Skip to contentSkip to conversation

FSCACHE support for AFS and NFS

PATCH In review Trond Myklebust 5 participants 30 messages Jul 6, 2006 – Jul 7, 2006

Subsystems (12)
lore
Share & tools
lore
#0 lore raw

cover FSCACHE support for AFS and NFS

The following is the latest incarnation of David Howells' fscache series.

Please note that the first 12 patches are really cleanups of the NFS client to prevent inode aliasing: the problem with the current client is that if the server exports two directories on the same filesystem, then those will be mapped to different superblocks on the client despite the fact that they may share files (due to hard links etc). David's patches fix this up by identifying mountpoints that refer to the same filesystem, and mapping them into the same superblock.

The fscache stuff proper begins at patch 13/21 and ends at 19/21.

Finally, patches 20 and 21 optimise generic_shutdown_super. The first fixes up an issue with the autofs code, while the second is the optimisation proper.

Cheers,
Trond

#1 lore raw

PATCH 1/21 NFS: Add dentry materialisation op

From: David Howells <[email protected]>

The attached patch adds a new directory cache management function that prepares a disconnected anonymous function to be connected into the dentry tree. The anonymous dentry is transferred the name and parentage from another dentry.

The following changes were made in [try #2]:

(*) d_materialise_dentry() now switches the parentage of the two nodes around
correctly when one or other of them is self-referential.

The following changes were made in [try #7]:

(*) d_instantiate_unique() has had the interior part split out as function
__d_instantiate_unique(). Callers of this latter function must be holding the appropriate locks.

(*) _d_rehash() has been added as a wrapper around __d_rehash() to call it
with the most obvious hash list (the one from the name). d_rehash() now calls _d_rehash().

(*) d_materialise_dentry() is now __d_materialise_dentry() and is static.

(*) d_materialise_unique() added to perform the combination of d_find_alias(),
d_materialise_dentry() and d_add_unique() that the NFS client was doing twice, all within a single dcache_lock critical section. This reduces the number of times two different spinlocks were being accessed.

The following further changes were made:

(*) Add the dentries onto their parents d_subdirs lists.

---

 fs/dcache.c            |  164 ++++++++++++++++++++++++++++++++++++++++++++----
 include/linux/dcache.h |    1 
 2 files changed, 151 insertions(+), 14 deletions(-)
fs/dcache.c +150 -14
index 1b4a3a3..17b392a 100644
@@ -828,17 +828,19 @@ void d_instantiate(struct dentry *entry,
  * (or otherwise set) by the caller to indicate that it is now
  * in use by the dcache.
  */
-struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
+static struct dentry *__d_instantiate_unique(struct dentry *entry,
+					     struct inode *inode)
 {
 	struct dentry *alias;
 	int len = entry->d_name.len;
 	const char *name = entry->d_name.name;
 	unsigned int hash = entry->d_name.hash;
 
-	BUG_ON(!list_empty(&entry->d_alias));
-	spin_lock(&dcache_lock);
-	if (!inode)
-		goto do_negative;
+	if (!inode) {
+		entry->d_inode = NULL;
+		return NULL;
+	}
+
 	list_for_each_entry(alias, &inode->i_dentry, d_alias) {
 		struct qstr *qstr = &alias->d_name;
 
@@ -851,19 +853,35 @@ struct dentry *d_instantiate_unique(stru
 		if (memcmp(qstr->name, name, len))
 			continue;
 		dget_locked(alias);
-		spin_unlock(&dcache_lock);
-		BUG_ON(!d_unhashed(alias));
-		iput(inode);
 		return alias;
 	}
+
 	list_add(&entry->d_alias, &inode->i_dentry);
-do_negative:
 	entry->d_inode = inode;
 	fsnotify_d_instantiate(entry, inode);
-	spin_unlock(&dcache_lock);
-	security_d_instantiate(entry, inode);
 	return NULL;
 }
+
+struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode)
+{
+	struct dentry *result;
+
+	BUG_ON(!list_empty(&entry->d_alias));
+
+	spin_lock(&dcache_lock);
+	result = __d_instantiate_unique(entry, inode);
+	spin_unlock(&dcache_lock);
+
+	if (!result) {
+		security_d_instantiate(entry, inode);
+		return NULL;
+	}
+
+	BUG_ON(!d_unhashed(result));
+	iput(inode);
+	return result;
+}
+
 EXPORT_SYMBOL(d_instantiate_unique);
 
 /**
@@ -1235,6 +1253,11 @@ static void __d_rehash(struct dentry * e
  	hlist_add_head_rcu(&entry->d_hash, list);
 }
 
+static void _d_rehash(struct dentry * entry)
+{
+	__d_rehash(entry, d_hash(entry->d_parent, entry->d_name.hash));
+}
+
 /**
  * d_rehash	- add an entry back to the hash
  * @entry: dentry to add to the hash
@@ -1244,11 +1267,9 @@ static void __d_rehash(struct dentry * e
  
 void d_rehash(struct dentry * entry)
 {
-	struct hlist_head *list = d_hash(entry->d_parent, entry->d_name.hash);
-
 	spin_lock(&dcache_lock);
 	spin_lock(&entry->d_lock);
-	__d_rehash(entry, list);
+	_d_rehash(entry);
 	spin_unlock(&entry->d_lock);
 	spin_unlock(&dcache_lock);
 }
@@ -1386,6 +1407,120 @@ already_unhashed:
 	spin_unlock(&dcache_lock);
 }
 
+/*
+ * Prepare an anonymous dentry for life in the superblock's dentry tree as a
+ * named dentry in place of the dentry to be replaced.
+ */
+static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
+{
+	struct dentry *dparent, *aparent;
+
+	switch_names(dentry, anon);
+	do_switch(dentry->d_name.len, anon->d_name.len);
+	do_switch(dentry->d_name.hash, anon->d_name.hash);
+
+	dparent = dentry->d_parent;
+	aparent = anon->d_parent;
+
+	dentry->d_parent = (aparent == anon) ? dentry : aparent;
+	list_del(&dentry->d_u.d_child);
+	if (!IS_ROOT(dentry))
+		list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&dentry->d_u.d_child);
+
+	anon->d_parent = (dparent == dentry) ? anon : dparent;
+	list_del(&anon->d_u.d_child);
+	if (!IS_ROOT(anon))
+		list_add(&anon->d_u.d_child, &anon->d_parent->d_subdirs);
+	else
+		INIT_LIST_HEAD(&anon->d_u.d_child);
+
+	anon->d_flags &= ~DCACHE_DISCONNECTED;
+}
+
+/**
+ * d_materialise_unique - introduce an inode into the tree
+ * @dentry: candidate dentry
+ * @inode: inode to bind to the dentry, to which aliases may be attached
+ *
+ * Introduces an dentry into the tree, substituting an extant disconnected
+ * root directory alias in its place if there is one
+ */
+struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode)
+{
+	struct dentry *alias, *actual;
+
+	BUG_ON(!d_unhashed(dentry));
+
+	spin_lock(&dcache_lock);
+
+	if (!inode) {
+		actual = dentry;
+		dentry->d_inode = NULL;
+		goto found_lock;
+	}
+
+	/* See if a disconnected directory already exists as an anonymous root
+	 * that we should splice into the tree instead */
+	if (S_ISDIR(inode->i_mode) && (alias = __d_find_alias(inode, 1))) {
+		spin_lock(&alias->d_lock);
+
+		/* Is this a mountpoint that we could splice into our tree? */
+		if (IS_ROOT(alias))
+			goto connect_mountpoint;
+
+		if (alias->d_name.len == dentry->d_name.len &&
+		    alias->d_parent == dentry->d_parent &&
+		    memcmp(alias->d_name.name,
+			   dentry->d_name.name,
+			   dentry->d_name.len) == 0)
+			goto replace_with_alias;
+
+		spin_unlock(&alias->d_lock);
+
+		/* Doh! Seem to be aliasing directories for some reason... */
+		dput(alias);
+	}
+
+	/* Add a unique reference */
+	actual = __d_instantiate_unique(dentry, inode);
+	if (!actual)
+		actual = dentry;
+	else if (unlikely(!d_unhashed(actual)))
+		goto shouldnt_be_hashed;
+
+found_lock:
+	spin_lock(&actual->d_lock);
+found:
+	_d_rehash(actual);
+	spin_unlock(&actual->d_lock);
+	spin_unlock(&dcache_lock);
+
+	if (actual == dentry) {
+		security_d_instantiate(dentry, inode);
+		return NULL;
+	}
+
+	iput(inode);
+	return actual;
+
+	/* Convert the anonymous/root alias into an ordinary dentry */
+connect_mountpoint:
+	__d_materialise_dentry(dentry, alias);
+
+	/* Replace the candidate dentry with the alias in the tree */
+replace_with_alias:
+	__d_drop(alias);
+	actual = alias;
+	goto found;
+
+shouldnt_be_hashed:
+	spin_unlock(&dcache_lock);
+	BUG();
+	goto shouldnt_be_hashed;
+}
+
 /**
  * d_path - return the path of a dentry
  * @dentry: dentry to report
@@ -1784,6 +1919,7 @@ EXPORT_SYMBOL(d_instantiate);
 EXPORT_SYMBOL(d_invalidate);
 EXPORT_SYMBOL(d_lookup);
 EXPORT_SYMBOL(d_move);
+EXPORT_SYMBOL_GPL(d_materialise_unique);
 EXPORT_SYMBOL(d_path);
 EXPORT_SYMBOL(d_prune_aliases);
 EXPORT_SYMBOL(d_rehash);
include/linux/dcache.h +1 -0
index 471781f..44605be 100644
@@ -221,6 +221,7 @@ static inline int dname_external(struct 
  */
 extern void d_instantiate(struct dentry *, struct inode *);
 extern struct dentry * d_instantiate_unique(struct dentry *, struct inode *);
+extern struct dentry * d_materialise_unique(struct dentry *, struct inode *);
 extern void d_delete(struct dentry *);
 
 /* allocate/de-allocate */
#2 lore raw

PATCH 2/21 NFS: Fix up split of fs/nfs/inode.c

In reply to Trond Myklebust

From: David Howells <[email protected]>

Fix ups for the splitting of the superblock stuff out of fs/nfs/inode.c, including:

(*) Move the callback tcpport module param into callback.c.

(*) Move the idmap cache timeout module param into idmap.c.

(*) Changes to internal.h:

(*) namespace-nfs4.c was renamed to nfs4namespace.c.

(*) nfs_stat_to_errno() is in nfs2xdr.c, not nfs4xdr.c.

(*) nfs4xdr.c is contingent on CONFIG_NFS_V4.

(*) nfs4_path() is only uses if CONFIG_NFS_V4 is set.

Plus also:

(*) The sec_flavours[] table should really be const.

---

 fs/nfs/callback.c |   15 +++++++++++++++
 fs/nfs/idmap.c    |   14 ++++++++++++++
 fs/nfs/internal.h |   12 ++++++------
 fs/nfs/super.c    |   40 ++++------------------------------------
 4 files changed, 39 insertions(+), 42 deletions(-)
fs/nfs/callback.c +15 -0
index fe0a6b8..d6c4bae 100644
@@ -36,6 +36,21 @@ static struct svc_program nfs4_callback_
 
 unsigned int nfs_callback_set_tcpport;
 unsigned short nfs_callback_tcpport;
+static const int nfs_set_port_min = 0;
+static const int nfs_set_port_max = 65535;
+
+static int param_set_port(const char *val, struct kernel_param *kp)
+{
+	char *endp;
+	int num = simple_strtol(val, &endp, 0);
+	if (endp == val || *endp || num < nfs_set_port_min || num > nfs_set_port_max)
+		return -EINVAL;
+	*((int *)kp->arg) = num;
+	return 0;
+}
+
+module_param_call(callback_tcpport, param_set_port, param_get_int,
+		 &nfs_callback_set_tcpport, 0644);
 
 /*
  * This is the callback kernel thread.
fs/nfs/idmap.c +14 -0
index b81e7ed..447ae91 100644
@@ -57,6 +57,20 @@ #define IDMAP_HASH_SZ          128
 /* Default cache timeout is 10 minutes */
 unsigned int nfs_idmap_cache_timeout = 600 * HZ;
 
+static int param_set_idmap_timeout(const char *val, struct kernel_param *kp)
+{
+	char *endp;
+	int num = simple_strtol(val, &endp, 0);
+	int jif = num * HZ;
+	if (endp == val || *endp || num < 0 || jif < num)
+		return -EINVAL;
+	*((int *)kp->arg) = jif;
+	return 0;
+}
+
+module_param_call(idmap_cache_timeout, param_set_idmap_timeout, param_get_int,
+		 &nfs_idmap_cache_timeout, 0644);
+
 struct idmap_hashent {
 	unsigned long ih_expires;
 	__u32 ih_id;
fs/nfs/internal.h +6 -6
index e4f4e5d..94a7870 100644
@@ -15,7 +15,7 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
-/* namespace-nfs4.c */
+/* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
 extern struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry);
 #else
@@ -46,6 +46,7 @@ #define nfs_destroy_directcache() do {} 
 #endif
 
 /* nfs2xdr.c */
+extern int nfs_stat_to_errno(int);
 extern struct rpc_procinfo nfs_procedures[];
 extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int);
 
@@ -54,8 +55,9 @@ extern struct rpc_procinfo nfs3_procedur
 extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int);
 
 /* nfs4xdr.c */
-extern int nfs_stat_to_errno(int);
+#ifdef CONFIG_NFS_V4
 extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus);
+#endif
 
 /* nfs4proc.c */
 #ifdef CONFIG_NFS_V4
@@ -94,15 +96,13 @@ extern char *nfs_path(const char *base, 
 /*
  * Determine the mount path as a string
  */
+#ifdef CONFIG_NFS_V4
 static inline char *
 nfs4_path(const struct dentry *dentry, char *buffer, ssize_t buflen)
 {
-#ifdef CONFIG_NFS_V4
 	return nfs_path(NFS_SB(dentry->d_sb)->mnt_path, dentry, buffer, buflen);
-#else
-	return NULL;
-#endif
 }
+#endif
 
 /*
  * Determine the device name as a string
fs/nfs/super.c +4 -36
index e8a9bee..1c20ff0 100644
@@ -187,40 +187,6 @@ static struct super_operations nfs4_sops
 };
 #endif
 
-#ifdef CONFIG_NFS_V4
-static const int nfs_set_port_min = 0;
-static const int nfs_set_port_max = 65535;
-
-static int param_set_port(const char *val, struct kernel_param *kp)
-{
-	char *endp;
-	int num = simple_strtol(val, &endp, 0);
-	if (endp == val || *endp || num < nfs_set_port_min || num > nfs_set_port_max)
-		return -EINVAL;
-	*((int *)kp->arg) = num;
-	return 0;
-}
-
-module_param_call(callback_tcpport, param_set_port, param_get_int,
-		 &nfs_callback_set_tcpport, 0644);
-#endif
-
-#ifdef CONFIG_NFS_V4
-static int param_set_idmap_timeout(const char *val, struct kernel_param *kp)
-{
-	char *endp;
-	int num = simple_strtol(val, &endp, 0);
-	int jif = num * HZ;
-	if (endp == val || *endp || num < 0 || jif < num)
-		return -EINVAL;
-	*((int *)kp->arg) = jif;
-	return 0;
-}
-
-module_param_call(idmap_cache_timeout, param_set_idmap_timeout, param_get_int,
-		 &nfs_idmap_cache_timeout, 0644);
-#endif
-
 /*
  * Register the NFS filesystems
  */
@@ -323,9 +289,12 @@ static int nfs_statfs(struct dentry *den
 
 }
 
+/*
+ * Map the security flavour number to a name
+ */
 static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour)
 {
-	static struct {
+	static const struct {
 		rpc_authflavor_t flavour;
 		const char *str;
 	} sec_flavours[] = {
@@ -1363,7 +1332,6 @@ static int nfs4_get_sb(struct file_syste
 	}
 
 	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);
-
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
 		goto out_free;
#3 lore raw

PATCH 3/21 NFS: Disambiguate nfs_stat_to_errno()

In reply to Trond Myklebust

From: David Howells <[email protected]>

Rename the NFS4 version of nfs_stat_to_errno() so that it doesn't conflict with the common one used by NFS2 and NFS3.

---

 fs/nfs/nfs4xdr.c |   14 +++++++-------
 1 files changed, 7 insertions(+), 7 deletions(-)
fs/nfs/nfs4xdr.c +7 -7
index 1750d99..14377f2 100644
@@ -58,7 +58,7 @@ #define NFSDBG_FACILITY		NFSDBG_XDR
 /* Mapping from NFS error code to "errno" error code. */
 #define errno_NFSERR_IO		EIO
 
-static int nfs_stat_to_errno(int);
+static int nfs4_stat_to_errno(int);
 
 /* NFSv4 COMPOUND tags are only wanted for debugging purposes */
 #ifdef DEBUG
@@ -2127,7 +2127,7 @@ static int decode_op_hdr(struct xdr_stre
 	}
 	READ32(nfserr);
 	if (nfserr != NFS_OK)
-		return -nfs_stat_to_errno(nfserr);
+		return -nfs4_stat_to_errno(nfserr);
 	return 0;
 }
 
@@ -3597,7 +3597,7 @@ static int decode_setclientid(struct xdr
 		READ_BUF(len);
 		return -NFSERR_CLID_INUSE;
 	} else
-		return -nfs_stat_to_errno(nfserr);
+		return -nfs4_stat_to_errno(nfserr);
 
 	return 0;
 }
@@ -4255,7 +4255,7 @@ static int nfs4_xdr_dec_fsinfo(struct rp
 	if (!status)
 		status = decode_fsinfo(&xdr, fsinfo);
 	if (!status)
-		status = -nfs_stat_to_errno(hdr.status);
+		status = -nfs4_stat_to_errno(hdr.status);
 	return status;
 }
 
@@ -4345,7 +4345,7 @@ static int nfs4_xdr_dec_setclientid(stru
 	if (!status)
 		status = decode_setclientid(&xdr, clp);
 	if (!status)
-		status = -nfs_stat_to_errno(hdr.status);
+		status = -nfs4_stat_to_errno(hdr.status);
 	return status;
 }
 
@@ -4367,7 +4367,7 @@ static int nfs4_xdr_dec_setclientid_conf
 	if (!status)
 		status = decode_fsinfo(&xdr, fsinfo);
 	if (!status)
-		status = -nfs_stat_to_errno(hdr.status);
+		status = -nfs4_stat_to_errno(hdr.status);
 	return status;
 }
 
@@ -4520,7 +4520,7 @@ static struct {
  * This one is used jointly by NFSv2 and NFSv3.
  */
 static int
-nfs_stat_to_errno(int stat)
+nfs4_stat_to_errno(int stat)
 {
 	int i;
 	for (i = 0; nfs_errtbl[i].stat != -1; i++) {
#4 lore raw

PATCH 4/21 NFS: Fix NFS4 callback up/down prototypes

In reply to Trond Myklebust

From: David Howells <[email protected]>

Make the nfs_callback_up()/down() prototypes just do nothing if NFS4 is not enabled. Also make the down function void type since we can't really do anything if it fails.

---

 fs/nfs/callback.c |    5 +----
 fs/nfs/callback.h |    7 ++++++-
 2 files changed, 7 insertions(+), 5 deletions(-)
fs/nfs/callback.c +1 -4
index d6c4bae..b1f7dc4 100644
@@ -149,10 +149,8 @@ out_err:
 /*
  * Kill the server process if it is not already up.
  */
-int nfs_callback_down(void)
+void nfs_callback_down(void)
 {
-	int ret = 0;
-
 	lock_kernel();
 	mutex_lock(&nfs_callback_mutex);
 	nfs_callback_info.users--;
@@ -164,7 +162,6 @@ int nfs_callback_down(void)
 	} while (wait_for_completion_timeout(&nfs_callback_info.stopped, 5*HZ) == 0);
 	mutex_unlock(&nfs_callback_mutex);
 	unlock_kernel();
-	return ret;
 }
 
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
fs/nfs/callback.h +6 -1
index b252e7f..5676163 100644
@@ -62,8 +62,13 @@ struct cb_recallargs {
 extern unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
 extern unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 
+#ifdef CONFIG_NFS_V4
 extern int nfs_callback_up(void);
-extern int nfs_callback_down(void);
+extern void nfs_callback_down(void);
+#else
+#define nfs_callback_up()	(0)
+#define nfs_callback_down()	do {} while(0)
+#endif
 
 extern unsigned int nfs_callback_set_tcpport;
 extern unsigned short nfs_callback_tcpport;
#5 lore raw

PATCH 5/21 NFS: Rename struct nfs4_client to struct nfs_client

In reply to Trond Myklebust

From: David Howells <[email protected]>

Rename struct nfs4_client to struct nfs_client so that it can become the basis for a general client record for NFS2 and NFS3 in addition to NFS4.

