From 44396f4b5cb8566f7118aec55eeac99be7ad94cb Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Tue, 31 May 2011 11:58:49 -0400
Subject: fs: add a DCACHE_NEED_LOOKUP flag for d_flags

Btrfs (and I'd venture most other fs's) stores its indexes in nice disk order
for readdir, but unfortunately in the case of anything that stats the files in
order that readdir spits back (like oh say ls) that means we still have to do
the normal lookup of the file, which means looking up our other index and then
looking up the inode.  What I want is a way to create dummy dentries when we
find them in readdir so that when ls or anything else subsequently does a
stat(), we already have the location information in the dentry and can go
straight to the inode itself.  The lookup stuff just assumes that if it finds a
dentry it is done, it doesn't perform a lookup.  So add a DCACHE_NEED_LOOKUP
flag so that the lookup code knows it still needs to run i_op->lookup() on the
parent to get the inode for the dentry.  I have tested this with btrfs and I
went from something that looks like this

http://people.redhat.com/jwhiter/ls-noreada.png

To this

http://people.redhat.com/jwhiter/ls-good.png

Thats a savings of 1300 seconds, or 22 minutes.  That is a significant savings.
Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/dcache.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include')

diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 19d90a55541d..5fa5bd33b979 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -216,6 +216,7 @@ struct dentry_operations {
 #define DCACHE_MOUNTED		0x10000	/* is a mountpoint */
 #define DCACHE_NEED_AUTOMOUNT	0x20000	/* handle automount on this dir */
 #define DCACHE_MANAGE_TRANSIT	0x40000	/* manage transit from this dirent */
+#define DCACHE_NEED_LOOKUP	0x80000 /* dentry requires i_op->lookup */
 #define DCACHE_MANAGED_DENTRY \
 	(DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT)
 
@@ -416,6 +417,12 @@ static inline bool d_mountpoint(struct dentry *dentry)
 	return dentry->d_flags & DCACHE_MOUNTED;
 }
 
+static inline bool d_need_lookup(struct dentry *dentry)
+{
+	return dentry->d_flags & DCACHE_NEED_LOOKUP;
+}
+
+extern void d_clear_need_lookup(struct dentry *dentry);
 extern struct dentry *lookup_create(struct nameidata *nd, int is_dir);
 
 extern int sysctl_vfs_cache_pressure;
-- 
cgit v1.2.3


From 43e15cdbefea4ce6d68113de98d4f61c0cf45687 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 3 Jun 2011 20:16:57 -0400
Subject: new helper: iterate_supers_type()

Call the given function for all superblocks of given type.  Function
gets a superblock (with s_umount locked shared) and (void *) argument
supplied by caller of iterator.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/super.c         | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/fs.h |  2 ++
 2 files changed, 38 insertions(+)

(limited to 'include')

diff --git a/fs/super.c b/fs/super.c
index ab3d672db0de..444da9579068 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -451,6 +451,42 @@ void iterate_supers(void (*f)(struct super_block *, void *), void *arg)
 	spin_unlock(&sb_lock);
 }
 
+/**
+ *	iterate_supers_type - call function for superblocks of given type
+ *	@type: fs type
+ *	@f: function to call
+ *	@arg: argument to pass to it
+ *
+ *	Scans the superblock list and calls given function, passing it
+ *	locked superblock and given argument.
+ */
+void iterate_supers_type(struct file_system_type *type,
+	void (*f)(struct super_block *, void *), void *arg)
+{
+	struct super_block *sb, *p = NULL;
+
+	spin_lock(&sb_lock);
+	list_for_each_entry(sb, &type->fs_supers, s_instances) {
+		sb->s_count++;
+		spin_unlock(&sb_lock);
+
+		down_read(&sb->s_umount);
+		if (sb->s_root)
+			f(sb, arg);
+		up_read(&sb->s_umount);
+
+		spin_lock(&sb_lock);
+		if (p)
+			__put_super(p);
+		p = sb;
+	}
+	if (p)
+		__put_super(p);
+	spin_unlock(&sb_lock);
+}
+
+EXPORT_SYMBOL(iterate_supers_type);
+
 /**
  *	get_super - get the superblock of a device
  *	@bdev: device to get the superblock for
diff --git a/include/linux/fs.h b/include/linux/fs.h
index b5b979247863..a8735e7e1b35 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2432,6 +2432,8 @@ extern struct super_block *get_active_super(struct block_device *bdev);
 extern struct super_block *user_get_super(dev_t);
 extern void drop_super(struct super_block *sb);
 extern void iterate_supers(void (*)(struct super_block *, void *), void *);
+extern void iterate_supers_type(struct file_system_type *,
+			        void (*)(struct super_block *, void *), void *);
 
