfs_readdir_array_cookie_in_range(array, desc->dir_cookie))
···
 
 	for (i = 0; i < array->size; i++) {
 		if (array->array[i].cookie == desc->dir_cookie) {
-			struct nfs_inode *nfsi = NFS_I(file_inode(desc->file));
-
-			new_pos = desc->current_index + i;
-			if (desc->attr_gencount != nfsi->attr_gencount ||
-			    !nfs_readdir_inode_mapping_valid(nfsi)) {
-				desc->duped = 0;
-				desc->attr_gencount = nfsi->attr_gencount;
-			} else if (new_pos < desc->prev_index) {
-				if (desc->duped > 0
-				    && desc->dup_cookie == desc->dir_cookie) {
-					if (printk_ratelimit()) {
-						pr_notice("NFS: directory %pD2 contains a readdir loop."
-								"Please contact your server vendor.  "
-								"The file: %s has duplicate cookie %llu\n",
-								desc->file, array->array[i].name, desc->dir_cookie);
-					}
-					status = -ELOOP;
-					goto out;
-				}
-				desc->dup_cookie = desc->dir_cookie;
-				desc->duped = -1;
-			}
 			if (nfs_readdir_use_cookie(desc->file))
 				desc->ctx->pos = desc->dir_cookie;
 			else
-				desc->ctx->pos = new_pos;
-			desc->prev_index = new_pos;
+				desc->ctx->pos = desc->current_index + i;
 			desc->cache_entry_index = i;
 			return 0;
 		}
···
 		status = -EBADCOOKIE;
 		if (desc->dir_cookie == array->last_cookie)
 			desc->eof = true;
-	}
-out:
+	} else
+		nfs_readdir_seek_next_array(array, desc);
 	return status;
 }
 
···
 	else
 		status = nfs_readdir_search_for_cookie(array, desc);
 
-	if (status == -EAGAIN) {
-		desc->last_cookie = array->last_cookie;
-		desc->current_index += array->size;
-		desc->page_index++;
-	}
 	kunmap_atomic(array);
 	return status;
 }
···
 		/* We requested READDIRPLUS, but the server doesn't grok it */
 		if (error == -ENOTSUPP && desc->plus) {
 			NFS_SERVER(inode)->caps &= ~NFS_CAP_READDIRPLUS;
-			clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags);
 			desc->plus = arg.plus = false;
 			goto again;
 		}
···
 	return 1;
 }
 
-static
-bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx)
+#define NFS_READDIR_CACHE_USAGE_THRESHOLD (8UL)
+
+static bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx,
+				unsigned int cache_hits,
+				unsigned int cache_misses)
 {
 	if (!nfs_server_capable(dir, NFS_CAP_READDIRPLUS))
 		return false;
-	if (test_and_clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(dir)->flags))
-		return true;
-	if (ctx->pos == 0)
+	if (ctx->pos == 0 ||
+	    cache_hits + cache_misses > NFS_READDIR_CACHE_USAGE_THRESHOLD)
 		return true;
 	return false;
 }
 
 /*
- * This function is called by the lookup and getattr code to request the
+ * This function is called by the getattr code to request the
  * use of readdirplus to accelerate any future lookups in the same
  * directory.
  */
-void nfs_advise_use_readdirplus(struct inode *dir)
+void nfs_readdir_record_entry_cache_hit(struct inode *dir)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
+	struct nfs_open_dir_context *ctx;
 
 	if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) &&
-	    !list_empty(&nfsi->open_files))
-		set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags);
+	    S_ISDIR(dir->i_mode)) {
+		rcu_read_lock();
+		list_for_each_entry_rcu (ctx, &nfsi->open_files, list)
+			atomic_inc(&ctx->cache_hits);
+		rcu_read_unlock();
+	}
 }
 
 /*
  * This function is mainly for use by nfs_getattr().
  *
  * If this is an 'ls -l', we want to force use of readdirplus.
- * Do this by checking if there is an active file descriptor
- * and calling nfs_advise_use_readdirplus, then forcing a
- * cache flush.
  */
-void nfs_force_use_readdirplus(struct inode *dir)
+void nfs_readdir_record_entry_cache_miss(struct inode *dir)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
+	struct nfs_open_dir_context *ctx;
 
 	if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) &&
-	    !list_empty(&nfsi->open_files)) {
-		set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags);
-		set_bit(NFS_INO_FORCE_READDIR, &nfsi->flags);
+	    S_ISDIR(dir->i_mode)) {
+		rcu_read_lock();
+		list_for_each_entry_rcu (ctx, &nfsi->open_files, list)
+			atomic_inc(&ctx->cache_misses);
+		rcu_read_unlock();
 	}
+}
+
+static void nfs_lookup_advise_force_readdirplus(struct inode *dir,
+						unsigned int flags)
+{
+	if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE))
+		return;
+	if (flags & (LOOKUP_EXCL | LOOKUP_PARENT | LOOKUP_REVAL))
+		return;
+	nfs_readdir_record_entry_cache_miss(dir);
 }
 
 static
···
 			status = nfs_refresh_inode(d_inode(dentry), entry->fattr);
 			if (!status)
 				nfs_setsecurity(d_inode(dentry), entry->fattr);
+			trace_nfs_readdir_lookup_revalidate(d_inode(parent),
+							    dentry, 0, status);
 			goto out;
 		} else {
+			trace_nfs_readdir_lookup_revalidate_failed(
+				d_inode(parent), dentry, 0);
 			d_invalidate(dentry);
 			dput(dentry);
 			dentry = NULL;
···
 		dentry = alias;
 	}
 	nfs_set_verifier(dentry, dir_verifier);
+	trace_nfs_readdir_lookup(d_inode(parent), dentry, 0);
 out:
 	dput(dentry);
+}
+
+static int nfs_readdir_entry_decode(struct nfs_readdir_descriptor *desc,
+				    struct nfs_entry *entry,
+				    struct xdr_stream *stream)
+{
+	int ret;
+
+	if (entry->fattr->label)
+		entry->fattr->label->len = NFS4_MAXLABELLEN;
+	ret = xdr_decode(desc, entry, stream);
+	if (ret || !desc->plus)
+		return ret;
+	nfs_prime_dcache(file_dentry(desc->file), entry, desc->dir_verifier);
+	return 0;
 }
 
 /* Perform conversion from xdr to cache array */
 static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc,
 				   struct nfs_entry *entry,
-				   struct page **xdr_pages,
-				   unsigned int buflen,
-				   struct page **arrays,
-				   size_t narrays)
+				   struct page **xdr_pages, unsigned int buflen,
+				   struct page **arrays, size_t narrays,
+				   u64 change_attr)
 {
 	struct address_space *mapping = desc->file->f_mapping;
 	struct xdr_stream stream;
 	struct xdr_buf buf;
 	struct page *scratch, *new, *page = *arrays;
+	u64 cookie;
 	int status;
 
 	scratch = alloc_page(GFP_KERNEL);
···
 	xdr_set_scratch_page(&stream, scratch);
 
 	do {
-		if (entry->fattr->label)
-			entry->fattr->label->len = NFS4_MAXLABELLEN;
-
-		status = xdr_decode(desc, entry, &stream);
+		status = nfs_readdir_entry_decode(desc, entry, &stream);
 		if (status != 0)
 			break;
 
-		if (desc->plus)
-			nfs_prime_dcache(file_dentry(desc->file), entry,
-					desc->dir_verifier);
-
-		status = nfs_readdir_add_to_array(entry, page);
+		status = nfs_readdir_page_array_append(page, entry, &cookie);
 		if (status != -ENOSPC)
 			continue;
 
 		if (page->mapping != mapping) {
 			if (!--narrays)
 				break;
-			new = nfs_readdir_page_array_alloc(entry->prev_cookie,
-							   GFP_KERNEL);
+			new = nfs_readdir_page_array_alloc(cookie, GFP_KERNEL);
 			if (!new)
 				break;
 			arrays++;
 			*arrays = page = new;
 		} else {
-			new = nfs_readdir_page_get_next(mapping,
-							page->index + 1,
-							entry->prev_cookie);
+			new = nfs_readdir_page_get_next(mapping, cookie,
+							change_attr);
 			if (!new)
 				break;
 			if (page != *arrays)
 				nfs_readdir_page_unlock_and_put(page);
 			page = new;
 		}
-		status = nfs_readdir_add_to_array(entry, page);
+		desc->page_index_max++;
+		status = nfs_readdir_page_array_append(page, entry, &cookie);
 	} while (!status && !entry->eof);
 
 	switch (status) {
 	case -EBADCOOKIE:
-		if (entry->eof) {
-			nfs_readdir_page_set_eof(page);
-			status = 0;
-		}
-		break;
-	case -ENOSPC:
+		if (!entry->eof)
+			break;
+		nfs_readdir_page_set_eof(page);
+		fallthrough;
 	case -EAGAIN:
 		status = 0;
 		break;
+	case -ENOSPC:
+		status = 0;
+		if (!desc->plus)
+			break;
+		while (!nfs_readdir_entry_decode(desc, entry, &stream))
+			;
 	}
 
 	if (page != *arrays)
···
 				    __be32 *verf_arg, __be32 *verf_res,
 				    struct page **arrays, size_t narrays)
 {
+	u64 change_attr;
 	struct page **pages;
 	struct page *page = *arrays;
 	struct nfs_entry *entry;
 	size_t array_size;
 	struct inode *inode = file_inode(desc->file);
-	size_t dtsize = NFS_SERVER(inode)->dtsize;
+	unsigned int dtsize = desc->dtsize;
+	unsigned int pglen;
 	int status = -ENOMEM;
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
···
 	if (!pages)
 		goto out;
 
-	do {
-		unsigned int pglen;
-		status = nfs_readdir_xdr_filler(desc, verf_arg, entry->cookie,
-						pages, dtsize,
-						verf_res);
-		if (status < 0)
-			break;
+	change_attr = inode_peek_iversion_raw(inode);
+	status = nfs_readdir_xdr_filler(desc, verf_arg, entry->cookie, pages,
+					dtsize, verf_res);
+	if (status < 0)
+		goto free_pages;
 
-		pglen = status;
-		if (pglen == 0) {
-			nfs_readdir_page_set_eof(page);
-			break;
-		}
-
-		verf_arg = verf_res;
-
+	pglen = status;
+	if (pglen != 0)
 		status = nfs_readdir_page_filler(desc, entry, pages, pglen,
-						 arrays, narrays);
-	} while (!status && nfs_readdir_page_needs_filling(page) &&
-		page_mapping(page));
+						 arrays, narrays, change_attr);
+	else
+		nfs_readdir_page_set_eof(page);
+	desc->buffer_fills++;
 
+free_pages:
 	nfs_readdir_free_pages(pages, array_size);
 out:
 	nfs_free_fattr(entry->fattr);
···
 static struct page *
 nfs_readdir_page_get_cached(struct nfs_readdir_descriptor *desc)
 {
-	return nfs_readdir_page_get_locked(desc->file->f_mapping,
-					   desc->page_index,
-					   desc->last_cookie);
+	struct address_space *mapping = desc->file->f_mapping;
+	u64 change_attr = inode_peek_iversion_raw(mapping->host);
+	u64 cookie = desc->last_cookie;
+	struct page *page;
+
+	page = nfs_readdir_page_get_locked(mapping, cookie, change_attr);
+	if (!page)
+		return NULL;
+	if (desc->clear_cache && !nfs_readdir_page_needs_filling(page))
+		nfs_readdir_page_reinit_array(page, cookie, change_attr);
+	return page;
 }
 
 /*
···
 	if (!desc->page)
 		return -ENOMEM;
 	if (nfs_readdir_page_needs_filling(desc->page)) {
+		/* Grow the dtsize if we had to go back for more pages */
+		if (desc->page_index == desc->page_index_max)
+			nfs_grow_dtsize(desc);
+		desc->page_index_max = desc->page_index;
+		trace_nfs_readdir_cache_fill(desc->file, nfsi->cookieverf,
+					     desc->last_cookie,
+					     desc->page->index, desc->dtsize);
 		res = nfs_readdir_xdr_to_array(desc, nfsi->cookieverf, verf,
 					       &desc->page, 1);
 		if (res < 0) {
 			nfs_readdir_page_unlock_and_put_cached(desc);
+			trace_nfs_readdir_cache_fill_done(inode, res);
 			if (res == -EBADCOOKIE || res == -ENOTSYNC) {
 				invalidate_inode_pages2(desc->file->f_mapping);
-				desc->page_index = 0;
+				nfs_readdir_rewind_search(desc);
+				trace_nfs_readdir_invalidate_cache_range(
+					inode, 0, MAX_LFS_FILESIZE);
 				return -EAGAIN;
 			}
 			return res;
···
 		/*
 		 * Set the cookie verifier if the page cache was empty
 		 */
-		if (desc->page_index == 0)
+		if (desc->last_cookie == 0 &&
+		    memcmp(nfsi->cookieverf, verf, sizeof(nfsi->cookieverf))) {
 			memcpy(nfsi->cookieverf, verf,
 			       sizeof(nfsi->cookieverf));
+			invalidate_inode_pages2_range(desc->file->f_mapping, 1,
+						      -1);
+			trace_nfs_readdir_invalidate_cache_range(
+				inode, 1, MAX_LFS_FILESIZE);
+		}
+		desc->clear_cache = false;
 	}
 	res = nfs_readdir_search_array(desc);
 	if (res == 0)
···
 	return res;
 }
 
-static bool nfs_readdir_dont_search_cache(struct nfs_readdir_descriptor *desc)
-{
-	struct address_space *mapping = desc->file->f_mapping;
-	struct inode *dir = file_inode(desc->file);
-	unsigned int dtsize = NFS_SERVER(dir)->dtsize;
-	loff_t size = i_size_read(dir);
-
-	/*
-	 * Default to uncached readdir if the page cache is empty, and
-	 * we're looking for a non-zero cookie in a large directory.
-	 */
-	return desc->dir_cookie != 0 && mapping->nrpages == 0 && size > dtsize;
-}
-
 /* Search for desc->dir_cookie from the beginning of the page cache */
 static int readdir_search_pagecache(struct nfs_readdir_descriptor *desc)
 {
 	int res;
 
-	if (nfs_readdir_dont_search_cache(desc))
-		return -EBADCOOKIE;
-
 	do {
-		if (desc->page_index == 0) {
-			desc->current_index = 0;
-			desc->prev_index = 0;
-			desc->last_cookie = 0;
-		}
 		res = find_and_lock_cache_page(desc);
 	} while (res == -EAGAIN);
 	return res;
···
 {
 	struct file	*file = desc->file;
 	struct nfs_cache_array *array;
-	unsigned int i = 0;
+	unsigned int i;
 
 	array = kmap(desc->page);
 	for (i = desc->cache_entry_index; i < array->size; i++) {
···
 			break;
 		}
 		memcpy(desc->verf, verf, sizeof(desc->verf));
-		if (i < (array->size-1))
-			desc->dir_cookie = array->array[i+1].cookie;
-		else
+		if (i == array->size - 1) {
 			desc->dir_cookie = array->last_cookie;
+			nfs_readdir_seek_next_array(array, desc);
+		} else {
+			desc->dir_cookie = array->array[i + 1].cookie;
+			desc->last_cookie = array->array[0].cookie;
+		}
 		if (nfs_readdir_use_cookie(file))
 			desc->ctx->pos = desc->dir_cookie;
 		else
 			desc->ctx->pos++;
-		if (desc->duped != 0)
-			desc->duped = 1;
 	}
 	if (array->page_is_eof)
 		desc->eof = !desc->eob;
···
 	desc->page_index = 0;
 	desc->cache_entry_index = 0;
 	desc->last_cookie = desc->dir_cookie;
-	desc->duped = 0;
+	desc->page_index_max = 0;
+
+	trace_nfs_readdir_uncached(desc->file, desc->verf, desc->last_cookie,
+				   -1, desc->dtsize);
 
 	status = nfs_readdir_xdr_to_array(desc, desc->verf, verf, arrays, sz);
+	if (status < 0) {
+		trace_nfs_readdir_uncached_done(file_inode(desc->file), status);
+		goto out_free;
+	}
 
 	for (i = 0; !desc->eob && i < sz && arrays[i]; i++) {
 		desc->page = arrays[i];
···
 	}
 	desc->page = NULL;
 
-
+	/*
+	 * Grow the dtsize if we have to go back for more pages,
+	 * or shrink it if we're reading too many.
+	 */
+	if (!desc->eof) {
+		if (!desc->eob)
+			nfs_grow_dtsize(desc);
+		else if (desc->buffer_fills == 1 &&
+			 i < (desc->page_index_max >> 1))
+			nfs_shrink_dtsize(desc);
+	}
+out_free:
 	for (i = 0; i < sz && arrays[i]; i++)
 		nfs_readdir_page_array_free(arrays[i]);
 out:
+	if (!nfs_readdir_use_cookie(desc->file))
+		nfs_readdir_rewind_search(desc);
+	desc->page_index_max = -1;
 	kfree(arrays);
 	dfprintk(DIRCACHE, "NFS: %s: returns %d\n", __func__, status);
 	return status;
+}
+
+#define NFS_READDIR_CACHE_MISS_THRESHOLD (16UL)
+
+static bool nfs_readdir_handle_cache_misses(struct inode *inode,
+					    struct nfs_readdir_descriptor *desc,
+					    unsigned int cache_misses,
+					    bool force_clear)
+{
+	if (desc->ctx->pos == 0 || !desc->plus)
+		return false;
+	if (cache_misses <= NFS_READDIR_CACHE_MISS_THRESHOLD && !force_clear)
+		return false;
+	trace_nfs_readdir_force_readdirplus(inode);
+	return true;
 }
 