---

 fs/nfs/callback.c         |    2 +-
 fs/nfs/callback_proc.c    |    4 ++--
 fs/nfs/delegation.c       |   24 +++++++++++-----------
 fs/nfs/delegation.h       |   10 +++++----
 fs/nfs/idmap.c            |   12 +++++------
 fs/nfs/nfs4_fs.h          |   30 ++++++++++++++-------------
 fs/nfs/nfs4proc.c         |   32 ++++++++++++++---------------
 fs/nfs/nfs4renewd.c       |    8 ++++---
 fs/nfs/nfs4state.c        |   50 +++++++++++++++++++++++----------------------
 fs/nfs/nfs4xdr.c          |   18 ++++++++--------
 fs/nfs/super.c            |    4 ++--
 include/linux/nfs_fs_sb.h |    2 +-
 include/linux/nfs_idmap.h |   14 ++++++-------
 13 files changed, 105 insertions(+), 105 deletions(-)
fs/nfs/callback.c +1 -1
index b1f7dc4..1b596b6 100644
@@ -167,7 +167,7 @@ void nfs_callback_down(void)
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 {
 	struct in_addr *addr = &rqstp->rq_addr.sin_addr;
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 
 	/* Don't talk to strangers */
 	clp = nfs4_find_client(addr);
fs/nfs/callback_proc.c +2 -2
index 7719483..55d6e2e 100644
@@ -15,7 +15,7 @@ #define NFSDBG_FACILITY NFSDBG_CALLBACK
  
 unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct nfs_delegation *delegation;
 	struct nfs_inode *nfsi;
 	struct inode *inode;
@@ -56,7 +56,7 @@ out:
 
 unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct inode *inode;
 	unsigned res;
 	
fs/nfs/delegation.c +12 -12
index 9540a31..5a1105c 100644
@@ -114,7 +114,7 @@ void nfs_inode_reclaim_delegation(struct
  */
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int status = 0;
@@ -176,7 +176,7 @@ static void nfs_msync_inode(struct inode
  */
 int __nfs_inode_return_delegation(struct inode *inode)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
@@ -208,7 +208,7 @@ int __nfs_inode_return_delegation(struct
  */
 void nfs_return_all_delegations(struct super_block *sb)
 {
-	struct nfs4_client *clp = NFS_SB(sb)->nfs4_state;
+	struct nfs_client *clp = NFS_SB(sb)->nfs4_state;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -232,7 +232,7 @@ restart:
 
 int nfs_do_expire_all_delegations(void *ptr)
 {
-	struct nfs4_client *clp = ptr;
+	struct nfs_client *clp = ptr;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -258,7 +258,7 @@ out:
 	module_put_and_exit(0);
 }
 
-void nfs_expire_all_delegations(struct nfs4_client *clp)
+void nfs_expire_all_delegations(struct nfs_client *clp)
 {
 	struct task_struct *task;
 
@@ -276,7 +276,7 @@ void nfs_expire_all_delegations(struct n
 /*
  * Return all delegations following an NFS4ERR_CB_PATH_DOWN error.
  */
-void nfs_handle_cb_pathdown(struct nfs4_client *clp)
+void nfs_handle_cb_pathdown(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	struct inode *inode;
@@ -299,7 +299,7 @@ restart:
 
 struct recall_threadargs {
 	struct inode *inode;
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	const nfs4_stateid *stateid;
 
 	struct completion started;
@@ -310,7 +310,7 @@ static int recall_thread(void *data)
 {
 	struct recall_threadargs *args = (struct recall_threadargs *)data;
 	struct inode *inode = igrab(args->inode);
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 
@@ -371,7 +371,7 @@ out_module_put:
 /*
  * Retrieve the inode associated with a delegation
  */
-struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle)
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle)
 {
 	struct nfs_delegation *delegation;
 	struct inode *res = NULL;
@@ -389,7 +389,7 @@ struct inode *nfs_delegation_find_inode(
 /*
  * Mark all delegations as needing to be reclaimed
  */
-void nfs_delegation_mark_reclaim(struct nfs4_client *clp)
+void nfs_delegation_mark_reclaim(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	spin_lock(&clp->cl_lock);
@@ -401,7 +401,7 @@ void nfs_delegation_mark_reclaim(struct 
 /*
  * Reap all unclaimed delegations after reboot recovery is done
  */
-void nfs_delegation_reap_unclaimed(struct nfs4_client *clp)
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation, *n;
 	LIST_HEAD(head);
@@ -423,7 +423,7 @@ void nfs_delegation_reap_unclaimed(struc
 
 int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode)
 {
-	struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
fs/nfs/delegation.h +5 -5
index 3858694..2cfd4b2 100644
@@ -29,13 +29,13 @@ void nfs_inode_reclaim_delegation(struct
 int __nfs_inode_return_delegation(struct inode *inode);
 int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
 
-struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle);
+struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
 void nfs_return_all_delegations(struct super_block *sb);
-void nfs_expire_all_delegations(struct nfs4_client *clp);
-void nfs_handle_cb_pathdown(struct nfs4_client *clp);
+void nfs_expire_all_delegations(struct nfs_client *clp);
+void nfs_handle_cb_pathdown(struct nfs_client *clp);
 
-void nfs_delegation_mark_reclaim(struct nfs4_client *clp);
-void nfs_delegation_reap_unclaimed(struct nfs4_client *clp);
+void nfs_delegation_mark_reclaim(struct nfs_client *clp);
+void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
 
 /* NFSv4 delegation-related procedures */
 int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid);
fs/nfs/idmap.c +6 -6
index 447ae91..b151053 100644
@@ -109,7 +109,7 @@ static struct rpc_pipe_ops idmap_upcall_
 };
 
 void
-nfs_idmap_new(struct nfs4_client *clp)
+nfs_idmap_new(struct nfs_client *clp)
 {
 	struct idmap *idmap;
 
@@ -138,7 +138,7 @@ nfs_idmap_new(struct nfs4_client *clp)
 }
 
 void
-nfs_idmap_delete(struct nfs4_client *clp)
+nfs_idmap_delete(struct nfs_client *clp)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
@@ -493,27 +493,27 @@ static unsigned int fnvhash32(const void
 	return (hash);
 }
 
-int nfs_map_name_to_uid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_id(idmap, &idmap->idmap_user_hash, name, namelen, uid);
 }
 
-int nfs_map_group_to_gid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid)
+int nfs_map_group_to_gid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_id(idmap, &idmap->idmap_group_hash, name, namelen, uid);
 }
 
-int nfs_map_uid_to_name(struct nfs4_client *clp, __u32 uid, char *buf)
+int nfs_map_uid_to_name(struct nfs_client *clp, __u32 uid, char *buf)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
 	return nfs_idmap_name(idmap, &idmap->idmap_user_hash, uid, buf);
 }
-int nfs_map_gid_to_group(struct nfs4_client *clp, __u32 uid, char *buf)
+int nfs_map_gid_to_group(struct nfs_client *clp, __u32 uid, char *buf)
 {
 	struct idmap *idmap = clp->cl_idmap;
 
fs/nfs/nfs4_fs.h +15 -15
index 9a10286..4e334cb 100644
@@ -43,9 +43,9 @@ enum nfs4_client_state {
 };
 
 /*
- * The nfs4_client identifies our client state to the server.
+ * The nfs_client identifies our client state to the server.
  */
-struct nfs4_client {
+struct nfs_client {
 	struct list_head	cl_servers;	/* Global list of servers */
 	struct in_addr		cl_addr;	/* Server identifier */
 	u64			cl_clientid;	/* constant */
@@ -127,7 +127,7 @@ static inline void nfs_confirm_seqid(str
 struct nfs4_state_owner {
 	spinlock_t	     so_lock;
 	struct list_head     so_list;	 /* per-clientid list of state_owners */
-	struct nfs4_client   *so_client;
+	struct nfs_client    *so_client;
 	u32                  so_id;      /* 32-bit identifier, unique */
 	atomic_t	     so_count;
 
@@ -210,10 +210,10 @@ extern ssize_t nfs4_listxattr(struct den
 
 /* nfs4proc.c */
 extern int nfs4_map_errors(int err);
-extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short, struct rpc_cred *);
-extern int nfs4_proc_setclientid_confirm(struct nfs4_client *, struct rpc_cred *);
-extern int nfs4_proc_async_renew(struct nfs4_client *, struct rpc_cred *);
-extern int nfs4_proc_renew(struct nfs4_client *, struct rpc_cred *);
+extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *);
+extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -231,19 +231,19 @@ extern const u32 nfs4_fsinfo_bitmap[2];
 extern const u32 nfs4_fs_locations_bitmap[2];
 
 /* nfs4renewd.c */
-extern void nfs4_schedule_state_renewal(struct nfs4_client *);
+extern void nfs4_schedule_state_renewal(struct nfs_client *);
 extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
-extern void nfs4_kill_renewd(struct nfs4_client *);
+extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(void *);
 
 /* nfs4state.c */
 extern void init_nfsv4_state(struct nfs_server *);
 extern void destroy_nfsv4_state(struct nfs_server *);
-extern struct nfs4_client *nfs4_get_client(struct in_addr *);
-extern void nfs4_put_client(struct nfs4_client *clp);
-extern struct nfs4_client *nfs4_find_client(struct in_addr *);
-struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp);
-extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *);
+extern struct nfs_client *nfs4_get_client(struct in_addr *);
+extern void nfs4_put_client(struct nfs_client *clp);
+extern struct nfs_client *nfs4_find_client(struct in_addr *);
+struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp);
+extern u32 nfs4_alloc_lockowner_id(struct nfs_client *);
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
 extern void nfs4_put_state_owner(struct nfs4_state_owner *);
@@ -252,7 +252,7 @@ extern struct nfs4_state * nfs4_get_open
 extern void nfs4_put_open_state(struct nfs4_state *);
 extern void nfs4_close_state(struct nfs4_state *, mode_t);
 extern void nfs4_state_set_mode_locked(struct nfs4_state *, mode_t);
-extern void nfs4_schedule_state_recovery(struct nfs4_client *);
+extern void nfs4_schedule_state_recovery(struct nfs_client *);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
fs/nfs/nfs4proc.c +16 -16
index e6ee97f..73f72a2 100644
@@ -64,7 +64,7 @@ static int nfs4_do_fsinfo(struct nfs_ser
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *);
 static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry);
 static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception);
-static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp);
+static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp);
 
 /* Prevent leaks of NFSv4 errors into userland */
 int nfs4_map_errors(int err)
@@ -195,7 +195,7 @@ static void nfs4_setup_readdir(u64 cooki
 
 static void renew_lease(const struct nfs_server *server, unsigned long timestamp)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	spin_lock(&clp->cl_lock);
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
@@ -792,7 +792,7 @@ out:
 
 int nfs4_recover_expired_lease(struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		nfs4_schedule_state_recovery(clp);
@@ -867,7 +867,7 @@ static int _nfs4_open_delegated(struct i
 {
 	struct nfs_delegation *delegation;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs4_state_owner *sp = NULL;
 	struct nfs4_state *state = NULL;
@@ -953,7 +953,7 @@ static int _nfs4_do_open(struct inode *d
 	struct nfs4_state_owner  *sp;
 	struct nfs4_state     *state = NULL;
 	struct nfs_server       *server = NFS_SERVER(dir);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs4_opendata *opendata;
 	int                     status;
 
@@ -2521,7 +2521,7 @@ static void nfs4_proc_commit_setup(struc
  */
 static void nfs4_renew_done(struct rpc_task *task, void *data)
 {
-	struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp;
+	struct nfs_client *clp = (struct nfs_client *)task->tk_msg.rpc_argp;
 	unsigned long timestamp = (unsigned long)data;
 
 	if (task->tk_status < 0) {
@@ -2543,7 +2543,7 @@ static const struct rpc_call_ops nfs4_re
 	.rpc_call_done = nfs4_renew_done,
 };
 
-int nfs4_proc_async_renew(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
@@ -2555,7 +2555,7 @@ int nfs4_proc_async_renew(struct nfs4_cl
 			&nfs4_renew_ops, (void *)jiffies);
 }
 
-int nfs4_proc_renew(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_RENEW],
@@ -2766,7 +2766,7 @@ static int nfs4_proc_set_acl(struct inod
 static int
 nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (!clp || task->tk_status >= 0)
 		return 0;
@@ -2803,7 +2803,7 @@ static int nfs4_wait_bit_interruptible(v
 	return 0;
 }
 
-static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp)
+static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp)
 {
 	sigset_t oldset;
 	int res;
@@ -2846,7 +2846,7 @@ static int nfs4_delay(struct rpc_clnt *c
  */
 int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	int ret = errorcode;
 
 	exception->retry = 0;
@@ -2873,7 +2873,7 @@ int nfs4_handle_exception(const struct n
 	return nfs4_map_errors(ret);
 }
 
-int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
+int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
 {
 	nfs4_verifier sc_verifier;
 	struct nfs4_setclientid setclientid = {
@@ -2920,7 +2920,7 @@ int nfs4_proc_setclientid(struct nfs4_cl
 	return status;
 }
 
-static int _nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred)
+static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	struct nfs_fsinfo fsinfo;
 	struct rpc_message msg = {
@@ -2944,7 +2944,7 @@ static int _nfs4_proc_setclientid_confir
 	return status;
 }
 
-int nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred)
+int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	long timeout;
 	int err;
@@ -3081,7 +3081,7 @@ static int _nfs4_proc_getlk(struct nfs4_
 {
 	struct inode *inode = state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct nfs_lockt_args arg = {
 		.fh = NFS_FH(inode),
 		.fl = request,
@@ -3488,7 +3488,7 @@ static int nfs4_lock_expired(struct nfs4
 
 static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request)
 {
-	struct nfs4_client *clp = state->owner->so_client;
+	struct nfs_client *clp = state->owner->so_client;
 	unsigned char fl_flags = request->fl_flags;
 	int status;
 
fs/nfs/nfs4renewd.c +4 -4
index 5d764d8..2087640 100644
@@ -61,7 +61,7 @@ #define NFSDBG_FACILITY	NFSDBG_PROC
 void
 nfs4_renew_state(void *data)
 {
-	struct nfs4_client *clp = (struct nfs4_client *)data;
+	struct nfs_client *clp = (struct nfs_client *)data;
 	struct rpc_cred *cred;
 	long lease, timeout;
 	unsigned long last, now;
@@ -108,7 +108,7 @@ out:
 
 /* Must be called with clp->cl_sem locked for writes */
 void
-nfs4_schedule_state_renewal(struct nfs4_client *clp)
+nfs4_schedule_state_renewal(struct nfs_client *clp)
 {
 	long timeout;
 
@@ -127,7 +127,7 @@ nfs4_schedule_state_renewal(struct nfs4_
 void
 nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 {
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 
 	if (!clp)
 		return;
@@ -140,7 +140,7 @@ nfs4_renewd_prepare_shutdown(struct nfs_
 
 /* Must be called with clp->cl_sem locked for writes */
 void
-nfs4_kill_renewd(struct nfs4_client *clp)
+nfs4_kill_renewd(struct nfs_client *clp)
 {
 	down_read(&clp->cl_sem);
 	if (!list_empty(&clp->cl_superblocks)) {
fs/nfs/nfs4state.c +25 -25

181 lines not on this page · raw message

fs/nfs/nfs4xdr.c +9 -9
index 14377f2..44ea44c 100644
@@ -1160,7 +1160,7 @@ static int encode_rename(struct xdr_stre
 	return 0;
 }
 
-static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client_stateid)
+static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid)
 {
 	uint32_t *p;
 
@@ -1246,7 +1246,7 @@ static int encode_setclientid(struct xdr
 	return 0;
 }
 
-static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_client *client_state)
+static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state)
 {
         uint32_t *p;
 
@@ -1945,7 +1945,7 @@ static int nfs4_xdr_enc_server_caps(stru
 /*
  * a RENEW request
  */
-static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp)
+static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -1975,7 +1975,7 @@ static int nfs4_xdr_enc_setclientid(stru
 /*
  * a SETCLIENTID_CONFIRM request
  */
-static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp)
+static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -2132,7 +2132,7 @@ static int decode_op_hdr(struct xdr_stre
 }
 
 /* Dummy routine */
-static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs4_client *clp)
+static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp)
 {
 	uint32_t *p;
 	unsigned int strlen;
@@ -2636,7 +2636,7 @@ static int decode_attr_nlink(struct xdr_
 	return 0;
 }
 
-static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *uid)
+static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *uid)
 {
 	uint32_t len, *p;
 
@@ -2660,7 +2660,7 @@ static int decode_attr_owner(struct xdr_
 	return 0;
 }
 
-static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *gid)
+static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *gid)
 {
 	uint32_t len, *p;
 
@@ -3564,7 +3564,7 @@ static int decode_setattr(struct xdr_str
 	return 0;
 }
 
-static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_client *clp)
+static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp)
 {
 	uint32_t *p;
 	uint32_t opnum;
@@ -4334,7 +4334,7 @@ static int nfs4_xdr_dec_renew(struct rpc
  * a SETCLIENTID request
  */
 static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p,
-		struct nfs4_client *clp)
+		struct nfs_client *clp)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
fs/nfs/super.c +2 -2
index 1c20ff0..b9a7c2b 100644
@@ -1094,7 +1094,7 @@ #ifdef CONFIG_NFS_V4
 static struct rpc_clnt *nfs4_create_client(struct nfs_server *server,
 	struct rpc_timeout *timeparms, int proto, rpc_authflavor_t flavor)
 {
-	struct nfs4_client *clp;
+	struct nfs_client *clp;
 	struct rpc_xprt *xprt = NULL;
 	struct rpc_clnt *clnt = NULL;
 	int err = -EIO;
@@ -1411,7 +1411,7 @@ static inline char *nfs4_dup_path(const 
 static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
 {
 	const struct dentry *dentry = data->dentry;
-	struct nfs4_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs4_state;
 	struct super_block *sb;
 
 	server->fsid = data->fattr->fsid;
include/linux/nfs_fs_sb.h +1 -1
index 6b4a13c..4db90df 100644
@@ -43,7 +43,7 @@ #ifdef CONFIG_NFS_V4
 	 */
 	char			ip_addr[16];
 	char *			mnt_path;
-	struct nfs4_client *	nfs4_state;	/* all NFSv4 state starts here */
+	struct nfs_client *	nfs4_state;	/* all NFSv4 state starts here */
 	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
 						 * that share the same clientid
 						 */
include/linux/nfs_idmap.h +7 -7
index 102e560..678fe68 100644
@@ -62,15 +62,15 @@ struct idmap_msg {
 #ifdef __KERNEL__
 
 /* Forward declaration to make this header independent of others */
-struct nfs4_client;
+struct nfs_client;
 
-void nfs_idmap_new(struct nfs4_client *);
-void nfs_idmap_delete(struct nfs4_client *);
+void nfs_idmap_new(struct nfs_client *);
+void nfs_idmap_delete(struct nfs_client *);
 
-int nfs_map_name_to_uid(struct nfs4_client *, const char *, size_t, __u32 *);
-int nfs_map_group_to_gid(struct nfs4_client *, const char *, size_t, __u32 *);
-int nfs_map_uid_to_name(struct nfs4_client *, __u32, char *);
-int nfs_map_gid_to_group(struct nfs4_client *, __u32, char *);
+int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *);
+int nfs_map_group_to_gid(struct nfs_client *, const char *, size_t, __u32 *);
+int nfs_map_uid_to_name(struct nfs_client *, __u32, char *);
+int nfs_map_gid_to_group(struct nfs_client *, __u32, char *);
 
 extern unsigned int nfs_idmap_cache_timeout;
 #endif /* __KERNEL__ */
#6 lore raw

PATCH 6/21 NFS: Rename nfs_server::nfs4_state

In reply to Trond Myklebust

From: David Howells <[email protected]>

Rename nfs_server::nfs4_state to nfs_client as it will be used to represent the client state for NFS2 and NFS3 also.

---

 fs/nfs/delegation.c       |   12 ++++++------
 fs/nfs/nfs4proc.c         |   26 +++++++++++++-------------
 fs/nfs/nfs4renewd.c       |    2 +-
 fs/nfs/nfs4state.c        |   10 +++++-----
 fs/nfs/nfs4xdr.c          |   10 +++++-----
 fs/nfs/super.c            |    6 +++---
 include/linux/nfs_fs_sb.h |    2 +-
 7 files changed, 34 insertions(+), 34 deletions(-)
fs/nfs/delegation.c +6 -6
index 5a1105c..cfe2397 100644
@@ -52,7 +52,7 @@ static int nfs_delegation_claim_locks(st
 			case -NFS4ERR_EXPIRED:
 				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
 			case -NFS4ERR_STALE_CLIENTID:
-				nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs4_state);
+				nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs_client);
 				goto out_err;
 		}
 	}
@@ -114,7 +114,7 @@ void nfs_inode_reclaim_delegation(struct
  */
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int status = 0;
@@ -176,7 +176,7 @@ static void nfs_msync_inode(struct inode
  */
 int __nfs_inode_return_delegation(struct inode *inode)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
@@ -208,7 +208,7 @@ int __nfs_inode_return_delegation(struct
  */
 void nfs_return_all_delegations(struct super_block *sb)
 {
-	struct nfs_client *clp = NFS_SB(sb)->nfs4_state;
+	struct nfs_client *clp = NFS_SB(sb)->nfs_client;
 	struct nfs_delegation *delegation;
 	struct inode *inode;
 
@@ -310,7 +310,7 @@ static int recall_thread(void *data)
 {
 	struct recall_threadargs *args = (struct recall_threadargs *)data;
 	struct inode *inode = igrab(args->inode);
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 
@@ -423,7 +423,7 @@ void nfs_delegation_reap_unclaimed(struc
 
 int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode)
 {
-	struct nfs_client *clp = NFS_SERVER(inode)->nfs4_state;
+	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_delegation *delegation;
 	int res = 0;
fs/nfs/nfs4proc.c +13 -13
index 73f72a2..eff6043 100644
@@ -195,7 +195,7 @@ static void nfs4_setup_readdir(u64 cooki
 
 static void renew_lease(const struct nfs_server *server, unsigned long timestamp)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	spin_lock(&clp->cl_lock);
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
@@ -252,7 +252,7 @@ static struct nfs4_opendata *nfs4_openda
 	atomic_inc(&sp->so_count);
 	p->o_arg.fh = NFS_FH(dir);
 	p->o_arg.open_flags = flags,
-	p->o_arg.clientid = server->nfs4_state->cl_clientid;
+	p->o_arg.clientid = server->nfs_client->cl_clientid;
 	p->o_arg.id = sp->so_id;
 	p->o_arg.name = &dentry->d_name;
 	p->o_arg.server = server;
@@ -550,7 +550,7 @@ int nfs4_open_delegation_recall(struct d
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
 				/* Don't recall a delegation if it was lost */
-				nfs4_schedule_state_recovery(server->nfs4_state);
+				nfs4_schedule_state_recovery(server->nfs_client);
 				return err;
 		}
 		err = nfs4_handle_exception(server, err, &exception);
@@ -792,7 +792,7 @@ out:
 
 int nfs4_recover_expired_lease(struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		nfs4_schedule_state_recovery(clp);
@@ -867,7 +867,7 @@ static int _nfs4_open_delegated(struct i
 {
 	struct nfs_delegation *delegation;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs4_state_owner *sp = NULL;
 	struct nfs4_state *state = NULL;
@@ -953,7 +953,7 @@ static int _nfs4_do_open(struct inode *d
 	struct nfs4_state_owner  *sp;
 	struct nfs4_state     *state = NULL;
 	struct nfs_server       *server = NFS_SERVER(dir);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_opendata *opendata;
 	int                     status;
 