 extern int dcache_dir_open(struct inode *, struct file *);
 extern int dcache_dir_close(struct inode *, struct file *);
-- 
cgit v1.2.3


From 1b5d783c94c328d406e801566f161adcfb018dda Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 19 Jun 2011 12:49:47 -0400
Subject: consolidate BINPRM_FLAGS_ENFORCE_NONDUMP handling

new helper: would_dump(bprm, file).  Checks if we are allowed to
read the file and if we are not - sets ENFORCE_NODUMP.  Exported,
used in places that previously open-coded the same logics.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/binfmt_elf.c         |  3 +--
 fs/binfmt_elf_fdpic.c   |  3 +--
 fs/binfmt_misc.c        |  3 +--
 fs/exec.c               | 14 +++++++++++---
 include/linux/binfmts.h |  1 +
 5 files changed, 15 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 303983fabfd6..dd0fdfc56d38 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -668,8 +668,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 			 * mm->dumpable = 0 regardless of the interpreter's
 			 * permissions.
 			 */
-			if (file_permission(interpreter, MAY_READ) < 0)
-				bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
+			would_dump(bprm, interpreter);
 
 			retval = kernel_read(interpreter, 0, bprm->buf,
 					     BINPRM_BUF_SIZE);
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 2bc5dc644b4c..30745f459faf 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -245,8 +245,7 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm,
 			 * mm->dumpable = 0 regardless of the interpreter's
 			 * permissions.
 			 */
-			if (file_permission(interpreter, MAY_READ) < 0)
-				bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
+			would_dump(bprm, interpreter);
 
 			retval = kernel_read(interpreter, 0, bprm->buf,
 					     BINPRM_BUF_SIZE);
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 1befe2ec8186..ba1a1ae4a18a 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -149,8 +149,7 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 
 		/* if the binary is not readable than enforce mm->dumpable=0
 		   regardless of the interpreter's permissions */
-		if (file_permission(bprm->file, MAY_READ))
-			bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
+		would_dump(bprm, bprm->file);
 
 		allow_write_access(bprm->file);
 		bprm->file = NULL;
diff --git a/fs/exec.c b/fs/exec.c
index 6075a1e727ae..f9f12ad299af 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1105,6 +1105,13 @@ out:
 }
 EXPORT_SYMBOL(flush_old_exec);
 
+void would_dump(struct linux_binprm *bprm, struct file *file)
+{
+	if (inode_permission(file->f_path.dentry->d_inode, MAY_READ) < 0)
+		bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
+}
+EXPORT_SYMBOL(would_dump);
+
 void setup_new_exec(struct linux_binprm * bprm)
 {
 	int i, ch;
@@ -1144,9 +1151,10 @@ void setup_new_exec(struct linux_binprm * bprm)
 	if (bprm->cred->uid != current_euid() ||
 	    bprm->cred->gid != current_egid()) {
 		current->pdeath_signal = 0;
-	} else if (file_permission(bprm->file, MAY_READ) ||
-		   bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP) {
-		set_dumpable(current->mm, suid_dumpable);
+	} else {
+		would_dump(bprm, bprm->file);
+		if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)
+			set_dumpable(current->mm, suid_dumpable);
 	}
 