 /* The file offset position represents the dirent entry number.  A
···
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_open_dir_context *dir_ctx = file->private_data;
 	struct nfs_readdir_descriptor *desc;
-	pgoff_t page_index;
+	unsigned int cache_hits, cache_misses;
+	bool force_clear;
 	int res;
 
 	dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n",
···
 	 * to either find the entry with the appropriate number or
 	 * revalidate the cookie.
 	 */
-	if (ctx->pos == 0 || nfs_attribute_cache_expired(inode)) {
-		res = nfs_revalidate_mapping(inode, file->f_mapping);
-		if (res < 0)
-			goto out;
-	}
+	nfs_revalidate_mapping(inode, file->f_mapping);
 
 	res = -ENOMEM;
 	desc = kzalloc(sizeof(*desc), GFP_KERNEL);
···
 		goto out;
 	desc->file = file;
 	desc->ctx = ctx;
-	desc->plus = nfs_use_readdirplus(inode, ctx);
+	desc->page_index_max = -1;
 
 	spin_lock(&file->f_lock);
 	desc->dir_cookie = dir_ctx->dir_cookie;
-	desc->dup_cookie = dir_ctx->dup_cookie;
-	desc->duped = dir_ctx->duped;
-	page_index = dir_ctx->page_index;
+	desc->page_index = dir_ctx->page_index;
+	desc->last_cookie = dir_ctx->last_cookie;
 	desc->attr_gencount = dir_ctx->attr_gencount;
 	desc->eof = dir_ctx->eof;
+	nfs_set_dtsize(desc, dir_ctx->dtsize);
 	memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf));
+	cache_hits = atomic_xchg(&dir_ctx->cache_hits, 0);
+	cache_misses = atomic_xchg(&dir_ctx->cache_misses, 0);
+	force_clear = dir_ctx->force_clear;
 	spin_unlock(&file->f_lock);
 
 	if (desc->eof) {
···
 		goto out_free;
 	}
 
-	if (test_and_clear_bit(NFS_INO_FORCE_READDIR, &nfsi->flags) &&
-	    list_is_singular(&nfsi->open_files))
-		invalidate_mapping_pages(inode->i_mapping, page_index + 1, -1);
+	desc->plus = nfs_use_readdirplus(inode, ctx, cache_hits, cache_misses);
+	force_clear = nfs_readdir_handle_cache_misses(inode, desc, cache_misses,
+						      force_clear);
+	desc->clear_cache = force_clear;
 
 	do {
 		res = readdir_search_pagecache(desc);
···
 			break;
 		}
 		if (res == -ETOOSMALL && desc->plus) {
-			clear_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags);
 			nfs_zap_caches(inode);
-			desc->page_index = 0;
 			desc->plus = false;
 			desc->eof = false;
 			continue;
···
 
 		nfs_do_filldir(desc, nfsi->cookieverf);
 		nfs_readdir_page_unlock_and_put_cached(desc);
+		if (desc->page_index == desc->page_index_max)
+			desc->clear_cache = force_clear;
 	} while (!desc->eob && !desc->eof);
 
 	spin_lock(&file->f_lock);
 	dir_ctx->dir_cookie = desc->dir_cookie;
-	dir_ctx->dup_cookie = desc->dup_cookie;
-	dir_ctx->duped = desc->duped;
+	dir_ctx->last_cookie = desc->last_cookie;
 	dir_ctx->attr_gencount = desc->attr_gencount;
 	dir_ctx->page_index = desc->page_index;
+	dir_ctx->force_clear = force_clear;
 	dir_ctx->eof = desc->eof;
+	dir_ctx->dtsize = desc->dtsize;
 	memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf));
 	spin_unlock(&file->f_lock);
 out_free:
···
 	}
 	if (offset != filp->f_pos) {
 		filp->f_pos = offset;
-		if (nfs_readdir_use_cookie(filp))
-			dir_ctx->dir_cookie = offset;
-		else
+		dir_ctx->page_index = 0;
+		if (!nfs_readdir_use_cookie(filp)) {
 			dir_ctx->dir_cookie = 0;
-		if (offset == 0)
-			memset(dir_ctx->verf, 0, sizeof(dir_ctx->verf));
-		dir_ctx->duped = 0;
+			dir_ctx->last_cookie = 0;
+		} else {
+			dir_ctx->dir_cookie = offset;
+			dir_ctx->last_cookie = offset;
+		}
 		dir_ctx->eof = false;
 	}
 	spin_unlock(&filp->f_lock);
···
 	if (flags & LOOKUP_REVAL)
 		goto out_force;
 out:
-	return (inode->i_nlink == 0) ? -ESTALE : 0;
+	if (inode->i_nlink > 0 ||
+	    (inode->i_nlink == 0 &&
+	     test_bit(NFS_INO_PRESERVE_UNLINKED, &NFS_I(inode)->flags)))
+		return 0;
+	else
+		return -ESTALE;
 out_force:
 	if (flags & LOOKUP_RCU)
 		return -ECHILD;
···
 {
 	switch (error) {
 	case 1:
-		dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is valid\n",
-			__func__, dentry);
-		return 1;
+		break;
 	case 0:
 		/*
 		 * We can't d_drop the root of a disconnected tree:
···
 		 * inodes on unmount and further oopses.
 		 */
 		if (inode && IS_ROOT(dentry))
-			return 1;
-		dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is invalid\n",
-				__func__, dentry);
-		return 0;
+			error = 1;
+		break;
 	}
-	dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) lookup returned error %d\n",
-				__func__, dentry, error);
+	trace_nfs_lookup_revalidate_exit(dir, dentry, 0, error);
 	return error;
 }
 
···
 	return nfs_lookup_revalidate_done(dir, dentry, inode, 1);
 }
 
-static int
-nfs_lookup_revalidate_dentry(struct inode *dir, struct dentry *dentry,
-			     struct inode *inode)
+static int nfs_lookup_revalidate_dentry(struct inode *dir,
+					struct dentry *dentry,
+					struct inode *inode, unsigned int flags)
 {
 	struct nfs_fh *fhandle;
 	struct nfs_fattr *fattr;
 	unsigned long dir_verifier;
 	int ret;
+
+	trace_nfs_lookup_revalidate_enter(dir, dentry, flags);
 
 	ret = -ENOMEM;
 	fhandle = nfs_alloc_fhandle();
···
 		}
 		goto out;
 	}
+
+	/* Request help from readdirplus */
+	nfs_lookup_advise_force_readdirplus(dir, flags);
+
 	ret = 0;
 	if (nfs_compare_fh(NFS_FH(inode), fhandle))
 		goto out;
···
 	nfs_setsecurity(inode, fattr);
 	nfs_set_verifier(dentry, dir_verifier);
 
-	/* set a readdirplus hint that we had a cache miss */
-	nfs_force_use_readdirplus(dir);
 	ret = 1;
 out:
 	nfs_free_fattr(fattr);
···
 				nfs_mark_dir_for_revalidate(dir);
 			goto out_bad;
 		}
-		nfs_advise_use_readdirplus(dir);
 		goto out_valid;
 	}
 
···
 	if (NFS_STALE(inode))
 		goto out_bad;
 
-	trace_nfs_lookup_revalidate_enter(dir, dentry, flags);
-	error = nfs_lookup_revalidate_dentry(dir, dentry, inode);
-	trace_nfs_lookup_revalidate_exit(dir, dentry, flags, error);
-	return error;
+	return nfs_lookup_revalidate_dentry(dir, dentry, inode, flags);
 out_valid:
 	return nfs_lookup_revalidate_done(dir, dentry, inode, 1);
 out_bad:
···
 		goto out;
 
 	/* Notify readdir to use READDIRPLUS */
-	nfs_force_use_readdirplus(dir);
+	nfs_lookup_advise_force_readdirplus(dir, flags);
 
 no_entry:
 	res = d_splice_alias(inode, dentry);
···
 reval_dentry:
 	if (flags & LOOKUP_RCU)
 		return -ECHILD;
-	return nfs_lookup_revalidate_dentry(dir, dentry, inode);
+	return nfs_lookup_revalidate_dentry(dir, dentry, inode, flags);
 
 full_reval:
 	return nfs_do_lookup_revalidate(dir, dentry, flags);
···
 
 	trace_nfs_unlink_enter(dir, dentry);
 	spin_lock(&dentry->d_lock);
-	if (d_count(dentry) > 1) {
+	if (d_count(dentry) > 1 && !test_bit(NFS_INO_PRESERVE_UNLINKED,
+					     &NFS_I(d_inode(dentry))->flags)) {
 		spin_unlock(&dentry->d_lock);
 		/* Start asynchronous writeout of the inode */
 		write_inode_now(d_inode(dentry), 0);
···
 	/*
 	 * Determine which access bits we want to ask for...
 	 */
-	cache.mask = NFS_ACCESS_READ | NFS_ACCESS_MODIFY | NFS_ACCESS_EXTEND;
-	if (nfs_server_capable(inode, NFS_CAP_XATTR)) {
-		cache.mask |= NFS_ACCESS_XAREAD | NFS_ACCESS_XAWRITE |
-		    NFS_ACCESS_XALIST;
-	}
+	cache.mask = NFS_ACCESS_READ | NFS_ACCESS_MODIFY | NFS_ACCESS_EXTEND |
+		     nfs_access_xattr_mask(NFS_SERVER(inode));
 	if (S_ISDIR(inode->i_mode))
 		cache.mask |= NFS_ACCESS_DELETE | NFS_ACCESS_LOOKUP;
 	else

fs/nfs/direct.c

···
 	VM_BUG_ON(iov_iter_count(iter) != PAGE_SIZE);
 
 	if (iov_iter_rw(iter) == READ)
-		return nfs_file_direct_read(iocb, iter);
-	return nfs_file_direct_write(iocb, iter);
+		return nfs_file_direct_read(iocb, iter, true);
+	return nfs_file_direct_write(iocb, iter, true);
 }
 
 static void nfs_direct_release_pages(struct page **pages, unsigned int npages)
···
  * nfs_file_direct_read - file direct read operation for NFS files
  * @iocb: target I/O control block
  * @iter: vector of user buffers into which to read data
+ * @swap: flag indicating this is swap IO, not O_DIRECT IO
  *
  * We use this function for direct reads instead of calling
  * generic_file_aio_read() in order to avoid gfar's check to see if
···
  * client must read the updated atime from the server back into its
  * cache.
  */
-ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter)
+ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter,
+			     bool swap)
 {
 	struct file *file = iocb->ki_filp;
 	struct address_space *mapping = file->f_mapping;
···
 	if (iter_is_iovec(iter))
 		dreq->flags = NFS_ODIRECT_SHOULD_DIRTY;
 
-	nfs_start_io_direct(inode);
+	if (!swap)
+		nfs_start_io_direct(inode);
 
 	NFS_I(inode)->read_io += count;
 	requested = nfs_direct_read_schedule_iovec(dreq, iter, iocb->ki_pos);
 
-	nfs_end_io_direct(inode);
+	if (!swap)
+		nfs_end_io_direct(inode);
 
 	if (requested > 0) {
 		result = nfs_direct_wait(dreq);
···
  */
 static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
 					       struct iov_iter *iter,
-					       loff_t pos)
+					       loff_t pos, int ioflags)
 {
 	struct nfs_pageio_descriptor desc;
 	struct inode *inode = dreq->inode;
···
 	size_t requested_bytes = 0;
 	size_t wsize = max_t(size_t, NFS_SERVER(inode)->wsize, PAGE_SIZE);
 
-	nfs_pageio_init_write(&desc, inode, FLUSH_COND_STABLE, false,
+	nfs_pageio_init_write(&desc, inode, ioflags, false,
 			      &nfs_direct_write_completion_ops);
 	desc.pg_dreq = dreq;
 	get_dreq(dreq);
···
  * nfs_file_direct_write - file direct write operation for NFS files
  * @iocb: target I/O control block
  * @iter: vector of user buffers from which to write data
+ * @swap: flag indicating this is swap IO, not O_DIRECT IO
  *
  * We use this function for direct writes instead of calling
  * generic_file_aio_write() in order to avoid taking the inode
···
  * Note that O_APPEND is not supported for NFS direct writes, as there
  * is no atomic O_APPEND write facility in the NFS protocol.
  */
-ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter)
+ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter,
+			      bool swap)
 {
 	ssize_t result, requested;
 	size_t count;
···
 	dfprintk(FILE, "NFS: direct write(%pD2, %zd@%Ld)\n",
 		file, iov_iter_count(iter), (long long) iocb->ki_pos);
 
-	result = generic_write_checks(iocb, iter);
+	if (swap)
+		/* bypass generic checks */
+		result =  iov_iter_count(iter);
+	else
+		result = generic_write_checks(iocb, iter);
 	if (result <= 0)
 		return result;
 	count = result;
···
 		dreq->iocb = iocb;
 	pnfs_init_ds_commit_info_ops(&dreq->ds_cinfo, inode);
 
-	nfs_start_io_direct(inode);
+	if (swap) {
+		requested = nfs_direct_write_schedule_iovec(dreq, iter, pos,
+							    FLUSH_STABLE);
+	} else {
+		nfs_start_io_direct(inode);
 
-	requested = nfs_direct_write_schedule_iovec(dreq, iter, pos);
+		requested = nfs_direct_write_schedule_iovec(dreq, iter, pos,
+							    FLUSH_COND_STABLE);
 
-	if (mapping->nrpages) {
-		invalidate_inode_pages2_range(mapping,
-					      pos >> PAGE_SHIFT, end);
+		if (mapping->nrpages) {
+			invalidate_inode_pages2_range(mapping,
+						      pos >> PAGE_SHIFT, end);
+		}
+
+		nfs_end_io_direct(inode);
 	}
-
-	nfs_end_io_direct(inode);
 
 	if (requested > 0) {
 		result = nfs_direct_wait(dreq);

fs/nfs/file.c

···
 
 static const struct vm_operations_struct nfs_file_vm_ops;
 
-/* Hack for future NFS swap support */
-#ifndef IS_SWAPFILE
-# define IS_SWAPFILE(inode)	(0)
-#endif
-
 int nfs_check_flags(int flags)
 {
 	if ((flags & (O_APPEND | O_DIRECT)) == (O_APPEND | O_DIRECT))
···
 	ssize_t result;
 
 	if (iocb->ki_flags & IOCB_DIRECT)
-		return nfs_file_direct_read(iocb, to);
+		return nfs_file_direct_read(iocb, to, false);
 
 	dprintk("NFS: read(%pD2, %zu@%lu)\n",
 		iocb->ki_filp,
···
 {
 	unsigned long blocks;
 	long long isize;
-	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
-	struct inode *inode = file->f_mapping->host;
+	struct inode *inode = file_inode(file);
+	struct rpc_clnt *clnt = NFS_CLIENT(inode);
+	struct nfs_client *cl = NFS_SERVER(inode)->nfs_client;
 
 	spin_lock(&inode->i_lock);
 	blocks = inode->i_blocks;
···
 
 	*span = sis->pages;
 
+
+	if (cl->rpc_ops->enable_swap)
+		cl->rpc_ops->enable_swap(inode);
+
 	return rpc_clnt_swap_activate(clnt);
 }
 
 static void nfs_swap_deactivate(struct file *file)
 {
-	struct rpc_clnt *clnt = NFS_CLIENT(file->f_mapping->host);
+	struct inode *inode = file_inode(file);
+	struct rpc_clnt *clnt = NFS_CLIENT(inode);
+	struct nfs_client *cl = NFS_SERVER(inode)->nfs_client;
 
 	rpc_clnt_swap_deactivate(clnt);
+	if (cl->rpc_ops->disable_swap)
+		cl->rpc_ops->disable_swap(file_inode(file));
 }
 
 const struct address_space_operations nfs_file_aops = {
 	.readpage = nfs_readpage,
-	.readpages = nfs_readpages,
+	.readahead = nfs_readahead,
 	.dirty_folio = filemap_dirty_folio,
 	.writepage = nfs_writepage,
 	.writepages = nfs_writepages,
···
 		return result;
 
 	if (iocb->ki_flags & IOCB_DIRECT)
-		return nfs_file_direct_write(iocb, from);
+		return nfs_file_direct_write(iocb, from, false);
 
 	dprintk("NFS: write(%pD2, %zu@%Ld)\n",
 		file, iov_iter_count(from), (long long) iocb->ki_pos);

fs/nfs/filelayout/filelayout.c

···
 	unsigned int size = (fl->stripe_type == STRIPE_SPARSE) ?
 		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
 
-	new = pnfs_alloc_commit_array(size, GFP_NOIO);
+	new = pnfs_alloc_commit_array(size, nfs_io_gfp_mask());
 	if (new) {
 		spin_lock(&inode->i_lock);
 		array = pnfs_add_commit_array(fl_cinfo, new, lseg);

fs/nfs/flexfilelayout/flexfilelayout.c

···
 	spin_unlock(&mirror->lock);
 
 	if (report)
-		pnfs_report_layoutstat(inode, GFP_KERNEL);
+		pnfs_report_layoutstat(inode, nfs_io_gfp_mask());
 }
 
 static void
···
 	spin_unlock(&mirror->lock);
 
 	if (report)
-		pnfs_report_layoutstat(inode, GFP_NOIO);
+		pnfs_report_layoutstat(inode, nfs_io_gfp_mask());
 }
 
 static void
···
 		      bool strict_iomode)
 {
 	pnfs_put_lseg(pgio->pg_lseg);
-	pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-					   nfs_req_openctx(req),
-					   req_offset(req),
-					   req->wb_bytes,
-					   IOMODE_READ,
-					   strict_iomode,
-					   GFP_KERNEL);
+	pgio->pg_lseg =
+		pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req),
+				   req_offset(req), req->wb_bytes, IOMODE_READ,
+				   strict_iomode, nfs_io_gfp_mask());
 	if (IS_ERR(pgio->pg_lseg)) {
 		pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 		pgio->pg_lseg = NULL;
···
 retry:
 	ff_layout_pg_check_layout(pgio, req);
 	if (!pgio->pg_lseg) {
-		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-						   nfs_req_openctx(req),
-						   req_offset(req),
-						   req->wb_bytes,
-						   IOMODE_RW,
-						   false,
-						   GFP_NOFS);
+		pgio->pg_lseg =
+			pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req),
+					   req_offset(req), req->wb_bytes,
+					   IOMODE_RW, false, nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
···
 				    struct nfs_page *req)
 {
 	if (!pgio->pg_lseg) {
-		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-						   nfs_req_openctx(req),
-						   req_offset(req),
-						   req->wb_bytes,
-						   IOMODE_RW,
-						   false,
-						   GFP_NOFS);
+		pgio->pg_lseg =
+			pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req),
+					   req_offset(req), req->wb_bytes,
+					   IOMODE_RW, false, nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
···
 	mirror = FF_LAYOUT_COMP(lseg, idx);
 	err = ff_layout_track_ds_error(FF_LAYOUT_FROM_HDR(lseg->pls_layout),
 				       mirror, offset, length, status, opnum,
-				       GFP_NOIO);
+				       nfs_io_gfp_mask());
 
 	switch (status) {
 	case NFS4ERR_DELAY:
···
 	struct inode *inode = lseg->pls_layout->plh_inode;
 	struct pnfs_commit_array *array, *new;
 
-	new = pnfs_alloc_commit_array(flseg->mirror_array_cnt, GFP_NOIO);
+	new = pnfs_alloc_commit_array(flseg->mirror_array_cnt,
+				      nfs_io_gfp_mask());
 	if (new) {
 		spin_lock(&inode->i_lock);
 		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
···
 	struct nfs4_flexfile_layoutreturn_args *ff_args;
 	struct nfs4_flexfile_layout *ff_layout = FF_LAYOUT_FROM_HDR(args->layout);
 
-	ff_args = kmalloc(sizeof(*ff_args), GFP_KERNEL);
+	ff_args = kmalloc(sizeof(*ff_args), nfs_io_gfp_mask());
 	if (!ff_args)
 		goto out_nomem;
-	ff_args->pages[0] = alloc_page(GFP_KERNEL);
+	ff_args->pages[0] = alloc_page(nfs_io_gfp_mask());
 	if (!ff_args->pages[0])
 		goto out_nomem_free;
 
···
 	if (list_empty(&head))
 		return;
 
-	errors = kmalloc_array(NFS42_LAYOUTERROR_MAX,
-			sizeof(*errors), GFP_NOFS);
+	errors = kmalloc_array(NFS42_LAYOUTERROR_MAX, sizeof(*errors),
+			       nfs_io_gfp_mask());
 	if (errors != NULL) {
 		const struct nfs4_ff_layout_ds_err *pos;
 		size_t n = 0;
···
 	const int dev_count = PNFS_LAYOUTSTATS_MAXDEV;
 