@@ -1133,7 +1133,7 @@ static void nfs4_close_done(struct rpc_t
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
-			nfs4_schedule_state_recovery(server->nfs4_state);
+			nfs4_schedule_state_recovery(server->nfs_client);
 			break;
 		default:
 			if (nfs4_async_handle_error(task, server) == -EAGAIN) {
@@ -2766,7 +2766,7 @@ static int nfs4_proc_set_acl(struct inod
 static int
 nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (!clp || task->tk_status >= 0)
 		return 0;
@@ -2846,7 +2846,7 @@ static int nfs4_delay(struct rpc_clnt *c
  */
 int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	int ret = errorcode;
 
 	exception->retry = 0;
@@ -3052,7 +3052,7 @@ int nfs4_proc_delegreturn(struct inode *
 		switch (err) {
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
-				nfs4_schedule_state_recovery(server->nfs4_state);
+				nfs4_schedule_state_recovery(server->nfs_client);
 			case 0:
 				return 0;
 		}
@@ -3081,7 +3081,7 @@ static int _nfs4_proc_getlk(struct nfs4_
 {
 	struct inode *inode = state->inode;
 	struct nfs_server *server = NFS_SERVER(inode);
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs_lockt_args arg = {
 		.fh = NFS_FH(inode),
 		.fl = request,
@@ -3206,7 +3206,7 @@ static void nfs4_locku_done(struct rpc_t
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_EXPIRED:
-			nfs4_schedule_state_recovery(calldata->server->nfs4_state);
+			nfs4_schedule_state_recovery(calldata->server->nfs_client);
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server) == -EAGAIN) {
@@ -3318,7 +3318,7 @@ static struct nfs4_lockdata *nfs4_alloc_
 	if (p->arg.lock_seqid == NULL)
 		goto out_free;
 	p->arg.lock_stateid = &lsp->ls_stateid;
-	p->arg.lock_owner.clientid = server->nfs4_state->cl_clientid;
+	p->arg.lock_owner.clientid = server->nfs_client->cl_clientid;
 	p->arg.lock_owner.id = lsp->ls_id;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
fs/nfs/nfs4renewd.c +1 -1
index 2087640..ff947ec 100644
@@ -127,7 +127,7 @@ nfs4_schedule_state_renewal(struct nfs_c
 void
 nfs4_renewd_prepare_shutdown(struct nfs_server *server)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 
 	if (!clp)
 		return;
fs/nfs/nfs4state.c +5 -5
index c0b6439..fa51a7d 100644
@@ -61,7 +61,7 @@ static LIST_HEAD(nfs4_clientid_list);
 void
 init_nfsv4_state(struct nfs_server *server)
 {
-	server->nfs4_state = NULL;
+	server->nfs_client = NULL;
 	INIT_LIST_HEAD(&server->nfs4_siblings);
 }
 
@@ -70,9 +70,9 @@ destroy_nfsv4_state(struct nfs_server *s
 {
 	kfree(server->mnt_path);
 	server->mnt_path = NULL;
-	if (server->nfs4_state) {
-		nfs4_put_client(server->nfs4_state);
-		server->nfs4_state = NULL;
+	if (server->nfs_client) {
+		nfs4_put_client(server->nfs_client);
+		server->nfs_client = NULL;
 	}
 }
 
@@ -306,7 +306,7 @@ nfs4_drop_state_owner(struct nfs4_state_
  */
 struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct rpc_cred *cred)
 {
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct nfs4_state_owner *sp, *new;
 
 	get_rpccred(cred);
fs/nfs/nfs4xdr.c +5 -5
index 44ea44c..992a713 100644
@@ -529,7 +529,7 @@ static int encode_attrs(struct xdr_strea
 	if (iap->ia_valid & ATTR_MODE)
 		len += 4;
 	if (iap->ia_valid & ATTR_UID) {
-		owner_namelen = nfs_map_uid_to_name(server->nfs4_state, iap->ia_uid, owner_name);
+		owner_namelen = nfs_map_uid_to_name(server->nfs_client, iap->ia_uid, owner_name);
 		if (owner_namelen < 0) {
 			printk(KERN_WARNING "nfs: couldn't resolve uid %d to string\n",
 			       iap->ia_uid);
@@ -541,7 +541,7 @@ static int encode_attrs(struct xdr_strea
 		len += 4 + (XDR_QUADLEN(owner_namelen) << 2);
 	}
 	if (iap->ia_valid & ATTR_GID) {
-		owner_grouplen = nfs_map_gid_to_group(server->nfs4_state, iap->ia_gid, owner_group);
+		owner_grouplen = nfs_map_gid_to_group(server->nfs_client, iap->ia_gid, owner_group);
 		if (owner_grouplen < 0) {
 			printk(KERN_WARNING "nfs4: couldn't resolve gid %d to string\n",
 			       iap->ia_gid);
@@ -3051,9 +3051,9 @@ static int decode_getfattr(struct xdr_st
 	fattr->mode |= fmode;
 	if ((status = decode_attr_nlink(xdr, bitmap, &fattr->nlink)) != 0)
 		goto xdr_error;
-	if ((status = decode_attr_owner(xdr, bitmap, server->nfs4_state, &fattr->uid)) != 0)
+	if ((status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid)) != 0)
 		goto xdr_error;
-	if ((status = decode_attr_group(xdr, bitmap, server->nfs4_state, &fattr->gid)) != 0)
+	if ((status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid)) != 0)
 		goto xdr_error;
 	if ((status = decode_attr_rdev(xdr, bitmap, &fattr->rdev)) != 0)
 		goto xdr_error;
@@ -3254,7 +3254,7 @@ static int decode_delegation(struct xdr_
 			if (decode_space_limit(xdr, &res->maxsize) < 0)
 				return -EIO;
 	}
-	return decode_ace(xdr, NULL, res->server->nfs4_state);
+	return decode_ace(xdr, NULL, res->server->nfs_client);
 }
 
 static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res)
fs/nfs/super.c +3 -3
index b9a7c2b..509fa99 100644
@@ -1136,7 +1136,7 @@ static struct rpc_clnt *nfs4_create_clie
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	clnt = rpc_clone_client(clp->cl_rpcclient);
 	if (!IS_ERR(clnt))
-		server->nfs4_state = clp;
+		server->nfs_client = clp;
 	up_write(&clp->cl_sem);
 	clp = NULL;
 
@@ -1146,7 +1146,7 @@ static struct rpc_clnt *nfs4_create_clie
 		return clnt;
 	}
 
-	if (server->nfs4_state->cl_idmap == NULL) {
+	if (server->nfs_client->cl_idmap == NULL) {
 		dprintk("%s: failed to create idmapper.\n", __FUNCTION__);
 		return ERR_PTR(-ENOMEM);
 	}
@@ -1411,7 +1411,7 @@ static inline char *nfs4_dup_path(const 
 static struct super_block *nfs4_clone_sb(struct nfs_server *server, struct nfs_clone_mount *data)
 {
 	const struct dentry *dentry = data->dentry;
-	struct nfs_client *clp = server->nfs4_state;
+	struct nfs_client *clp = server->nfs_client;
 	struct super_block *sb;
 
 	server->fsid = data->fattr->fsid;
include/linux/nfs_fs_sb.h +1 -1
index 4db90df..fc20d6b 100644
@@ -43,7 +43,7 @@ #ifdef CONFIG_NFS_V4
 	 */
 	char			ip_addr[16];
 	char *			mnt_path;
-	struct nfs_client *	nfs4_state;	/* all NFSv4 state starts here */
+	struct nfs_client *	nfs_client;	/* all NFSv4 state starts here */
 	struct list_head	nfs4_siblings;	/* List of other nfs_server structs
 						 * that share the same clientid
 						 */
#7 lore raw

PATCH 7/21 NFS: Return an error when starting the idmapping pipe

In reply to Trond Myklebust

From: David Howells <[email protected]>

Return an error when starting the idmapping pipe so that we can detect it failing.

---

 fs/nfs/idmap.c            |   12 ++++++++----
 fs/nfs/super.c            |    3 ++-
 include/linux/nfs_idmap.h |    2 +-
 3 files changed, 11 insertions(+), 6 deletions(-)
fs/nfs/idmap.c +8 -4
index b151053..cd80d89 100644
@@ -108,15 +108,17 @@ static struct rpc_pipe_ops idmap_upcall_
         .destroy_msg    = idmap_pipe_destroy_msg,
 };
 
-void
+int
 nfs_idmap_new(struct nfs_client *clp)
 {
 	struct idmap *idmap;
+	int error;
 
 	if (clp->cl_idmap != NULL)
-		return;
+		return 0;
+
         if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL)
-                return;
+                return -ENOMEM;
 
 	snprintf(idmap->idmap_path, sizeof(idmap->idmap_path),
 	    "%s/idmap", clp->cl_rpcclient->cl_pathname);
@@ -124,8 +126,9 @@ nfs_idmap_new(struct nfs_client *clp)
         idmap->idmap_dentry = rpc_mkpipe(idmap->idmap_path,
 	    idmap, &idmap_upcall_ops, 0);
         if (IS_ERR(idmap->idmap_dentry)) {
+		error = PTR_ERR(idmap->idmap_dentry);
 		kfree(idmap);
-		return;
+		return error;
 	}
 
         mutex_init(&idmap->idmap_lock);
@@ -135,6 +138,7 @@ nfs_idmap_new(struct nfs_client *clp)
 	idmap->idmap_group_hash.h_type = IDMAP_TYPE_GROUP;
 
 	clp->cl_idmap = idmap;
+	return 0;
 }
 
 void
fs/nfs/super.c +2 -1
index 509fa99..0fbb75e 100644
@@ -1131,7 +1131,8 @@ static struct rpc_clnt *nfs4_create_clie
 		clnt->cl_softrtry = 1;
 		clp->cl_rpcclient = clnt;
 		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
-		nfs_idmap_new(clp);
+		if (nfs_idmap_new(clp) < 0)
+			goto out_fail;
 	}
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	clnt = rpc_clone_client(clp->cl_rpcclient);
include/linux/nfs_idmap.h +1 -1
index 678fe68..15a9f3b 100644
@@ -64,7 +64,7 @@ #ifdef __KERNEL__
 /* Forward declaration to make this header independent of others */
 struct nfs_client;
 
-void nfs_idmap_new(struct nfs_client *);
+int nfs_idmap_new(struct nfs_client *);
 void nfs_idmap_delete(struct nfs_client *);
 
 int nfs_map_name_to_uid(struct nfs_client *, const char *, size_t, __u32 *);
#8 lore raw

PATCH 8/21 NFS: Add a lookupfh NFS RPC op

In reply to Trond Myklebust

From: David Howells <[email protected]>

Add a lookup filehandle NFS RPC op so that a file handle can be looked up without requiring dentries and inodes and other VFS stuff when doing an NFS4 pathwalk during mounting.

---

 fs/nfs/nfs4proc.c       |   47 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs_xdr.h |    3 +++
 2 files changed, 50 insertions(+), 0 deletions(-)
fs/nfs/nfs4proc.c +47 -0
index eff6043..7fa2938 100644
@@ -1583,6 +1583,52 @@ nfs4_proc_setattr(struct dentry *dentry,
 	return status;
 }
 
+static int _nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
+		struct qstr *name, struct nfs_fh *fhandle,
+		struct nfs_fattr *fattr)
+{
+	int		       status;
+	struct nfs4_lookup_arg args = {
+		.bitmask = server->attr_bitmask,
+		.dir_fh = dirfh,
+		.name = name,
+	};
+	struct nfs4_lookup_res res = {
+		.server = server,
+		.fattr = fattr,
+		.fh = fhandle,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+
+	nfs_fattr_init(fattr);
+
+	dprintk("NFS call  lookupfh %s\n", name->name);
+	status = rpc_call_sync(server->client, &msg, 0);
+	dprintk("NFS reply lookupfh: %d\n", status);
+	if (status == -NFS4ERR_MOVED)
+		status = -EREMOTE;
+	return status;
+}
+
+static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
+			      struct qstr *name, struct nfs_fh *fhandle,
+			      struct nfs_fattr *fattr)
+{
+	struct nfs4_exception exception = { };
+	int err;
+	do {
+		err = nfs4_handle_exception(server,
+				_nfs4_proc_lookupfh(server, dirfh, name,
+						    fhandle, fattr),
+				&exception);
+	} while (exception.retry);
+	return err;
+}
+
 static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name,
 		struct nfs_fh *fhandle, struct nfs_fattr *fattr)
 {
@@ -3698,6 +3744,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.getroot	= nfs4_proc_get_root,
 	.getattr	= nfs4_proc_getattr,
 	.setattr	= nfs4_proc_setattr,
+	.lookupfh	= nfs4_proc_lookupfh,
 	.lookup		= nfs4_proc_lookup,
 	.access		= nfs4_proc_access,
 	.readlink	= nfs4_proc_readlink,
include/linux/nfs_xdr.h +3 -0
index 2d3fb64..be80310 100644
@@ -770,6 +770,9 @@ struct nfs_rpc_ops {
 
 	int	(*getroot) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fsinfo *);
+	int	(*lookupfh)(struct nfs_server *, struct nfs_fh *,
+			    struct qstr *, struct nfs_fh *,
+			    struct nfs_fattr *);
 	int	(*getattr) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fattr *);
 	int	(*setattr) (struct dentry *, struct nfs_fattr *,
#9 lore raw

Re: [PATCH 08/21] NFS: Add a lookupfh NFS RPC op

Quoted 32 lines · On Thu, 06 Jul 2006 11:10:39 -0400 Trond Myklebust…

From: David Howells <[email protected]>

Add a lookup filehandle NFS RPC op so that a file handle can be looked up without requiring dentries and inodes and other VFS stuff when doing an NFS4 pathwalk during mounting.

---

 fs/nfs/nfs4proc.c       |   47 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs_xdr.h |    3 +++
 2 files changed, 50 insertions(+), 0 deletions(-)
fs/nfs/nfs4proc.c +10 -0
index eff6043..7fa2938 100644
@@ -1583,6 +1583,52 @@ nfs4_proc_setattr(struct dentry *dentry,
 	return status;
 }
 
+static int _nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh,
+		struct qstr *name, struct nfs_fh *fhandle,
+		struct nfs_fattr *fattr)
+{
+	int		       status;
+	struct nfs4_lookup_arg args = {
+		.bitmask = server->attr_bitmask,
+		.dir_fh = dirfh,
+		.name = name,
+	};

fs/nfs/nfs4proc.c nfs4_proc_setattr(struct dentry *dentry, · line 1605

Quoted 10 lines
fs/nfs/nfs4proc.c +10 -0
+	struct nfs4_lookup_res res = {
+		.server = server,
+		.fattr = fattr,
+		.fh = fhandle,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};

If this is called with any frequency then we might want to take a look at the code generation quality. gcc doesn't do this well.

#10 lore raw

PATCH 9/21 NFS: Add a lookupfh NFS RPC op

In reply to Trond Myklebust

From: David Howells <[email protected]>

Add a set_capabilities NFS RPC op so that the server capabilities can be set.

---

 fs/nfs/nfs4proc.c       |    1 +
 include/linux/nfs_xdr.h |    1 +
 2 files changed, 2 insertions(+), 0 deletions(-)
fs/nfs/nfs4proc.c +1 -0
index 7fa2938..b433810 100644
@@ -3765,6 +3765,7 @@ struct nfs_rpc_ops	nfs_v4_clientops = {
 	.statfs		= nfs4_proc_statfs,
 	.fsinfo		= nfs4_proc_fsinfo,
 	.pathconf	= nfs4_proc_pathconf,
+	.set_capabilities = nfs4_server_capabilities,
 	.decode_dirent	= nfs4_decode_dirent,
 	.read_setup	= nfs4_proc_read_setup,
 	.read_done	= nfs4_read_done,
include/linux/nfs_xdr.h +1 -0
index be80310..24ceac1 100644
@@ -809,6 +809,7 @@ struct nfs_rpc_ops {
 			    struct nfs_fsinfo *);
 	int	(*pathconf) (struct nfs_server *, struct nfs_fh *,
 			     struct nfs_pathconf *);
+	int	(*set_capabilities)(struct nfs_server *, struct nfs_fh *);
 	u32 *	(*decode_dirent)(u32 *, struct nfs_entry *, int plus);
 	void	(*read_setup)   (struct nfs_read_data *);
 	int	(*read_done)  (struct rpc_task *, struct nfs_read_data *);
#11 lore raw

PATCH 10/21 NFS: Generalise the nfs_client structure

In reply to Trond Myklebust

From: David Howells <[email protected]>

Generalise the nfs_client structure by:

(1) Moving nfs_client to a more general place (nfs_fs_sb.h).

(2) Renaming its maintenance routines to be non-NFS4 specific.

(3) Move those maintenance routines to a new non-NFS4 specific file (client.c) and move the declarations to internal.h.

(4) Make nfs_find/get_client() take a full sockaddr_in to include the port number (will be required for NFS2/3).

(5) Make nfs_find/get_client() take the NFS protocol version (again will be required to differentiate NFS2, 3 & 4 client records).

Also:

(6) Make nfs_client construction proceed akin to inodes, marking them as under construction and providing a function to indicate completion.

(7) Make nfs_get_client() wait interruptibly if it finds a client that it can share, but that client is currently being constructed.

(8) Make nfs4_create_client() use (6) and (7) instead of locking cl_sem.

---

 fs/nfs/Makefile           |    6 -
 fs/nfs/callback.c         |    9 +
 fs/nfs/callback_proc.c    |    9 +
 fs/nfs/client.c           |  305 +++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/delegation.c       |    9 +
 fs/nfs/internal.h         |    6 +
 fs/nfs/nfs4_fs.h          |   52 --------
 fs/nfs/nfs4proc.c         |    2 
 fs/nfs/nfs4state.c        |  128 +------------------
 fs/nfs/super.c            |   31 ++---
 include/linux/nfs_fs.h    |    1 
 include/linux/nfs_fs_sb.h |   56 ++++++++
 12 files changed, 406 insertions(+), 208 deletions(-)
fs/nfs/Makefile +3 -3
index 0b572a0..3b993a6 100644
@@ -4,9 +4,9 @@ #
 
 obj-$(CONFIG_NFS_FS) += nfs.o
 
-nfs-y 			:= dir.o file.o inode.o super.o nfs2xdr.o pagelist.o \
-			   proc.o read.o symlink.o unlink.o write.o \
-			   namespace.o
+nfs-y 			:= client.o dir.o file.o inode.o super.o nfs2xdr.o \
+			   pagelist.o proc.o read.o symlink.o unlink.o \
+			   write.o namespace.o
 nfs-$(CONFIG_ROOT_NFS)	+= nfsroot.o mount_clnt.o      
 nfs-$(CONFIG_NFS_V3)	+= nfs3proc.o nfs3xdr.o
 nfs-$(CONFIG_NFS_V3_ACL)	+= nfs3acl.o
fs/nfs/callback.c +5 -4
index 1b596b6..a3ee113 100644
@@ -19,6 +19,7 @@ #include <net/inet_sock.h>
 
 #include "nfs4_fs.h"
 #include "callback.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
 
@@ -166,15 +167,15 @@ void nfs_callback_down(void)
 
 static int nfs_callback_authenticate(struct svc_rqst *rqstp)
 {
-	struct in_addr *addr = &rqstp->rq_addr.sin_addr;
+	struct sockaddr_in *addr = &rqstp->rq_addr;
 	struct nfs_client *clp;
 
 	/* Don't talk to strangers */
-	clp = nfs4_find_client(addr);
+	clp = nfs_find_client(addr, 4);
 	if (clp == NULL)
 		return SVC_DROP;
-	dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr));
-	nfs4_put_client(clp);
+	dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr->sin_addr));
+	nfs_put_client(clp);
 	switch (rqstp->rq_authop->flavour) {
 		case RPC_AUTH_NULL:
 			if (rqstp->rq_proc != CB_NULL)
fs/nfs/callback_proc.c +5 -4
index 55d6e2e..97cf8f7 100644
@@ -10,6 +10,7 @@ #include <linux/nfs_fs.h>
 #include "nfs4_fs.h"
 #include "callback.h"
 #include "delegation.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
  
@@ -22,7 +23,7 @@ unsigned nfs4_callback_getattr(struct cb
 	
 	res->bitmap[0] = res->bitmap[1] = 0;
 	res->status = htonl(NFS4ERR_BADHANDLE);
-	clp = nfs4_find_client(&args->addr->sin_addr);
+	clp = nfs_find_client(args->addr, 4);
 	if (clp == NULL)
 		goto out;
 	inode = nfs_delegation_find_inode(clp, &args->fh);
@@ -48,7 +49,7 @@ out_iput:
 	up_read(&nfsi->rwsem);
 	iput(inode);
 out_putclient:
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 out:
 	dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res->status));
 	return res->status;
@@ -61,7 +62,7 @@ unsigned nfs4_callback_recall(struct cb_
 	unsigned res;
 	
 	res = htonl(NFS4ERR_BADHANDLE);
-	clp = nfs4_find_client(&args->addr->sin_addr);
+	clp = nfs_find_client(args->addr, 4);
 	if (clp == NULL)
 		goto out;
 	inode = nfs_delegation_find_inode(clp, &args->fh);
@@ -80,7 +81,7 @@ unsigned nfs4_callback_recall(struct cb_
 	}
 	iput(inode);
 out_putclient:
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 out:
 	dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res));
 	return res;
fs/nfs/client.c new file +305 -0

305 lines not on this page · raw message

fs/nfs/delegation.c +5 -4
index cfe2397..5713367 100644
@@ -18,6 +18,7 @@ #include <linux/nfs_xdr.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
+#include "internal.h"
 
 static struct nfs_delegation *nfs_alloc_delegation(void)
 {
@@ -145,7 +146,7 @@ int nfs_inode_set_delegation(struct inod
 					sizeof(delegation->stateid)) != 0 ||
 				delegation->type != nfsi->delegation->type) {
 			printk("%s: server %u.%u.%u.%u, handed out a duplicate delegation!\n",
-					__FUNCTION__, NIPQUAD(clp->cl_addr));
+					__FUNCTION__, NIPQUAD(clp->cl_addr.sin_addr));
 			status = -EIO;
 		}
 	}
@@ -254,7 +255,7 @@ restart:
 	}
 out:
 	spin_unlock(&clp->cl_lock);
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put_and_exit(0);
 }
 
@@ -266,10 +267,10 @@ void nfs_expire_all_delegations(struct n
 	atomic_inc(&clp->cl_count);
 	task = kthread_run(nfs_do_expire_all_delegations, clp,
 			"%u.%u.%u.%u-delegreturn",
-			NIPQUAD(clp->cl_addr));
+			NIPQUAD(clp->cl_addr.sin_addr));
 	if (!IS_ERR(task))
 		return;
-	nfs4_put_client(clp);
+	nfs_put_client(clp);
 	module_put(THIS_MODULE);
 }
 
fs/nfs/internal.h +6 -0
index 94a7870..5ae1306 100644
@@ -15,6 +15,12 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
+/* client.c */
+extern void nfs_put_client(struct nfs_client *);
+extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int);
+extern struct nfs_client *nfs_get_client(const char *, const struct sockaddr_in *, int);
+extern void nfs_mark_client_ready(struct nfs_client *, int);
+
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
 extern struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry);
fs/nfs/nfs4_fs.h +0 -52