 	/*
diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index 8845613fd7e3..fd88a3945aa1 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -111,6 +111,7 @@ extern int __must_check remove_arg_zero(struct linux_binprm *);
 extern int search_binary_handler(struct linux_binprm *, struct pt_regs *);
 extern int flush_old_exec(struct linux_binprm * bprm);
 extern void setup_new_exec(struct linux_binprm * bprm);
+extern void would_dump(struct linux_binprm *, struct file *);
 
 extern int suid_dumpable;
 #define SUID_DUMP_DISABLE	0	/* No setuid dumping */
-- 
cgit v1.2.3


From 3bfa784a6539f91a27d7ffdd408efdb638e3bebd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 19 Jun 2011 12:55:10 -0400
Subject: kill file_permission() completely

convert the last remaining caller to inode_permission()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c         | 18 ------------------
 include/linux/fs.h |  1 -
 kernel/cgroup.c    |  3 ++-
 3 files changed, 2 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index cf2554635a1c..4ad2b781a65c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -303,23 +303,6 @@ int inode_permission(struct inode *inode, int mask)
 	return security_inode_permission(inode, mask);
 }
 
-/**
- * file_permission  -  check for additional access rights to a given file
- * @file:	file to check access rights for
- * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
- *
- * Used to check for read/write/execute permissions on an already opened
- * file.
- *
- * Note:
- *	Do not use this function in new code.  All access checks should
- *	be done using inode_permission().
- */
-int file_permission(struct file *file, int mask)
-{
-	return inode_permission(file->f_path.dentry->d_inode, mask);
-}
-
 /*
  * get_write_access() gets write permission for a file.
  * put_write_access() releases this write permission.
@@ -3405,7 +3388,6 @@ EXPORT_SYMBOL(kern_path_parent);
 EXPORT_SYMBOL(kern_path);
 EXPORT_SYMBOL(vfs_path_lookup);
 EXPORT_SYMBOL(inode_permission);
-EXPORT_SYMBOL(file_permission);
 EXPORT_SYMBOL(unlock_rename);
 EXPORT_SYMBOL(vfs_create);
 EXPORT_SYMBOL(vfs_follow_link);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index a8735e7e1b35..d04e55586a17 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1490,7 +1490,6 @@ extern void dentry_unhash(struct dentry *dentry);
 /*
  * VFS file helper functions.
  */
-extern int file_permission(struct file *, int);
 extern void inode_init_owner(struct inode *inode, const struct inode *dir,
 			mode_t mode);
 /*
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 2731d115d725..e1c72c0f512b 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -3542,7 +3542,8 @@ static int cgroup_write_event_control(struct cgroup *cgrp, struct cftype *cft,
 	}
 
 	/* the process need read permission on control file */
-	ret = file_permission(cfile, MAY_READ);
+	/* AV: shouldn't we check that it's been opened for read instead? */
+	ret = inode_permission(cfile->f_path.dentry->d_inode, MAY_READ);
 	if (ret < 0)
 		goto fail;
 
-- 
cgit v1.2.3


From 07b8ce1ee87d291ff564c02cf878fae973317a52 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 10:52:57 -0400
Subject: lockless get_write_access/deny_write_access

new helpers: atomic_inc_unless_negative()/atomic_dec_unless_positive()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c             | 46 ----------------------------------------------
 include/linux/atomic.h | 26 ++++++++++++++++++++++++++
 include/linux/fs.h     | 29 ++++++++++++++++++++++++++---
 3 files changed, 52 insertions(+), 49 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index e04f15ae4b07..d286cbc3f3e5 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -341,52 +341,6 @@ ok:
 	return security_inode_exec_permission(inode, flags);
 }
 