 	/* For now, send at most PNFS_LAYOUTSTATS_MAXDEV statistics */
-	args->devinfo = kmalloc_array(dev_count, sizeof(*args->devinfo), GFP_NOIO);
+	args->devinfo = kmalloc_array(dev_count, sizeof(*args->devinfo),
+				      nfs_io_gfp_mask());
 	if (!args->devinfo)
 		return -ENOMEM;
 

fs/nfs/fs_context.c

···
 	Opt_source,
 	Opt_tcp,
 	Opt_timeo,
+	Opt_trunkdiscovery,
 	Opt_udp,
 	Opt_v,
 	Opt_vers,
···
 	fsparam_string("source",	Opt_source),
 	fsparam_flag  ("tcp",		Opt_tcp),
 	fsparam_u32   ("timeo",		Opt_timeo),
+	fsparam_flag_no("trunkdiscovery", Opt_trunkdiscovery),
 	fsparam_flag  ("udp",		Opt_udp),
 	fsparam_flag  ("v2",		Opt_v),
 	fsparam_flag  ("v3",		Opt_v),
···
 			ctx->flags |= NFS_MOUNT_NOCTO;
 		else
 			ctx->flags &= ~NFS_MOUNT_NOCTO;
+		break;
+	case Opt_trunkdiscovery:
+		if (result.negated)
+			ctx->flags &= ~NFS_MOUNT_TRUNK_DISCOVERY;
+		else
+			ctx->flags |= NFS_MOUNT_TRUNK_DISCOVERY;
 		break;
 	case Opt_ac:
 		if (result.negated)

fs/nfs/fscache.c

···
 #include "internal.h"
 #include "iostat.h"
 #include "fscache.h"
-
-#define NFSDBG_FACILITY		NFSDBG_FSCACHE
+#include "nfstrace.h"
 
 #define NFS_MAX_KEY_LEN 1000
 
···
 	vcookie = fscache_acquire_volume(key,
 					 NULL, /* preferred_cache */
 					 NULL, 0 /* coherency_data */);
-	dfprintk(FSCACHE, "NFS: get superblock cookie (0x%p/0x%p)\n",
-		 nfss, vcookie);
 	if (IS_ERR(vcookie)) {
 		if (vcookie != ERR_PTR(-EBUSY)) {
 			kfree(key);
···
 {
 	struct nfs_server *nfss = NFS_SB(sb);
 
-	dfprintk(FSCACHE, "NFS: releasing superblock cookie (0x%p/0x%p)\n",
-		 nfss, nfss->fscache);
-
 	fscache_relinquish_volume(nfss->fscache, NULL, false);
 	nfss->fscache = NULL;
 	kfree(nfss->fscache_uniq);
···
 	if (!(nfss->fscache && S_ISREG(inode->i_mode)))
 		return;
 
-	nfs_fscache_update_auxdata(&auxdata, nfsi);
+	nfs_fscache_update_auxdata(&auxdata, inode);
 
 	nfsi->fscache = fscache_acquire_cookie(NFS_SB(inode->i_sb)->fscache,
 					       0,
···
 					       nfsi->fh.size,
 					       &auxdata,      /* aux_data */
 					       sizeof(auxdata),
-					       i_size_read(&nfsi->vfs_inode));
+					       i_size_read(inode));
 }
 
 /*
···
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct fscache_cookie *cookie = nfs_i_fscache(inode);
-
-	dfprintk(FSCACHE, "NFS: clear cookie (0x%p/0x%p)\n", nfsi, cookie);
 
 	fscache_relinquish_cookie(cookie, false);
 	nfsi->fscache = NULL;
···
 void nfs_fscache_open_file(struct inode *inode, struct file *filp)
 {
 	struct nfs_fscache_inode_auxdata auxdata;
-	struct nfs_inode *nfsi = NFS_I(inode);
 	struct fscache_cookie *cookie = nfs_i_fscache(inode);
 	bool open_for_write = inode_is_open_for_write(inode);
 
···
 
 	fscache_use_cookie(cookie, open_for_write);
 	if (open_for_write) {
-		dfprintk(FSCACHE, "NFS: nfsi 0x%p disabling cache\n", nfsi);
-		nfs_fscache_update_auxdata(&auxdata, nfsi);
+		nfs_fscache_update_auxdata(&auxdata, inode);
 		fscache_invalidate(cookie, &auxdata, i_size_read(inode),
 				   FSCACHE_INVAL_DIO_WRITE);
 	}
···
 void nfs_fscache_release_file(struct inode *inode, struct file *filp)
 {
 	struct nfs_fscache_inode_auxdata auxdata;
-	struct nfs_inode *nfsi = NFS_I(inode);
 	struct fscache_cookie *cookie = nfs_i_fscache(inode);
 
 	if (fscache_cookie_valid(cookie)) {
-		nfs_fscache_update_auxdata(&auxdata, nfsi);
+		nfs_fscache_update_auxdata(&auxdata, inode);
 		fscache_unuse_cookie(cookie, &auxdata, NULL);
 	}
 }
···
 /*
  * Retrieve a page from fscache
  */
-int __nfs_readpage_from_fscache(struct inode *inode, struct page *page)
+int __nfs_fscache_read_page(struct inode *inode, struct page *page)
 {
 	int ret;
 
-	dfprintk(FSCACHE,
-		 "NFS: readpage_from_fscache(fsc:%p/p:%p(i:%lx f:%lx)/0x%p)\n",
-		 nfs_i_fscache(inode), page, page->index, page->flags, inode);
-
+	trace_nfs_fscache_read_page(inode, page);
 	if (PageChecked(page)) {
-		dfprintk(FSCACHE, "NFS:    readpage_from_fscache: PageChecked\n");
 		ClearPageChecked(page);
-		return 1;
+		ret = 1;
+		goto out;
 	}
 
 	ret = fscache_fallback_read_page(inode, page);
 	if (ret < 0) {
 		nfs_inc_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_FAIL);
-		dfprintk(FSCACHE,
-			 "NFS:    readpage_from_fscache failed %d\n", ret);
 		SetPageChecked(page);
-		return ret;
+		goto out;
 	}
 
 	/* Read completed synchronously */
-	dfprintk(FSCACHE, "NFS:    readpage_from_fscache: read successful\n");
 	nfs_inc_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_READ_OK);
 	SetPageUptodate(page);
-	return 0;
+	ret = 0;
+out:
+	trace_nfs_fscache_read_page_exit(inode, page, ret);
+	return ret;
 }
 
 /*
  * Store a newly fetched page in fscache.  We can be certain there's no page
  * stored in the cache as yet otherwise we would've read it from there.
  */
-void __nfs_readpage_to_fscache(struct inode *inode, struct page *page)
+void __nfs_fscache_write_page(struct inode *inode, struct page *page)
 {
 	int ret;
 
-	dfprintk(FSCACHE,
-		 "NFS: readpage_to_fscache(fsc:%p/p:%p(i:%lx f:%lx))\n",
-		 nfs_i_fscache(inode), page, page->index, page->flags);
+	trace_nfs_fscache_write_page(inode, page);
 
 	ret = fscache_fallback_write_page(inode, page, true);
-
-	dfprintk(FSCACHE,
-		 "NFS:     readpage_to_fscache: p:%p(i:%lu f:%lx) ret %d\n",
-		 page, page->index, page->flags, ret);
 
 	if (ret != 0) {
 		nfs_inc_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_WRITTEN_FAIL);
···
 	} else {
 		nfs_inc_fscache_stats(inode, NFSIOS_FSCACHE_PAGES_WRITTEN_OK);
 	}
+	trace_nfs_fscache_write_page_exit(inode, page, ret);
 }

fs/nfs/fscache.h

···
 extern void nfs_fscache_open_file(struct inode *, struct file *);
 extern void nfs_fscache_release_file(struct inode *, struct file *);
 
-extern int __nfs_readpage_from_fscache(struct inode *, struct page *);
-extern void __nfs_read_completion_to_fscache(struct nfs_pgio_header *hdr,
-					     unsigned long bytes);
-extern void __nfs_readpage_to_fscache(struct inode *, struct page *);
+extern int __nfs_fscache_read_page(struct inode *, struct page *);
+extern void __nfs_fscache_write_page(struct inode *, struct page *);
 
 static inline int nfs_fscache_release_page(struct page *page, gfp_t gfp)
 {
···
 /*
  * Retrieve a page from an inode data storage object.
  */
-static inline int nfs_readpage_from_fscache(struct inode *inode,
-					    struct page *page)
+static inline int nfs_fscache_read_page(struct inode *inode, struct page *page)
 {
-	if (NFS_I(inode)->fscache)
-		return __nfs_readpage_from_fscache(inode, page);
+	if (nfs_i_fscache(inode))
+		return __nfs_fscache_read_page(inode, page);
 	return -ENOBUFS;
 }
 
···
  * Store a page newly fetched from the server in an inode data storage object
  * in the cache.
  */
-static inline void nfs_readpage_to_fscache(struct inode *inode,
+static inline void nfs_fscache_write_page(struct inode *inode,
 					   struct page *page)
 {
-	if (NFS_I(inode)->fscache)
-		__nfs_readpage_to_fscache(inode, page);
+	if (nfs_i_fscache(inode))
+		__nfs_fscache_write_page(inode, page);
 }
 
 static inline void nfs_fscache_update_auxdata(struct nfs_fscache_inode_auxdata *auxdata,
-					      struct nfs_inode *nfsi)
+					      struct inode *inode)
 {
 	memset(auxdata, 0, sizeof(*auxdata));
-	auxdata->mtime_sec  = nfsi->vfs_inode.i_mtime.tv_sec;
-	auxdata->mtime_nsec = nfsi->vfs_inode.i_mtime.tv_nsec;
-	auxdata->ctime_sec  = nfsi->vfs_inode.i_ctime.tv_sec;
-	auxdata->ctime_nsec = nfsi->vfs_inode.i_ctime.tv_nsec;
+	auxdata->mtime_sec  = inode->i_mtime.tv_sec;
+	auxdata->mtime_nsec = inode->i_mtime.tv_nsec;
+	auxdata->ctime_sec  = inode->i_ctime.tv_sec;
+	auxdata->ctime_nsec = inode->i_ctime.tv_nsec;
 
-	if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4)
-		auxdata->change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode);
+	if (NFS_SERVER(inode)->nfs_client->rpc_ops->version == 4)
+		auxdata->change_attr = inode_peek_iversion_raw(inode);
 }
 
 /*
···
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	if (nfsi->fscache) {
-		nfs_fscache_update_auxdata(&auxdata, nfsi);
+		nfs_fscache_update_auxdata(&auxdata, inode);
 		fscache_invalidate(nfsi->fscache, &auxdata,
-				   i_size_read(&nfsi->vfs_inode), flags);
+				   i_size_read(inode), flags);
 	}
 }
 
···
 {
 	return 1; /* True: may release page */
 }
-static inline int nfs_readpage_from_fscache(struct inode *inode,
-					    struct page *page)
+static inline int nfs_fscache_read_page(struct inode *inode, struct page *page)
 {
 	return -ENOBUFS;
 }
-static inline void nfs_readpage_to_fscache(struct inode *inode,
-					   struct page *page) {}
-
-
+static inline void nfs_fscache_write_page(struct inode *inode, struct page *page) {}
 static inline void nfs_fscache_invalidate(struct inode *inode, int flags) {}
 
 static inline const char *nfs_server_fscache_state(struct nfs_server *server)

fs/nfs/inode.c

···
 				   NFS_INO_INVALID_OTHER |
 				   NFS_INO_INVALID_XATTR);
 		flags &= ~(NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE);
-	} else if (flags & NFS_INO_REVAL_PAGECACHE)
-		flags |= NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE;
+	}
 
 	if (!nfs_has_xattr_cache(nfsi))
 		flags &= ~NFS_INO_INVALID_XATTR;
 	if (flags & NFS_INO_INVALID_DATA)
 		nfs_fscache_invalidate(inode, 0);
-	flags &= ~(NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED);
+	flags &= ~NFS_INO_REVAL_FORCED;
 
 	nfsi->cache_validity |= flags;
 
···
 	nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
 	nfsi->attrtimeo_timestamp = jiffies;
 
-	if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) {
-		nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR
-					| NFS_INO_INVALID_DATA
-					| NFS_INO_INVALID_ACCESS
-					| NFS_INO_INVALID_ACL
-					| NFS_INO_INVALID_XATTR
-					| NFS_INO_REVAL_PAGECACHE);
-	} else
-		nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR
-					| NFS_INO_INVALID_ACCESS
-					| NFS_INO_INVALID_ACL
-					| NFS_INO_INVALID_XATTR
-					| NFS_INO_REVAL_PAGECACHE);
+	if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))
+		nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR |
+						     NFS_INO_INVALID_DATA |
+						     NFS_INO_INVALID_ACCESS |
+						     NFS_INO_INVALID_ACL |
+						     NFS_INO_INVALID_XATTR);
+	else
+		nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR |
+						     NFS_INO_INVALID_ACCESS |
+						     NFS_INO_INVALID_ACL |
+						     NFS_INO_INVALID_XATTR);
 	nfs_zap_label_cache_locked(nfsi);
 }
 
···
 			inode->i_gid = fattr->gid;
 		else if (fattr_supported & NFS_ATTR_FATTR_GROUP)
 			nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER);
-		if (nfs_server_capable(inode, NFS_CAP_XATTR))
-			nfs_set_cache_invalid(inode, NFS_INO_INVALID_XATTR);
 		if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED)
 			inode->i_blocks = fattr->du.nfs2.blocks;
 		else if (fattr_supported & NFS_ATTR_FATTR_BLOCKS_USED &&
···
 }
 EXPORT_SYMBOL_GPL(nfs_setattr_update_inode);
 
+/*
+ * Don't request help from readdirplus if the file is being written to,
+ * or if attribute caching is turned off
+ */
+static bool nfs_getattr_readdirplus_enable(const struct inode *inode)
+{
+	return nfs_server_capable(inode, NFS_CAP_READDIRPLUS) &&
+	       !nfs_have_writebacks(inode) && NFS_MAXATTRTIMEO(inode) > 5 * HZ;
+}
+
 static void nfs_readdirplus_parent_cache_miss(struct dentry *dentry)
 {
-	struct dentry *parent;
-
-	if (!nfs_server_capable(d_inode(dentry), NFS_CAP_READDIRPLUS))
-		return;
-	parent = dget_parent(dentry);
-	nfs_force_use_readdirplus(d_inode(parent));
-	dput(parent);
+	if (!IS_ROOT(dentry)) {
+		struct dentry *parent = dget_parent(dentry);
+		nfs_readdir_record_entry_cache_miss(d_inode(parent));
+		dput(parent);
+	}
 }
 
 static void nfs_readdirplus_parent_cache_hit(struct dentry *dentry)
 {
-	struct dentry *parent;
-
-	if (!nfs_server_capable(d_inode(dentry), NFS_CAP_READDIRPLUS))
-		return;
-	parent = dget_parent(dentry);
-	nfs_advise_use_readdirplus(d_inode(parent));
-	dput(parent);
+	if (!IS_ROOT(dentry)) {
+		struct dentry *parent = dget_parent(dentry);
+		nfs_readdir_record_entry_cache_hit(d_inode(parent));
+		dput(parent);
+	}
 }
 
 static u32 nfs_get_valid_attrmask(struct inode *inode)
···
 	int err = 0;
 	bool force_sync = query_flags & AT_STATX_FORCE_SYNC;
 	bool do_update = false;
+	bool readdirplus_enabled = nfs_getattr_readdirplus_enable(inode);
 
 	trace_nfs_getattr_enter(inode);
 
···
 			STATX_INO | STATX_SIZE | STATX_BLOCKS;
 
 	if ((query_flags & AT_STATX_DONT_SYNC) && !force_sync) {
-		nfs_readdirplus_parent_cache_hit(path->dentry);
+		if (readdirplus_enabled)
+			nfs_readdirplus_parent_cache_hit(path->dentry);
 		goto out_no_revalidate;
 	}
 
···
 		do_update |= cache_validity & NFS_INO_INVALID_BLOCKS;
 
 	if (do_update) {
-		/* Update the attribute cache */
-		if (!(server->flags & NFS_MOUNT_NOAC))
+		if (readdirplus_enabled)
 			nfs_readdirplus_parent_cache_miss(path->dentry);
-		else
-			nfs_readdirplus_parent_cache_hit(path->dentry);
 		err = __nfs_revalidate_inode(server, inode);
 		if (err)
 			goto out;
-	} else
+	} else if (readdirplus_enabled)
 		nfs_readdirplus_parent_cache_hit(path->dentry);
 out_no_revalidate:
 	/* Only return attributes that were revalidated. */
···
 	res = __nfs_find_lock_context(ctx);
 	rcu_read_unlock();
 	if (res == NULL) {
-		new = kmalloc(sizeof(*new), GFP_KERNEL);
+		new = kmalloc(sizeof(*new), GFP_KERNEL_ACCOUNT);
 		if (new == NULL)
 			return ERR_PTR(-ENOMEM);
 		nfs_init_lock_context(new);
···
 {
 	struct nfs_open_context *ctx;
 
-	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
+	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL_ACCOUNT);
 	if (!ctx)
 		return ERR_PTR(-ENOMEM);
 	nfs_sb_active(dentry->d_sb);
···
 {
 	struct nfs_fattr *fattr;
 
-	fattr = kmalloc(sizeof(*fattr), GFP_NOFS);
+	fattr = kmalloc(sizeof(*fattr), GFP_KERNEL);
 	if (fattr != NULL) {
 		nfs_fattr_init(fattr);
 		fattr->label = NULL;
···
 	if (!fattr)
 		return NULL;
 
-	fattr->label = nfs4_label_alloc(server, GFP_NOFS);
+	fattr->label = nfs4_label_alloc(server, GFP_KERNEL);
 	if (IS_ERR(fattr->label)) {
 		kfree(fattr);
 		return NULL;
···
 {
 	struct nfs_fh *fh;
 
-	fh = kmalloc(sizeof(struct nfs_fh), GFP_NOFS);
+	fh = kmalloc(sizeof(struct nfs_fh), GFP_KERNEL);
 	if (fh != NULL)
 		fh->size = 0;
 	return fh;

fs/nfs/internal.h

···
 			   const struct nfs_client_initdata *);
 
 /* dir.c */
-extern void nfs_advise_use_readdirplus(struct inode *dir);
-extern void nfs_force_use_readdirplus(struct inode *dir);
+extern void nfs_readdir_record_entry_cache_hit(struct inode *dir);
+extern void nfs_readdir_record_entry_cache_miss(struct inode *dir);
 extern unsigned long nfs_access_cache_count(struct shrinker *shrink,
 					    struct shrink_control *sc);
 extern unsigned long nfs_access_cache_scan(struct shrinker *shrink,
···
 	      dev_t);
 int nfs_rename(struct user_namespace *, struct inode *, struct dentry *,
 	       struct inode *, struct dentry *, unsigned int);
+
+#ifdef CONFIG_NFS_V4_2
+static inline __u32 nfs_access_xattr_mask(const struct nfs_server *server)
+{
+	if (!(server->caps & NFS_CAP_XATTR))
+		return 0;
+	return NFS4_ACCESS_XAREAD | NFS4_ACCESS_XAWRITE | NFS4_ACCESS_XALIST;
+}
+#else
+static inline __u32 nfs_access_xattr_mask(const struct nfs_server *server)
+{
+	return 0;
+}
+#endif
 
 /* file.c */
 int nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync);
···
 {
 	return verf->committed > NFS_UNSTABLE &&
 		!nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier);
+}
+
+static inline gfp_t nfs_io_gfp_mask(void)
+{
+	if (current->flags & PF_WQ_WORKER)
+		return GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN;
+	return GFP_KERNEL;
 }
 
 /* unlink.c */

fs/nfs/nfs2xdr.c

···
 
 	error = decode_filename_inline(xdr, &entry->name, &entry->len);
 	if (unlikely(error))
-		return error;
+		return -EAGAIN;
 