64 lines not on this page · raw message

fs/nfs/nfs4proc.c +1 -1
index b433810..2925d37 100644
@@ -2943,7 +2943,7 @@ int nfs4_proc_setclientid(struct nfs_cli
 	for(;;) {
 		setclientid.sc_name_len = scnprintf(setclientid.sc_name,
 				sizeof(setclientid.sc_name), "%s/%u.%u.%u.%u %s %u",
-				clp->cl_ipaddr, NIPQUAD(clp->cl_addr.s_addr),
+				clp->cl_ipaddr, NIPQUAD(clp->cl_addr.sin_addr),
 				cred->cr_ops->cr_name,
 				clp->cl_id_uniquifier);
 		setclientid.sc_netid_len = scnprintf(setclientid.sc_netid,
fs/nfs/nfs4state.c +6 -122

168 lines not on this page · raw message

fs/nfs/super.c +13 -18
index 0fbb75e..8b9d3aa 100644
@@ -1099,19 +1099,18 @@ static struct rpc_clnt *nfs4_create_clie
 	struct rpc_clnt *clnt = NULL;
 	int err = -EIO;
 
-	clp = nfs4_get_client(&server->addr.sin_addr);
+	clp = nfs_get_client(server->hostname, &server->addr, 4);
 	if (!clp) {
 		dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__);
 		return ERR_PTR(err);
 	}
 
 	/* Now create transport and client */
-	down_write(&clp->cl_sem);
 	if (IS_ERR(clp->cl_rpcclient)) {
 		xprt = xprt_create_proto(proto, &server->addr, timeparms);
 		if (IS_ERR(xprt)) {
-			up_write(&clp->cl_sem);
 			err = PTR_ERR(xprt);
+			nfs_mark_client_ready(clp, err);
 			dprintk("%s: cannot create RPC transport. Error = %d\n",
 					__FUNCTION__, err);
 			goto out_fail;
@@ -1121,8 +1120,8 @@ static struct rpc_clnt *nfs4_create_clie
 		clnt = rpc_create_client(xprt, server->hostname, &nfs_program,
 				server->rpc_ops->version, flavor);
 		if (IS_ERR(clnt)) {
-			up_write(&clp->cl_sem);
 			err = PTR_ERR(clnt);
+			nfs_mark_client_ready(clp, err);
 			dprintk("%s: cannot create RPC client. Error = %d\n",
 					__FUNCTION__, err);
 			goto out_fail;
@@ -1131,14 +1130,20 @@ static struct rpc_clnt *nfs4_create_clie
 		clnt->cl_softrtry = 1;
 		clp->cl_rpcclient = clnt;
 		memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr));
-		if (nfs_idmap_new(clp) < 0)
+		err = nfs_idmap_new(clp);
+		if (err < 0) {
+			nfs_mark_client_ready(clp, err);
 			goto out_fail;
+		}
 	}
 	list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks);
 	clnt = rpc_clone_client(clp->cl_rpcclient);
-	if (!IS_ERR(clnt))
+	if (!IS_ERR(clnt)) {
 		server->nfs_client = clp;
-	up_write(&clp->cl_sem);
+		nfs_mark_client_ready(clp, 0);
+	} else {
+		nfs_mark_client_ready(clp, PTR_ERR(clnt));
+	}
 	clp = NULL;
 
 	if (IS_ERR(clnt)) {
@@ -1165,7 +1170,7 @@ static struct rpc_clnt *nfs4_create_clie
 
  out_fail:
 	if (clp)
-		nfs4_put_client(clp);
+		nfs_put_client(clp);
 	return ERR_PTR(err);
 }
 
@@ -1324,14 +1329,6 @@ static int nfs4_get_sb(struct file_syste
 		goto out_free;
 	}
 
-	/* Fire up rpciod if not yet running */
-	error = rpciod_up();
-	if (error < 0) {
-		dprintk("%s: couldn't start rpciod! Error = %d\n",
-				__FUNCTION__, error);
-		goto out_free;
-	}
-
 	s = sget(fs_type, nfs4_compare_super, nfs_set_super, server);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
@@ -1378,8 +1375,6 @@ static void nfs4_kill_super(struct super
 
 	destroy_nfsv4_state(server);
 
-	rpciod_down();
-
 	nfs_free_iostats(server->io_stats);
 	kfree(server->hostname);
 	kfree(server);
include/linux/nfs_fs.h +1 -0
index 55ea853..9616419 100644
@@ -581,6 +581,7 @@ #define NFSDBG_XDR		0x0020
 #define NFSDBG_FILE		0x0040
 #define NFSDBG_ROOT		0x0080
 #define NFSDBG_CALLBACK		0x0100
+#define NFSDBG_CLIENT		0x0200
 #define NFSDBG_ALL		0xFFFF
 
 #ifdef __KERNEL__
include/linux/nfs_fs_sb.h +56 -0

62 lines not on this page · raw message

#12 lore raw

PATCH 12/21 NFS: Add server and volume lists to /proc

In reply to Trond Myklebust

From: David Howells <[email protected]>

Make two new proc files available:

/proc/fs/nfsfs/servers
/proc/fs/nfsfs/volumes

The first lists the servers with which we are currently dealing (struct nfs_client), and the second lists the volumes we have on those servers (struct nfs_server).

---

 fs/nfs/client.c           |  310 +++++++++++++++++++++++++++++++++++++++++++--
 fs/nfs/inode.c            |    7 +
 fs/nfs/internal.h         |   12 ++
 include/linux/nfs_fs_sb.h |    4 +
 4 files changed, 320 insertions(+), 13 deletions(-)
fs/nfs/client.c +297 -13
index fcbf2f3..8ae3f74 100644
@@ -113,11 +113,15 @@ static struct nfs_client *nfs_alloc_clie
 	if (error < 0) {
 		dprintk("%s: couldn't start rpciod! Error = %d\n",
 				__FUNCTION__, error);
+		__set_bit(NFS_CS_RPCIOD, &clp->cl_svcs_state);
 		goto error_1;
 	}
 
-	if (nfsversion == 4 && nfs_callback_up() < 0)
-		goto error_2;
+	if (nfsversion == 4) {
+		if (nfs_callback_up() < 0)
+			goto error_2;
+		__set_bit(NFS_CS_CALLBACK, &clp->cl_svcs_state);
+	}
 
 	atomic_set(&clp->cl_count, 1);
 	clp->cl_cons_state = NFS_CS_INITING;
@@ -150,8 +154,10 @@ #endif
 
 error_3:
 	nfs_callback_down();
+	__clear_bit(NFS_CS_CALLBACK, &clp->cl_svcs_state);
 error_2:
 	rpciod_down();
+	__clear_bit(NFS_CS_RPCIOD, &clp->cl_svcs_state);
 error_1:
 	kfree(clp);
 error_0:
@@ -163,10 +169,10 @@ error_0:
  */
 static void nfs_free_client(struct nfs_client *clp)
 {
-	dprintk("--> nfs_free_client()\n");
+	dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion);
 
 #ifdef CONFIG_NFS_V4
-	if (clp->cl_nfsversion == 4) {
+	if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_svcs_state)) {
 		while (!list_empty(&clp->cl_unused)) {
 			struct nfs4_state_owner *sp;
 
@@ -185,9 +191,10 @@ #endif
 	if (!IS_ERR(clp->cl_rpcclient))
 		rpc_shutdown_client(clp->cl_rpcclient);
 
-	if (clp->cl_nfsversion == 4)
+	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_svcs_state))
 		nfs_callback_down();
 
+	if (__test_and_clear_bit(NFS_CS_RPCIOD, &clp->cl_svcs_state))
 	rpciod_down();
 
 	kfree(clp->cl_hostname);
@@ -867,18 +874,11 @@ static int nfs4_init_client(struct nfs_c
 	if (error < 0)
 		goto error;
 
-	error = nfs_callback_up();
-	if (error < 0) {
-		dprintk("%s: failed to initialise callback channel. Error = %d\n",
-			__FUNCTION__, error);
-		goto error;
-	}
-
 	error = nfs_idmap_new(clp);
 	if (error < 0) {
 		dprintk("%s: failed to create idmapper. Error = %d\n",
 			__FUNCTION__, error);
-		nfs_callback_down();
+		__set_bit(NFS_CS_IDMAP, &clp->cl_svcs_state);
 		goto error;
 	}
 
@@ -1148,3 +1148,287 @@ out_free_server:
 	dprintk("<-- nfs_clone_server() = error %d\n", error);
 	return ERR_PTR(error);
 }
+
+#ifdef CONFIG_PROC_FS
+static struct proc_dir_entry *proc_fs_nfs;
+
+static int nfs_server_list_open(struct inode *inode, struct file *file);
+static void *nfs_server_list_start(struct seq_file *p, loff_t *pos);
+static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos);
+static void nfs_server_list_stop(struct seq_file *p, void *v);
+static int nfs_server_list_show(struct seq_file *m, void *v);
+
+static struct seq_operations nfs_server_list_ops = {
+	.start	= nfs_server_list_start,
+	.next	= nfs_server_list_next,
+	.stop	= nfs_server_list_stop,
+	.show	= nfs_server_list_show,
+};
+
+static struct file_operations nfs_server_list_fops = {
+	.open		= nfs_server_list_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int nfs_volume_list_open(struct inode *inode, struct file *file);
+static void *nfs_volume_list_start(struct seq_file *p, loff_t *pos);
+static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos);
+static void nfs_volume_list_stop(struct seq_file *p, void *v);
+static int nfs_volume_list_show(struct seq_file *m, void *v);
+
+static struct seq_operations nfs_volume_list_ops = {
+	.start	= nfs_volume_list_start,
+	.next	= nfs_volume_list_next,
+	.stop	= nfs_volume_list_stop,
+	.show	= nfs_volume_list_show,
+};
+
+static struct file_operations nfs_volume_list_fops = {
+	.open		= nfs_volume_list_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+/*
+ * open "/proc/fs/nfsfs/servers" which provides a summary of servers with which
+ * we're dealing
+ */
+static int nfs_server_list_open(struct inode *inode, struct file *file)
+{
+	struct seq_file *m;
+	int ret;
+
+	ret = seq_open(file, &nfs_server_list_ops);
+	if (ret < 0)
+		return ret;
+
+	m = file->private_data;
+	m->private = PDE(inode)->data;
+
+	return 0;
+}
+
+/*
+ * set up the iterator to start reading from the server list and return the first item
+ */
+static void *nfs_server_list_start(struct seq_file *m, loff_t *_pos)
+{
+	struct list_head *_p;
+	loff_t pos = *_pos;
+
+	/* lock the list against modification */
+	spin_lock(&nfs_client_lock);
+
+	/* allow for the header line */
+	if (!pos)
+		return SEQ_START_TOKEN;
+	pos--;
+
+	/* find the n'th element in the list */
+	list_for_each(_p, &nfs_client_list)
+		if (!pos--)
+			break;
+
+	return _p != &nfs_client_list ? _p : NULL;
+}
+
+/*
+ * move to next server
+ */
+static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos)
+{
+	struct list_head *_p;
+
+	(*pos)++;
+
+	_p = v;
+	_p = (v == SEQ_START_TOKEN) ? nfs_client_list.next : _p->next;
+
+	return _p != &nfs_client_list ? _p : NULL;
+}
+
+/*
+ * clean up after reading from the transports list
+ */
+static void nfs_server_list_stop(struct seq_file *p, void *v)
+{
+	spin_unlock(&nfs_client_lock);
+}
+
+/*
+ * display a header line followed by a load of call lines
+ */
+static int nfs_server_list_show(struct seq_file *m, void *v)
+{
+	struct nfs_client *clp;
+
+	/* display header on line 1 */
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(m, "NV SERVER   PORT USE HOSTNAME\n");
+		return 0;
+	}
+
+	/* display one transport per line on subsequent lines */
+	clp = list_entry(v, struct nfs_client, cl_share_link);
+
+	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %3d %s\n",
+		   clp->cl_nfsversion,
+		   NIPQUAD(clp->cl_addr.sin_addr),
+		   ntohs(clp->cl_addr.sin_port),
+		   atomic_read(&clp->cl_count),
+		   clp->cl_hostname);
+
+	return 0;
+}
+
+/*
+ * open "/proc/fs/nfsfs/volumes" which provides a summary of extant volumes
+ */
+static int nfs_volume_list_open(struct inode *inode, struct file *file)
+{
+	struct seq_file *m;
+	int ret;
+
+	ret = seq_open(file, &nfs_volume_list_ops);
+	if (ret < 0)
+		return ret;
+
+	m = file->private_data;
+	m->private = PDE(inode)->data;
+
+	return 0;
+}
+
+/*
+ * set up the iterator to start reading from the volume list and return the first item
+ */
+static void *nfs_volume_list_start(struct seq_file *m, loff_t *_pos)
+{
+	struct list_head *_p;
+	loff_t pos = *_pos;
+
+	/* lock the list against modification */
+	spin_lock(&nfs_client_lock);
+
+	/* allow for the header line */
+	if (!pos)
+		return SEQ_START_TOKEN;
+	pos--;
+
+	/* find the n'th element in the list */
+	list_for_each(_p, &nfs_volume_list)
+		if (!pos--)
+			break;
+
+	return _p != &nfs_volume_list ? _p : NULL;
+}
+
+/*
+ * move to next volume
+ */
+static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos)
+{
+	struct list_head *_p;
+
+	(*pos)++;
+
+	_p = v;
+	_p = (v == SEQ_START_TOKEN) ? nfs_volume_list.next : _p->next;
+
+	return _p != &nfs_volume_list ? _p : NULL;
+}
+
+/*
+ * clean up after reading from the transports list
+ */
+static void nfs_volume_list_stop(struct seq_file *p, void *v)
+{
+	spin_unlock(&nfs_client_lock);
+}
+
+/*
+ * display a header line followed by a load of call lines
+ */
+static int nfs_volume_list_show(struct seq_file *m, void *v)
+{
+	struct nfs_server *server;
+	struct nfs_client *clp;
+	char dev[8], fsid[17];
+
+	/* display header on line 1 */
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(m, "NV SERVER   PORT DEV     FSID\n");
+		return 0;
+	}
+	/* display one transport per line on subsequent lines */
+	server = list_entry(v, struct nfs_server, master_link);
+	clp = server->nfs_client;
+
+	snprintf(dev, 8, "%u:%u",
+		 MAJOR(server->s_dev), MINOR(server->s_dev));
+
+	snprintf(fsid, 17, "%llx:%llx",
+		 server->fsid.major, server->fsid.minor);
+
+	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s\n",
+		   clp->cl_nfsversion,
+		   NIPQUAD(clp->cl_addr.sin_addr),
+		   ntohs(clp->cl_addr.sin_port),
+		   dev,
+		   fsid);
+
+	return 0;
+}
+
+/*
+ * initialise the /proc/fs/nfsfs/ directory
+ */
+int __init nfs_fs_proc_init(void)
+{
+	struct proc_dir_entry *p;
+
+	proc_fs_nfs = proc_mkdir("nfsfs", proc_root_fs);
+	if (!proc_fs_nfs)
+		goto error_0;
+
+	proc_fs_nfs->owner = THIS_MODULE;
+
+	/* a file of servers with which we're dealing */
+	p = create_proc_entry("servers", S_IFREG|S_IRUGO, proc_fs_nfs);
+	if (!p)
+		goto error_1;
+
+	p->proc_fops = &nfs_server_list_fops;
+	p->owner = THIS_MODULE;
+
+	/* a file of volumes that we have mounted */
+	p = create_proc_entry("volumes", S_IFREG|S_IRUGO, proc_fs_nfs);
+	if (!p)
+		goto error_2;
+
+	p->proc_fops = &nfs_volume_list_fops;
+	p->owner = THIS_MODULE;
+	return 0;
+
+error_2:
+	remove_proc_entry("servers", proc_fs_nfs);
+error_1:
+	remove_proc_entry("nfsfs", proc_root_fs);
+error_0:
+	return -ENOMEM;
+}
+
+/*
+ * clean up the /proc/fs/nfsfs/ directory
+ */
+void nfs_fs_proc_exit(void)
+{
+	remove_proc_entry("volumes", proc_fs_nfs);
+	remove_proc_entry("servers", proc_fs_nfs);
+	remove_proc_entry("nfsfs", proc_root_fs);
+}
+
+#endif /* CONFIG_PROC_FS */
fs/nfs/inode.c +7 -0
index 261411a..7665b73 100644
@@ -1144,6 +1144,10 @@ static int __init init_nfs_fs(void)
 {
 	int err;
 
+	err = nfs_fs_proc_init();
+	if (err)
+		goto out5;
+
 	err = nfs_init_nfspagecache();
 	if (err)
 		goto out4;
@@ -1184,6 +1188,8 @@ out2:
 out3:
 	nfs_destroy_nfspagecache();
 out4:
+	nfs_fs_proc_exit();
+out5:
 	return err;
 }
 
@@ -1198,6 +1204,7 @@ #ifdef CONFIG_PROC_FS
 	rpc_proc_unregister("nfs");
 #endif
 	unregister_nfs_fs();
+	nfs_fs_proc_exit();
 }
 
 /* Not quite true; I just maintain it */
fs/nfs/internal.h +12 -0
index 3d51eee..1a1312e 100644
@@ -47,6 +47,18 @@ extern void nfs_free_server(struct nfs_s
 extern struct nfs_server *nfs_clone_server(struct nfs_server *,
 					   struct nfs_fh *,
 					   struct nfs_fattr *);
+#ifdef CONFIG_PROC_FS
+extern int __init nfs_fs_proc_init(void);
+extern void nfs_fs_proc_exit(void);
+#else
+static inline int nfs_fs_proc_init(void)
+{
+	return 0;
+}
+static inline void nfs_fs_proc_exit(void)
+{
+}
+#endif
 
 /* nfs4namespace.c */
 #ifdef CONFIG_NFS_V4
include/linux/nfs_fs_sb.h +4 -0
index 4e13a6c..62b7ca9 100644
@@ -16,6 +16,10 @@ #define NFS_CS_READY		0		/* ready to be 
 #define NFS_CS_INITING		1		/* busy initialising */
 	int			cl_nfsversion;	/* NFS protocol version */
 	rpc_authflavor_t	cl_authflavour;
+	unsigned long		cl_svcs_state;	/* NFS services state */
+#define NFS_CS_RPCIOD		0		/* - rpciod started */
+#define NFS_CS_CALLBACK		1		/* - callback started */
+#define NFS_CS_IDMAP		2		/* - idmap started */
 	struct sockaddr_in	cl_addr;	/* server identifier */
 	char *			cl_hostname;	/* hostname of server */
 	struct list_head	cl_share_link;	/* link in global client list */
#13 lore raw

PATCH 13/21 FS-Cache: Provide a filesystem-specific sync'able page bit

In reply to Trond Myklebust

From: David Howells <[email protected]>

The attached patch provides a filesystem-specific page bit that a filesystem can synchronise upon. This can be used, for example, by a netfs to synchronise with CacheFS writing its pages to disk.

The PG_checked bit is replaced with PG_fs_misc, and various operations are provided based upon that. The *PageChecked() macros still exist, though now they just convert to *PageFsMisc() macros. The name of the "checked" macros seems appropriate as they're used for metadata page validation by various filesystems.