-/*
- * get_write_access() gets write permission for a file.
- * put_write_access() releases this write permission.
- * This is used for regular files.
- * We cannot support write (and maybe mmap read-write shared) accesses and
- * MAP_DENYWRITE mmappings simultaneously. The i_writecount field of an inode
- * can have the following values:
- * 0: no writers, no VM_DENYWRITE mappings
- * < 0: (-i_writecount) vm_area_structs with VM_DENYWRITE set exist
- * > 0: (i_writecount) users are writing to the file.
- *
- * Normally we operate on that counter with atomic_{inc,dec} and it's safe
- * except for the cases where we don't hold i_writecount yet. Then we need to
- * use {get,deny}_write_access() - these functions check the sign and refuse
- * to do the change if sign is wrong. Exclusion between them is provided by
- * the inode->i_lock spinlock.
- */
-
-int get_write_access(struct inode * inode)
-{
-	spin_lock(&inode->i_lock);
-	if (atomic_read(&inode->i_writecount) < 0) {
-		spin_unlock(&inode->i_lock);
-		return -ETXTBSY;
-	}
-	atomic_inc(&inode->i_writecount);
-	spin_unlock(&inode->i_lock);
-
-	return 0;
-}
-
-int deny_write_access(struct file * file)
-{
-	struct inode *inode = file->f_path.dentry->d_inode;
-
-	spin_lock(&inode->i_lock);
-	if (atomic_read(&inode->i_writecount) > 0) {
-		spin_unlock(&inode->i_lock);
-		return -ETXTBSY;
-	}
-	atomic_dec(&inode->i_writecount);
-	spin_unlock(&inode->i_lock);
-
-	return 0;
-}
-
 /**
  * path_get - get a reference to a path
  * @path: path to get the reference to
diff --git a/include/linux/atomic.h b/include/linux/atomic.h
index ee456c79b0e6..bc6615d4132b 100644
--- a/include/linux/atomic.h
+++ b/include/linux/atomic.h
@@ -34,6 +34,32 @@ static inline int atomic_inc_not_zero_hint(atomic_t *v, int hint)
 }
 #endif
 
+#ifndef atomic_inc_unless_negative
+static inline int atomic_inc_unless_negative(atomic_t *p)
+{
+	int v, v1;
+	for (v = 0; v >= 0; v = v1) {
+		v1 = atomic_cmpxchg(p, v, v + 1);
+		if (likely(v1 == v))
+			return 1;
+	}
+	return 0;
+}
+#endif
+
+#ifndef atomic_dec_unless_positive
+static inline int atomic_dec_unless_positive(atomic_t *p)
+{
+	int v, v1;
+	for (v = 0; v <= 0; v = v1) {
+		v1 = atomic_cmpxchg(p, v, v - 1);
+		if (likely(v1 == v))
+			return 1;
+	}
+	return 0;
+}
+#endif
+
 #ifndef CONFIG_ARCH_HAS_ATOMIC_OR
 static inline void atomic_or(int i, atomic_t *v)
 {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d04e55586a17..8c84ed930389 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -392,8 +392,8 @@ struct inodes_stat_t {
 #include <linux/semaphore.h>
 #include <linux/fiemap.h>
 #include <linux/rculist_bl.h>
+#include <linux/atomic.h>
 
-#include <asm/atomic.h>
 #include <asm/byteorder.h>
 
 struct export_operations;
@@ -2195,8 +2195,31 @@ static inline bool execute_ok(struct inode *inode)
 	return (inode->i_mode & S_IXUGO) || S_ISDIR(inode->i_mode);
 }
 
-extern int get_write_access(struct inode *);
-extern int deny_write_access(struct file *);
+/*
+ * get_write_access() gets write permission for a file.
+ * put_write_access() releases this write permission.
+ * This is used for regular files.
+ * We cannot support write (and maybe mmap read-write shared) accesses and
+ * MAP_DENYWRITE mmappings simultaneously. The i_writecount field of an inode
+ * can have the following values:
+ * 0: no writers, no VM_DENYWRITE mappings
+ * < 0: (-i_writecount) vm_area_structs with VM_DENYWRITE set exist
+ * > 0: (i_writecount) users are writing to the file.
+ *
+ * Normally we operate on that counter with atomic_{inc,dec} and it's safe
+ * except for the cases where we don't hold i_writecount yet. Then we need to
+ * use {get,deny}_write_access() - these functions check the sign and refuse
+ * to do the change if sign is wrong.
+ */
+static inline int get_write_access(struct inode *inode)
+{
+	return atomic_inc_unless_negative(&inode->i_writecount) ? 0 : -ETXTBSY;
+}
+static inline int deny_write_access(struct file *file)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	return atomic_dec_unless_positive(&inode->i_writecount) ? 0 : -ETXTBSY;
+}
 static inline void put_write_access(struct inode * inode)
 {
 	atomic_dec(&inode->i_writecount);
-- 
cgit v1.2.3