 	/*
 	 * The type (size and byte order) of nfscookie isn't defined in
 	 * RFC 1094.  This implementation assumes that it's an XDR uint32.
 	 */
-	entry->prev_cookie = entry->cookie;
 	p = xdr_inline_decode(xdr, 4);
 	if (unlikely(!p))
 		return -EAGAIN;

fs/nfs/nfs3xdr.c

···
 static void encode_readdirplus3args(struct xdr_stream *xdr,
 				    const struct nfs3_readdirargs *args)
 {
+	uint32_t dircount = args->count;
+	uint32_t maxcount = args->count;
 	__be32 *p;
 
 	encode_nfs_fh3(xdr, args->fh);
···
 	 * readdirplus: need dircount + buffer size.
 	 * We just make sure we make dircount big enough
 	 */
-	*p++ = cpu_to_be32(args->count >> 3);
-
-	*p = cpu_to_be32(args->count);
+	*p++ = cpu_to_be32(dircount);
+	*p = cpu_to_be32(maxcount);
 }
 
 static void nfs3_xdr_enc_readdirplus3args(struct rpc_rqst *req,
···
 		       bool plus)
 {
 	struct user_namespace *userns = rpc_userns(entry->server->client);
-	struct nfs_entry old = *entry;
 	__be32 *p;
 	int error;
 	u64 new_cookie;
···
 
 	error = decode_fileid3(xdr, &entry->ino);
 	if (unlikely(error))
-		return error;
+		return -EAGAIN;
 
 	error = decode_inline_filename3(xdr, &entry->name, &entry->len);
 	if (unlikely(error))
-		return error;
+		return -EAGAIN;
 
 	error = decode_cookie3(xdr, &new_cookie);
 	if (unlikely(error))
-		return error;
+		return -EAGAIN;
 
 	entry->d_type = DT_UNKNOWN;
 
···
 		entry->fattr->valid = 0;
 		error = decode_post_op_attr(xdr, entry->fattr, userns);
 		if (unlikely(error))
-			return error;
+			return -EAGAIN;
 		if (entry->fattr->valid & NFS_ATTR_FATTR_V3)
 			entry->d_type = nfs_umode_to_dtype(entry->fattr->mode);
 
···
 			return -EAGAIN;
 		if (*p != xdr_zero) {
 			error = decode_nfs_fh3(xdr, entry->fh);
-			if (unlikely(error)) {
-				if (error == -E2BIG)
-					goto out_truncated;
-				return error;
-			}
+			if (unlikely(error))
+				return -EAGAIN;
 		} else
 			zero_nfs_fh3(entry->fh);
 	}
 
-	entry->prev_cookie = entry->cookie;
 	entry->cookie = new_cookie;
 
 	return 0;
-
-out_truncated:
-	dprintk("NFS: directory entry contains invalid file handle\n");
-	*entry = old;
-	return -EAGAIN;
 }
 
 /*
···
 	/* ignore properties */
 	result->lease_time = 0;
 	result->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED;
+	result->xattr_support = 0;
 	return 0;
 }
 

fs/nfs/nfs42proc.c

···
 			     nfs4_stateid *src_stateid,
 			     bool *restart)
 {
-	struct nfs4_copy_state *copy, *tmp_copy;
+	struct nfs4_copy_state *copy, *tmp_copy = NULL, *iter;
 	int status = NFS4_OK;
-	bool found_pending = false;
 	struct nfs_open_context *dst_ctx = nfs_file_open_context(dst);
 	struct nfs_open_context *src_ctx = nfs_file_open_context(src);
 
-	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_NOFS);
+	copy = kzalloc(sizeof(struct nfs4_copy_state), GFP_KERNEL);
 	if (!copy)
 		return -ENOMEM;
 
 	spin_lock(&dst_server->nfs_client->cl_lock);
-	list_for_each_entry(tmp_copy,
+	list_for_each_entry(iter,
 				&dst_server->nfs_client->pending_cb_stateids,
 				copies) {
-		if (memcmp(&res->write_res.stateid, &tmp_copy->stateid,
+		if (memcmp(&res->write_res.stateid, &iter->stateid,
 				NFS4_STATEID_SIZE))
 			continue;
-		found_pending = true;
-		list_del(&tmp_copy->copies);
+		tmp_copy = iter;
+		list_del(&iter->copies);
 		break;
 	}
-	if (found_pending) {
+	if (tmp_copy) {
 		spin_unlock(&dst_server->nfs_client->cl_lock);
 		kfree(copy);
 		copy = tmp_copy;
···
 	struct nfs_commitres cres;
 	int status = -ENOMEM;
 
-	cres.verf = kzalloc(sizeof(struct nfs_writeverf), GFP_NOFS);
+	cres.verf = kzalloc(sizeof(struct nfs_writeverf), GFP_KERNEL);
 	if (!cres.verf)
 		goto out;
 
···
 	res->commit_res.verf = NULL;
 	if (args->sync) {
 		res->commit_res.verf =
-			kzalloc(sizeof(struct nfs_writeverf), GFP_NOFS);
+			kzalloc(sizeof(struct nfs_writeverf), GFP_KERNEL);
 		if (!res->commit_res.verf)
 			return -ENOMEM;
 	}
···
 	if (!(dst_server->caps & NFS_CAP_OFFLOAD_CANCEL))
 		return -EOPNOTSUPP;
 
-	data = kzalloc(sizeof(struct nfs42_offloadcancel_data), GFP_NOFS);
+	data = kzalloc(sizeof(struct nfs42_offloadcancel_data), GFP_KERNEL);
 	if (data == NULL)
 		return -ENOMEM;
 
···
 
 	ctx = get_nfs_open_context(nfs_file_open_context(src));
 	l_ctx = nfs_get_lock_context(ctx);
-	if (IS_ERR(l_ctx))
-		return PTR_ERR(l_ctx);
+	if (IS_ERR(l_ctx)) {
+		status = PTR_ERR(l_ctx);
+		goto out;
+	}
 
 	status = nfs4_set_rw_stateid(&args->cna_src_stateid, ctx, l_ctx,
 				     FMODE_READ);
···
 	if (status) {
 		if (status == -EAGAIN)
 			status = -NFS4ERR_BAD_STATEID;
-		return status;
+		goto out;
 	}
 
 	status = nfs4_call_sync(src_server->client, src_server, &msg,
···
 	if (status == -ENOTSUPP)
 		src_server->caps &= ~NFS_CAP_COPY_NOTIFY;
 
+out:
 	put_nfs_open_context(nfs_file_open_context(src));
 	return status;
 }
···
 	if (!(src_server->caps & NFS_CAP_COPY_NOTIFY))
 		return -EOPNOTSUPP;
 
-	args = kzalloc(sizeof(struct nfs42_copy_notify_args), GFP_NOFS);
+	args = kzalloc(sizeof(struct nfs42_copy_notify_args), GFP_KERNEL);
 	if (args == NULL)
 		return -ENOMEM;
 
···
 		return -EOPNOTSUPP;
 	if (n > NFS42_LAYOUTERROR_MAX)
 		return -EINVAL;
-	data = nfs42_alloc_layouterror_data(lseg, GFP_NOFS);
+	data = nfs42_alloc_layouterror_data(lseg, nfs_io_gfp_mask());
 	if (!data)
 		return -ENOMEM;
 	for (i = 0; i < n; i++) {

fs/nfs/nfs42xattr.c

···
 		flags = NFS4_XATTR_ENTRY_EXTVAL;
 	}
 
-	buf = kmalloc(alloclen, GFP_KERNEL_ACCOUNT | GFP_NOFS);
+	buf = kmalloc(alloclen, GFP_KERNEL);
 	if (buf == NULL)
 		return NULL;
 	entry = (struct nfs4_xattr_entry *)buf;
···
 
 
 	if (flags & NFS4_XATTR_ENTRY_EXTVAL) {
-		valp = kvmalloc(len, GFP_KERNEL_ACCOUNT | GFP_NOFS);
+		valp = kvmalloc(len, GFP_KERNEL);
 		if (valp == NULL) {
 			kfree(buf);
 			return NULL;
···
 {
 	struct nfs4_xattr_cache *cache;
 
-	cache = kmem_cache_alloc(nfs4_xattr_cache_cachep,
-	    GFP_KERNEL_ACCOUNT | GFP_NOFS);
+	cache = kmem_cache_alloc(nfs4_xattr_cache_cachep, GFP_KERNEL);
 	if (cache == NULL)
 		return NULL;
 

fs/nfs/nfs4_fs.h

···
 	NFS4CLNT_LEASE_MOVED,
 	NFS4CLNT_DELEGATION_EXPIRED,
 	NFS4CLNT_RUN_MANAGER,
+	NFS4CLNT_MANAGER_AVAILABLE,
 	NFS4CLNT_RECALL_RUNNING,
 	NFS4CLNT_RECALL_ANY_LAYOUT_READ,
 	NFS4CLNT_RECALL_ANY_LAYOUT_RW,

fs/nfs/nfs4file.c

···
 		if (sync)
 			return -EOPNOTSUPP;
 		cn_resp = kzalloc(sizeof(struct nfs42_copy_notify_res),
-				GFP_NOFS);
+				  GFP_KERNEL);
 		if (unlikely(cn_resp == NULL))
 			return -ENOMEM;
 
···
 	ret = nfs42_proc_copy(file_in, pos_in, file_out, pos_out, count,
 				nss, cnrs, sync);
 out:
-	if (!nfs42_files_from_same_server(file_in, file_out))
-		kfree(cn_resp);
+	kfree(cn_resp);
+
 	if (ret == -EAGAIN)
 		goto retry;
 	return ret;
···
 
 	res = ERR_PTR(-ENOMEM);
 	len = strlen(SSC_READ_NAME_BODY) + 16;
-	read_name = kzalloc(len, GFP_NOFS);
+	read_name = kzalloc(len, GFP_KERNEL);
 	if (read_name == NULL)
 		goto out;
 	snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++);

fs/nfs/nfs4proc.c

···
 	case NFS4_OPEN_CLAIM_FH:
 		p->o_arg.access = NFS4_ACCESS_READ | NFS4_ACCESS_MODIFY |
 				  NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE |
-				  NFS4_ACCESS_EXECUTE;
-#ifdef CONFIG_NFS_V4_2
-		if (!(server->caps & NFS_CAP_XATTR))
-			break;
-		p->o_arg.access |= NFS4_ACCESS_XAREAD | NFS4_ACCESS_XAWRITE |
-				   NFS4_ACCESS_XALIST;
-#endif
+				  NFS4_ACCESS_EXECUTE |
+				  nfs_access_xattr_mask(server);
 	}
 	p->o_arg.clientid = server->nfs_client->cl_clientid;
 	p->o_arg.id.create_time = ktime_to_ns(sp->so_seqid.create_time);
···
 		set_bit(NFS_STATE_POSIX_LOCKS, &state->flags);
 	if (opendata->o_res.rflags & NFS4_OPEN_RESULT_MAY_NOTIFY_LOCK)
 		set_bit(NFS_STATE_MAY_NOTIFY_LOCK, &state->flags);
+	if (opendata->o_res.rflags & NFS4_OPEN_RESULT_PRESERVE_UNLINKED)
+		set_bit(NFS_INO_PRESERVE_UNLINKED, &NFS_I(state->inode)->flags);
 
 	dentry = opendata->dentry;
 	if (d_really_is_negative(dentry)) {
···
 		buflen = server->rsize;
 
 	npages = DIV_ROUND_UP(buflen, PAGE_SIZE) + 1;
-	pages = kmalloc_array(npages, sizeof(struct page *), GFP_NOFS);
+	pages = kmalloc_array(npages, sizeof(struct page *), GFP_KERNEL);
 	if (!pages)
 		return -ENOMEM;
 
···
 	};
 	int status = 0;
 
-	data = kzalloc(sizeof(*data), GFP_NOFS);
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (data == NULL)
 		return -ENOMEM;
 
···
 	struct nfs4_state *state = lsp->ls_state;
 	struct inode *inode = state->inode;
 
-	p = kzalloc(sizeof(*p), GFP_NOFS);
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
 	if (p == NULL)
 		return NULL;
 	p->arg.fh = NFS_FH(inode);
···
 		task_setup_data.flags |= RPC_TASK_MOVEABLE;
 
 	data = nfs4_alloc_lockdata(fl, nfs_file_open_context(fl->fl_file),
-			fl->fl_u.nfs4_fl.owner,
-			recovery_type == NFS_LOCK_NEW ? GFP_KERNEL : GFP_NOFS);
+				   fl->fl_u.nfs4_fl.owner, GFP_KERNEL);
 	if (data == NULL)
 		return -ENOMEM;
 	if (IS_SETLKW(cmd))
···
 	if (server->nfs_client->cl_mvops->minor_version != 0)
 		return;
 
-	data = kmalloc(sizeof(*data), GFP_NOFS);
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return;
 	data->lsp = lsp;
···
 		.rpc_resp	= &res,
 		.rpc_cred	= cred,
 	};
+	struct nfs4_call_sync_data data = {
+		.seq_server = server,
+		.seq_args = &args.seq_args,
+		.seq_res = &res.seq_res,
+	};
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clnt,
+		.rpc_message = &msg,
+		.callback_ops = server->nfs_client->cl_mvops->call_sync_ops,
+		.callback_data = &data,
+		.flags = RPC_TASK_NO_ROUND_ROBIN,
+	};
 	int status;
 
 	nfs_fattr_init(&locations->fattr);
···
 	locations->nlocations = 0;
 
 	nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1);
-	status = nfs4_call_sync_sequence(clnt, server, &msg,
-					&args.seq_args, &res.seq_res);
+	status = nfs4_call_sync_custom(&task_setup_data);
 	if (status == NFS4_OK &&
 	    res.seq_res.sr_status_flags & SEQ4_STATUS_LEASE_MOVED)
 		status = -NFS4ERR_LEASE_MOVED;
···
 	case -NFS4ERR_DEADSESSION:
 		nfs4_schedule_session_recovery(clp->cl_session,
 				task->tk_status);
+		return;
 	}
 	if (args->dir == NFS4_CDFC4_FORE_OR_BOTH &&
 			res->dir != NFS4_CDFS4_BOTH) {
···
 		goto out_err;
 
 	ret = ERR_PTR(-ENOMEM);
-	calldata = kzalloc(sizeof(*calldata), GFP_NOFS);
+	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
 	if (calldata == NULL)
 		goto out_put_clp;
 	nfs4_init_sequence(&calldata->args, &calldata->res, 0, is_privileged);
···
 		&task_setup.rpc_client, &msg);
 
 	dprintk("NFS call  free_stateid %p\n", stateid);
-	data = kmalloc(sizeof(*data), GFP_NOFS);
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 	data->server = server;
···
 	return error + error2 + error3;
 }
 
+static void nfs4_enable_swap(struct inode *inode)
+{
+	/* The state manager thread must always be running.
+	 * It will notice the client is a swapper, and stay put.
+	 */
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
+
+	nfs4_schedule_state_manager(clp);
+}
+
+static void nfs4_disable_swap(struct inode *inode)
+{
+	/* The state manager thread will now exit once it is
+	 * woken.
+	 */
+	wake_up_var(&NFS_SERVER(inode)->nfs_client->cl_state);
+}
+
 static const struct inode_operations nfs4_dir_inode_operations = {
 	.create		= nfs_create,
 	.lookup		= nfs_lookup,
···
 	.create_server	= nfs4_create_server,
 	.clone_server	= nfs_clone_server,
 	.discover_trunking = nfs4_discover_trunking,
+	.enable_swap	= nfs4_enable_swap,
+	.disable_swap	= nfs4_disable_swap,
 };
 
 static const struct xattr_handler nfs4_xattr_nfs4_acl_handler = {

fs/nfs/nfs4state.c

···
 #include <linux/workqueue.h>
 #include <linux/bitops.h>
 #include <linux/jiffies.h>
+#include <linux/sched/mm.h>
 
 #include <linux/sunrpc/clnt.h>
 
···
 {
 	struct nfs4_state *state;
 
-	state = kzalloc(sizeof(*state), GFP_NOFS);
+	state = kzalloc(sizeof(*state), GFP_KERNEL_ACCOUNT);
 	if (!state)
 		return NULL;
 	refcount_set(&state->count, 1);
···
 
 void nfs4_close_state(struct nfs4_state *state, fmode_t fmode)
 {
-	__nfs4_close(state, fmode, GFP_NOFS, 0);
+	__nfs4_close(state, fmode, GFP_KERNEL, 0);
 }
 
 void nfs4_close_sync(struct nfs4_state *state, fmode_t fmode)
···
 	struct nfs4_lock_state *lsp;
 	struct nfs_server *server = state->owner->so_server;
 
-	lsp = kzalloc(sizeof(*lsp), GFP_NOFS);
+	lsp = kzalloc(sizeof(*lsp), GFP_KERNEL_ACCOUNT);
 	if (lsp == NULL)
 		return NULL;
 	nfs4_init_seqid_counter(&lsp->ls_seqid);
 	refcount_set(&lsp->ls_count, 1);
 	lsp->ls_state = state;
 	lsp->ls_owner = fl_owner;
-	lsp->ls_seqid.owner_id = ida_simple_get(&server->lockowner_id, 0, 0, GFP_NOFS);
+	lsp->ls_seqid.owner_id = ida_simple_get(&server->lockowner_id,
+						0, 0, GFP_KERNEL_ACCOUNT);
 	if (lsp->ls_seqid.owner_id < 0)
 		goto out_free;
 	INIT_LIST_HEAD(&lsp->ls_locks);
···
 {
 	struct task_struct *task;
 	char buf[INET6_ADDRSTRLEN + sizeof("-manager") + 1];
+	struct rpc_clnt *cl = clp->cl_rpcclient;
+
+	while (cl != cl->cl_parent)
+		cl = cl->cl_parent;
 
 	set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state);
-	if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0)
+	if (test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state) != 0) {
+		wake_up_var(&clp->cl_state);
 		return;
+	}
+	set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state);
 	__module_get(THIS_MODULE);
 	refcount_inc(&clp->cl_count);
 
···
 		printk(KERN_ERR "%s: kthread_run: %ld\n",
 			__func__, PTR_ERR(task));
 		nfs4_clear_state_manager_bit(clp);
+		clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state);
 		nfs_put_client(clp);
 		module_put(THIS_MODULE);
 	}
···
 
 static void nfs4_state_manager(struct nfs_client *clp)
 {
+	unsigned int memflags;
 	int status = 0;
 	const char *section = "", *section_sep = "";
+
+	/*
+	 * State recovery can deadlock if the direct reclaim code tries
+	 * start NFS writeback. So ensure memory allocations are all
+	 * GFP_NOFS.
+	 */
+	memflags = memalloc_nofs_save();
 
 	/* Ensure exclusive access to NFSv4 state */
 	do {
···
 			clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
 		}
 
+		memalloc_nofs_restore(memflags);
 		nfs4_end_drain_session(clp);
 		nfs4_clear_state_manager_bit(clp);
 
···
 			clear_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state);
 		}
 
-		/* Did we race with an attempt to give us more work? */
-		if (!test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state))
-			return;
-		if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0)
-			return;
+		return;
+
 	} while (refcount_read(&clp->cl_count) > 1 && !signalled());
 	goto out_drain;
 