---

 fs/afs/dir.c               |    5 +----
 fs/ext2/dir.c              |    6 +++---
 fs/ext3/inode.c            |   10 +++++-----
 fs/freevxfs/vxfs_subr.c    |    2 +-
 fs/reiserfs/inode.c        |   10 +++++-----
 fs/ufs/dir.c               |    6 +++---
 include/linux/page-flags.h |   15 ++++++++++-----
 include/linux/pagemap.h    |   11 +++++++++++
 mm/filemap.c               |   17 +++++++++++++++++
 mm/migrate.c               |    4 ++--
 mm/page_alloc.c            |    2 +-
 11 files changed, 59 insertions(+), 29 deletions(-)
fs/afs/dir.c +1 -4
index 2fc9987..3b78e70 100644
@@ -155,11 +155,9 @@ #endif
 		}
 	}
 
-	SetPageChecked(page);
 	return;
 
  error:
-	SetPageChecked(page);
 	SetPageError(page);
 
 } /* end afs_dir_check_page() */
@@ -191,8 +189,7 @@ static struct page *afs_dir_get_page(str
 		kmap(page);
 		if (!PageUptodate(page))
 			goto fail;
-		if (!PageChecked(page))
-			afs_dir_check_page(dir, page);
+		afs_dir_check_page(dir, page);
 		if (PageError(page))
 			goto fail;
 	}
fs/ext2/dir.c +3 -3
index 92ea826..b0e7d9f 100644
@@ -112,7 +112,7 @@ static void ext2_check_page(struct page 
 	if (offs != limit)
 		goto Eend;
 out:
-	SetPageChecked(page);
+	SetPageFsMisc(page);
 	return;
 
 	/* Too bad, we had an error */
@@ -152,7 +152,7 @@ Eend:
 		dir->i_ino, (page->index<<PAGE_CACHE_SHIFT)+offs,
 		(unsigned long) le32_to_cpu(p->inode));
 fail:
-	SetPageChecked(page);
+	SetPageFsMisc(page);
 	SetPageError(page);
 }
 
@@ -165,7 +165,7 @@ static struct page * ext2_get_page(struc
 		kmap(page);
 		if (!PageUptodate(page))
 			goto fail;
-		if (!PageChecked(page))
+		if (!PageFsMisc(page))
 			ext2_check_page(page);
 		if (PageError(page))
 			goto fail;
fs/ext3/inode.c +5 -5
index f804d5e..8073fc9 100644
@@ -1527,12 +1527,12 @@ static int ext3_journalled_writepage(str
 		goto no_write;
 	}
 
-	if (!page_has_buffers(page) || PageChecked(page)) {
+	if (!page_has_buffers(page) || PageFsMisc(page)) {
 		/*
 		 * It's mmapped pagecache.  Add buffers and journal it.  There
 		 * doesn't seem much point in redirtying the page here.
 		 */
-		ClearPageChecked(page);
+		ClearPageFsMisc(page);
 		ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE,
 					ext3_get_block);
 		if (ret != 0) {
@@ -1589,7 +1589,7 @@ static void ext3_invalidatepage(struct p
 	 * If it's a full truncate we just forget about the pending dirtying
 	 */
 	if (offset == 0)
-		ClearPageChecked(page);
+		ClearPageFsMisc(page);
 
 	journal_invalidatepage(journal, page, offset);
 }
@@ -1598,7 +1598,7 @@ static int ext3_releasepage(struct page 
 {
 	journal_t *journal = EXT3_JOURNAL(page->mapping->host);
 
-	WARN_ON(PageChecked(page));
+	WARN_ON(PageFsMisc(page));
 	if (!page_has_buffers(page))
 		return 0;
 	return journal_try_to_free_buffers(journal, page, wait);
@@ -1694,7 +1694,7 @@ out:
  */
 static int ext3_journalled_set_page_dirty(struct page *page)
 {
-	SetPageChecked(page);
+	SetPageFsMisc(page);
 	return __set_page_dirty_nobuffers(page);
 }
 
fs/freevxfs/vxfs_subr.c +1 -1
index decac62..805bbb2 100644
@@ -78,7 +78,7 @@ vxfs_get_page(struct address_space *mapp
 		kmap(pp);
 		if (!PageUptodate(pp))
 			goto fail;
-		/** if (!PageChecked(pp)) **/
+		/** if (!PageFsMisc(pp)) **/
 			/** vxfs_check_page(pp); **/
 		if (PageError(pp))
 			goto fail;
fs/reiserfs/inode.c +5 -5
index 12dfdcf..df64f04 100644
@@ -2351,7 +2351,7 @@ static int reiserfs_write_full_page(stru
 	struct buffer_head *head, *bh;
 	int partial = 0;
 	int nr = 0;
-	int checked = PageChecked(page);
+	int checked = PageFsMisc(page);
 	struct reiserfs_transaction_handle th;
 	struct super_block *s = inode->i_sb;
 	int bh_per_page = PAGE_CACHE_SIZE / s->s_blocksize;
@@ -2420,7 +2420,7 @@ static int reiserfs_write_full_page(stru
 	 * blocks we're going to log
 	 */
 	if (checked) {
-		ClearPageChecked(page);
+		ClearPageFsMisc(page);
 		reiserfs_write_lock(s);
 		error = journal_begin(&th, s, bh_per_page + 1);
 		if (error) {
@@ -2801,7 +2801,7 @@ static void reiserfs_invalidatepage(stru
 	BUG_ON(!PageLocked(page));
 
 	if (offset == 0)
-		ClearPageChecked(page);
+		ClearPageFsMisc(page);
 
 	if (!page_has_buffers(page))
 		goto out;
@@ -2842,7 +2842,7 @@ static int reiserfs_set_page_dirty(struc
 {
 	struct inode *inode = page->mapping->host;
 	if (reiserfs_file_data_log(inode)) {
-		SetPageChecked(page);
+		SetPageFsMisc(page);
 		return __set_page_dirty_nobuffers(page);
 	}
 	return __set_page_dirty_buffers(page);
@@ -2865,7 +2865,7 @@ static int reiserfs_releasepage(struct p
 	struct buffer_head *bh;
 	int ret = 1;
 
-	WARN_ON(PageChecked(page));
+	WARN_ON(PageFsMisc(page));
 	spin_lock(&j->j_dirty_buffers_lock);
 	head = page_buffers(page);
 	bh = head;
fs/ufs/dir.c +3 -3
index 7f0a0aa..e04327c 100644
@@ -135,7 +135,7 @@ static void ufs_check_page(struct page *
 	if (offs != limit)
 		goto Eend;
 out:
-	SetPageChecked(page);
+	SetPageFsMisc(page);
 	return;
 
 	/* Too bad, we had an error */
@@ -173,7 +173,7 @@ Eend:
 		   "offset=%lu",
 		   dir->i_ino, (page->index<<PAGE_CACHE_SHIFT)+offs);
 fail:
-	SetPageChecked(page);
+	SetPageFsMisc(page);
 	SetPageError(page);
 }
 
@@ -187,7 +187,7 @@ static struct page *ufs_get_page(struct 
 		kmap(page);
 		if (!PageUptodate(page))
 			goto fail;
-		if (!PageChecked(page))
+		if (!PageFsMisc(page))
 			ufs_check_page(page);
 		if (PageError(page))
 			goto fail;
include/linux/page-flags.h +10 -5
index 5748642..6e017b7 100644
@@ -71,7 +71,7 @@ #define PG_lru			 5
 #define PG_active		 6
 #define PG_slab			 7	/* slab debug (Suparna wants this) */
 
-#define PG_checked		 8	/* kill me in 2.5.<early>. */
+#define PG_fs_misc		 8
 #define PG_arch_1		 9
 #define PG_reserved		10
 #define PG_private		11	/* Has something at ->private */
@@ -161,10 +161,6 @@ #else
 #define PageHighMem(page)	0 /* needed to optimize away at compile time */
 #endif
 
-#define PageChecked(page)	test_bit(PG_checked, &(page)->flags)
-#define SetPageChecked(page)	set_bit(PG_checked, &(page)->flags)
-#define ClearPageChecked(page)	clear_bit(PG_checked, &(page)->flags)
-
 #define PageReserved(page)	test_bit(PG_reserved, &(page)->flags)
 #define SetPageReserved(page)	set_bit(PG_reserved, &(page)->flags)
 #define ClearPageReserved(page)	clear_bit(PG_reserved, &(page)->flags)
@@ -263,4 +259,13 @@ static inline void set_page_writeback(st
 	test_set_page_writeback(page);
 }
 
+/*
+ * Filesystem-specific page bit testing
+ */
+#define PageFsMisc(page)		test_bit(PG_fs_misc, &(page)->flags)
+#define SetPageFsMisc(page)		set_bit(PG_fs_misc, &(page)->flags)
+#define TestSetPageFsMisc(page)		test_and_set_bit(PG_fs_misc, &(page)->flags)
+#define ClearPageFsMisc(page)		clear_bit(PG_fs_misc, &(page)->flags)
+#define TestClearPageFsMisc(page)	test_and_clear_bit(PG_fs_misc, &(page)->flags)
+
 #endif	/* PAGE_FLAGS_H */
include/linux/pagemap.h +11 -0
index 0a2f5d2..82b2753 100644
@@ -170,6 +170,17 @@ static inline void wait_on_page_writebac
 extern void end_page_writeback(struct page *page);
 
 /*
+ * Wait for filesystem-specific page synchronisation to complete
+ */
+static inline void wait_on_page_fs_misc(struct page *page)
+{
+	if (PageFsMisc(page))
+		wait_on_page_bit(page, PG_fs_misc);
+}
+
+extern void fastcall end_page_fs_misc(struct page *page);
+
+/*
  * Fault a userspace page into pagetables.  Return non-zero on a fault.
  *
  * This assumes that two userspace pages are always sufficient.  That's
mm/filemap.c +17 -0
index d087fc3..c71d8fd 100644
@@ -577,6 +577,23 @@ void fastcall __lock_page(struct page *p
 }
 EXPORT_SYMBOL(__lock_page);
 
+/*
+ * Note completion of filesystem specific page synchronisation
+ *
+ * This is used to allow a page to be written to a filesystem cache in the
+ * background without holding up the completion of readpage
+ */
+void fastcall end_page_fs_misc(struct page *page)
+{
+	smp_mb__before_clear_bit();
+	if (!TestClearPageFsMisc(page))
+		BUG();
+	smp_mb__after_clear_bit();
+	__wake_up_bit(page_waitqueue(page), &page->flags, PG_fs_misc);
+}
+
+EXPORT_SYMBOL(end_page_fs_misc);
+
 /**
  * find_get_page - find and get a page reference
  * @mapping: the address_space to search
mm/migrate.c +2 -2
index 3f1e0c2..08c9fff 100644
@@ -348,8 +348,8 @@ static void migrate_page_copy(struct pag
 		SetPageUptodate(newpage);
 	if (PageActive(page))
 		SetPageActive(newpage);
-	if (PageChecked(page))
-		SetPageChecked(newpage);
+	if (PageFsMisc(page))
+		SetPageFsMisc(newpage);
 	if (PageMappedToDisk(page))
 		SetPageMappedToDisk(newpage);
 
mm/page_alloc.c +1 -1
index 54a4f53..8bb003b 100644
@@ -550,7 +550,7 @@ static int prep_new_page(struct page *pa
 
 	page->flags &= ~(1 << PG_uptodate | 1 << PG_error |
 			1 << PG_referenced | 1 << PG_arch_1 |
-			1 << PG_checked | 1 << PG_mappedtodisk);
+			1 << PG_fs_misc | 1 << PG_mappedtodisk);
 	set_page_private(page, 0);
 	set_page_refcounted(page);
 	kernel_map_pages(page, 1 << order, 1);
#14 lore raw

PATCH 14/21 FS-Cache: Avoid ENFILE checking for kernel-specific open files

In reply to Trond Myklebust

From: David Howells <[email protected]>

Make it possible to avoid ENFILE checking for kernel specific open files, such as are used by the CacheFiles module.

After, for example, tarring up a kernel source tree over the network, the CacheFiles module may easily have 20000+ files open in the backing filesystem, thus causing all non-root processes to be given error ENFILE when they try to open a file, socket, pipe, etc..

---

 Documentation/sysctl/fs.txt |    6 ++++-
 fs/file_table.c             |   48 +++++++++++++++++++++++++++++++++++--------
 fs/open.c                   |   20 ++++++++++++++++++
 include/linux/file.h        |    1 -
 include/linux/fs.h          |   10 +++++++++
 include/linux/sysctl.h      |    1 +
 kernel/sysctl.c             |   11 ++++++++++
 7 files changed, 86 insertions(+), 11 deletions(-)
Documentation/sysctl/fs.txt +5 -1
index 0b62c62..ead15f0 100644
@@ -71,7 +71,7 @@ you might want to raise the limit.
 
 ==============================================================
 
-file-max & file-nr:
+file-max, file-nr & file-kernel:
 
 The kernel allocates file handles dynamically, but as yet it
 doesn't free them again.
@@ -88,6 +88,10 @@ close to the maximum, but the number of 
 significantly greater than 0, you've encountered a peak in your 
 usage of file handles and you don't need to increase the maximum.
 
+The value in file-kernel denotes the number of internal file handles
+that the kernel has open.  These do not contribute to ENFILE
+accounting.
+
 ==============================================================
 
 inode-max, inode-nr & inode-state:
fs/file_table.c +39 -9

107 lines not on this page · raw message

fs/open.c +20 -0
index 303f06d..3cf6c00 100644
@@ -974,6 +974,26 @@ struct file *dentry_open(struct dentry *
 EXPORT_SYMBOL(dentry_open);
 
 /*
+ * open a specifically in-kernel file
+ */
+struct file *dentry_open_kernel(struct dentry *dentry, struct vfsmount *mnt, int flags)
+{
+	int error;
+	struct file *f;
+
+	error = -ENFILE;
+	f = get_empty_kernel_filp(FKFLAGS_NO_ENFILE);
+	if (f == NULL) {
+		dput(dentry);
+		mntput(mnt);
+		return ERR_PTR(error);
+	}
+
+	return __dentry_open(dentry, mnt, flags, f, NULL);
+}
+EXPORT_SYMBOL_GPL(dentry_open_kernel);
+
+/*
  * Find an empty file descriptor entry, and mark it busy.
  */
 int get_unused_fd(void)
include/linux/file.h +0 -1
index 9f7c251..da7be8f 100644
@@ -79,7 +79,6 @@ extern void FASTCALL(set_close_on_exec(u
 extern void put_filp(struct file *);
 extern int get_unused_fd(void);
 extern void FASTCALL(put_unused_fd(unsigned int fd));
-struct kmem_cache;
 
 extern struct file ** alloc_fd_array(int);
 extern void free_fd_array(struct file **, int);
include/linux/fs.h +10 -0
index 43aef9b..623df73 100644
@@ -33,7 +33,11 @@ struct files_stat_struct {
 	int nr_free_files;	/* read only */
 	int max_files;		/* tunable */
 };
+struct files_kernel_stat_struct {
+	int nr_kernel_files;	/* read only */
+};
 extern struct files_stat_struct files_stat;
+extern struct files_kernel_stat_struct files_kernel_stat;
 extern int get_max_files(void);
 
 struct inodes_stat_t {
@@ -69,6 +73,8 @@ #define FMODE_PWRITE	FMODE_PREAD	/* Thes
    behavior for cross-node execution/opening_for_writing of files */
 #define FMODE_EXEC	16
 
+#define FKFLAGS_NO_ENFILE	1	/* kernel internal file (ignored for ENFILE accounting) */
+
 #define RW_MASK		1
 #define RWA_MASK	2
 #define READ 0
@@ -678,6 +684,7 @@ struct file {
 	atomic_t		f_count;
 	unsigned int 		f_flags;
 	mode_t			f_mode;
+	unsigned short		f_kernel_flags;
 	loff_t			f_pos;
 	struct fown_struct	f_owner;
 	unsigned int		f_uid, f_gid;
@@ -1419,6 +1426,7 @@ extern long do_sys_open(int fdf, const c
 			int mode);
 extern struct file *filp_open(const char *, int, int);
 extern struct file * dentry_open(struct dentry *, struct vfsmount *, int);
+extern struct file * dentry_open_kernel(struct dentry *, struct vfsmount *, int);
 extern int filp_close(struct file *, fl_owner_t id);
 extern char * getname(const char __user *);
 
@@ -1622,6 +1630,7 @@ static inline void insert_inode_hash(str
 }
 
 extern struct file * get_empty_filp(void);
+extern struct file * get_empty_kernel_filp(unsigned short fkflags);
 extern void file_move(struct file *f, struct list_head *list);
 extern void file_kill(struct file *f);
 struct bio;
@@ -1647,6 +1656,7 @@ extern ssize_t generic_file_direct_write
 		unsigned long *, loff_t, loff_t *, size_t, size_t);
 extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *,
 		unsigned long, loff_t, loff_t *, size_t, ssize_t);
+extern int generic_file_buffered_write_one_kernel_page(struct file *, pgoff_t, struct page *);
 extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos);
 extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos);
 ssize_t generic_file_write_nolock(struct file *file, const struct iovec *iov,
include/linux/sysctl.h +1 -0
index e4b1a4d..5bbbc79 100644
@@ -794,6 +794,7 @@ enum
 	FS_AIO_NR=18,	/* current system-wide number of aio requests */
 	FS_AIO_MAX_NR=19,	/* system-wide maximum number of aio requests */
 	FS_INOTIFY=20,	/* inotify submenu */
+	FS_FILE_KERNEL=21,	/* int: number of internal kernel files */
 };
 
 /* /proc/sys/fs/quota/ */
kernel/sysctl.c +11 -0
index 362a0cc..32043a2 100644
@@ -52,6 +52,9 @@ #include <asm/processor.h>
 extern int proc_nr_files(ctl_table *table, int write, struct file *filp,
                      void __user *buffer, size_t *lenp, loff_t *ppos);
 
+extern int proc_files_kernel(ctl_table *table, int write, struct file *filp,
+                     void __user *buffer, size_t *lenp, loff_t *ppos);
+
 #if defined(CONFIG_SYSCTL)
 
 /* External variables not in a header file. */
@@ -993,6 +996,14 @@ static ctl_table fs_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 	{
+		.ctl_name	= FS_FILE_KERNEL,
+		.procname	= "file-kernel",
+		.data		= &files_stat,
+		.maxlen		= 1*sizeof(int),
+		.mode		= 0444,
+		.proc_handler	= &proc_files_kernel,
+	},
+	{
 		.ctl_name	= FS_DENTRY,
 		.procname	= "dentry-state",
 		.data		= &dentry_stat,
#15 lore raw

Re: [PATCH 14/21] FS-Cache: Avoid ENFILE checking for kernel-specific open files

fs/file_table.c int proc_nr_files(ctl_table *table, int · line 118

Quoted 7 lines · On Thu, 06 Jul 2006 11:10:53 -0400 Trond Myklebust…
fs/file_table.c +4 -1
-	if (get_nr_files() >= files_stat.max_files && !capable(CAP_SYS_ADMIN)) {
+	if (!(kflags & FKFLAGS_NO_ENFILE) &&
+	    get_nr_files() >= files_stat.max_files &&
+	    !capable(CAP_SYS_ADMIN)
+	    ) {

Somebody's just going to put this back to kernel-style later anyway. We might as well get it correct up-front.

#16 lore raw

Re: [PATCH 14/21] FS-Cache: Avoid ENFILE checking for kernel-specific open files

In reply to Trond Myklebust
Quoted 9 lines · Trond Myklebust

From: David Howells <[email protected]>

Make it possible to avoid ENFILE checking for kernel specific open files, such as are used by the CacheFiles module.

After, for example, tarring up a kernel source tree over the network, the CacheFiles module may easily have 20000+ files open in the backing filesystem, thus causing all non-root processes to be given error ENFILE when they try to open a file, socket, pipe, etc..

Big fat NACK/. If you don't want file accouting don't use files. The actual users of this unfortunately wasn't posted, but it's using files only as the optional arguments to ->readpage, to call ->flush and for ->setattr. You could do all that aswell at the fs level. Besides that unposted patch adds various odd exports, another silly write something variant in filemap.c, etc.. Also the documentation claims it needs FIBMAP support from userspace which is a big no-way. Can we please get a clear description why all this bloat is needed?

Also David, please try to follow the kernel codingstyle. No silly

	} /* end silly function */

comments. No

/*****************************************************************************/

lines before b lock comments, please follow kernel doc style instead.

Also please don't add _MODULE ifdefs in headers.

maintainer
#17 lore raw

Re: [PATCH 14/21] FS-Cache: Avoid ENFILE checking for kernel-specific open files

Quoted 9 lines · Christoph Hellwig
Quoted 7 lines

Make it possible to avoid ENFILE checking for kernel specific open files, such as are used by the CacheFiles module.

After, for example, tarring up a kernel source tree over the network, the CacheFiles module may easily have 20000+ files open in the backing filesystem, thus causing all non-root processes to be given error ENFILE when they try to open a file, socket, pipe, etc..

Big fat NACK/. If you don't want file accouting don't use files.

File accounting is to prevent *userspace* from executing a DoS attack.

"struct file" is the interface I have to use.

I'd prefer to only have to deal with inodes or maybe dentries and inodes - but various interfaces seem to require it. I'm trying to keep the memory usage down.

Quoted 1 line

The actual users of this unfortunately wasn't posted,

That's not my fault. Trond or the mailing list seems to have lost patch 18/21 somewhere.

Quoted 2 lines

but it's using files only as the optional arguments to ->readpage, to call ->flush and for ->setattr. You could do all that aswell at the fs level.

No, I couldn't. I'm trying to keep the VFS API changes down for the moment.

Besides, you're list of ops that seem to require it is incomplete.

Quoted 1 line

Besides that unposted patch adds various odd exports,

Yes, and?

Quoted 1 line

another silly write something variant in filemap.c, etc..

And how am I supposed to do this otherwise? I don't want to muck about with kiocb's and iovecs aren't any use, also I'm going to write one page and exactly one page, so I can optimise it quite well.

I could move this function into cachefiles, but judging from past performance you'd blow a gasket if I did. Also, I'm not sure that all the functions it calls are globally available.

Quoted 2 lines

Also the documentation claims it needs FIBMAP support from userspace which is a big no-way.

Actually, I don't use FIBMAP at all. I was just using that to guide people to what I meant by bmap().

There isn't any way around using bmap() at the moment. The current AIO interface just isn't usable for this as is, and I have to detect holes somehow (I'd *like* to add O_NOHOLE and use O_DIRECT, but that looks like a major upheaval will be required in the VM/VFS).

This works for the moment, permitting the other bits to be tested more widely.

Quoted 1 line

Can we please get a clear description why all this bloat is needed?

You have one. Read the description. Then go and re-read the discussion back on the 20th of April.

I'm sorry you don't have a new copy of the CacheFiles patch. You can find it in:

http://people.redhat.com/~dhowells/nfs/nfs+fscache.tar.bz2

Quoted 1 line

Also please don't add _MODULE ifdefs in headers.

What choice do I have? CONFIG_FSCACHE is not defined if CONFIG_FSCACHE_MODULE is.

David

#18 lore raw

Re: [PATCH 14/21] FS-Cache: Avoid ENFILE checking for kernel-specific open files

In reply to Christoph Hellwig
Quoted 8 lines · Christoph Hellwig

Big fat NACK/. If you don't want file accouting don't use files. The actual users of this unfortunately wasn't posted, but it's using files only as the optional arguments to ->readpage, to call ->flush and for ->setattr. You could do all that aswell at the fs level. Besides that unposted patch adds various odd exports, another silly write something variant in filemap.c, etc.. Also the documentation claims it needs FIBMAP support from userspace which is a big no-way. Can we please get a clear description why all this bloat is needed?

I'm not sure why some of the mails got lost. I checked that they were sent, so presumably it would be some size limit on the linux-fsdevel list or something like that caused them to be censored.

The full set of patches can be viewed on the linux-nfs.org website. See

http://client.linux-nfs.org/Linux-2.6.x/2.6.18-rc1/

Cheers,
Trond

#19 lore raw

PATCH 16/21 FS-Cache: Release page->private in failed readahead

In reply to Trond Myklebust

From: David Howells <[email protected]>

The attached patch causes read_cache_pages() to release page-private data on a page for which add_to_page_cache() fails or the filler function fails. This permits pages with caching references associated with them to be cleaned up.

Further changes [try #9] that have been made:

(*) The try_to_release_page() is called instead of calling the releasepage()
op directly.

(*) The page is locked before try_to_release_page() is called.

(*) The call to try_to_release_page() and page_cache_release() have been
abstracted out into a helper function as this bit of code occurs twice..

Further changes [try #10] that have been made:

(*) The comment header on the helper function is much expanded. This states
why there's a need to call the releasepage() op in the event of an error.

(*) BUG() if the page is already locked when we try and lock it.

(*) Don't set the page mapping pointer until we've locked the page.

(*) The page is unlocked after try_to_release_page() is called.