From 178ea73521d64ba41d7aa5488fb9f549c6d4507d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 11:31:30 -0400
Subject: kill check_acl callback of generic_permission()

its value depends only on inode and does not change; we might as
well store it in ->i_op->check_acl and be done with that.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/afs/security.c              |  2 +-
 fs/btrfs/inode.c               |  7 ++++++-
 fs/ceph/inode.c                |  2 +-
 fs/cifs/cifsfs.c               |  2 +-
 fs/fuse/dir.c                  |  4 ++--
 fs/gfs2/inode.c                |  5 ++++-
 fs/hostfs/hostfs_kern.c        |  2 +-
 fs/hpfs/namei.c                |  2 +-
 fs/namei.c                     | 17 +++++++----------
 fs/nfs/dir.c                   |  2 +-
 fs/nilfs2/inode.c              |  2 +-
 fs/ocfs2/file.c                |  4 +++-
 fs/ocfs2/namei.c               |  1 +
 fs/proc/base.c                 |  2 +-
 fs/reiserfs/file.c             |  1 +
 fs/reiserfs/namei.c            |  4 +++-
 fs/reiserfs/xattr.c            | 18 ++++++++----------
 fs/sysfs/inode.c               |  2 +-
 include/linux/fs.h             |  3 +--
 include/linux/reiserfs_xattr.h |  2 ++
 20 files changed, 47 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/fs/afs/security.c b/fs/afs/security.c
index f44b9d355377..745ee654165f 100644
--- a/fs/afs/security.c
+++ b/fs/afs/security.c
@@ -350,7 +350,7 @@ int afs_permission(struct inode *inode, int mask, unsigned int flags)
 	}
 
 	key_put(key);
-	ret = generic_permission(inode, mask, flags, NULL);
+	ret = generic_permission(inode, mask, flags);
 	_leave(" = %d", ret);
 	return ret;
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 3601f0aebddf..f0bd87371566 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7339,7 +7339,7 @@ static int btrfs_permission(struct inode *inode, int mask, unsigned int flags)
 		return -EROFS;
 	if ((BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) && (mask & MAY_WRITE))
 		return -EACCES;
-	return generic_permission(inode, mask, flags, btrfs_check_acl);
+	return generic_permission(inode, mask, flags);
 }
 
 static const struct inode_operations btrfs_dir_inode_operations = {
@@ -7359,10 +7359,12 @@ static const struct inode_operations btrfs_dir_inode_operations = {
 	.listxattr	= btrfs_listxattr,
 	.removexattr	= btrfs_removexattr,
 	.permission	= btrfs_permission,
+	.check_acl	= btrfs_check_acl,
 };
 static const struct inode_operations btrfs_dir_ro_inode_operations = {
 	.lookup		= btrfs_lookup,
 	.permission	= btrfs_permission,
+	.check_acl	= btrfs_check_acl,
 };
 
 static const struct file_operations btrfs_dir_file_operations = {
@@ -7431,6 +7433,7 @@ static const struct inode_operations btrfs_file_inode_operations = {
 	.removexattr	= btrfs_removexattr,
 	.permission	= btrfs_permission,
 	.fiemap		= btrfs_fiemap,
+	.check_acl	= btrfs_check_acl,
 };
 static const struct inode_operations btrfs_special_inode_operations = {
 	.getattr	= btrfs_getattr,
@@ -7440,6 +7443,7 @@ static const struct inode_operations btrfs_special_inode_operations = {
 	.getxattr	= btrfs_getxattr,
 	.listxattr	= btrfs_listxattr,
 	.removexattr	= btrfs_removexattr,
+	.check_acl	= btrfs_check_acl,
 };
 static const struct inode_operations btrfs_symlink_inode_operations = {
 	.readlink	= generic_readlink,
@@ -7451,6 +7455,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = {
 	.getxattr	= btrfs_getxattr,
 	.listxattr	= btrfs_listxattr,
 	.removexattr	= btrfs_removexattr,
+	.check_acl	= btrfs_check_acl,
 };
 