···
 			clp->cl_hostname, -status);
 	ssleep(1);
 out_drain:
+	memalloc_nofs_restore(memflags);
 	nfs4_end_drain_session(clp);
 	nfs4_clear_state_manager_bit(clp);
 }
···
 static int nfs4_run_state_manager(void *ptr)
 {
 	struct nfs_client *clp = ptr;
+	struct rpc_clnt *cl = clp->cl_rpcclient;
+
+	while (cl != cl->cl_parent)
+		cl = cl->cl_parent;
 
 	allow_signal(SIGKILL);
+again:
+	set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state);
 	nfs4_state_manager(clp);
+	if (atomic_read(&cl->cl_swapper)) {
+		wait_var_event_interruptible(&clp->cl_state,
+					     test_bit(NFS4CLNT_RUN_MANAGER,
+						      &clp->cl_state));
+		if (atomic_read(&cl->cl_swapper) &&
+		    test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state))
+			goto again;
+		/* Either no longer a swapper, or were signalled */
+	}
+	clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state);
+
+	if (refcount_read(&clp->cl_count) > 1 && !signalled() &&
+	    test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state) &&
+	    !test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state))
+		goto again;
+
 	nfs_put_client(clp);
 	return 0;
 }

fs/nfs/nfs4xdr.c

···
 		FATTR4_WORD0_RDATTR_ERROR,
 		FATTR4_WORD1_MOUNTED_ON_FILEID,
 	};
-	uint32_t dircount = readdir->count >> 1;
+	uint32_t dircount = readdir->count;
+	uint32_t maxcount = readdir->count;
 	__be32 *p, verf[2];
 	uint32_t attrlen = 0;
 	unsigned int i;
···
 			FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
 			FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
 		attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
-		dircount >>= 1;
 	}
 	/* Use mounted_on_fileid only if the server supports it */
 	if (!(readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID))
···
 	encode_nfs4_verifier(xdr, &readdir->verifier);
 	p = reserve_space(xdr, 12 + (attrlen << 2));
 	*p++ = cpu_to_be32(dircount);
-	*p++ = cpu_to_be32(readdir->count);
+	*p++ = cpu_to_be32(maxcount);
 	*p++ = cpu_to_be32(attrlen);
 	for (i = 0; i < attrlen; i++)
 		*p++ = cpu_to_be32(attrs[i]);
···
 	if (entry->fattr->valid & NFS_ATTR_FATTR_TYPE)
 		entry->d_type = nfs_umode_to_dtype(entry->fattr->mode);
 
-	entry->prev_cookie = entry->cookie;
 	entry->cookie = new_cookie;
 
 	return 0;

fs/nfs/nfstrace.h

···
 			{ NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \
 			{ NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \
 			{ NFS_INO_INVALID_ACL, "INVALID_ACL" }, \
-			{ NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \
 			{ NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \
 			{ NFS_INO_INVALID_LABEL, "INVALID_LABEL" }, \
 			{ NFS_INO_INVALID_CHANGE, "INVALID_CHANGE" }, \
···
 
 #define nfs_show_nfsi_flags(v) \
 	__print_flags(v, "|", \
-			{ BIT(NFS_INO_ADVISE_RDPLUS), "ADVISE_RDPLUS" }, \
 			{ BIT(NFS_INO_STALE), "STALE" }, \
 			{ BIT(NFS_INO_ACL_LRU_SET), "ACL_LRU_SET" }, \
 			{ BIT(NFS_INO_INVALIDATING), "INVALIDATING" }, \
···
 DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit);
 DEFINE_NFS_INODE_EVENT(nfs_access_enter);
 DEFINE_NFS_INODE_EVENT_DONE(nfs_set_cache_invalid);
+DEFINE_NFS_INODE_EVENT(nfs_readdir_force_readdirplus);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_readdir_cache_fill_done);
+DEFINE_NFS_INODE_EVENT_DONE(nfs_readdir_uncached_done);
 
 TRACE_EVENT(nfs_access_exit,
 		TP_PROTO(
···
 DEFINE_NFS_UPDATE_SIZE_EVENT(update);
 DEFINE_NFS_UPDATE_SIZE_EVENT(grow);
 
+DECLARE_EVENT_CLASS(nfs_inode_range_event,
+		TP_PROTO(
+			const struct inode *inode,
+			loff_t range_start,
+			loff_t range_end
+		),
+
+		TP_ARGS(inode, range_start, range_end),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(u64, version)
+			__field(loff_t, range_start)
+			__field(loff_t, range_end)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->fileid = nfsi->fileid;
+			__entry->version = inode_peek_iversion_raw(inode);
+			__entry->range_start = range_start;
+			__entry->range_end = range_end;
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu "
+			"range=[%lld, %lld]",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle, __entry->version,
+			__entry->range_start, __entry->range_end
+		)
+);
+
+#define DEFINE_NFS_INODE_RANGE_EVENT(name) \
+	DEFINE_EVENT(nfs_inode_range_event, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				loff_t range_start, \
+				loff_t range_end \
+			), \
+			TP_ARGS(inode, range_start, range_end))
+
+DEFINE_NFS_INODE_RANGE_EVENT(nfs_readdir_invalidate_cache_range);
+
+DECLARE_EVENT_CLASS(nfs_readdir_event,
+		TP_PROTO(
+			const struct file *file,
+			const __be32 *verifier,
+			u64 cookie,
+			pgoff_t page_index,
+			unsigned int dtsize
+		),
+
+		TP_ARGS(file, verifier, cookie, page_index, dtsize),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(u64, version)
+			__array(char, verifier, NFS4_VERIFIER_SIZE)
+			__field(u64, cookie)
+			__field(pgoff_t, index)
+			__field(unsigned int, dtsize)
+		),
+
+		TP_fast_assign(
+			const struct inode *dir = file_inode(file);
+			const struct nfs_inode *nfsi = NFS_I(dir);
+
+			__entry->dev = dir->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
+			__entry->version = inode_peek_iversion_raw(dir);
+			if (cookie != 0)
+				memcpy(__entry->verifier, verifier,
+				       NFS4_VERIFIER_SIZE);
+			else
+				memset(__entry->verifier, 0,
+				       NFS4_VERIFIER_SIZE);
+			__entry->cookie = cookie;
+			__entry->index = page_index;
+			__entry->dtsize = dtsize;
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu "
+			"cookie=%s:0x%llx cache_index=%lu dtsize=%u",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid, __entry->fhandle,
+			__entry->version, show_nfs4_verifier(__entry->verifier),
+			(unsigned long long)__entry->cookie, __entry->index,
+			__entry->dtsize
+		)
+);
+
+#define DEFINE_NFS_READDIR_EVENT(name) \
+	DEFINE_EVENT(nfs_readdir_event, name, \
+			TP_PROTO( \
+				const struct file *file, \
+				const __be32 *verifier, \
+				u64 cookie, \
+				pgoff_t page_index, \
+				unsigned int dtsize \
+				), \
+			TP_ARGS(file, verifier, cookie, page_index, dtsize))
+
+DEFINE_NFS_READDIR_EVENT(nfs_readdir_cache_fill);
+DEFINE_NFS_READDIR_EVENT(nfs_readdir_uncached);
+
 DECLARE_EVENT_CLASS(nfs_lookup_event,
 		TP_PROTO(
 			const struct inode *dir,
···
 DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit);
 DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter);
 DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit);
+DEFINE_NFS_LOOKUP_EVENT(nfs_readdir_lookup);
+DEFINE_NFS_LOOKUP_EVENT(nfs_readdir_lookup_revalidate_failed);
+DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_readdir_lookup_revalidate);
 
 TRACE_EVENT(nfs_atomic_open_enter,
 		TP_PROTO(
···
 TRACE_EVENT(nfs_aop_readahead,
 		TP_PROTO(
 			const struct inode *inode,
-			struct page *page,
+			loff_t pos,
 			unsigned int nr_pages
 		),
 
-		TP_ARGS(inode, page, nr_pages),
+		TP_ARGS(inode, pos, nr_pages),
 
 		TP_STRUCT__entry(
 			__field(dev_t, dev)
···
 			__entry->fileid = nfsi->fileid;
 			__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
 			__entry->version = inode_peek_iversion_raw(inode);
-			__entry->offset = page_index(page) << PAGE_SHIFT;
+			__entry->offset = pos;
 			__entry->nr_pages = nr_pages;
 		),
 
···
 			__entry->eof ? " eof" : ""
 		)
 );
+
+DECLARE_EVENT_CLASS(nfs_fscache_page_event,
+		TP_PROTO(
+			const struct inode *inode,
+			struct page *page
+		),
+
+		TP_ARGS(inode, page),
+
+		TP_STRUCT__entry(
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = &nfsi->fh;
+
+			__entry->offset = page_index(page) << PAGE_SHIFT;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%lld",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			(long long)__entry->offset
+		)
+);
+DECLARE_EVENT_CLASS(nfs_fscache_page_event_done,
+		TP_PROTO(
+			const struct inode *inode,
+			struct page *page,
+			int error
+		),
+
+		TP_ARGS(inode, page, error),
+
+		TP_STRUCT__entry(
+			__field(int, error)
+			__field(dev_t, dev)
+			__field(u32, fhandle)
+			__field(u64, fileid)
+			__field(loff_t, offset)
+		),
+
+		TP_fast_assign(
+			const struct nfs_inode *nfsi = NFS_I(inode);
+			const struct nfs_fh *fh = &nfsi->fh;
+
+			__entry->offset = page_index(page) << PAGE_SHIFT;
+			__entry->dev = inode->i_sb->s_dev;
+			__entry->fileid = nfsi->fileid;
+			__entry->fhandle = nfs_fhandle_hash(fh);
+			__entry->error = error;
+		),
+
+		TP_printk(
+			"fileid=%02x:%02x:%llu fhandle=0x%08x "
+			"offset=%lld error=%d",
+			MAJOR(__entry->dev), MINOR(__entry->dev),
+			(unsigned long long)__entry->fileid,
+			__entry->fhandle,
+			(long long)__entry->offset, __entry->error
+		)
+);
+#define DEFINE_NFS_FSCACHE_PAGE_EVENT(name) \
+	DEFINE_EVENT(nfs_fscache_page_event, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				struct page *page \
+			), \
+			TP_ARGS(inode, page))
+#define DEFINE_NFS_FSCACHE_PAGE_EVENT_DONE(name) \
+	DEFINE_EVENT(nfs_fscache_page_event_done, name, \
+			TP_PROTO( \
+				const struct inode *inode, \
+				struct page *page, \
+				int error \
+			), \
+			TP_ARGS(inode, page, error))
+DEFINE_NFS_FSCACHE_PAGE_EVENT(nfs_fscache_read_page);
+DEFINE_NFS_FSCACHE_PAGE_EVENT_DONE(nfs_fscache_read_page_exit);
+DEFINE_NFS_FSCACHE_PAGE_EVENT(nfs_fscache_write_page);
+DEFINE_NFS_FSCACHE_PAGE_EVENT_DONE(nfs_fscache_write_page_exit);
 
 TRACE_EVENT(nfs_pgio_error,
 	TP_PROTO(

fs/nfs/pagelist.c

···
 	}
 }
 
-static inline struct nfs_page *
-nfs_page_alloc(void)
+static inline struct nfs_page *nfs_page_alloc(void)
 {
-	struct nfs_page	*p = kmem_cache_zalloc(nfs_page_cachep, GFP_KERNEL);
+	struct nfs_page *p =
+		kmem_cache_zalloc(nfs_page_cachep, nfs_io_gfp_mask());
 	if (p)
 		INIT_LIST_HEAD(&p->wb_list);
 	return p;
···
 	struct nfs_commit_info cinfo;
 	struct nfs_page_array *pg_array = &hdr->page_array;
 	unsigned int pagecount, pageused;
-	gfp_t gfp_flags = GFP_KERNEL;
+	gfp_t gfp_flags = nfs_io_gfp_mask();
 
 	pagecount = nfs_page_array_len(mirror->pg_base, mirror->pg_count);
 	pg_array->npages = pagecount;
···
 	desc->pg_mirrors_dynamic = NULL;
 	if (mirror_count == 1)
 		return desc->pg_mirrors_static;
-	ret = kmalloc_array(mirror_count, sizeof(*ret), GFP_KERNEL);
+	ret = kmalloc_array(mirror_count, sizeof(*ret), nfs_io_gfp_mask());
 	if (ret != NULL) {
 		for (i = 0; i < mirror_count; i++)
 			nfs_pageio_mirror_init(&ret[i], desc->pg_bsize);
···
 
 	do {
 		list_splice_init(&mirror->pg_list, &head);
+		mirror->pg_recoalesce = 0;
 
 		while (!list_empty(&head)) {
 			struct nfs_page *req;

fs/nfs/pnfs.c

···
 	return local;
 }
 
+const struct pnfs_layoutdriver_type *pnfs_find_layoutdriver(u32 id)
+{
+	return find_pnfs_driver(id);
+}
+
+void pnfs_put_layoutdriver(const struct pnfs_layoutdriver_type *ld)
+{
+	if (ld)
+		module_put(ld->owner);
+}
+
 void
 unset_pnfs_layoutdriver(struct nfs_server *nfss)
 {
···
 	int status = 0;
 
 	*pcred = NULL;
-	lrp = kzalloc(sizeof(*lrp), GFP_NOFS);
+	lrp = kzalloc(sizeof(*lrp), nfs_io_gfp_mask());
 	if (unlikely(lrp == NULL)) {
 		status = -ENOMEM;
 		spin_lock(&ino->i_lock);
···
 	struct pnfs_layout_hdr *lo;
 
 	spin_lock(&ino->i_lock);
-	lo = pnfs_find_alloc_layout(ino, ctx, GFP_KERNEL);
+	lo = pnfs_find_alloc_layout(ino, ctx, nfs_io_gfp_mask());
 	if (!lo)
 		goto out_unlock;
 	if (!test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags))
···
 	lo = _pnfs_grab_empty_layout(ino, ctx);
 	if (!lo)
 		return;
-	lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &current_stateid,
-					     &rng, GFP_KERNEL);
+	lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &current_stateid, &rng,
+					     nfs_io_gfp_mask());
 	if (!lgp) {
 		pnfs_clear_first_layoutget(lo);
 		nfs_layoutget_end(lo);
···
 	};
 	struct nfs4_layoutget *lgp;
 
-	lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &current_stateid,
-					     &rng, GFP_KERNEL);
+	lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &current_stateid, &rng,
+					     nfs_io_gfp_mask());
 	if (!lgp)
 		return;
 	data->lgp = lgp;
···
 		else
 			rd_size = nfs_dreq_bytes_left(pgio->pg_dreq);
 
-		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-						   nfs_req_openctx(req),
-						   req_offset(req),
-						   rd_size,
-						   IOMODE_READ,
-						   false,
-						   GFP_KERNEL);
+		pgio->pg_lseg =
+			pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req),
+					   req_offset(req), rd_size,
+					   IOMODE_READ, false,
+					   nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
···
 	pnfs_generic_pg_check_layout(pgio);
 	pnfs_generic_pg_check_range(pgio, req);
 	if (pgio->pg_lseg == NULL) {
-		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
-						   nfs_req_openctx(req),
-						   req_offset(req),
-						   wb_size,
-						   IOMODE_RW,
-						   false,
-						   GFP_KERNEL);
+		pgio->pg_lseg =
+			pnfs_update_layout(pgio->pg_inode, nfs_req_openctx(req),
+					   req_offset(req), wb_size, IOMODE_RW,
+					   false, nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
···
 
 	status = -ENOMEM;
 	/* Note kzalloc ensures data->res.seq_res.sr_slot == NULL */
-	data = kzalloc(sizeof(*data), GFP_NOFS);
+	data = kzalloc(sizeof(*data), nfs_io_gfp_mask());
 	if (!data)
 		goto clear_layoutcommitting;
 
···
 {
 	struct nfs4_threshold *thp;
 
-	thp = kzalloc(sizeof(*thp), GFP_NOFS);
+	thp = kzalloc(sizeof(*thp), nfs_io_gfp_mask());
 	if (!thp) {
 		dprintk("%s mdsthreshold allocation failed\n", __func__);
 		return NULL;

fs/nfs/pnfs.h

···
 
 extern int pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *);
 extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *);
+extern const struct pnfs_layoutdriver_type *pnfs_find_layoutdriver(u32 id);
+extern void pnfs_put_layoutdriver(const struct pnfs_layoutdriver_type *ld);
 
 /* nfs4proc.c */
 extern size_t max_response_pages(struct nfs_server *server);

fs/nfs/pnfs_nfs.c

···
 pnfs_bucket_fetch_commitdata(struct pnfs_commit_bucket *bucket,
 			     struct nfs_commit_info *cinfo)
 {
-	struct nfs_commit_data *data = nfs_commitdata_alloc(false);
+	struct nfs_commit_data *data = nfs_commitdata_alloc();
 
 	if (!data)
 		return NULL;
···
 	unsigned int nreq = 0;
 
 	if (!list_empty(mds_pages)) {
-		data = nfs_commitdata_alloc(true);
+		data = nfs_commitdata_alloc();
+		if (!data) {
+			nfs_retry_commit(mds_pages, NULL, cinfo, -1);
+			return -ENOMEM;
+		}
 		data->ds_commit_index = -1;
 		list_splice_init(mds_pages, &data->pages);
 		list_add_tail(&data->list, &list);

fs/nfs/proc.c

···
 	info->maxfilesize = 0x7FFFFFFF;
 	info->lease_time = 0;
 	info->change_attr_type = NFS4_CHANGE_TYPE_IS_UNDEFINED;
+	info->xattr_support = 0;
 	return 0;
 }
 

fs/nfs/read.c

···
 		struct address_space *mapping = page_file_mapping(page);
 
 		if (PageUptodate(page))
-			nfs_readpage_to_fscache(inode, page);
+			nfs_fscache_write_page(inode, page);
 		else if (!PageError(page) && !PagePrivate(page))
 			generic_error_remove_page(mapping, page);
 		unlock_page(page);
···
 			      const struct nfs_rpc_ops *rpc_ops,
 			      struct rpc_task_setup *task_setup_data, int how)
 {
-	struct inode *inode = hdr->inode;
-	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
-
-	task_setup_data->flags |= swap_flags;
 	rpc_ops->read_setup(hdr, msg);
 	trace_nfs_initiate_read(hdr);
 }
···
 }
 
 static int
-readpage_async_filler(void *data, struct page *page)
+readpage_async_filler(struct nfs_readdesc *desc, struct page *page)
 {
-	struct nfs_readdesc *desc = data;
 	struct inode *inode = page_file_mapping(page)->host;
 	unsigned int rsize = NFS_SERVER(inode)->rsize;
 	struct nfs_page *new;
···
 	aligned_len = min_t(unsigned int, ALIGN(len, rsize), PAGE_SIZE);
 
 	if (!IS_SYNC(page->mapping->host)) {
-		error = nfs_readpage_from_fscache(page->mapping->host, page);
+		error = nfs_fscache_read_page(page->mapping->host, page);
 		if (error == 0)
 			goto out_unlock;
 	}
···
 	return ret;
 }
 
-int nfs_readpages(struct file *file, struct address_space *mapping,
-		struct list_head *pages, unsigned nr_pages)
+void nfs_readahead(struct readahead_control *ractl)
 {
+	unsigned int nr_pages = readahead_count(ractl);
+	struct file *file = ractl->file;
 	struct nfs_readdesc desc;
-	struct inode *inode = mapping->host;
+	struct inode *inode = ractl->mapping->host;
+	struct page *page;
 	int ret;
 
-	trace_nfs_aop_readahead(inode, lru_to_page(pages), nr_pages);
+	trace_nfs_aop_readahead(inode, readahead_pos(ractl), nr_pages);
 	nfs_inc_stats(inode, NFSIOS_VFSREADPAGES);
 
 	ret = -ESTALE;
···
 	nfs_pageio_init_read(&desc.pgio, inode, false,
 			     &nfs_async_read_completion_ops);
 