---

 mm/readahead.c |   25 +++++++++++++++++++++++--
 1 files changed, 23 insertions(+), 2 deletions(-)
mm/readahead.c +23 -2
index aa7ec42..86a40db 100644
@@ -14,6 +14,7 @@ #include <linux/module.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
+#include <linux/buffer_head.h>
 
 void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
 {
@@ -117,6 +118,26 @@ static inline unsigned long get_next_ra_
 
 #define list_to_page(head) (list_entry((head)->prev, struct page, lru))
 
+/*
+ * see if a page needs releasing upon read_cache_pages() failure
+ * - the caller of read_cache_pages() may have set PG_private before calling,
+ *   such as the NFS fs marking pages that are cached locally on disk, thus we
+ *   need to give the fs a chance to clean up in the event of an error
+ */
+static inline void read_cache_pages_release_page(struct address_space *mapping,
+						 struct page *page)
+{
+	if (PagePrivate(page)) {
+		if (TestSetPageLocked(page))
+			BUG();
+		page->mapping = mapping;
+		try_to_release_page(page, GFP_KERNEL);
+		page->mapping = NULL;
+		unlock_page(page);
+	}
+	page_cache_release(page);
+}
+
 /**
  * read_cache_pages - populate an address space with some pages & start reads against them
  * @mapping: the address_space
@@ -140,7 +161,7 @@ int read_cache_pages(struct address_spac
 		page = list_to_page(pages);
 		list_del(&page->lru);
 		if (add_to_page_cache(page, mapping, page->index, GFP_KERNEL)) {
-			page_cache_release(page);
+			read_cache_pages_release_page(mapping, page);
 			continue;
 		}
 		ret = filler(data, page);
@@ -152,7 +173,7 @@ int read_cache_pages(struct address_spac
 
 				victim = list_to_page(pages);
 				list_del(&victim->lru);
-				page_cache_release(victim);
+				read_cache_pages_release_page(mapping, victim);
 			}
 			break;
 		}
#20 lore raw

Re: [PATCH 16/21] FS-Cache: Release page->private in failed readahead

mm/readahead.c static inline unsigned long get_next_ra_ · line 128

Quoted 2 lines · On Thu, 06 Jul 2006 11:10:58 -0400 Trond Myklebust…
mm/readahead.c +2 -0
+static inline void read_cache_pages_release_page(struct address_space *mapping,
+						 struct page *page)

This has two callsites and is too large to inline.

I've already fixed this once:

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17-rc4/2.6.17-rc4-mm1/broken-out/fs-cache-release-page-private-in-failed-readahead-uninlining.patch

Did any other patches got lost? (and signed-off-bys..)

maintainer
#21 lore raw

Re: [PATCH 16/21] FS-Cache: Release page->private in failed readahead

Quoted 1 line · Andrew Morton

This has two callsites and is too large to inline.

The compiler may inline it anyway, but that's up to it.

I think this is the only bit of your extra patches that I've missed. I'll add it.

Quoted 1 line

Did any other patches got lost? (and signed-off-bys..)

Not at as far as I know.

David

#22 lore raw

PATCH 17/21 FS-Cache: Make kAFS use FS-Cache

In reply to Trond Myklebust

From: David Howells <[email protected]>

The attached patch makes the kAFS filesystem in fs/afs/ use FS-Cache, and through it any attached caches. The kAFS filesystem will use caching automatically if it's available.

---

 fs/Kconfig         |    7 +
 fs/afs/cache.h     |   27 ------
 fs/afs/cell.c      |  109 ++++++++++++++---------
 fs/afs/cell.h      |   16 +--
 fs/afs/cmservice.c |    2 
 fs/afs/dir.c       |   10 +-
 fs/afs/file.c      |  246 ++++++++++++++++++++++++++++++++++------------------
 fs/afs/fsclient.c  |    4 +
 fs/afs/inode.c     |   45 +++++++--
 fs/afs/internal.h  |   25 ++---
 fs/afs/main.c      |   24 ++---
 fs/afs/mntpt.c     |   12 +--
 fs/afs/proc.c      |    1 
 fs/afs/server.c    |    3 -
 fs/afs/vlocation.c |  179 +++++++++++++++++++++++---------------
 fs/afs/vnode.c     |  248 +++++++++++++++++++++++++++++++++++++++++++---------
 fs/afs/vnode.h     |   10 +-
 fs/afs/volume.c    |   78 ++++++----------
 fs/afs/volume.h    |   28 +-----
 19 files changed, 662 insertions(+), 412 deletions(-)
fs/Kconfig +7 -0
index fae41c0..aecaa6e 100644
@@ -1921,6 +1921,13 @@ # for fs/nls/Config.in
 
 	  If unsure, say N.
 
+config AFS_FSCACHE
+	bool "Provide AFS client caching support"
+	depends on AFS_FS && FSCACHE && EXPERIMENTAL
+	help
+	  Say Y here if you want AFS data to be cached locally on through the
+	  generic filesystem cache manager
+
 config RXRPC
 	tristate
 
fs/afs/cache.h deleted +0 -27
deleted file mode 100644
index 9eb7722..0000000
@@ -1,27 +0,0 @@
-/* cache.h: AFS local cache management interface
- *
- * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells ([email protected])
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#ifndef _LINUX_AFS_CACHE_H
-#define _LINUX_AFS_CACHE_H
-
-#undef AFS_CACHING_SUPPORT
-
-#include <linux/mm.h>
-#ifdef AFS_CACHING_SUPPORT
-#include <linux/cachefs.h>
-#endif
-#include "types.h"
-
-#ifdef __KERNEL__
-
-#endif /* __KERNEL__ */
-
-#endif /* _LINUX_AFS_CACHE_H */
fs/afs/cell.c +65 -44

146 lines not on this page · raw message

fs/afs/cell.h +3 -13
index 4834910..d670502 100644
@@ -13,7 +13,7 @@ #ifndef _LINUX_AFS_CELL_H
 #define _LINUX_AFS_CELL_H
 
 #include "types.h"
-#include "cache.h"
+#include <linux/fscache.h>
 
 #define AFS_CELL_MAX_ADDRS 15
 
@@ -21,16 +21,6 @@ extern volatile int afs_cells_being_purg
 
 /*****************************************************************************/
 /*
- * entry in the cached cell catalogue
- */
-struct afs_cache_cell
-{
-	char			name[64];	/* cell name (padded with NULs) */
-	struct in_addr		vl_servers[15];	/* cached cell VL servers */
-};
-
-/*****************************************************************************/
-/*
  * AFS cell record
  */
 struct afs_cell
@@ -39,8 +29,8 @@ struct afs_cell
 	struct list_head	link;		/* main cell list link */
 	struct list_head	proc_link;	/* /proc cell list link */
 	struct proc_dir_entry	*proc_dir;	/* /proc dir for this cell */
-#ifdef AFS_CACHING_SUPPORT
-	struct cachefs_cookie	*cache;		/* caching cookie */
+#ifdef CONFIG_AFS_FSCACHE
+	struct fscache_cookie	*cache;		/* caching cookie */
 #endif
 
 	/* server record management */
fs/afs/cmservice.c +1 -1
index 3d097fd..f87d5a7 100644
@@ -24,7 +24,7 @@ #include "cmservice.h"
 #include "internal.h"
 
 static unsigned afscm_usage;		/* AFS cache manager usage count */
-static struct rw_semaphore afscm_sem;	/* AFS cache manager start/stop semaphore */
+static DECLARE_RWSEM(afscm_sem);	/* AFS cache manager start/stop semaphore */
 
 static int afscm_new_call(struct rxrpc_call *call);
 static void afscm_attention(struct rxrpc_call *call);
fs/afs/dir.c +6 -4
index 3b78e70..9800a07 100644
@@ -145,7 +145,7 @@ #endif
 	qty /= sizeof(union afs_dir_block);
 
 	/* check them */
-	dbuf = page_address(page);
+	dbuf = kmap_atomic(page, KM_USER0);
 	for (tmp = 0; tmp < qty; tmp++) {
 		if (dbuf->blocks[tmp].pagehdr.magic != AFS_DIR_MAGIC) {
 			printk("kAFS: %s(%lu): bad magic %d/%d is %04hx\n",
@@ -154,10 +154,12 @@ #endif
 			goto error;
 		}
 	}
+	kunmap_atomic(dbuf, KM_USER0);
 
 	return;
 
  error:
+	kunmap_atomic(dbuf, KM_USER0);
 	SetPageError(page);
 
 } /* end afs_dir_check_page() */
@@ -168,7 +170,6 @@ #endif
  */
 static inline void afs_dir_put_page(struct page *page)
 {
-	kunmap(page);
 	page_cache_release(page);
 
 } /* end afs_dir_put_page() */
@@ -186,7 +187,6 @@ static struct page *afs_dir_get_page(str
 	page = read_mapping_page(dir->i_mapping, index, NULL);
 	if (!IS_ERR(page)) {
 		wait_on_page_locked(page);
-		kmap(page);
 		if (!PageUptodate(page))
 			goto fail;
 		afs_dir_check_page(dir, page);
@@ -354,7 +354,7 @@ static int afs_dir_iterate(struct inode 
 
 		limit = blkoff & ~(PAGE_SIZE - 1);
 
-		dbuf = page_address(page);
+		dbuf = kmap_atomic(page, KM_USER0);
 
 		/* deal with the individual blocks stashed on this page */
 		do {
@@ -363,6 +363,7 @@ static int afs_dir_iterate(struct inode 
 			ret = afs_dir_iterate_block(fpos, dblock, blkoff,
 						    cookie, filldir);
 			if (ret != 1) {
+				kunmap_atomic(dbuf, KM_USER0);
 				afs_dir_put_page(page);
 				goto out;
 			}
@@ -371,6 +372,7 @@ static int afs_dir_iterate(struct inode 
 
 		} while (*fpos < dir->i_size && blkoff < limit);
 
+		kunmap_atomic(dbuf, KM_USER0);
 		afs_dir_put_page(page);
 		ret = 0;
 	}
fs/afs/file.c +164 -82

384 lines not on this page · raw message

fs/afs/fsclient.c +3 -1
index 61bc371..c88c41a 100644
@@ -398,6 +398,8 @@ int afs_rxfs_fetch_file_status(struct af
 		bp++; /* spare6 */
 	}
 
+	_debug("Data Version %llx\n", vnode->status.version);
+
 	/* success */
 	ret = 0;
 
@@ -408,7 +410,7 @@ int afs_rxfs_fetch_file_status(struct af
  out_put_conn:
 	afs_server_release_callslot(server, &callslot);
  out:
-	_leave("");
+	_leave(" = %d", ret);
 	return ret;
 
  abort:
fs/afs/inode.c +31 -14

91 lines not on this page · raw message

fs/afs/internal.h +10 -15
index e88b3b6..482dbd1 100644
@@ -16,15 +16,17 @@ #include <linux/compiler.h>
 #include <linux/kernel.h>
 #include <linux/fs.h>
 #include <linux/pagemap.h>
+#include <linux/fscache.h>
 
 /*
  * debug tracing
  */
-#define kenter(FMT, a...)	printk("==> %s("FMT")\n",__FUNCTION__ , ## a)
-#define kleave(FMT, a...)	printk("<== %s()"FMT"\n",__FUNCTION__ , ## a)
-#define kdebug(FMT, a...)	printk(FMT"\n" , ## a)
-#define kproto(FMT, a...)	printk("### "FMT"\n" , ## a)
-#define knet(FMT, a...)		printk(FMT"\n" , ## a)
+#define __kdbg(FMT, a...)	printk("[%05d] "FMT"\n", current->pid , ## a)
+#define kenter(FMT, a...)	__kdbg("==> %s("FMT")", __FUNCTION__ , ## a)
+#define kleave(FMT, a...)	__kdbg("<== %s()"FMT, __FUNCTION__ , ## a)
+#define kdebug(FMT, a...)	__kdbg(FMT , ## a)
+#define kproto(FMT, a...)	__kdbg("### "FMT , ## a)
+#define knet(FMT, a...)		__kdbg(FMT , ## a)
 
 #ifdef __KDEBUG
 #define _enter(FMT, a...)	kenter(FMT , ## a)
@@ -56,9 +58,6 @@ static inline void afs_discard_my_signal
  */
 extern struct rw_semaphore afs_proc_cells_sem;
 extern struct list_head afs_proc_cells;
-#ifdef AFS_CACHING_SUPPORT
-extern struct cachefs_index_def afs_cache_cell_index_def;
-#endif
 
 /*
  * dir.c
@@ -71,11 +70,7 @@ extern const struct file_operations afs_
  */
 extern const struct address_space_operations afs_fs_aops;
 extern struct inode_operations afs_file_inode_operations;
-
-#ifdef AFS_CACHING_SUPPORT
-extern int afs_cache_get_page_cookie(struct page *page,
-				     struct cachefs_page **_page_cookie);
-#endif
+extern const struct file_operations afs_file_file_operations;
 
 /*
  * inode.c
@@ -97,8 +92,8 @@ #endif
 /*
  * main.c
  */
-#ifdef AFS_CACHING_SUPPORT
-extern struct cachefs_netfs afs_cache_netfs;
+#ifdef CONFIG_AFS_FSCACHE
+extern struct fscache_netfs afs_cache_netfs;
 #endif
 
 /*
fs/afs/main.c +11 -13
index 913c689..5840bb2 100644
@@ -1,6 +1,6 @@
 /* main.c: AFS client file system
  *
- * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2002,5 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells ([email protected])
  *
  * This program is free software; you can redistribute it and/or
@@ -14,11 +14,11 @@ #include <linux/moduleparam.h>
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/completion.h>
+#include <linux/fscache.h>
 #include <rxrpc/rxrpc.h>
 #include <rxrpc/transport.h>
 #include <rxrpc/call.h>
 #include <rxrpc/peer.h>
-#include "cache.h"
 #include "cell.h"
 #include "server.h"
 #include "fsclient.h"
@@ -51,12 +51,11 @@ static struct rxrpc_peer_ops afs_peer_op
 struct list_head afs_cb_hash_tbl[AFS_CB_HASH_COUNT];
 DEFINE_SPINLOCK(afs_cb_hash_lock);
 
-#ifdef AFS_CACHING_SUPPORT
-static struct cachefs_netfs_operations afs_cache_ops = {
-	.get_page_cookie	= afs_cache_get_page_cookie,
+#ifdef CONFIG_AFS_FSCACHE
+static struct fscache_netfs_operations afs_cache_ops = {
 };
 
-struct cachefs_netfs afs_cache_netfs = {
+struct fscache_netfs afs_cache_netfs = {
 	.name			= "afs",
 	.version		= 0,
 	.ops			= &afs_cache_ops,
@@ -83,10 +82,9 @@ static int __init afs_init(void)
 	if (ret < 0)
 		return ret;
 
-#ifdef AFS_CACHING_SUPPORT
+#ifdef CONFIG_AFS_FSCACHE
 	/* we want to be able to cache */
-	ret = cachefs_register_netfs(&afs_cache_netfs,
-				     &afs_cache_cell_index_def);
+	ret = fscache_register_netfs(&afs_cache_netfs);
 	if (ret < 0)
 		goto error;
 #endif
@@ -137,8 +135,8 @@ #ifdef CONFIG_KEYS_TURNED_OFF
 	afs_key_unregister();
  error_cache:
 #endif
-#ifdef AFS_CACHING_SUPPORT
-	cachefs_unregister_netfs(&afs_cache_netfs);
+#ifdef CONFIG_AFS_FSCACHE
+	fscache_unregister_netfs(&afs_cache_netfs);
  error:
 #endif
 	afs_cell_purge();
@@ -167,8 +165,8 @@ static void __exit afs_exit(void)
 #ifdef CONFIG_KEYS_TURNED_OFF
 	afs_key_unregister();
 #endif
-#ifdef AFS_CACHING_SUPPORT
-	cachefs_unregister_netfs(&afs_cache_netfs);
+#ifdef CONFIG_AFS_FSCACHE
+	fscache_unregister_netfs(&afs_cache_netfs);
 #endif
 	afs_proc_cleanup();
 
fs/afs/mntpt.c +6 -6
index 99785a7..2a53d51 100644
@@ -78,7 +78,7 @@ int afs_mntpt_check_symlink(struct afs_v
 
 	ret = -EIO;
 	wait_on_page_locked(page);
-	buf = kmap(page);
+	buf = kmap_atomic(page, KM_USER0);
 	if (!PageUptodate(page))
 		goto out_free;
 	if (PageError(page))
@@ -101,7 +101,7 @@ int afs_mntpt_check_symlink(struct afs_v
 	ret = 0;
 
  out_free:
-	kunmap(page);
+	kunmap_atomic(buf, KM_USER0);
 	page_cache_release(page);
  out:
 	_leave(" = %d", ret);
@@ -188,9 +188,9 @@ static struct vfsmount *afs_mntpt_do_aut
 	if (!PageUptodate(page) || PageError(page))
 		goto error;
 
-	buf = kmap(page);
+	buf = kmap_atomic(page, KM_USER0);
 	memcpy(devname, buf, size);
-	kunmap(page);
+	kunmap_atomic(buf, KM_USER0);
 	page_cache_release(page);
 	page = NULL;
 
@@ -269,12 +269,12 @@ static void *afs_mntpt_follow_link(struc
  */
 static void afs_mntpt_expiry_timed_out(struct afs_timer *timer)
 {
-	kenter("");
+//	kenter("");
 
 	mark_mounts_for_expiry(&afs_vfsmounts);
 
 	afs_kafstimod_add_timer(&afs_mntpt_expiry_timer,
 				afs_mntpt_expiry_timeout * HZ);
 
-	kleave("");
+//	kleave("");
 } /* end afs_mntpt_expiry_timed_out() */
fs/afs/proc.c +1 -0
index 101d21b..db58488 100644
@@ -177,6 +177,7 @@ int afs_proc_init(void)
  */
 void afs_proc_cleanup(void)
 {
+	remove_proc_entry("rootcell", proc_afs);
 	remove_proc_entry("cells", proc_afs);
 
 	remove_proc_entry("fs/afs", NULL);
fs/afs/server.c +1 -2
index 22afaae..e94628c 100644
@@ -375,7 +375,6 @@ int afs_server_request_callslot(struct a
 	else if (list_empty(&server->fs_callq)) {
 		/* no one waiting */
 		server->fs_conn_cnt[nconn]++;
-		spin_unlock(&server->fs_lock);
 	}
 	else {
 		/* someone's waiting - dequeue them and wake them up */
@@ -393,9 +392,9 @@ int afs_server_request_callslot(struct a
 		}
 		pcallslot->ready = 1;
 		wake_up_process(pcallslot->task);
-		spin_unlock(&server->fs_lock);
 	}
 
+	spin_unlock(&server->fs_lock);
 	rxrpc_put_connection(callslot->conn);
 	callslot->conn = NULL;
 
fs/afs/vlocation.c +110 -69

237 lines not on this page · raw message

fs/afs/vnode.c +204 -44

287 lines not on this page · raw message

fs/afs/vnode.h +5 -5
index b86a971..3f0602d 100644
@@ -13,9 +13,9 @@ #ifndef _LINUX_AFS_VNODE_H
 #define _LINUX_AFS_VNODE_H
 
 #include <linux/fs.h>
+#include <linux/fscache.h>
 #include "server.h"
 #include "kafstimod.h"
-#include "cache.h"
 
 #ifdef __KERNEL__
 
@@ -32,8 +32,8 @@ struct afs_cache_vnode
 	afs_dataversion_t	data_version;	/* data version */
 };
 
-#ifdef AFS_CACHING_SUPPORT
-extern struct cachefs_index_def afs_vnode_cache_index_def;
+#ifdef CONFIG_AFS_FSCACHE
+extern struct fscache_cookie_def afs_vnode_cache_index_def;
 #endif
 
 /*****************************************************************************/
@@ -47,8 +47,8 @@ struct afs_vnode
 	struct afs_volume	*volume;	/* volume on which vnode resides */
 	struct afs_fid		fid;		/* the file identifier for this inode */
 	struct afs_file_status	status;		/* AFS status info for this file */
-#ifdef AFS_CACHING_SUPPORT
-	struct cachefs_cookie	*cache;		/* caching cookie */
+#ifdef CONFIG_AFS_FSCACHE
+	struct fscache_cookie	*cache;		/* caching cookie */
 #endif
 
 	wait_queue_head_t	update_waitq;	/* status fetch waitqueue */
fs/afs/volume.c +29 -49

116 lines not on this page · raw message

fs/afs/volume.h +5 -23
index bfdcf19..fc9895a 100644
@@ -12,11 +12,11 @@
 #ifndef _LINUX_AFS_VOLUME_H
 #define _LINUX_AFS_VOLUME_H
 
+#include <linux/fscache.h>
 #include "types.h"
 #include "fsclient.h"
 #include "kafstimod.h"
 #include "kafsasyncd.h"
-#include "cache.h"
 
 typedef enum {
 	AFS_VLUPD_SLEEP,		/* sleeping waiting for update timer to fire */
@@ -45,24 +45,6 @@ #define AFS_VOL_VTM_BAK	0x04 /* backup v
 	time_t			rtime;		/* last retrieval time */
 };
 
-#ifdef AFS_CACHING_SUPPORT
-extern struct cachefs_index_def afs_vlocation_cache_index_def;
-#endif
-
-/*****************************************************************************/
-/*
- * volume -> vnode hash table entry
- */
-struct afs_cache_vhash
-{
-	afs_voltype_t		vtype;		/* which volume variation */
-	uint8_t			hash_bucket;	/* which hash bucket this represents */
-} __attribute__((packed));
-
-#ifdef AFS_CACHING_SUPPORT
-extern struct cachefs_index_def afs_volume_cache_index_def;
-#endif
-
 /*****************************************************************************/
 /*
  * AFS volume location record
@@ -73,8 +55,8 @@ struct afs_vlocation
 	struct list_head	link;		/* link in cell volume location list */
 	struct afs_timer	timeout;	/* decaching timer */
 	struct afs_cell		*cell;		/* cell to which volume belongs */
-#ifdef AFS_CACHING_SUPPORT
-	struct cachefs_cookie	*cache;		/* caching cookie */
+#ifdef CONFIG_AFS_FSCACHE
+	struct fscache_cookie	*cache;		/* caching cookie */
 #endif
 	struct afs_cache_vlocation vldb;	/* volume information DB record */
 	struct afs_volume	*vols[3];	/* volume access record pointer (index by type) */
@@ -109,8 +91,8 @@ struct afs_volume
 	atomic_t		usage;
 	struct afs_cell		*cell;		/* cell to which belongs (unrefd ptr) */
 	struct afs_vlocation	*vlocation;	/* volume location */
-#ifdef AFS_CACHING_SUPPORT
-	struct cachefs_cookie	*cache;		/* caching cookie */
+#ifdef CONFIG_AFS_FSCACHE
+	struct fscache_cookie	*cache;		/* caching cookie */
 #endif
 	afs_volid_t		vid;		/* volume ID */
 	afs_voltype_t		type;		/* type of volume */
#23 lore raw

PATCH 19/21 NFS: Use local caching

In reply to Trond Myklebust

From: David Howells <[email protected]>

The attached patch makes it possible for the NFS filesystem to make use of the network filesystem local caching service (FS-Cache).