 const struct dentry_operations btrfs_dentry_operations = {
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index d8858e96ab18..beb5d55d6fd2 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1805,7 +1805,7 @@ int ceph_permission(struct inode *inode, int mask, unsigned int flags)
 	err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
 
 	if (!err)
-		err = generic_permission(inode, mask, flags, NULL);
+		err = generic_permission(inode, mask, flags);
 	return err;
 }
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index bc4b12ca537b..b79804fa410f 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -239,7 +239,7 @@ static int cifs_permission(struct inode *inode, int mask, unsigned int flags)
 		on the client (above and beyond ACL on servers) for
 		servers which do not support setting and viewing mode bits,
 		so allowing client to check permissions is useful */
-		return generic_permission(inode, mask, flags, NULL);
+		return generic_permission(inode, mask, flags);
 }
 
 static struct kmem_cache *cifs_inode_cachep;
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index d50160714595..0a2fcd860ad6 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -1018,7 +1018,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 	}
 
 	if (fc->flags & FUSE_DEFAULT_PERMISSIONS) {
-		err = generic_permission(inode, mask, flags, NULL);
+		err = generic_permission(inode, mask, flags);
 
 		/* If permission is denied, try to refresh file
 		   attributes.  This is also needed, because the root
@@ -1027,7 +1027,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 			err = fuse_perm_getattr(inode, flags);
 			if (!err)
 				err = generic_permission(inode, mask,
-							flags, NULL);
+							flags);
 		}
 
 		/* Note: the opposite of the above test does not
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 03e0c529063e..d5f0f4ea25dc 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -1564,7 +1564,7 @@ int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
 	if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode))
 		error = -EACCES;
 	else
-		error = generic_permission(inode, mask, flags, gfs2_check_acl);
+		error = generic_permission(inode, mask, flags);
 	if (unlock)
 		gfs2_glock_dq_uninit(&i_gh);
 
@@ -1854,6 +1854,7 @@ const struct inode_operations gfs2_file_iops = {
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
 	.fiemap = gfs2_fiemap,
+	.check_acl = gfs2_check_acl,
 };
 
 const struct inode_operations gfs2_dir_iops = {
@@ -1874,6 +1875,7 @@ const struct inode_operations gfs2_dir_iops = {
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
 	.fiemap = gfs2_fiemap,
+	.check_acl = gfs2_check_acl,
 };
 
 const struct inode_operations gfs2_symlink_iops = {
@@ -1888,5 +1890,6 @@ const struct inode_operations gfs2_symlink_iops = {
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
 	.fiemap = gfs2_fiemap,
+	.check_acl = gfs2_check_acl,
 };
 
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 2638c834ed28..a98d0d1aef65 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -770,7 +770,7 @@ int hostfs_permission(struct inode *ino, int desired, unsigned int flags)
 		err = access_file(name, r, w, x);
 	__putname(name);
 	if (!err)
-		err = generic_permission(ino, desired, flags, NULL);
+		err = generic_permission(ino, desired, flags);
 	return err;
 }
 
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index acf95dab2aac..bd2ce7dd8df3 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -398,7 +398,7 @@ again:
 			hpfs_unlock(dir->i_sb);
 			return -ENOSPC;
 		}
-		if (generic_permission(inode, MAY_WRITE, 0, NULL) ||
+		if (generic_permission(inode, MAY_WRITE, 0) ||
 		    !S_ISREG(inode->i_mode) ||
 		    get_write_access(inode)) {
 			d_rehash(dentry);
diff --git a/fs/namei.c b/fs/namei.c
index d286cbc3f3e5..c5af0f37e67d 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -176,9 +176,9 @@ EXPORT_SYMBOL(putname);
 /*
  * This does basic POSIX ACL permission checking
  */
-static int acl_permission_check(struct inode *inode, int mask, unsigned int flags,
-		int (*check_acl)(struct inode *inode, int mask, unsigned int flags))
+static int acl_permission_check(struct inode *inode, int mask, unsigned int flags)
 {
+	int (*check_acl)(struct inode *inode, int mask, unsigned int flags);
 	unsigned int mode = inode->i_mode;
 