-	ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);
+	while ((page = readahead_page(ractl)) != NULL) {
+		ret = readpage_async_filler(&desc, page);
+		put_page(page);
+		if (ret)
+			break;
+	}
 
 	nfs_pageio_complete_read(&desc.pgio);
 
 	put_nfs_open_context(desc.ctx);
 out:
 	trace_nfs_aop_readahead_done(inode, nr_pages, ret);
-	return ret;
 }
 
 int __init nfs_init_readpagecache(void)

fs/nfs/write.c

···
 static struct kmem_cache *nfs_cdata_cachep;
 static mempool_t *nfs_commit_mempool;
 
-struct nfs_commit_data *nfs_commitdata_alloc(bool never_fail)
+struct nfs_commit_data *nfs_commitdata_alloc(void)
 {
 	struct nfs_commit_data *p;
 
-	if (never_fail)
-		p = mempool_alloc(nfs_commit_mempool, GFP_NOIO);
-	else {
-		/* It is OK to do some reclaim, not no safe to wait
-		 * for anything to be returned to the pool.
-		 * mempool_alloc() cannot handle that particular combination,
-		 * so we need two separate attempts.
-		 */
+	p = kmem_cache_zalloc(nfs_cdata_cachep, nfs_io_gfp_mask());
+	if (!p) {
 		p = mempool_alloc(nfs_commit_mempool, GFP_NOWAIT);
 		if (!p)
-			p = kmem_cache_alloc(nfs_cdata_cachep, GFP_NOIO |
-					     __GFP_NOWARN | __GFP_NORETRY);
-		if (!p)
 			return NULL;
+		memset(p, 0, sizeof(*p));
 	}
-
-	memset(p, 0, sizeof(*p));
 	INIT_LIST_HEAD(&p->pages);
 	return p;
 }
···
 
 static struct nfs_pgio_header *nfs_writehdr_alloc(void)
 {
-	struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_KERNEL);
+	struct nfs_pgio_header *p;
 
-	memset(p, 0, sizeof(*p));
+	p = kmem_cache_zalloc(nfs_wdata_cachep, nfs_io_gfp_mask());
+	if (!p) {
+		p = mempool_alloc(nfs_wdata_mempool, GFP_NOWAIT);
+		if (!p)
+			return NULL;
+		memset(p, 0, sizeof(*p));
+	}
 	p->rw_mode = FMODE_WRITE;
 	return p;
 }
···
 	/* Force file size revalidation */
 	spin_lock(&inode->i_lock);
 	nfs_set_cache_invalid(inode, NFS_INO_REVAL_FORCED |
-					     NFS_INO_REVAL_PAGECACHE |
+					     NFS_INO_INVALID_CHANGE |
 					     NFS_INO_INVALID_SIZE);
 	spin_unlock(&inode->i_lock);
 }
···
 	struct address_space *mapping = page_file_mapping(page);
 
 	SetPageError(page);
-	mapping_set_error(mapping, error);
+	filemap_set_wb_err(mapping, error);
+	if (mapping->host)
+		errseq_set(&mapping->host->i_sb->s_wb_err,
+			   error == -ENOSPC ? -ENOSPC : -EIO);
 	nfs_set_pageerror(mapping);
 }
 
···
 {
 	int priority = flush_task_priority(how);
 
+	if (IS_SWAPFILE(hdr->inode))
+		task_setup_data->flags |= RPC_TASK_SWAPPER;
 	task_setup_data->priority = priority;
 	rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client);
 	trace_nfs_initiate_write(hdr);
···
 	if (list_empty(head))
 		return 0;
 
-	data = nfs_commitdata_alloc(true);
+	data = nfs_commitdata_alloc();
+	if (!data) {
+		nfs_retry_commit(head, NULL, cinfo, -1);
+		return -ENOMEM;
+	}
 
 	/* Set up the argument struct */
 	nfs_init_commit(data, head, NULL, cinfo);

include/linux/nfs_fs.h

···
 #define NFS_MAX_TRANSPORTS 16
 
 /*
- * These are the default flags for swap requests
- */
-#define NFS_RPC_SWAPFLAGS		(RPC_TASK_SWAPPER|RPC_TASK_ROOTCREDS)
-
-/*
  * Size of the NFS directory verifier
  */
 #define NFS_DIR_VERIFIER_SIZE		2
···
 
 struct nfs_open_dir_context {
 	struct list_head list;
+	atomic_t cache_hits;
+	atomic_t cache_misses;
 	unsigned long attr_gencount;
 	__be32	verf[NFS_DIR_VERIFIER_SIZE];
 	__u64 dir_cookie;
-	__u64 dup_cookie;
+	__u64 last_cookie;
 	pgoff_t page_index;
-	signed char duped;
+	unsigned int dtsize;
+	bool force_clear;
 	bool eof;
+	struct rcu_head rcu_head;
 };
 
 /*
···
 #define NFS_INO_INVALID_ATIME	BIT(2)		/* cached atime is invalid */
 #define NFS_INO_INVALID_ACCESS	BIT(3)		/* cached access cred invalid */
 #define NFS_INO_INVALID_ACL	BIT(4)		/* cached acls are invalid */
-#define NFS_INO_REVAL_PAGECACHE	BIT(5)		/* must revalidate pagecache */
 #define NFS_INO_REVAL_FORCED	BIT(6)		/* force revalidation ignoring a delegation */
 #define NFS_INO_INVALID_LABEL	BIT(7)		/* cached label is invalid */
 #define NFS_INO_INVALID_CHANGE	BIT(8)		/* cached change is invalid */
···
 /*
  * Bit offsets in flags field
  */
-#define NFS_INO_ADVISE_RDPLUS	(0)		/* advise readdirplus */
 #define NFS_INO_STALE		(1)		/* possible stale inode */
 #define NFS_INO_ACL_LRU_SET	(2)		/* Inode is on the LRU list */
 #define NFS_INO_INVALIDATING	(3)		/* inode is being invalidated */
+#define NFS_INO_PRESERVE_UNLINKED (4)		/* preserve file if removed while open */
 #define NFS_INO_FSCACHE		(5)		/* inode can be cached by FS-Cache */
-#define NFS_INO_FORCE_READDIR	(7)		/* force readdirplus */
 #define NFS_INO_LAYOUTCOMMIT	(9)		/* layoutcommit required */
 #define NFS_INO_LAYOUTCOMMITTING (10)		/* layoutcommit inflight */
 #define NFS_INO_LAYOUTSTATS	(11)		/* layoutstats inflight */
···
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	spin_lock(&inode->i_lock);
-	nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE
-		| NFS_INO_INVALID_ACCESS
-		| NFS_INO_INVALID_ACL
-		| NFS_INO_INVALID_CHANGE
-		| NFS_INO_INVALID_CTIME;
+	nfsi->cache_validity |= NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL |
+				NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME |
+				NFS_INO_INVALID_SIZE;
 	if (S_ISDIR(inode->i_mode))
 		nfsi->cache_validity |= NFS_INO_INVALID_DATA;
 	spin_unlock(&inode->i_lock);
 }
 
-static inline int nfs_server_capable(struct inode *inode, int cap)
+static inline int nfs_server_capable(const struct inode *inode, int cap)
 {
 	return NFS_SERVER(inode)->caps & cap;
 }
···
  * linux/fs/nfs/direct.c
  */
 extern ssize_t nfs_direct_IO(struct kiocb *, struct iov_iter *);
-extern ssize_t nfs_file_direct_read(struct kiocb *iocb,
-			struct iov_iter *iter);
-extern ssize_t nfs_file_direct_write(struct kiocb *iocb,
-			struct iov_iter *iter);
+ssize_t nfs_file_direct_read(struct kiocb *iocb,
+			     struct iov_iter *iter, bool swap);
+ssize_t nfs_file_direct_write(struct kiocb *iocb,
+			      struct iov_iter *iter, bool swap);
 
 /*
  * linux/fs/nfs/dir.c
···
 extern int nfs_wb_page(struct inode *inode, struct page *page);
 int nfs_wb_folio_cancel(struct inode *inode, struct folio *folio);
 extern int  nfs_commit_inode(struct inode *, int);
-extern struct nfs_commit_data *nfs_commitdata_alloc(bool never_fail);
+extern struct nfs_commit_data *nfs_commitdata_alloc(void);
 extern void nfs_commit_free(struct nfs_commit_data *data);
 bool nfs_commit_end(struct nfs_mds_commit_info *cinfo);
 
-static inline int
-nfs_have_writebacks(struct inode *inode)
+static inline bool nfs_have_writebacks(const struct inode *inode)
 {
 	if (S_ISREG(inode->i_mode))
 		return atomic_long_read(&NFS_I(inode)->nrequests) != 0;
-	return 0;
+	return false;
 }
 
 /*
  * linux/fs/nfs/read.c
  */
 extern int  nfs_readpage(struct file *, struct page *);
-extern int  nfs_readpages(struct file *, struct address_space *,
-		struct list_head *, unsigned);
+void nfs_readahead(struct readahead_control *);
 
 /*
  * inline functions

include/linux/nfs_fs_sb.h

···
 #define NFS_MOUNT_SOFTREVAL		0x800000
 #define NFS_MOUNT_WRITE_EAGER		0x01000000
 #define NFS_MOUNT_WRITE_WAIT		0x02000000
+#define NFS_MOUNT_TRUNK_DISCOVERY	0x04000000
 
 	unsigned int		fattr_valid;	/* Valid attributes */
 	unsigned int		caps;		/* server capabilities */

include/linux/nfs_xdr.h

···
  */
 struct nfs_entry {
 	__u64			ino;
-	__u64			cookie,
-				prev_cookie;
+	__u64			cookie;
 	const char *		name;
 	unsigned int		len;
 	int			eof;
···
 	struct nfs_server *(*clone_server)(struct nfs_server *, struct nfs_fh *,
 					   struct nfs_fattr *, rpc_authflavor_t);
 	int	(*discover_trunking)(struct nfs_server *, struct nfs_fh *);
+	void	(*enable_swap)(struct inode *inode);
+	void	(*disable_swap)(struct inode *inode);
 };
 
 /*

include/linux/sunrpc/auth.h

···
 
 /* Flags for rpcauth_lookupcred() */
 #define RPCAUTH_LOOKUP_NEW		0x01	/* Accept an uninitialised cred */
+#define RPCAUTH_LOOKUP_ASYNC		0x02	/* Don't block waiting for memory */
 
 /*
  * Client authentication ops

include/linux/sunrpc/sched.h

···
 #define RPC_TASK_MOVEABLE	0x0004		/* nfs4.1+ rpc tasks */
 #define RPC_TASK_NULLCREDS	0x0010		/* Use AUTH_NULL credential */
 #define RPC_CALL_MAJORSEEN	0x0020		/* major timeout seen */
-#define RPC_TASK_ROOTCREDS	0x0040		/* force root creds */
 #define RPC_TASK_DYNAMIC	0x0080		/* task was kmalloc'ed */
 #define	RPC_TASK_NO_ROUND_ROBIN	0x0100		/* send requests on "main" xprt */
 #define RPC_TASK_SOFT		0x0200		/* Use soft timeouts */
···
 extern struct workqueue_struct *rpciod_workqueue;
 extern struct workqueue_struct *xprtiod_workqueue;
 void		rpc_prepare_task(struct rpc_task *task);
+gfp_t		rpc_task_gfp_mask(void);
 
 static inline int rpc_wait_for_completion_task(struct rpc_task *task)
 {

include/linux/sunrpc/xprt.h

···
 	void		(*rpcbind)(struct rpc_task *task);
 	void		(*set_port)(struct rpc_xprt *xprt, unsigned short port);
 	void		(*connect)(struct rpc_xprt *xprt, struct rpc_task *task);
+	int		(*get_srcaddr)(struct rpc_xprt *xprt, char *buf,
+				       size_t buflen);
+	unsigned short	(*get_srcport)(struct rpc_xprt *xprt);
 	int		(*buf_alloc)(struct rpc_task *task);
 	void		(*buf_free)(struct rpc_task *task);
 	void		(*prepare_request)(struct rpc_rqst *req);

include/linux/sunrpc/xprtsock.h

···
 
 int		init_socket_xprt(void);
 void		cleanup_socket_xprt(void);
-unsigned short	get_srcport(struct rpc_xprt *);
 
 #define RPC_MIN_RESVPORT	(1U)
 #define RPC_MAX_RESVPORT	(65535U)
···
 #define XPRT_SOCK_WAKE_WRITE	(5)
 #define XPRT_SOCK_WAKE_PENDING	(6)
 #define XPRT_SOCK_WAKE_DISCONNECT	(7)
+#define XPRT_SOCK_CONNECT_SENT	(8)
+#define XPRT_SOCK_NOSPACE	(9)
 
 #endif /* _LINUX_SUNRPC_XPRTSOCK_H */

include/trace/events/sunrpc.h

···
 		{ RPC_TASK_MOVEABLE, "MOVEABLE" },			\
 		{ RPC_TASK_NULLCREDS, "NULLCREDS" },			\
 		{ RPC_CALL_MAJORSEEN, "MAJORSEEN" },			\
-		{ RPC_TASK_ROOTCREDS, "ROOTCREDS" },			\
 		{ RPC_TASK_DYNAMIC, "DYNAMIC" },			\
 		{ RPC_TASK_NO_ROUND_ROBIN, "NO_ROUND_ROBIN" },		\
 		{ RPC_TASK_SOFT, "SOFT" },				\

include/uapi/linux/nfs4.h

···
 
 #define NFS4_OPEN_RESULT_CONFIRM		0x0002
 #define NFS4_OPEN_RESULT_LOCKTYPE_POSIX		0x0004
+#define NFS4_OPEN_RESULT_PRESERVE_UNLINKED	0x0008
 #define NFS4_OPEN_RESULT_MAY_NOTIFY_LOCK	0x0020
 
 #define NFS4_SHARE_ACCESS_MASK	0x000F

include/uapi/linux/nfs_fs.h

···
 #define NFSDBG_CALLBACK		0x0100
 #define NFSDBG_CLIENT		0x0200
 #define NFSDBG_MOUNT		0x0400
-#define NFSDBG_FSCACHE		0x0800
+#define NFSDBG_FSCACHE		0x0800 /* unused */
 #define NFSDBG_PNFS		0x1000
 #define NFSDBG_PNFS_LD		0x2000
 #define NFSDBG_STATE		0x4000

net/sunrpc/auth.c

···
 	};
 	struct rpc_cred *ret;
 
+	if (RPC_IS_ASYNC(task))
+		lookupflags |= RPCAUTH_LOOKUP_ASYNC;
 	ret = auth->au_ops->lookup_cred(auth, &acred, lookupflags);
 	put_cred(acred.cred);
 	return ret;
···
 
 	if (!acred.principal)
 		return NULL;
+	if (RPC_IS_ASYNC(task))
+		lookupflags |= RPCAUTH_LOOKUP_ASYNC;
 	return auth->au_ops->lookup_cred(auth, &acred, lookupflags);
 }
 
···
 	};
 
 	if (flags & RPC_TASK_ASYNC)
-		lookupflags |= RPCAUTH_LOOKUP_NEW;
+		lookupflags |= RPCAUTH_LOOKUP_NEW | RPCAUTH_LOOKUP_ASYNC;
 	if (task->tk_op_cred)
 		/* Task must use exactly this rpc_cred */
 		new = get_rpccred(task->tk_op_cred);
···
 	/* If machine cred couldn't be bound, try a root cred */
 	if (new)
 		;
-	else if (cred == &machine_cred || (flags & RPC_TASK_ROOTCREDS))
+	else if (cred == &machine_cred)
 		new = rpcauth_bind_root_cred(task, lookupflags);
 	else if (flags & RPC_TASK_NULLCREDS)
 		new = authnull_ops.lookup_cred(NULL, NULL, 0);

net/sunrpc/auth_gss/auth_gss.c

···
 {
 	struct gss_cl_ctx *ctx;
 
-	ctx = kzalloc(sizeof(*ctx), GFP_NOFS);
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
 	if (ctx != NULL) {
 		ctx->gc_proc = RPC_GSS_PROC_DATA;
 		ctx->gc_seq = 1;	/* NetApp 6.4R1 doesn't accept seq. no. 0 */
···
 		p = ERR_PTR(-EFAULT);
 		goto err;
 	}
-	ret = gss_import_sec_context(p, seclen, gm, &ctx->gc_gss_ctx, NULL, GFP_NOFS);
+	ret = gss_import_sec_context(p, seclen, gm, &ctx->gc_gss_ctx, NULL, GFP_KERNEL);
 	if (ret < 0) {
 		trace_rpcgss_import_ctx(ret);
 		p = ERR_PTR(ret);
···
 	int vers;
 	int err = -ENOMEM;
 
-	gss_msg = kzalloc(sizeof(*gss_msg), GFP_NOFS);
+	gss_msg = kzalloc(sizeof(*gss_msg), GFP_KERNEL);
 	if (gss_msg == NULL)
 		goto err;
 	vers = get_pipe_version(gss_auth->net);
···
 	gss_msg->auth = gss_auth;
 	kref_get(&gss_auth->kref);
 	if (service_name) {
-		gss_msg->service_name = kstrdup_const(service_name, GFP_NOFS);
+		gss_msg->service_name = kstrdup_const(service_name, GFP_KERNEL);
 		if (!gss_msg->service_name) {
 			err = -ENOMEM;
 			goto err_put_pipe_version;
···
 	if (mlen > MSG_BUF_MAXSIZE)
 		goto out;
 	err = -ENOMEM;
-	buf = kmalloc(mlen, GFP_NOFS);
+	buf = kmalloc(mlen, GFP_KERNEL);
 	if (!buf)
 		goto out;
 
···
 	struct gss_cred *new;
 
 	/* Make a copy of the cred so that we can reference count it */
-	new = kzalloc(sizeof(*gss_cred), GFP_NOFS);
+	new = kzalloc(sizeof(*gss_cred), GFP_KERNEL);
 	if (new) {
 		struct auth_cred acred = {
 			.cred = gss_cred->gc_base.cr_cred,
···
 static struct rpc_cred *
 gss_lookup_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags)
 {
-	return rpcauth_lookup_credcache(auth, acred, flags, GFP_NOFS);
+	gfp_t gfp = GFP_KERNEL;
+
+	if (flags & RPCAUTH_LOOKUP_ASYNC)
+		gfp = GFP_NOWAIT | __GFP_NOWARN;
+	return rpcauth_lookup_credcache(auth, acred, flags, gfp);
 }
 
 static struct rpc_cred *
···
 	if (!p)
 		goto validate_failed;
 