To be able to use this, an updated mount program is required. This can be obtained from:

http://people.redhat.com/steved/cachefs/util-linux/

To mount an NFS filesystem to use caching, add an "fsc" option to the mount:

mount warthog:/ /a -o fsc

---

 fs/Kconfig                   |    7 +
 fs/cachefiles/cf-interface.c |    6 -
 fs/nfs/Makefile              |    1 
 fs/nfs/client.c              |   11 +
 fs/nfs/file.c                |   33 +++
 fs/nfs/fscache.c             |  349 +++++++++++++++++++++++++++++++
 fs/nfs/fscache.h             |  466 ++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/inode.c               |   22 ++
 fs/nfs/internal.h            |   32 +++
 fs/nfs/pagelist.c            |    3 
 fs/nfs/read.c                |   30 +++
 fs/nfs/super.c               |    1 
 fs/nfs/sysctl.c              |   43 ++++
 fs/nfs/write.c               |   11 +
 include/linux/nfs4_mount.h   |    1 
 include/linux/nfs_fs.h       |    5 
 include/linux/nfs_fs_sb.h    |    5 
 include/linux/nfs_mount.h    |    1 
 18 files changed, 1017 insertions(+), 10 deletions(-)
fs/Kconfig +7 -0
index 4743bcc..12e77a1 100644
@@ -1505,6 +1505,13 @@ config NFS_V4
 
 	  If unsure, say N.
 
+config NFS_FSCACHE
+	bool "Provide NFS client caching support (EXPERIMENTAL)"
+	depends on NFS_FS && FSCACHE && EXPERIMENTAL
+	help
+	  Say Y here if you want NFS data to be cached locally on disc through
+	  the general filesystem cache manager
+
 config NFS_DIRECTIO
 	bool "Allow direct I/O on NFS files (EXPERIMENTAL)"
 	depends on NFS_FS && EXPERIMENTAL
fs/cachefiles/cf-interface.c +4 -2
index 131a4be..32ad844 100644
@@ -635,8 +635,10 @@ success:
 out:
 	if (backpage)
 		page_cache_release(backpage);
-	fscache_put_context(object->fscache.cookie, monitor->context);
-	kfree(monitor);
+	if (monitor) {
+		fscache_put_context(object->fscache.cookie, monitor->context);
+		kfree(monitor);
+	}
 
 	_leave(" = %d", ret);
 	return ret;
fs/nfs/Makefile +1 -0
index f4580b4..2af6f22 100644
@@ -16,4 +16,5 @@ nfs-$(CONFIG_NFS_V4)	+= nfs4proc.o nfs4x
 			   nfs4namespace.o
 nfs-$(CONFIG_NFS_DIRECTIO) += direct.o
 nfs-$(CONFIG_SYSCTL) += sysctl.o
+nfs-$(CONFIG_NFS_FSCACHE) += fscache.o
 nfs-objs		:= $(nfs-y)
fs/nfs/client.c +8 -3
index 8ae3f74..bf701f1 100644
@@ -150,6 +150,8 @@ #ifdef CONFIG_NFS_V4
 	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
 #endif
 
+	nfs_fscache_get_client_cookie(clp);
+
 	return clp;
 
 error_3:
@@ -187,6 +189,8 @@ #ifdef CONFIG_NFS_V4
 	}
 #endif
 
+	nfs_fscache_release_client_cookie(clp);
+
 	/* -EIO all pending I/O */
 	if (!IS_ERR(clp->cl_rpcclient))
 		rpc_shutdown_client(clp->cl_rpcclient);
@@ -1360,7 +1364,7 @@ static int nfs_volume_list_show(struct s
 
 	/* display header on line 1 */
 	if (v == SEQ_START_TOKEN) {
-		seq_puts(m, "NV SERVER   PORT DEV     FSID\n");
+		seq_puts(m, "NV SERVER   PORT DEV     FSID              FSC\n");
 		return 0;
 	}
 	/* display one transport per line on subsequent lines */
@@ -1373,12 +1377,13 @@ static int nfs_volume_list_show(struct s
 	snprintf(fsid, 17, "%llx:%llx",
 		 server->fsid.major, server->fsid.minor);
 
-	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s\n",
+	seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s %s\n",
 		   clp->cl_nfsversion,
 		   NIPQUAD(clp->cl_addr.sin_addr),
 		   ntohs(clp->cl_addr.sin_port),
 		   dev,
-		   fsid);
+		   fsid,
+		   nfs_server_fscache_state(server));
 
 	return 0;
 }
fs/nfs/file.c +32 -1
index 52f161d..855bb97 100644
@@ -27,12 +27,14 @@ #include <linux/mm.h>
 #include <linux/slab.h>
 #include <linux/pagemap.h>
 #include <linux/smp_lock.h>
+#include <linux/buffer_head.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
 
 #include "delegation.h"
 #include "iostat.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY		NFSDBG_FILE
 
@@ -249,6 +251,10 @@ nfs_file_mmap(struct file * file, struct
 	status = nfs_revalidate_mapping(inode, file->f_mapping);
 	if (!status)
 		status = generic_file_mmap(file, vma);
+
+	if (status == 0)
+		nfs_fscache_install_vm_ops(inode, vma);
+
 	return status;
 }
 
@@ -308,13 +314,35 @@ static void nfs_invalidate_page(struct p
 	/* Cancel any unstarted writes on this page */
 	if (offset == 0)
 		nfs_sync_inode_wait(inode, page->index, 1, FLUSH_INVALIDATE);
+
+	nfs_fscache_invalidate_page(page, inode, offset);
+
+	/* we can do this here as the bits are only set with the page lock
+	 * held, and our caller is holding that */
+	if (!page->private)
+		ClearPagePrivate(page);
 }
 
 static int nfs_release_page(struct page *page, gfp_t gfp)
 {
-	return !nfs_wb_page(page->mapping->host, page);
+	int error = nfs_wb_page(page->mapping->host, page);
+
+	if (error == 0) {
+		nfs_fscache_release_page(page);
+
+		/* may have been set due to either caching or writing */
+		ClearPagePrivate(page);
+	}
+
+	/* releasepage() returns true/false */
+	return (error == 0) ? 1 : 0;
 }
 
+/*
+ * Since we use page->private for our own nefarious purposes when using
+ * fscache, we have to override extra address space ops to prevent fs/buffer.c
+ * from getting confused, even though we may not have asked its opinion
+ */
 const struct address_space_operations nfs_file_aops = {
 	.readpage = nfs_readpage,
 	.readpages = nfs_readpages,
@@ -328,6 +356,9 @@ const struct address_space_operations nf
 #ifdef CONFIG_NFS_DIRECTIO
 	.direct_IO = nfs_direct_IO,
 #endif
+#ifdef CONFIG_NFS_FSCACHE
+	.sync_page	= block_sync_page,
+#endif
 };
 
 /* 
fs/nfs/fscache.c new file +349 -0

349 lines not on this page · raw message

fs/nfs/fscache.h new file +466 -0

466 lines not on this page · raw message

fs/nfs/inode.c +22 -0
index 7665b73..dcf916b 100644
@@ -88,6 +88,8 @@ void nfs_clear_inode(struct inode *inode
 	cred = nfsi->cache_access.cred;
 	if (cred)
 		put_rpccred(cred);
+
+	nfs_fscache_release_fh_cookie(NFS_SERVER(inode), nfsi);
 	BUG_ON(atomic_read(&nfsi->data_updates) != 0);
 }
 
@@ -134,6 +136,8 @@ void nfs_zap_caches(struct inode *inode)
 	spin_lock(&inode->i_lock);
 	nfs_zap_caches_locked(inode);
 	spin_unlock(&inode->i_lock);
+
+	nfs_fscache_zap_fh_cookie(NFS_SERVER(inode), NFS_I(inode));
 }
 
 static void nfs_zap_acl_cache(struct inode *inode)
@@ -212,6 +216,7 @@ nfs_fhget(struct super_block *sb, struct
 	};
 	struct inode *inode = ERR_PTR(-ENOENT);
 	unsigned long hash;
+	int maycache = 1;
 
 	if ((fattr->valid & NFS_ATTR_FATTR) == 0)
 		goto out_no_inode;
@@ -260,6 +265,7 @@ nfs_fhget(struct super_block *sb, struct
 				else
 					inode->i_op = &nfs_mountpoint_inode_operations;
 				inode->i_fop = NULL;
+				maycache = 0;
 			}
 		} else if (S_ISLNK(inode->i_mode))
 			inode->i_op = &nfs_symlink_inode_operations;
@@ -292,6 +298,8 @@ nfs_fhget(struct super_block *sb, struct
 		memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf));
 		nfsi->cache_access.cred = NULL;
 
+		nfs_fscache_get_fh_cookie(sb, nfsi, maycache);
+
 		unlock_new_inode(inode);
 	} else
 		nfs_refresh_inode(inode, fattr);
@@ -374,6 +382,7 @@ void nfs_setattr_update_inode(struct ino
 	if ((attr->ia_valid & ATTR_SIZE) != 0) {
 		nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC);
 		inode->i_size = attr->ia_size;
+		nfs_fscache_set_size(NFS_SERVER(inode), NFS_I(inode), inode->i_size);
 		vmtruncate(inode, attr->ia_size);
 	}
 }
@@ -556,6 +565,8 @@ int nfs_open(struct inode *inode, struct
 	ctx->mode = filp->f_mode;
 	nfs_file_set_open_context(filp, ctx);
 	put_nfs_open_context(ctx);
+	if ((filp->f_flags & O_ACCMODE) != O_RDONLY)
+		nfs_fscache_disable_fh_cookie(inode);
 	return 0;
 }
 
@@ -694,6 +705,8 @@ int nfs_revalidate_mapping(struct inode 
 		}
 		spin_unlock(&inode->i_lock);
 
+		nfs_fscache_renew_fh_cookie(NFS_SERVER(inode), nfsi);
+
 		dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n",
 				inode->i_sb->s_id,
 				(long long)NFS_FILEID(inode));
@@ -927,11 +940,13 @@ static int nfs_update_inode(struct inode
 			if (data_stable) {
 				inode->i_size = new_isize;
 				invalid |= NFS_INO_INVALID_DATA;
+				nfs_fscache_set_size(NFS_SERVER(inode), nfsi, inode->i_size);
 			}
 			invalid |= NFS_INO_INVALID_ATTR;
 		} else if (new_isize > cur_isize) {
 			inode->i_size = new_isize;
 			invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA;
+			nfs_fscache_set_size(NFS_SERVER(inode), nfsi, inode->i_size);
 		}
 		nfsi->cache_change_attribute = jiffies;
 		dprintk("NFS: isize change on server for file %s/%ld\n",
@@ -1144,6 +1159,10 @@ static int __init init_nfs_fs(void)
 {
 	int err;
 
+	err = nfs_fscache_register();
+	if (err < 0)
+		goto out6;
+
 	err = nfs_fs_proc_init();
 	if (err)
 		goto out5;
@@ -1190,6 +1209,8 @@ out3:
 out4:
 	nfs_fs_proc_exit();
 out5:
+	nfs_fscache_unregister();
+out6:
 	return err;
 }
 
@@ -1200,6 +1221,7 @@ static void __exit exit_nfs_fs(void)
 	nfs_destroy_readpagecache();
 	nfs_destroy_inodecache();
 	nfs_destroy_nfspagecache();
+	nfs_fscache_unregister();
 #ifdef CONFIG_PROC_FS
 	rpc_proc_unregister("nfs");
 #endif
fs/nfs/internal.h +32 -0
index 1a1312e..18febb4 100644
@@ -4,6 +4,30 @@
 
 #include <linux/mount.h>
 
+#define NFS_PAGE_WRITING	0
+#define NFS_PAGE_CACHED		1
+
+#define PageNfsBit(bit, page)		test_bit(bit, &(page)->private)
+
+#define SetPageNfsBit(bit, page)		\
+do {						\
+	SetPagePrivate((page));			\
+	set_bit(bit, &(page)->private);		\
+} while(0)
+
+#define ClearPageNfsBit(bit, page)		\
+do {						\
+	clear_bit(bit, &(page)->private);	\
+} while(0)
+
+#define PageNfsWriting(page)		PageNfsBit(NFS_PAGE_WRITING, (page))
+#define SetPageNfsWriting(page)		SetPageNfsBit(NFS_PAGE_WRITING, (page))
+#define ClearPageNfsWriting(page)	ClearPageNfsBit(NFS_PAGE_WRITING, (page))
+
+#define PageNfsCached(page)		PageNfsBit(NFS_PAGE_CACHED, (page))
+#define SetPageNfsCached(page)		SetPageNfsBit(NFS_PAGE_CACHED, (page))
+#define ClearPageNfsCached(page)	ClearPageNfsBit(NFS_PAGE_CACHED, (page))
+
 struct nfs_string;
 struct nfs_mount_data;
 struct nfs4_mount_data;
@@ -27,6 +51,11 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
+/*
+ * include filesystem caching stuff here
+ */
+#include "fscache.h"
+
 /* client.c */
 extern struct rpc_program nfs_program;
 
@@ -150,6 +179,9 @@ extern int nfs4_path_walk(struct nfs_ser
 			  const char *path);
 #endif
 
+/* read.c */
+extern int nfs_readpage_async(struct nfs_open_context *, struct inode *, struct page *);
+
 /*
  * Determine the device name as a string
  */
fs/nfs/pagelist.c +2 -1
index 36e902a..c45f724 100644
@@ -17,6 +17,7 @@ #include <linux/nfs4.h>
 #include <linux/nfs_page.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_mount.h>
+#include "internal.h"
 
 #define NFS_PARANOIA 1
 
@@ -84,7 +85,7 @@ nfs_create_request(struct nfs_open_conte
 	atomic_set(&req->wb_complete, 0);
 	req->wb_index	= page->index;
 	page_cache_get(page);
-	BUG_ON(PagePrivate(page));
+	BUG_ON(PageNfsWriting(page));
 	BUG_ON(!PageLocked(page));
 	BUG_ON(page->mapping->host != inode);
 	req->wb_offset  = offset;
fs/nfs/read.c +29 -1
index 3093527..13ff66a 100644
@@ -26,11 +26,13 @@ #include <linux/pagemap.h>
 #include <linux/sunrpc/clnt.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
+#include <linux/nfs_mount.h>
 #include <linux/smp_lock.h>
 
 #include <asm/system.h>
 
 #include "iostat.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PAGECACHE
 
@@ -200,13 +202,18 @@ static int nfs_readpage_sync(struct nfs_
 		SetPageUptodate(page);
 	result = 0;
 
+	nfs_readpage_to_fscache(inode, page, 1);
+	unlock_page(page);
+
+	return result;
+
 io_error:
 	unlock_page(page);
 	nfs_readdata_free(rdata);
 	return result;
 }
 
-static int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
+int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 		struct page *page)
 {
 	LIST_HEAD(one_request);
@@ -231,6 +238,11 @@ static int nfs_readpage_async(struct nfs
 
 static void nfs_readpage_release(struct nfs_page *req)
 {
+	struct inode *d_inode = req->wb_context->dentry->d_inode;
+
+	if (PageUptodate(req->wb_page))
+		nfs_readpage_to_fscache(d_inode, req->wb_page, 0);
+
 	unlock_page(req->wb_page);
 
 	dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
@@ -613,6 +625,10 @@ int nfs_readpage(struct file *file, stru
 		ctx = get_nfs_open_context((struct nfs_open_context *)
 				file->private_data);
 	if (!IS_SYNC(inode)) {
+		error = nfs_readpage_from_fscache(ctx, inode, page);
+		if (error == 0)
+			goto out;
+
 		error = nfs_readpage_async(ctx, inode, page);
 		goto out;
 	}
@@ -643,6 +659,7 @@ readpage_async_filler(void *data, struct
 	unsigned int len;
 
 	nfs_wb_page(inode, page);
+
 	len = nfs_page_length(inode, page);
 	if (len == 0)
 		return nfs_return_empty_page(page);
@@ -682,6 +699,17 @@ int nfs_readpages(struct file *filp, str
 	} else
 		desc.ctx = get_nfs_open_context((struct nfs_open_context *)
 				filp->private_data);
+
+	/* attempt to read as many of the pages as possible from the cache
+	 * - this returns -ENOBUFS immediately if the cookie is negative
+	 */
+	ret = nfs_readpages_from_fscache(desc.ctx, inode, mapping,
+					 pages, &nr_pages);
+	if (ret == 0) {
+		put_nfs_open_context(desc.ctx);
+		return ret; /* all read */
+	}
+
 	ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);
 	if (!list_empty(&head)) {
 		int err = nfs_pagein_list(&head, server->rpages);
fs/nfs/super.c +1 -0
index b5f4c4f..db4e072 100644
@@ -286,6 +286,7 @@ static void nfs_show_mount_options(struc
 		{ NFS_MOUNT_NOAC, ",noac", "" },
 		{ NFS_MOUNT_NONLM, ",nolock", "" },
 		{ NFS_MOUNT_NOACL, ",noacl", "" },
+		{ NFS_MOUNT_FSCACHE, ",fsc", "" },
 		{ 0, NULL, NULL }
 	};
 	const struct proc_nfs_info *nfs_infop;
fs/nfs/sysctl.c +43 -0

55 lines not on this page · raw message

fs/nfs/write.c +9 -2
index a3f3f04..cd8d972 100644
@@ -63,6 +63,7 @@ #include <linux/smp_lock.h>
 
 #include "delegation.h"
 #include "iostat.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PAGECACHE
 
@@ -163,6 +164,9 @@ static void nfs_grow_file(struct page *p
 		return;
 	nfs_inc_stats(inode, NFSIOS_EXTENDWRITE);
 	i_size_write(inode, end);
+#ifdef FSCACHE_WRITE_SUPPORT
+	nfs_set_fscsize(NFS_SERVER(inode), NFS_I(inode), end);
+#endif
 }
 
 /* We can set the PG_uptodate flag if we see that a write request
@@ -342,6 +346,9 @@ do_it:
 		err = -EBADF;
 		goto out;
 	}
+
+	nfs_writepage_to_fscache(inode, page);
+
 	lock_kernel();
 	if (!IS_SYNC(inode) && inode_referenced) {
 		err = nfs_writepage_async(ctx, inode, page, 0, offset);
@@ -424,7 +431,7 @@ static int nfs_inode_add_request(struct 
 		if (nfs_have_delegation(inode, FMODE_WRITE))
 			nfsi->change_attr++;
 	}
-	SetPagePrivate(req->wb_page);
+	SetPageNfsWriting(req->wb_page);
 	nfsi->npages++;
 	atomic_inc(&req->wb_count);
 	return 0;
@@ -441,7 +448,7 @@ static void nfs_inode_remove_request(str
 	BUG_ON (!NFS_WBACK_BUSY(req));
 
 	spin_lock(&nfsi->req_lock);
-	ClearPagePrivate(req->wb_page);
+	ClearPageNfsWriting(req->wb_page);
 	radix_tree_delete(&nfsi->nfs_page_tree, req->wb_index);
 	nfsi->npages--;
 	if (!nfsi->npages) {
include/linux/nfs4_mount.h +1 -0
index 26b4c83..15199cc 100644
@@ -65,6 +65,7 @@ #define NFS4_MOUNT_INTR		0x0002	/* 1 */
 #define NFS4_MOUNT_NOCTO	0x0010	/* 1 */
 #define NFS4_MOUNT_NOAC		0x0020	/* 1 */
 #define NFS4_MOUNT_STRICTLOCK	0x1000	/* 1 */
+#define NFS4_MOUNT_FSCACHE	0x4000	/* 1 */
 #define NFS4_MOUNT_FLAGMASK	0xFFFF
 
 #endif
include/linux/nfs_fs.h +5 -0
index b7b4371..71cf935 100644
@@ -29,6 +29,7 @@ #include <linux/nfs_fs_sb.h>
 
 #include <linux/rwsem.h>
 #include <linux/mempool.h>
+#include <linux/fscache.h>
 
 /*
  * Enable debugging support for nfs client.
@@ -180,6 +181,9 @@ #ifdef CONFIG_NFS_V4
 	int			 delegation_state;
 	struct rw_semaphore	rwsem;
 #endif /* CONFIG_NFS_V4*/
+#ifdef CONFIG_NFS_FSCACHE
+	struct fscache_cookie	*fscache;
+#endif
 	struct inode		vfs_inode;
 };
 
@@ -582,6 +586,7 @@ #define NFSDBG_FILE		0x0040
 #define NFSDBG_ROOT		0x0080
 #define NFSDBG_CALLBACK		0x0100
 #define NFSDBG_CLIENT		0x0200
+#define NFSDBG_FSCACHE		0x0400
 #define NFSDBG_ALL		0xFFFF
 
 #ifdef __KERNEL__
include/linux/nfs_fs_sb.h +5 -0
index 62b7ca9..d0c926f 100644
@@ -3,6 +3,7 @@ #define _NFS_FS_SB
 
 #include <linux/list.h>
 #include <linux/backing-dev.h>
+#include <linux/fscache.h>
 
 struct nfs_iostats;
 
@@ -67,6 +68,10 @@ #ifdef CONFIG_NFS_V4
 	char			cl_ipaddr[16];
 	unsigned char		cl_id_uniquifier;
 #endif
+
+#ifdef CONFIG_NFS_FSCACHE
+	struct fscache_cookie	*fscache;	/* client index cache cookie */
+#endif
 };
 
 /*
include/linux/nfs_mount.h +1 -0
index 659c754..278bb4e 100644
@@ -61,6 +61,7 @@ #define NFS_MOUNT_BROKEN_SUID	0x0400	/* 
 #define NFS_MOUNT_NOACL		0x0800	/* 4 */
 #define NFS_MOUNT_STRICTLOCK	0x1000	/* reserved for NFSv4 */
 #define NFS_MOUNT_SECFLAVOUR	0x2000	/* 5 */
+#define NFS_MOUNT_FSCACHE	0x4000
 #define NFS_MOUNT_FLAGMASK	0xFFFF
 
 #endif
#24 lore raw

PATCH 20/21 AUTOFS: Make sure all dentries refs are released before calling kill_anon_super()

In reply to Trond Myklebust

From: David Howells <[email protected]>

Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true.

What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super().

This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine.