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
@@ -189,6 +189,7 @@ static int acl_permission_check(struct inode *inode, int mask, unsigned int flag
 	if (current_fsuid() == inode->i_uid)
 		mode >>= 6;
 	else {
+		check_acl = inode->i_op->check_acl;
 		if (IS_POSIXACL(inode) && (mode & S_IRWXG) && check_acl) {
 			int error = check_acl(inode, mask, flags);
 			if (error != -EAGAIN)
@@ -212,7 +213,6 @@ other_perms:
  * generic_permission -  check for access rights on a Posix-like filesystem
  * @inode:	inode to check access rights for
  * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
- * @check_acl:	optional callback to check for Posix ACLs
  * @flags:	IPERM_FLAG_ flags.
  *
  * Used to check for read/write/execute permissions on a file.
@@ -224,15 +224,14 @@ other_perms:
  * request cannot be satisfied (eg. requires blocking or too much complexity).
  * It would then be called again in ref-walk mode.
  */
-int generic_permission(struct inode *inode, int mask, unsigned int flags,
-	int (*check_acl)(struct inode *inode, int mask, unsigned int flags))
+int generic_permission(struct inode *inode, int mask, unsigned int flags)
 {
 	int ret;
 
 	/*
 	 * Do the basic POSIX ACL permission checks.
 	 */
-	ret = acl_permission_check(inode, mask, flags, check_acl);
+	ret = acl_permission_check(inode, mask, flags);
 	if (ret != -EACCES)
 		return ret;
 
@@ -290,8 +289,7 @@ int inode_permission(struct inode *inode, int mask)
 	if (inode->i_op->permission)
 		retval = inode->i_op->permission(inode, mask, 0);
 	else
-		retval = generic_permission(inode, mask, 0,
-				inode->i_op->check_acl);
+		retval = generic_permission(inode, mask, 0);
 
 	if (retval)
 		return retval;
@@ -326,8 +324,7 @@ static inline int exec_permission(struct inode *inode, unsigned int flags)
 		if (likely(!ret))
 			goto ok;
 	} else {
-		ret = acl_permission_check(inode, MAY_EXEC, flags,
-				inode->i_op->check_acl);
+		ret = acl_permission_check(inode, MAY_EXEC, flags);
 		if (likely(!ret))
 			goto ok;
 		if (ret != -EACCES)
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index ededdbd0db38..0485dca34fb1 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -2328,7 +2328,7 @@ out:
 out_notsup:
 	res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
 	if (res == 0)
-		res = generic_permission(inode, mask, flags, NULL);
+		res = generic_permission(inode, mask, flags);
 	goto out;
 }
 
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index b9b45fc2903e..650aa7755003 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -806,7 +806,7 @@ int nilfs_permission(struct inode *inode, int mask, unsigned int flags)
 	    root->cno != NILFS_CPTREE_CURRENT_CNO)
 		return -EROFS; /* snapshot is not writable */
 
-	return generic_permission(inode, mask, flags, NULL);
+	return generic_permission(inode, mask, flags);
 }
 
 int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh)
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index b1e35a392ca5..d058cb7e12d4 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1293,7 +1293,7 @@ int ocfs2_permission(struct inode *inode, int mask, unsigned int flags)
 		goto out;
 	}
 
-	ret = generic_permission(inode, mask, flags, ocfs2_check_acl);
+	ret = generic_permission(inode, mask, flags);
 
 	ocfs2_inode_unlock(inode, 0);
 out:
@@ -2593,12 +2593,14 @@ const struct inode_operations ocfs2_file_iops = {
 	.listxattr	= ocfs2_listxattr,
 	.removexattr	= generic_removexattr,
 	.fiemap		= ocfs2_fiemap,
+	.check_acl	= ocfs2_check_acl,
 };
 
 const struct inode_operations ocfs2_special_file_iops = {
 	.setattr	= ocfs2_setattr,
 	.getattr	= ocfs2_getattr,
 	.permission	= ocfs2_permission,
+	.check_acl	= ocfs2_check_acl,
 };
 