-	seq = kmalloc(4, GFP_NOFS);
+	seq = kmalloc(4, GFP_KERNEL);
 	if (!seq)
 		goto validate_failed;
 	*seq = cpu_to_be32(task->tk_rqstp->rq_seqno);
···
 	rqstp->rq_enc_pages
 		= kmalloc_array(rqstp->rq_enc_pages_num,
 				sizeof(struct page *),
-				GFP_NOFS);
+				GFP_KERNEL);
 	if (!rqstp->rq_enc_pages)
 		goto out;
 	for (i=0; i < rqstp->rq_enc_pages_num; i++) {
-		rqstp->rq_enc_pages[i] = alloc_page(GFP_NOFS);
+		rqstp->rq_enc_pages[i] = alloc_page(GFP_KERNEL);
 		if (rqstp->rq_enc_pages[i] == NULL)
 			goto out_free;
 	}
···
 	if (offset + len > rcv_buf->len)
 		goto unwrap_failed;
 	mic.len = len;
-	mic.data = kmalloc(len, GFP_NOFS);
+	mic.data = kmalloc(len, GFP_KERNEL);
 	if (!mic.data)
 		goto unwrap_failed;
 	if (read_bytes_from_xdr_buf(rcv_buf, offset, mic.data, mic.len))

net/sunrpc/auth_gss/auth_gss_internal.h

···
 	if (unlikely(q > end || q < p))
 		return ERR_PTR(-EFAULT);
 	if (len) {
-		dest->data = kmemdup(p, len, GFP_NOFS);
+		dest->data = kmemdup(p, len, GFP_KERNEL);
 		if (unlikely(dest->data == NULL))
 			return ERR_PTR(-ENOMEM);
 	} else

net/sunrpc/auth_gss/gss_krb5_crypto.c

···
 		return GSS_S_FAILURE;
 	}
 
-	checksumdata = kmalloc(GSS_KRB5_MAX_CKSUM_LEN, GFP_NOFS);
+	checksumdata = kmalloc(GSS_KRB5_MAX_CKSUM_LEN, GFP_KERNEL);
 	if (checksumdata == NULL)
 		return GSS_S_FAILURE;
 
···
 	if (IS_ERR(tfm))
 		goto out_free_cksum;
 
-	req = ahash_request_alloc(tfm, GFP_NOFS);
+	req = ahash_request_alloc(tfm, GFP_KERNEL);
 	if (!req)
 		goto out_free_ahash;
 
···
 		return GSS_S_FAILURE;
 	}
 
-	checksumdata = kmalloc(GSS_KRB5_MAX_CKSUM_LEN, GFP_NOFS);
+	checksumdata = kmalloc(GSS_KRB5_MAX_CKSUM_LEN, GFP_KERNEL);
 	if (!checksumdata)
 		return GSS_S_FAILURE;
 
···
 	if (IS_ERR(tfm))
 		goto out_free_cksum;
 
-	req = ahash_request_alloc(tfm, GFP_NOFS);
+	req = ahash_request_alloc(tfm, GFP_KERNEL);
 	if (!req)
 		goto out_free_ahash;
 
···
 		WARN_ON(0);
 		return -ENOMEM;
 	}
-	data = kmalloc(GSS_KRB5_MAX_BLOCKSIZE * 2, GFP_NOFS);
+	data = kmalloc(GSS_KRB5_MAX_BLOCKSIZE * 2, GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 

net/sunrpc/auth_gss/gss_krb5_seqnum.c

···
 	unsigned char *plain;
 	s32 code;
 
-	plain = kmalloc(8, GFP_NOFS);
+	plain = kmalloc(8, GFP_KERNEL);
 	if (!plain)
 		return -ENOMEM;
 
···
 
 	dprintk("RPC:       krb5_get_seq_num:\n");
 
-	plain = kmalloc(8, GFP_NOFS);
+	plain = kmalloc(8, GFP_KERNEL);
 	if (!plain)
 		return -ENOMEM;
 

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)
 {
-	u8		*ptr, *plainhdr;
+	u8		*ptr;
 	time64_t	now;
 	u8		flags = 0x00;
 	__be16		*be16ptr;
···
 		return GSS_S_FAILURE;
 
 	/* construct gss token header */
-	ptr = plainhdr = buf->head[0].iov_base + offset;
+	ptr = buf->head[0].iov_base + offset;
 	*ptr++ = (unsigned char) ((KG2_TOK_WRAP>>8) & 0xff);
 	*ptr++ = (unsigned char) (KG2_TOK_WRAP & 0xff);
 

net/sunrpc/auth_unix.c

···
 /*
  * Lookup AUTH_UNIX creds for current process
  */
-static struct rpc_cred *
-unx_lookup_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags)
+static struct rpc_cred *unx_lookup_cred(struct rpc_auth *auth,
+					struct auth_cred *acred, int flags)
 {
-	struct rpc_cred *ret = mempool_alloc(unix_pool, GFP_NOFS);
+	struct rpc_cred *ret;
 
+	ret = kmalloc(sizeof(*ret), rpc_task_gfp_mask());
+	if (!ret) {
+		if (!(flags & RPCAUTH_LOOKUP_ASYNC))
+			return ERR_PTR(-ENOMEM);
+		ret = mempool_alloc(unix_pool, GFP_NOWAIT);
+		if (!ret)
+			return ERR_PTR(-ENOMEM);
+	}
 	rpcauth_init_cred(ret, acred, auth, &unix_credops);
 	ret->cr_flags = 1UL << RPCAUTH_CRED_UPTODATE;
 	return ret;

net/sunrpc/backchannel_rqst.c

···
 	return 0;
 }
 
-static
-struct rpc_rqst *xprt_alloc_bc_req(struct rpc_xprt *xprt, gfp_t gfp_flags)
+static struct rpc_rqst *xprt_alloc_bc_req(struct rpc_xprt *xprt)
 {
+	gfp_t gfp_flags = GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN;
 	struct rpc_rqst *req;
 
 	/* Pre-allocate one backchannel rpc_rqst */
···
 	INIT_LIST_HEAD(&tmp_list);
 	for (i = 0; i < min_reqs; i++) {
 		/* Pre-allocate one backchannel rpc_rqst */
-		req = xprt_alloc_bc_req(xprt, GFP_KERNEL);
+		req = xprt_alloc_bc_req(xprt);
 		if (req == NULL) {
 			printk(KERN_ERR "Failed to create bc rpc_rqst\n");
 			goto out_free;
···
 			break;
 		} else if (req)
 			break;
-		new = xprt_alloc_bc_req(xprt, GFP_KERNEL);
+		new = xprt_alloc_bc_req(xprt);
 	} while (new);
 	return req;
 }

net/sunrpc/clnt.c

···
 static
 void rpc_task_set_transport(struct rpc_task *task, struct rpc_clnt *clnt)
 {
-	if (task->tk_xprt)
+	if (task->tk_xprt &&
+			!(test_bit(XPRT_OFFLINE, &task->tk_xprt->state) &&
+                        (task->tk_flags & RPC_TASK_MOVEABLE)))
 		return;
 	if (task->tk_flags & RPC_TASK_NO_ROUND_ROBIN)
 		task->tk_xprt = rpc_task_get_first_xprt(clnt);
···
 		task->tk_flags |= RPC_TASK_TIMEOUT;
 	if (clnt->cl_noretranstimeo)
 		task->tk_flags |= RPC_TASK_NO_RETRANS_TIMEOUT;
-	if (atomic_read(&clnt->cl_swapper))
-		task->tk_flags |= RPC_TASK_SWAPPER;
 	/* Add to the client's list of all tasks */
 	spin_lock(&clnt->cl_lock);
 	list_add_tail(&task->tk_task, &clnt->cl_tasks);
···
 			break;
 		task->tk_cred_retry--;
 		trace_rpc_retry_refresh_status(task);
+		return;
+	case -ENOMEM:
+		rpc_delay(task, HZ >> 4);
 		return;
 	}
 	trace_rpc_refresh_status(task);
···
 		return -EINVAL;
 	}
 
-	data = kmalloc(sizeof(*data), GFP_NOFS);
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 	data->xps = xprt_switch_get(xps);
···
 int
 rpc_clnt_swap_activate(struct rpc_clnt *clnt)
 {
+	while (clnt != clnt->cl_parent)
+		clnt = clnt->cl_parent;
 	if (atomic_inc_return(&clnt->cl_swapper) == 1)
 		return rpc_clnt_iterate_for_each_xprt(clnt,
 				rpc_clnt_swap_activate_callback, NULL);

net/sunrpc/rpcb_clnt.c

···
 		goto bailout_nofree;
 	}
 
-	map = kzalloc(sizeof(struct rpcbind_args), GFP_NOFS);
+	map = kzalloc(sizeof(struct rpcbind_args), rpc_task_gfp_mask());
 	if (!map) {
 		status = -ENOMEM;
 		goto bailout_release_client;
···
 	case RPCBVERS_4:
 	case RPCBVERS_3:
 		map->r_netid = xprt->address_strings[RPC_DISPLAY_NETID];
-		map->r_addr = rpc_sockaddr2uaddr(sap, GFP_NOFS);
+		map->r_addr = rpc_sockaddr2uaddr(sap, rpc_task_gfp_mask());
 		if (!map->r_addr) {
 			status = -ENOMEM;
 			goto bailout_free_args;

net/sunrpc/sched.c

···
 struct workqueue_struct *xprtiod_workqueue __read_mostly;
 EXPORT_SYMBOL_GPL(xprtiod_workqueue);
 
+gfp_t rpc_task_gfp_mask(void)
+{
+	if (current->flags & PF_WQ_WORKER)
+		return GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN;
+	return GFP_KERNEL;
+}
+
 unsigned long
 rpc_task_timeout(const struct rpc_task *task)
 {
···
 
 /*
  * Add new request to wait queue.
- *
- * Swapper tasks always get inserted at the head of the queue.
- * This should avoid many nasty memory deadlocks and hopefully
- * improve overall performance.
- * Everyone else gets appended to the queue to ensure proper FIFO behavior.
  */
 static void __rpc_add_wait_queue(struct rpc_wait_queue *queue,
 		struct rpc_task *task,
···
 	INIT_LIST_HEAD(&task->u.tk_wait.timer_list);
 	if (RPC_IS_PRIORITY(queue))
 		__rpc_add_wait_queue_priority(queue, task, queue_priority);
-	else if (RPC_IS_SWAPPER(task))
-		list_add(&task->u.tk_wait.list, &queue->tasks[0]);
 	else
 		list_add_tail(&task->u.tk_wait.list, &queue->tasks[0]);
 	task->tk_waitqueue = queue;
···
 		ops->rpc_release(calldata);
 }
 
+static bool xprt_needs_memalloc(struct rpc_xprt *xprt, struct rpc_task *tk)
+{
+	if (!xprt)
+		return false;
+	if (!atomic_read(&xprt->swapper))
+		return false;
+	return test_bit(XPRT_LOCKED, &xprt->state) && xprt->snd_task == tk;
+}
+
 /*
  * This is the RPC `scheduler' (or rather, the finite state machine).
  */
···
 	struct rpc_wait_queue *queue;
 	int task_is_async = RPC_IS_ASYNC(task);
 	int status = 0;
+	unsigned long pflags = current->flags;
 
 	WARN_ON_ONCE(RPC_IS_QUEUED(task));
 	if (RPC_IS_QUEUED(task))
···
 		}
 		if (!do_action)
 			break;
+		if (RPC_IS_SWAPPER(task) ||
+		    xprt_needs_memalloc(task->tk_xprt, task))
+			current->flags |= PF_MEMALLOC;
+
 		trace_rpc_task_run_action(task, do_action);
 		do_action(task);
 
···
 		rpc_clear_running(task);
 		spin_unlock(&queue->lock);
 		if (task_is_async)
-			return;
+			goto out;
 
 		/* sync task: sleep here */
 		trace_rpc_task_sync_sleep(task, task->tk_action);
···
 
 	/* Release all resources associated with the task */
 	rpc_release_task(task);
+out:
+	current_restore_flags(pflags, PF_MEMALLOC);
 }
 
 /*
···
 	struct rpc_rqst *rqst = task->tk_rqstp;
 	size_t size = rqst->rq_callsize + rqst->rq_rcvsize;
 	struct rpc_buffer *buf;
-	gfp_t gfp = GFP_NOFS;
-
-	if (RPC_IS_SWAPPER(task))
-		gfp = __GFP_MEMALLOC | GFP_NOWAIT | __GFP_NOWARN;
+	gfp_t gfp = rpc_task_gfp_mask();
 
 	size += sizeof(struct rpc_buffer);
-	if (size <= RPC_BUFFER_MAXSIZE)
-		buf = mempool_alloc(rpc_buffer_mempool, gfp);
-	else
+	if (size <= RPC_BUFFER_MAXSIZE) {
+		buf = kmem_cache_alloc(rpc_buffer_slabp, gfp);
+		/* Reach for the mempool if dynamic allocation fails */
+		if (!buf && RPC_IS_ASYNC(task))
+			buf = mempool_alloc(rpc_buffer_mempool, GFP_NOWAIT);
+	} else
 		buf = kmalloc(size, gfp);
 
 	if (!buf)
···
 	rpc_init_task_statistics(task);
 }
 
-static struct rpc_task *
-rpc_alloc_task(void)
+static struct rpc_task *rpc_alloc_task(void)
 {
-	return (struct rpc_task *)mempool_alloc(rpc_task_mempool, GFP_NOFS);
+	struct rpc_task *task;
+
+	task = kmem_cache_alloc(rpc_task_slabp, rpc_task_gfp_mask());
+	if (task)
+		return task;
+	return mempool_alloc(rpc_task_mempool, GFP_NOWAIT);
 }
 
 /*

net/sunrpc/socklib.c

···
 #include <linux/pagemap.h>
 #include <linux/udp.h>
 #include <linux/sunrpc/msg_prot.h>
+#include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/xdr.h>
 #include <linux/export.h>
 
···
 {
 	int err;
 
-	err = xdr_alloc_bvec(xdr, GFP_KERNEL);
+	err = xdr_alloc_bvec(xdr, rpc_task_gfp_mask());
 	if (err < 0)
 		return err;
 

net/sunrpc/sysfs.c

···
 	struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj);
 	ssize_t ret;
 
-	if (!xprt)
-		return 0;
+	if (!xprt) {
+		ret = sprintf(buf, "<closed>\n");
+		goto out;
+	}
 	ret = sprintf(buf, "%s\n", xprt->address_strings[RPC_DISPLAY_ADDR]);
 	xprt_put(xprt);
-	return ret + 1;
+out:
+	return ret;
 }
 
 static ssize_t rpc_sysfs_xprt_srcaddr_show(struct kobject *kobj,
···
 					   char *buf)
 {
 	struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj);
-	struct sockaddr_storage saddr;
-	struct sock_xprt *sock;
-	ssize_t ret = -1;
-
-	if (!xprt || !xprt_connected(xprt)) {
-		xprt_put(xprt);
-		return -ENOTCONN;
-	}
-
-	sock = container_of(xprt, struct sock_xprt, xprt);
-	mutex_lock(&sock->recv_mutex);
-	if (sock->sock == NULL ||
-	    kernel_getsockname(sock->sock, (struct sockaddr *)&saddr) < 0)
-		goto out;
-
-	ret = sprintf(buf, "%pISc\n", &saddr);
-out:
-	mutex_unlock(&sock->recv_mutex);
-	xprt_put(xprt);
-	return ret + 1;
-}
-
-static ssize_t rpc_sysfs_xprt_info_show(struct kobject *kobj,
-					struct kobj_attribute *attr,
-					char *buf)
-{
-	struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj);
+	size_t buflen = PAGE_SIZE;
 	ssize_t ret;
 
 	if (!xprt || !xprt_connected(xprt)) {
-		xprt_put(xprt);
-		return -ENOTCONN;
+		ret = sprintf(buf, "<closed>\n");
+	} else if (xprt->ops->get_srcaddr) {
+		ret = xprt->ops->get_srcaddr(xprt, buf, buflen);
+		if (ret > 0) {
+			if (ret < buflen - 1) {
+				buf[ret] = '\n';
+				ret++;
+				buf[ret] = '\0';
+			}
+		} else
+			ret = sprintf(buf, "<closed>\n");
+	} else
+		ret = sprintf(buf, "<not a socket>\n");
+	xprt_put(xprt);
+	return ret;
+}
+
+static ssize_t rpc_sysfs_xprt_info_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct rpc_xprt *xprt = rpc_sysfs_xprt_kobj_get_xprt(kobj);
+	unsigned short srcport = 0;
+	size_t buflen = PAGE_SIZE;
+	ssize_t ret;
+
+	if (!xprt || !xprt_connected(xprt)) {
+		ret = sprintf(buf, "<closed>\n");
+		goto out;
 	}
 
-	ret = sprintf(buf, "last_used=%lu\ncur_cong=%lu\ncong_win=%lu\n"
+	if (xprt->ops->get_srcport)
+		srcport = xprt->ops->get_srcport(xprt);
+
+	ret = snprintf(buf, buflen,
+		       "last_used=%lu\ncur_cong=%lu\ncong_win=%lu\n"
 		       "max_num_slots=%u\nmin_num_slots=%u\nnum_reqs=%u\n"
 		       "binding_q_len=%u\nsending_q_len=%u\npending_q_len=%u\n"
 		       "backlog_q_len=%u\nmain_xprt=%d\nsrc_port=%u\n"
···
 		       xprt->last_used, xprt->cong, xprt->cwnd, xprt->max_reqs,
 		       xprt->min_reqs, xprt->num_reqs, xprt->binding.qlen,
 		       xprt->sending.qlen, xprt->pending.qlen,
-		       xprt->backlog.qlen, xprt->main,
-		       (xprt->xprt_class->ident == XPRT_TRANSPORT_TCP) ?
-		       get_srcport(xprt) : 0,
+		       xprt->backlog.qlen, xprt->main, srcport,
 		       atomic_long_read(&xprt->queuelen),
-		       (xprt->xprt_class->ident == XPRT_TRANSPORT_TCP) ?
-				xprt->address_strings[RPC_DISPLAY_PORT] : "0");
+		       xprt->address_strings[RPC_DISPLAY_PORT]);
+out:
 	xprt_put(xprt);
-	return ret + 1;
+	return ret;
 }
 
 static ssize_t rpc_sysfs_xprt_state_show(struct kobject *kobj,
···
 	int locked, connected, connecting, close_wait, bound, binding,
 	    closing, congested, cwnd_wait, write_space, offline, remove;
 
-	if (!xprt)
-		return 0;
-
-	if (!xprt->state) {
+	if (!(xprt && xprt->state)) {
 		ret = sprintf(buf, "state=CLOSED\n");
 	} else {
 		locked = test_bit(XPRT_LOCKED, &xprt->state);
···
 	}
 
 	xprt_put(xprt);
-	return ret + 1;
+	return ret;
 }
 
 static ssize_t rpc_sysfs_xprt_switch_info_show(struct kobject *kobj,
···
 		      xprt_switch->xps_nunique_destaddr_xprts,
 		      atomic_long_read(&xprt_switch->xps_queuelen));
 	xprt_switch_put(xprt_switch);
-	return ret + 1;
+	return ret;
 }
 
 static ssize_t rpc_sysfs_xprt_dstaddr_store(struct kobject *kobj,

net/sunrpc/xprt.c

···
 				INIT_LIST_HEAD(&req->rq_xmit2);
 				goto out;
 			}
-		} else if (RPC_IS_SWAPPER(task)) {
-			list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
-				if (pos->rq_cong || pos->rq_bytes_sent)
-					continue;
-				if (RPC_IS_SWAPPER(pos->rq_task))
-					continue;
-				/* Note: req is added _before_ pos */
-				list_add_tail(&req->rq_xmit, &pos->rq_xmit);
-				INIT_LIST_HEAD(&req->rq_xmit2);
-				goto out;
-			}
 		} else if (!req->rq_seqno) {
 			list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
 				if (pos->rq_task->tk_owner != task->tk_owner)
···
 		return false;
 
 	}
+	if (atomic_read(&xprt->swapper))
+		/* This will be clear in __rpc_execute */
+		current->flags |= PF_MEMALLOC;
 	return true;
 }
 