---

 fs/autofs4/autofs_i.h |    3 +--
 fs/autofs4/init.c     |    2 +-
 fs/autofs4/inode.c    |   22 ++++------------------
 fs/autofs4/waitq.c    |    1 -
 4 files changed, 6 insertions(+), 22 deletions(-)
fs/autofs4/autofs_i.h +1 -2
index d6603d0..47e38f3 100644
@@ -96,7 +96,6 @@ #define AUTOFS_TYPE_OFFSET       0x0004
 
 struct autofs_sb_info {
 	u32 magic;
-	struct dentry *root;
 	int pipefd;
 	struct file *pipe;
 	pid_t oz_pgrp;
@@ -231,4 +230,4 @@ out:
 }
 
 void autofs4_dentry_release(struct dentry *);
-
+extern void autofs4_kill_sb(struct super_block *);
fs/autofs4/init.c +1 -1
index 5d91933..723a1c5 100644
@@ -24,7 +24,7 @@ static struct file_system_type autofs_fs
 	.owner		= THIS_MODULE,
 	.name		= "autofs",
 	.get_sb		= autofs_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= autofs4_kill_sb,
 };
 
 static int __init init_autofs4_fs(void)
fs/autofs4/inode.c +4 -18
index fde78b1..1bf68c5 100644
@@ -95,7 +95,7 @@ void autofs4_free_ino(struct autofs_info
  */
 static void autofs4_force_release(struct autofs_sb_info *sbi)
 {
-	struct dentry *this_parent = sbi->root;
+	struct dentry *this_parent = sbi->sb->s_root;
 	struct list_head *next;
 
 	spin_lock(&dcache_lock);
@@ -126,7 +126,7 @@ resume:
 		spin_lock(&dcache_lock);
 	}
 
-	if (this_parent != sbi->root) {
+	if (this_parent != sbi->sb->s_root) {
 		struct dentry *dentry = this_parent;
 
 		next = this_parent->d_u.d_child.next;
@@ -139,15 +139,9 @@ resume:
 		goto resume;
 	}
 	spin_unlock(&dcache_lock);
-
-	dput(sbi->root);
-	sbi->root = NULL;
-	shrink_dcache_sb(sbi->sb);
-
-	return;
 }
 
-static void autofs4_put_super(struct super_block *sb)
+void autofs4_kill_sb(struct super_block *sb)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(sb);
 
@@ -162,6 +156,7 @@ static void autofs4_put_super(struct sup
 	kfree(sbi);
 
 	DPRINTK("shutting down");
+	kill_anon_super(sb);
 }
 
 static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt)
@@ -188,7 +183,6 @@ static int autofs4_show_options(struct s
 }
 
 static struct super_operations autofs4_sops = {
-	.put_super	= autofs4_put_super,
 	.statfs		= simple_statfs,
 	.show_options	= autofs4_show_options,
 };
@@ -314,7 +308,6 @@ int autofs4_fill_super(struct super_bloc
 
 	s->s_fs_info = sbi;
 	sbi->magic = AUTOFS_SBI_MAGIC;
-	sbi->root = NULL;
 	sbi->pipefd = -1;
 	sbi->catatonic = 0;
 	sbi->exp_timeout = 0;
@@ -396,13 +389,6 @@ int autofs4_fill_super(struct super_bloc
 	sbi->pipefd = pipefd;
 
 	/*
-	 * Take a reference to the root dentry so we get a chance to
-	 * clean up the dentry tree on umount.
-	 * See autofs4_force_release.
-	 */
-	sbi->root = dget(root);
-
-	/*
 	 * Success! Install the root dentry now to indicate completion.
 	 */
 	s->s_root = root;
fs/autofs4/waitq.c +0 -1
index ce103e7..c0a6c8d 100644
@@ -45,7 +45,6 @@ void autofs4_catatonic_mode(struct autof
 		fput(sbi->pipe);	/* Close the pipe */
 		sbi->pipe = NULL;
 	}
-	shrink_dcache_sb(sbi->sb);
 }
 
 static int autofs4_write(struct file *file, const void *addr, int bytes)
#25 lore raw

Re: [PATCH 20/21] AUTOFS: Make sure all dentries refs are released before calling kill_anon_super()

Quoted 6 lines · On Thu, 06 Jul 2006 11:11:07 -0400 Trond Myklebust…

From: David Howells <[email protected]>

Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true.

Quoted 8 lines

What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super().

This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine.

AFAIK, Ian is still off thinking about this one?

Quoted 127 lines
 fs/autofs4/autofs_i.h |    3 +--
 fs/autofs4/init.c     |    2 +-
 fs/autofs4/inode.c    |   22 ++++------------------
 fs/autofs4/waitq.c    |    1 -
 4 files changed, 6 insertions(+), 22 deletions(-)
fs/autofs4/autofs_i.h +1 -2
index d6603d0..47e38f3 100644
@@ -96,7 +96,6 @@ #define AUTOFS_TYPE_OFFSET       0x0004
 
 struct autofs_sb_info {
 	u32 magic;
-	struct dentry *root;
 	int pipefd;
 	struct file *pipe;
 	pid_t oz_pgrp;
@@ -231,4 +230,4 @@ out:
 }
 
 void autofs4_dentry_release(struct dentry *);
-
+extern void autofs4_kill_sb(struct super_block *);
fs/autofs4/init.c +1 -1
index 5d91933..723a1c5 100644
@@ -24,7 +24,7 @@ static struct file_system_type autofs_fs
 	.owner		= THIS_MODULE,
 	.name		= "autofs",
 	.get_sb		= autofs_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= autofs4_kill_sb,
 };
 
 static int __init init_autofs4_fs(void)
fs/autofs4/inode.c +4 -18
index fde78b1..1bf68c5 100644
@@ -95,7 +95,7 @@ void autofs4_free_ino(struct autofs_info
  */
 static void autofs4_force_release(struct autofs_sb_info *sbi)
 {
-	struct dentry *this_parent = sbi->root;
+	struct dentry *this_parent = sbi->sb->s_root;
 	struct list_head *next;
 
 	spin_lock(&dcache_lock);
@@ -126,7 +126,7 @@ resume:
 		spin_lock(&dcache_lock);
 	}
 
-	if (this_parent != sbi->root) {
+	if (this_parent != sbi->sb->s_root) {
 		struct dentry *dentry = this_parent;
 
 		next = this_parent->d_u.d_child.next;
@@ -139,15 +139,9 @@ resume:
 		goto resume;
 	}
 	spin_unlock(&dcache_lock);
-
-	dput(sbi->root);
-	sbi->root = NULL;
-	shrink_dcache_sb(sbi->sb);
-
-	return;
 }
 
-static void autofs4_put_super(struct super_block *sb)
+void autofs4_kill_sb(struct super_block *sb)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(sb);
 
@@ -162,6 +156,7 @@ static void autofs4_put_super(struct sup
 	kfree(sbi);
 
 	DPRINTK("shutting down");
+	kill_anon_super(sb);
 }
 
 static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt)
@@ -188,7 +183,6 @@ static int autofs4_show_options(struct s
 }
 
 static struct super_operations autofs4_sops = {
-	.put_super	= autofs4_put_super,
 	.statfs		= simple_statfs,
 	.show_options	= autofs4_show_options,
 };
@@ -314,7 +308,6 @@ int autofs4_fill_super(struct super_bloc
 
 	s->s_fs_info = sbi;
 	sbi->magic = AUTOFS_SBI_MAGIC;
-	sbi->root = NULL;
 	sbi->pipefd = -1;
 	sbi->catatonic = 0;
 	sbi->exp_timeout = 0;
@@ -396,13 +389,6 @@ int autofs4_fill_super(struct super_bloc
 	sbi->pipefd = pipefd;
 
 	/*
-	 * Take a reference to the root dentry so we get a chance to
-	 * clean up the dentry tree on umount.
-	 * See autofs4_force_release.
-	 */
-	sbi->root = dget(root);
-
-	/*
 	 * Success! Install the root dentry now to indicate completion.
 	 */
 	s->s_root = root;
fs/autofs4/waitq.c +0 -1
index ce103e7..c0a6c8d 100644
@@ -45,7 +45,6 @@ void autofs4_catatonic_mode(struct autof
 		fput(sbi->pipe);	/* Close the pipe */
 		sbi->pipe = NULL;
 	}
-	shrink_dcache_sb(sbi->sb);
 }
 
 static int autofs4_write(struct file *file, const void *addr, int bytes)
#26 lore raw

Re: [PATCH 20/21] AUTOFS: Make sure all dentries refs are released before calling kill_anon_super()

Quoted 12 lines · Andrew Morton
Quoted 9 lines · On Thu, 06 Jul 2006 11:11:07 -0400 Trond Myklebust…

From: David Howells <[email protected]>

Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true.

What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super().

Quoted 7 lines
Quoted 5 lines

This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine.

AFAIK, Ian is still off thinking about this one?

The patch looks fine to me but I wanted to give it a basic sanity test before ACKing it. I haven't got to that yet partly because I wasn't sure if the two patches from David would stand alone. Will they or will I need to apply Davids whole patch set?

Anyway, I'll have a go at it this afternoon.

Quoted 128 lines
Quoted 128 lines

 fs/autofs4/autofs_i.h |    3 +--
 fs/autofs4/init.c     |    2 +-
 fs/autofs4/inode.c    |   22 ++++------------------
 fs/autofs4/waitq.c    |    1 -
 4 files changed, 6 insertions(+), 22 deletions(-)
fs/autofs4/autofs_i.h +1 -2
index d6603d0..47e38f3 100644
@@ -96,7 +96,6 @@ #define AUTOFS_TYPE_OFFSET       0x0004
 
 struct autofs_sb_info {
 	u32 magic;
-	struct dentry *root;
 	int pipefd;
 	struct file *pipe;
 	pid_t oz_pgrp;
@@ -231,4 +230,4 @@ out:
 }
 
 void autofs4_dentry_release(struct dentry *);
-
+extern void autofs4_kill_sb(struct super_block *);
fs/autofs4/init.c +1 -1
index 5d91933..723a1c5 100644
@@ -24,7 +24,7 @@ static struct file_system_type autofs_fs
 	.owner		= THIS_MODULE,
 	.name		= "autofs",
 	.get_sb		= autofs_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= autofs4_kill_sb,
 };
 
 static int __init init_autofs4_fs(void)
fs/autofs4/inode.c +4 -18
index fde78b1..1bf68c5 100644
@@ -95,7 +95,7 @@ void autofs4_free_ino(struct autofs_info
  */
 static void autofs4_force_release(struct autofs_sb_info *sbi)
 {
-	struct dentry *this_parent = sbi->root;
+	struct dentry *this_parent = sbi->sb->s_root;
 	struct list_head *next;
 
 	spin_lock(&dcache_lock);
@@ -126,7 +126,7 @@ resume:
 		spin_lock(&dcache_lock);
 	}
 
-	if (this_parent != sbi->root) {
+	if (this_parent != sbi->sb->s_root) {
 		struct dentry *dentry = this_parent;
 
 		next = this_parent->d_u.d_child.next;
@@ -139,15 +139,9 @@ resume:
 		goto resume;
 	}
 	spin_unlock(&dcache_lock);
-
-	dput(sbi->root);
-	sbi->root = NULL;
-	shrink_dcache_sb(sbi->sb);
-
-	return;
 }
 
-static void autofs4_put_super(struct super_block *sb)
+void autofs4_kill_sb(struct super_block *sb)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(sb);
 
@@ -162,6 +156,7 @@ static void autofs4_put_super(struct sup
 	kfree(sbi);
 
 	DPRINTK("shutting down");
+	kill_anon_super(sb);
 }
 
 static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt)
@@ -188,7 +183,6 @@ static int autofs4_show_options(struct s
 }
 
 static struct super_operations autofs4_sops = {
-	.put_super	= autofs4_put_super,
 	.statfs		= simple_statfs,
 	.show_options	= autofs4_show_options,
 };
@@ -314,7 +308,6 @@ int autofs4_fill_super(struct super_bloc
 
 	s->s_fs_info = sbi;
 	sbi->magic = AUTOFS_SBI_MAGIC;
-	sbi->root = NULL;
 	sbi->pipefd = -1;
 	sbi->catatonic = 0;
 	sbi->exp_timeout = 0;
@@ -396,13 +389,6 @@ int autofs4_fill_super(struct super_bloc
 	sbi->pipefd = pipefd;
 
 	/*
-	 * Take a reference to the root dentry so we get a chance to
-	 * clean up the dentry tree on umount.
-	 * See autofs4_force_release.
-	 */
-	sbi->root = dget(root);
-
-	/*
 	 * Success! Install the root dentry now to indicate completion.
 	 */
 	s->s_root = root;
fs/autofs4/waitq.c +0 -1
index ce103e7..c0a6c8d 100644
@@ -45,7 +45,6 @@ void autofs4_catatonic_mode(struct autof
 		fput(sbi->pipe);	/* Close the pipe */
 		sbi->pipe = NULL;
 	}
-	shrink_dcache_sb(sbi->sb);
 }
 
 static int autofs4_write(struct file *file, const void *addr, int bytes)
#27 lore raw

Re: [PATCH 20/21] AUTOFS: Make sure all dentries refs are released before calling kill_anon_super()

Quoted 19 lines · On Fri, 07 Jul 2006 11:15:55 +0800 Ian Kent <[email protected]> wrote:
Quoted 18 lines · Andrew Morton
Quoted 14 lines · On Thu, 06 Jul 2006 11:11:07 -0400 Trond Myklebust…

From: David Howells <[email protected]>

Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true.

What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super().

This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine.

Quoted 6 lines
Quoted 1 line

AFAIK, Ian is still off thinking about this one?

The patch looks fine to me but I wanted to give it a basic sanity test before ACKing it. I haven't got to that yet partly because I wasn't sure if the two patches from David would stand alone. Will they or will I need to apply Davids whole patch set?

I expect this patch is standalone, but it's not really useful or interesting to test it unless [21/21] is also applied, because the two patches will tend to stick together. Plus this one is supposed to fix a umount/shutdown-time BUG in the other.

Quoted 1 line

Anyway, I'll have a go at it this afternoon.

Thanks.

#28 lore raw

Re: [PATCH 20/21] AUTOFS: Make sure all dentries refs are released before calling kill_anon_super()

Quoted 29 lines · Andrew Morton
Quoted 19 lines · On Fri, 07 Jul 2006 11:15:55 +0800 Ian Kent <[email protected]> wrote:
Quoted 18 lines · Andrew Morton
Quoted 14 lines · On Thu, 06 Jul 2006 11:11:07 -0400 Trond Myklebust…

From: David Howells <[email protected]>

Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true.

What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super().

This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine.

Quoted 6 lines
Quoted 1 line

AFAIK, Ian is still off thinking about this one?

The patch looks fine to me but I wanted to give it a basic sanity test before ACKing it. I haven't got to that yet partly because I wasn't sure if the two patches from David would stand alone. Will they or will I need to apply Davids whole patch set?

Quoted 4 lines

I expect this patch is standalone, but it's not really useful or interesting to test it unless [21/21] is also applied, because the two patches will tend to stick together. Plus this one is supposed to fix a umount/shutdown-time BUG in the other.

OK. I'm with it now.

David copied me on two patches, the other was:
[PATCH 1/2] VFS: Destroy the dentries contributed by a superblock on unmounting

I see that these correspond to patches 20 and 21 of Tronds 2.6.18-rc1 submission.

I was expecting that both would be required at least, so I'll try with these two before I try anything else.

Thanks.
Ian

#29 lore raw

PATCH 21/21 VFS: Destroy the dentries contributed by a superblock on unmounting

In reply to Trond Myklebust

From: David Howells <[email protected]>

The attached patch destroys all the dentries attached to a superblock in one go by:

(1) Destroying the tree rooted at s_root.

(2) Destroying every entry in the anon list, one at a time.

(3) Each entry in the anon list has its subtree consumed from the leaves inwards.

This reduces the amount of work generic_shutdown_super() does, and avoids iterating through the dentry_unused list.

Note that locking is almost entirely absent in the shrink_dcache_for_umount*() functions added by this patch. This is because:

(1) at the point the filesystem calls generic_shutdown_super(), it is not permitted to further touch the superblock's set of dentries, and nor may it remove aliases from inodes;

(2) the dcache memory shrinker now skips dentries that are being unmounted; and

(3) the superblock no longer has any external references through which the VFS can reach it.

Given these points, the only locking we need to do is when we remove dentries from the unused list and the name hashes, which we do a directory's worth at a time.

We also don't need to guard against reference counts going to zero unexpectedly and removing bits of the tree we're working on as nothing else can call dput().

A cut down version of dentry_iput() has been folded into shrink_dcache_for_umount_subtree() function. Apart from not needing to unlock things, it also doesn't need to check for inotify watches.

In this version of the patch, the complaint about a dentry still being in use has been expanded from a single BUG_ON() and now gives much more information.

---

 fs/dcache.c            |  133 ++++++++++++++++++++++++++++++++++++++++++++++++
 fs/super.c             |   12 ++--
 include/linux/dcache.h |    1 
 3 files changed, 140 insertions(+), 6 deletions(-)
fs/dcache.c +133 -0
index 17b392a..780f014 100644
@@ -547,6 +547,139 @@ repeat:
 }
 
 /*
+ * destroy a single subtree of dentries for unmount
+ * - see the comments on shrink_dcache_for_umount() for a description of the
+ *   locking
+ */
+static void shrink_dcache_for_umount_subtree(struct dentry *dentry)
+{
+	struct dentry *parent;
+
+	BUG_ON(!IS_ROOT(dentry));
+
+	/* detach this root from the system */
+	spin_lock(&dcache_lock);
+	if (!list_empty(&dentry->d_lru)) {
+		dentry_stat.nr_unused--;
+		list_del_init(&dentry->d_lru);
+	}
+	__d_drop(dentry);
+	spin_unlock(&dcache_lock);
+
+	for (;;) {
+		/* descend to the first leaf in the current subtree */
+		while (!list_empty(&dentry->d_subdirs)) {
+			struct dentry *loop;
+
+			/* this is a branch with children - detach all of them
+			 * from the system in one go */
+			spin_lock(&dcache_lock);
+			list_for_each_entry(loop, &dentry->d_subdirs,
+					    d_u.d_child) {
+				if (!list_empty(&loop->d_lru)) {
+					dentry_stat.nr_unused--;
+					list_del_init(&loop->d_lru);
+				}
+
+				__d_drop(loop);
+				cond_resched_lock(&dcache_lock);
+			}
+			spin_unlock(&dcache_lock);
+
+			/* move to the first child */
+			dentry = list_entry(dentry->d_subdirs.next,
+					    struct dentry, d_u.d_child);
+		}
+
+		/* consume the dentries from this leaf up through its parents
+		 * until we find one with children or run out altogether */
+		do {
+			struct inode *inode;
+
+			if (atomic_read(&dentry->d_count) != 0) {
+				printk(KERN_ERR
+				       "BUG: Dentry %p{i=%lx,n=%s}"
+				       " still in use (%d)"
+				       " [unmount of %s %s]\n",
+				       dentry,
+				       dentry->d_inode ?
+				       dentry->d_inode->i_ino : 0UL,
+				       dentry->d_name.name,
+				       atomic_read(&dentry->d_count),
+				       dentry->d_sb->s_type->name,
+				       dentry->d_sb->s_id);
+				BUG();
+			}
+
+			parent = dentry->d_parent;
+			if (parent == dentry)
+				parent = NULL;
+			else
+				atomic_dec(&parent->d_count);
+
+			list_del(&dentry->d_u.d_child);
+			dentry_stat.nr_dentry--;	/* For d_free, below */
+
+			inode = dentry->d_inode;
+			if (inode) {
+#ifdef CONFIG_INOTIFY
+				BUG_ON(!list_empty(&inode->inotify_watches));
+#endif
+				dentry->d_inode = NULL;
+				list_del_init(&dentry->d_alias);
+				if (dentry->d_op && dentry->d_op->d_iput)
+					dentry->d_op->d_iput(dentry, inode);
+				else
+					iput(inode);
+			}
+
+			d_free(dentry);
+
+			/* finished when we fall off the top of the tree,
+			 * otherwise we ascend to the parent and move to the
+			 * next sibling if there is one */
+			if (!parent)
+				return;
+
+			dentry = parent;
+
+		} while (list_empty(&dentry->d_subdirs));
+
+		dentry = list_entry(dentry->d_subdirs.next,
+				    struct dentry, d_u.d_child);
+	}
+}
+
+/*
+ * destroy the dentries attached to a superblock on unmounting
+ * - we don't need to use dentry->d_lock, and only need dcache_lock when
+ *   removing the dentry from the system lists and hashes because:
+ *   - the superblock is detached from all mountings and open files, so the
+ *     dentry trees will not be rearranged by the VFS
+ *   - s_umount is write-locked, so the memory pressure shrinker will ignore
+ *     any dentries belonging to this superblock that it comes across
+ *   - the filesystem itself is no longer permitted to rearrange the dentries
+ *     in this superblock
+ */
+void shrink_dcache_for_umount(struct super_block *sb)
+{
+	struct dentry *dentry;
+
+	if (down_read_trylock(&sb->s_umount))
+		BUG();
+
+	dentry = sb->s_root;
+	sb->s_root = NULL;
+	atomic_dec(&dentry->d_count);
+	shrink_dcache_for_umount_subtree(dentry);
+
+	while (!hlist_empty(&sb->s_anon)) {
+		dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash);
+		shrink_dcache_for_umount_subtree(dentry);
+	}
+}
+
+/*
  * Search for at least 1 mount point in the dentry's subdirs.
  * We descend to the next level whenever the d_subdirs
  * list is non-empty and continue searching.
fs/super.c +6 -6
index 6d4e817..3bf8e5f 100644
@@ -228,17 +228,17 @@ static int grab_super(struct super_block
  *	that need destruction out of superblock, call generic_shutdown_super()
  *	and release aforementioned objects.  Note: dentries and inodes _are_
  *	taken care of and do not need specific handling.
+ *
+ *	Upon calling this function, the filesystem may no longer alter or
+ *	rearrange the set of dentries belonging to this super_block, nor may it
+ *	change the attachments of dentries to inodes.
  */
 void generic_shutdown_super(struct super_block *sb)
 {
-	struct dentry *root = sb->s_root;
 	struct super_operations *sop = sb->s_op;
 
-	if (root) {
-		sb->s_root = NULL;
-		shrink_dcache_parent(root);
-		shrink_dcache_sb(sb);
-		dput(root);
+	if (sb->s_root) {
+		shrink_dcache_for_umount(sb);
 		fsync_super(sb);
 		lock_super(sb);
 		sb->s_flags &= ~MS_ACTIVE;
include/linux/dcache.h +1 -0
index 44605be..63f64a9 100644
@@ -230,6 +230,7 @@ extern struct dentry * d_alloc_anon(stru
 extern struct dentry * d_splice_alias(struct inode *, struct dentry *);
 extern void shrink_dcache_sb(struct super_block *);
 extern void shrink_dcache_parent(struct dentry *);
+extern void shrink_dcache_for_umount(struct super_block *);
 extern int d_invalidate(struct dentry *);
 
 /* only used at mount-time */