···
 		goto out;
 	++xprt->num_reqs;
 	spin_unlock(&xprt->reserve_lock);
-	req = kzalloc(sizeof(struct rpc_rqst), GFP_NOFS);
+	req = kzalloc(sizeof(*req), rpc_task_gfp_mask());
 	spin_lock(&xprt->reserve_lock);
 	if (req != NULL)
 		goto out;
···
 	 */
 	wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_UNINTERRUPTIBLE);
 
+	/*
+	 * xprt_schedule_autodisconnect() can run after XPRT_LOCKED
+	 * is cleared.  We use ->transport_lock to ensure the mod_timer()
+	 * can only run *before* del_time_sync(), never after.
+	 */
+	spin_lock(&xprt->transport_lock);
 	del_timer_sync(&xprt->timer);
+	spin_unlock(&xprt->transport_lock);
 
 	/*
 	 * Destroy sockets etc from the system workqueue so they can

net/sunrpc/xprtrdma/frwr_ops.c

···
 	if (IS_ERR(frmr))
 		goto out_mr_err;
 
-	sg = kmalloc_array(depth, sizeof(*sg), GFP_NOFS);
+	sg = kmalloc_array(depth, sizeof(*sg), GFP_KERNEL);
 	if (!sg)
 		goto out_list_err;
 

net/sunrpc/xprtrdma/transport.c

···
 	struct rpcrdma_xprt *r_xprt = container_of(work, struct rpcrdma_xprt,
 						   rx_connect_worker.work);
 	struct rpc_xprt *xprt = &r_xprt->rx_xprt;
+	unsigned int pflags = current->flags;
 	int rc;
 
+	if (atomic_read(&xprt->swapper))
+		current->flags |= PF_MEMALLOC;
 	rc = rpcrdma_xprt_connect(r_xprt);
 	xprt_clear_connecting(xprt);
 	if (!rc) {
···
 		rpcrdma_xprt_disconnect(r_xprt);
 	xprt_unlock_connect(xprt, r_xprt);
 	xprt_wake_pending_tasks(xprt, rc);
+	current_restore_flags(pflags, PF_MEMALLOC);
 }
 
 /**
···
 	return;
 
 out_sleep:
-	task->tk_status = -EAGAIN;
+	task->tk_status = -ENOMEM;
 	xprt_add_backlog(xprt, task);
 }
 
···
 	gfp_t flags;
 
 	flags = RPCRDMA_DEF_GFP;
-	if (RPC_IS_SWAPPER(task))
-		flags = __GFP_MEMALLOC | GFP_NOWAIT | __GFP_NOWARN;
+	if (RPC_IS_ASYNC(task))
+		flags = GFP_NOWAIT | __GFP_NOWARN;
 
 	if (!rpcrdma_check_regbuf(r_xprt, req->rl_sendbuf, rqst->rq_callsize,
 				  flags))

net/sunrpc/xprtrdma/verbs.c

···
 	struct rpcrdma_ep *ep;
 	int rc;
 
-	ep = kzalloc(sizeof(*ep), GFP_NOFS);
+	ep = kzalloc(sizeof(*ep), GFP_KERNEL);
 	if (!ep)
 		return -ENOTCONN;
 	ep->re_xprt = &r_xprt->rx_xprt;
···
 		struct rpcrdma_mr *mr;
 		int rc;
 
-		mr = kzalloc(sizeof(*mr), GFP_NOFS);
+		mr = kzalloc(sizeof(*mr), GFP_KERNEL);
 		if (!mr)
 			break;
 

net/sunrpc/xprtsock.c

···
 #include "sunrpc.h"
 
 static void xs_close(struct rpc_xprt *xprt);
+static void xs_set_srcport(struct sock_xprt *transport, struct socket *sock);
 static void xs_tcp_set_socket_timeouts(struct rpc_xprt *xprt,
 		struct socket *sock);
 
···
 		offset += want;
 	}
 
-	want = xs_alloc_sparse_pages(buf,
-			min_t(size_t, count - offset, buf->page_len),
-			GFP_KERNEL);
+	want = xs_alloc_sparse_pages(
+		buf, min_t(size_t, count - offset, buf->page_len),
+		GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
 	if (seek < want) {
 		ret = xs_read_bvec(sock, msg, flags, buf->bvec,
 				xdr_buf_pagecount(buf),
···
 /**
  * xs_nospace - handle transmit was incomplete
  * @req: pointer to RPC request
+ * @transport: pointer to struct sock_xprt
  *
  */
-static int xs_nospace(struct rpc_rqst *req)
+static int xs_nospace(struct rpc_rqst *req, struct sock_xprt *transport)
 {
-	struct rpc_xprt *xprt = req->rq_xprt;
-	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
+	struct rpc_xprt *xprt = &transport->xprt;
 	struct sock *sk = transport->inet;
 	int ret = -EAGAIN;
 
···
 	/* Don't race with disconnect */
 	if (xprt_connected(xprt)) {
 		/* wait for more buffer space */
+		set_bit(XPRT_SOCK_NOSPACE, &transport->sock_state);
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
 		sk->sk_write_pending++;
 		xprt_wait_for_buffer_space(xprt);
 	} else
 		ret = -ENOTCONN;
 
 	spin_unlock(&xprt->transport_lock);
+	return ret;
+}
 
-	/* Race breaker in case memory is freed before above code is called */
-	if (ret == -EAGAIN) {
-		struct socket_wq *wq;
+static int xs_sock_nospace(struct rpc_rqst *req)
+{
+	struct sock_xprt *transport =
+		container_of(req->rq_xprt, struct sock_xprt, xprt);
+	struct sock *sk = transport->inet;
+	int ret = -EAGAIN;
 
-		rcu_read_lock();
-		wq = rcu_dereference(sk->sk_wq);
-		set_bit(SOCKWQ_ASYNC_NOSPACE, &wq->flags);
-		rcu_read_unlock();
+	lock_sock(sk);
+	if (!sock_writeable(sk))
+		ret = xs_nospace(req, transport);
+	release_sock(sk);
+	return ret;
+}
 
-		sk->sk_write_space(sk);
-	}
+static int xs_stream_nospace(struct rpc_rqst *req, bool vm_wait)
+{
+	struct sock_xprt *transport =
+		container_of(req->rq_xprt, struct sock_xprt, xprt);
+	struct sock *sk = transport->inet;
+	int ret = -EAGAIN;
+
+	if (vm_wait)
+		return -ENOBUFS;
+	lock_sock(sk);
+	if (!sk_stream_memory_free(sk))
+		ret = xs_nospace(req, transport);
+	release_sock(sk);
 	return ret;
 }
 
···
 xs_stream_prepare_request(struct rpc_rqst *req)
 {
 	xdr_free_bvec(&req->rq_rcv_buf);
-	req->rq_task->tk_status = xdr_alloc_bvec(&req->rq_rcv_buf, GFP_KERNEL);
+	req->rq_task->tk_status = xdr_alloc_bvec(
+		&req->rq_rcv_buf, GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
 }
 
 /*
···
 	struct msghdr msg = {
 		.msg_flags	= XS_SENDMSG_FLAGS,
 	};
+	bool vm_wait;
 	unsigned int sent;
 	int status;
 
···
 	xs_pktdump("packet data:",
 			req->rq_svec->iov_base, req->rq_svec->iov_len);
 
+	vm_wait = sk_stream_is_writeable(transport->inet) ? true : false;
+
 	req->rq_xtime = ktime_get();
 	status = xprt_sock_sendmsg(transport->sock, &msg, xdr,
 				   transport->xmit.offset, rm, &sent);
 	dprintk("RPC:       %s(%u) = %d\n",
 			__func__, xdr->len - transport->xmit.offset, status);
-
-	if (status == -EAGAIN && sock_writeable(transport->inet))
-		status = -ENOBUFS;
 
 	if (likely(sent > 0) || status == 0) {
 		transport->xmit.offset += sent;
···
 			return 0;
 		}
 		status = -EAGAIN;
+		vm_wait = false;
 	}
 
 	switch (status) {
-	case -ENOBUFS:
-		break;
 	case -EAGAIN:
-		status = xs_nospace(req);
+		status = xs_stream_nospace(req, vm_wait);
 		break;
 	default:
 		dprintk("RPC:       sendmsg returned unrecognized error %d\n",
···
 		/* Should we call xs_close() here? */
 		break;
 	case -EAGAIN:
-		status = xs_nospace(req);
+		status = xs_sock_nospace(req);
 		break;
 	case -ENETUNREACH:
 	case -ENOBUFS:
···
 	struct msghdr msg = {
 		.msg_flags	= XS_SENDMSG_FLAGS,
 	};
-	bool vm_wait = false;
+	bool vm_wait;
 	unsigned int sent;
 	int status;
 
···
 	if (test_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state))
 		xs_tcp_set_socket_timeouts(xprt, transport->sock);
 
+	xs_set_srcport(transport, transport->sock);
+
 	/* Continue transmitting the packet/record. We must be careful
 	 * to cope with writespace callbacks arriving _after_ we have
 	 * called sendmsg(). */
 	req->rq_xtime = ktime_get();
 	tcp_sock_set_cork(transport->inet, true);
-	while (1) {
+
+	vm_wait = sk_stream_is_writeable(transport->inet) ? true : false;
+
+	do {
 		status = xprt_sock_sendmsg(transport->sock, &msg, xdr,
 					   transport->xmit.offset, rm, &sent);
 
···
 
 		WARN_ON_ONCE(sent == 0 && status == 0);
 
-		if (status == -EAGAIN ) {
-			/*
-			 * Return EAGAIN if we're sure we're hitting the
-			 * socket send buffer limits.
-			 */
-			if (test_bit(SOCK_NOSPACE, &transport->sock->flags))
-				break;
-			/*
-			 * Did we hit a memory allocation failure?
-			 */
-			if (sent == 0) {
-				status = -ENOBUFS;
-				if (vm_wait)
-					break;
-				/* Retry, knowing now that we're below the
-				 * socket send buffer limit
-				 */
-				vm_wait = true;
-			}
-			continue;
-		}
-		if (status < 0)
-			break;
-		vm_wait = false;
-	}
+		if (sent > 0)
+			vm_wait = false;
+
+	} while (status == 0);
 
 	switch (status) {
 	case -ENOTSOCK:
···
 		/* Should we call xs_close() here? */
 		break;
 	case -EAGAIN:
-		status = xs_nospace(req);
+		status = xs_stream_nospace(req, vm_wait);
 		break;
 	case -ECONNRESET:
 	case -ECONNREFUSED:
···
 	clear_bit(XPRT_SOCK_WAKE_ERROR, &transport->sock_state);
 	clear_bit(XPRT_SOCK_WAKE_WRITE, &transport->sock_state);
 	clear_bit(XPRT_SOCK_WAKE_DISCONNECT, &transport->sock_state);
+	clear_bit(XPRT_SOCK_NOSPACE, &transport->sock_state);
 }
 
 static void xs_run_error_worker(struct sock_xprt *transport, unsigned int nr)
···
 
 static void xs_write_space(struct sock *sk)
 {
-	struct socket_wq *wq;
 	struct sock_xprt *transport;
 	struct rpc_xprt *xprt;
 
···
 	if (unlikely(!(xprt = xprt_from_sock(sk))))
 		return;
 	transport = container_of(xprt, struct sock_xprt, xprt);
-	rcu_read_lock();
-	wq = rcu_dereference(sk->sk_wq);
-	if (!wq || test_and_clear_bit(SOCKWQ_ASYNC_NOSPACE, &wq->flags) == 0)
-		goto out;
-
+	if (!test_and_clear_bit(XPRT_SOCK_NOSPACE, &transport->sock_state))
+		return;
 	xs_run_error_worker(transport, XPRT_SOCK_WAKE_WRITE);
 	sk->sk_write_pending--;
-out:
-	rcu_read_unlock();
 }
 
 /**
···
 	return port;
 }
 
-unsigned short get_srcport(struct rpc_xprt *xprt)
+static unsigned short xs_sock_srcport(struct rpc_xprt *xprt)
 {
 	struct sock_xprt *sock = container_of(xprt, struct sock_xprt, xprt);
 	unsigned short ret = 0;
···
 	mutex_unlock(&sock->recv_mutex);
 	return ret;
 }
-EXPORT_SYMBOL(get_srcport);
+
+static int xs_sock_srcaddr(struct rpc_xprt *xprt, char *buf, size_t buflen)
+{
+	struct sock_xprt *sock = container_of(xprt, struct sock_xprt, xprt);
+	union {
+		struct sockaddr sa;
+		struct sockaddr_storage st;
+	} saddr;
+	int ret = -ENOTCONN;
+
+	mutex_lock(&sock->recv_mutex);
+	if (sock->sock) {
+		ret = kernel_getsockname(sock->sock, &saddr.sa);
+		if (ret >= 0)
+			ret = snprintf(buf, buflen, "%pISc", &saddr.sa);
+	}
+	mutex_unlock(&sock->recv_mutex);
+	return ret;
+}
 
 static unsigned short xs_next_srcport(struct sock_xprt *transport, unsigned short port)
 {
···
 		sk->sk_user_data = xprt;
 		sk->sk_data_ready = xs_data_ready;
 		sk->sk_write_space = xs_udp_write_space;
-		sock_set_flag(sk, SOCK_FASYNC);
 		sk->sk_error_report = xs_error_report;
 
 		xprt_clear_connected(xprt);
···
 
 #if IS_ENABLED(CONFIG_SUNRPC_SWAP)
 /*
- * Note that this should be called with XPRT_LOCKED held (or when we otherwise
- * know that we have exclusive access to the socket), to guard against
- * races with xs_reset_transport.
+ * Note that this should be called with XPRT_LOCKED held, or recv_mutex
+ * held, or when we otherwise know that we have exclusive access to the
+ * socket, to guard against races with xs_reset_transport.
  */
 static void xs_set_memalloc(struct rpc_xprt *xprt)
 {
···
 {
 	struct sock_xprt *xs = container_of(xprt, struct sock_xprt, xprt);
 
-	if (atomic_inc_return(&xprt->swapper) != 1)
-		return 0;
-	if (wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_KILLABLE))
-		return -ERESTARTSYS;
-	if (xs->inet)
+	mutex_lock(&xs->recv_mutex);
+	if (atomic_inc_return(&xprt->swapper) == 1 &&
+	    xs->inet)
 		sk_set_memalloc(xs->inet);
-	xprt_release_xprt(xprt, NULL);
+	mutex_unlock(&xs->recv_mutex);
 	return 0;
 }
 
···
 {
 	struct sock_xprt *xs = container_of(xprt, struct sock_xprt, xprt);
 
-	if (!atomic_dec_and_test(&xprt->swapper))
-		return;
-	if (wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_KILLABLE))
-		return;
-	if (xs->inet)
+	mutex_lock(&xs->recv_mutex);
+	if (atomic_dec_and_test(&xprt->swapper) &&
+	    xs->inet)
 		sk_clear_memalloc(xs->inet);
-	xprt_release_xprt(xprt, NULL);
+	mutex_unlock(&xs->recv_mutex);
 }
 #else
 static void xs_set_memalloc(struct rpc_xprt *xprt)
···
 		sk->sk_user_data = xprt;
 		sk->sk_data_ready = xs_data_ready;
 		sk->sk_write_space = xs_udp_write_space;
-		sock_set_flag(sk, SOCK_FASYNC);
 
 		xprt_set_connected(xprt);
 
···
 	struct rpc_xprt *xprt = &transport->xprt;
 	struct socket *sock;
 	int status = -EIO;
+	unsigned int pflags = current->flags;
 
+	if (atomic_read(&xprt->swapper))
+		current->flags |= PF_MEMALLOC;
 	sock = xs_create_sock(xprt, transport,
 			xs_addr(xprt)->sa_family, SOCK_DGRAM,
 			IPPROTO_UDP, false);
···
 	xprt_clear_connecting(xprt);
 	xprt_unlock_connect(xprt, transport);
 	xprt_wake_pending_tasks(xprt, status);
+	current_restore_flags(pflags, PF_MEMALLOC);
 }
 
 /**
···
 		sk->sk_data_ready = xs_data_ready;
 		sk->sk_state_change = xs_tcp_state_change;
 		sk->sk_write_space = xs_tcp_write_space;
-		sock_set_flag(sk, SOCK_FASYNC);
 		sk->sk_error_report = xs_error_report;
 
 		/* socket options */
···
 	struct socket *sock = transport->sock;
 	struct rpc_xprt *xprt = &transport->xprt;
 	int status;
+	unsigned int pflags = current->flags;
 
-	if (!sock) {
-		sock = xs_create_sock(xprt, transport,
-				xs_addr(xprt)->sa_family, SOCK_STREAM,
-				IPPROTO_TCP, true);
+	if (atomic_read(&xprt->swapper))
+		current->flags |= PF_MEMALLOC;
+
+	if (xprt_connected(xprt))
+		goto out;
+	if (test_and_clear_bit(XPRT_SOCK_CONNECT_SENT,
+			       &transport->sock_state) ||
+	    !sock) {
+		xs_reset_transport(transport);
+		sock = xs_create_sock(xprt, transport, xs_addr(xprt)->sa_family,
+				      SOCK_STREAM, IPPROTO_TCP, true);
 		if (IS_ERR(sock)) {
 			xprt_wake_pending_tasks(xprt, PTR_ERR(sock));
 			goto out;
···
 			sock->sk->sk_state);
 	switch (status) {
 	case 0:
-		xs_set_srcport(transport, sock);
-		fallthrough;
 	case -EINPROGRESS:
 		/* SYN_SENT! */
+		set_bit(XPRT_SOCK_CONNECT_SENT, &transport->sock_state);
 		if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO)
 			xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO;
 		fallthrough;
···
 	xprt_clear_connecting(xprt);
 out_unlock:
 	xprt_unlock_connect(xprt, transport);
+	current_restore_flags(pflags, PF_MEMALLOC);
 }
 
 /**
···
 
 	WARN_ON_ONCE(!xprt_lock_connect(xprt, task, transport));
 
-	if (transport->sock != NULL && !xprt_connecting(xprt)) {
+	if (transport->sock != NULL) {
 		dprintk("RPC:       xs_connect delayed xprt %p for %lu "
-				"seconds\n",
-				xprt, xprt->reestablish_timeout / HZ);
-
-		/* Start by resetting any existing state */
-		xs_reset_transport(transport);
+			"seconds\n", xprt, xprt->reestablish_timeout / HZ);
 
 		delay = xprt_reconnect_delay(xprt);
 		xprt_reconnect_backoff(xprt, XS_TCP_INIT_REEST_TO);
···
 		return -EINVAL;
 	}
 
-	page = alloc_page(GFP_KERNEL);
+	page = alloc_page(GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
 	if (!page)
 		return -ENOMEM;
 
···
 	.rpcbind		= rpcb_getport_async,
 	.set_port		= xs_set_port,
 	.connect		= xs_connect,
+	.get_srcaddr		= xs_sock_srcaddr,
+	.get_srcport		= xs_sock_srcport,
 	.buf_alloc		= rpc_malloc,
 	.buf_free		= rpc_free,
 	.send_request		= xs_udp_send_request,
···
 	.rpcbind		= rpcb_getport_async,
 	.set_port		= xs_set_port,
 	.connect		= xs_connect,
+	.get_srcaddr		= xs_sock_srcaddr,
+	.get_srcport		= xs_sock_srcport,
 	.buf_alloc		= rpc_malloc,
 	.buf_free		= rpc_free,
 	.prepare_request	= xs_stream_prepare_request,