 /*
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index e5d738cd9cc0..33889dc52dd7 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -2498,4 +2498,5 @@ const struct inode_operations ocfs2_dir_iops = {
 	.listxattr	= ocfs2_listxattr,
 	.removexattr	= generic_removexattr,
 	.fiemap         = ocfs2_fiemap,
+	.check_acl	= ocfs2_check_acl,
 };
diff --git a/fs/proc/base.c b/fs/proc/base.c
index fc5bc2767692..8b8470113576 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2169,7 +2169,7 @@ static const struct file_operations proc_fd_operations = {
  */
 static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags)
 {
-	int rv = generic_permission(inode, mask, flags, NULL);
+	int rv = generic_permission(inode, mask, flags);
 	if (rv == 0)
 		return 0;
 	if (task_pid(current) == proc_pid(inode))
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 91f080cc76c8..bbf31003d308 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -312,4 +312,5 @@ const struct inode_operations reiserfs_file_inode_operations = {
 	.listxattr = reiserfs_listxattr,
 	.removexattr = reiserfs_removexattr,
 	.permission = reiserfs_permission,
+	.check_acl = reiserfs_check_acl,
 };
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 118662690cdf..551f1b79dbc4 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -1529,6 +1529,7 @@ const struct inode_operations reiserfs_dir_inode_operations = {
 	.listxattr = reiserfs_listxattr,
 	.removexattr = reiserfs_removexattr,
 	.permission = reiserfs_permission,
+	.check_acl = reiserfs_check_acl,
 };
 
 /*
@@ -1545,6 +1546,7 @@ const struct inode_operations reiserfs_symlink_inode_operations = {
 	.listxattr = reiserfs_listxattr,
 	.removexattr = reiserfs_removexattr,
 	.permission = reiserfs_permission,
+	.check_acl = reiserfs_check_acl,
 
 };
 
@@ -1558,5 +1560,5 @@ const struct inode_operations reiserfs_special_inode_operations = {
 	.listxattr = reiserfs_listxattr,
 	.removexattr = reiserfs_removexattr,
 	.permission = reiserfs_permission,
-
+	.check_acl = reiserfs_check_acl,
 };
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index d78089690965..ddc5301d2986 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -868,11 +868,17 @@ out:
 	return err;
 }
 
-static int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags)
+int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags)
 {
 	struct posix_acl *acl;
 	int error = -EAGAIN; /* do regular unix permission checks by default */
 
+	/*
+	 * Stat data v1 doesn't support ACLs.
+	 */
+	if (get_inode_sd_version(inode) == STAT_DATA_V1)
+		return -EAGAIN;
+
 	if (flags & IPERM_FLAG_RCU)
 		return -ECHILD;
 
@@ -961,15 +967,7 @@ int reiserfs_permission(struct inode *inode, int mask, unsigned int flags)
 	if (IS_PRIVATE(inode))
 		return 0;
 
-#ifdef CONFIG_REISERFS_FS_XATTR
-	/*
-	 * Stat data v1 doesn't support ACLs.
-	 */
-	if (get_inode_sd_version(inode) != STAT_DATA_V1)
-		return generic_permission(inode, mask, flags,
-					reiserfs_check_acl);
-#endif
-	return generic_permission(inode, mask, flags, NULL);
+	return generic_permission(inode, mask, flags);
 }
 
 static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd)
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 0a12eb89cd32..a37165c64757 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -362,5 +362,5 @@ int sysfs_permission(struct inode *inode, int mask, unsigned int flags)
 	sysfs_refresh_inode(sd, inode);
 	mutex_unlock(&sysfs_mutex);
 
-	return generic_permission(inode, mask, flags, NULL);
+	return generic_permission(inode, mask, flags);
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8c84ed930389..0c15d5e459d5 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2187,8 +2187,7 @@ extern sector_t bmap(struct inode *, sector_t);
 #endif
 extern int notify_change(struct dentry *, struct iattr *);
 extern int inode_permission(struct inode *, int);
-extern int generic_permission(struct inode *, int, unsigned int,
-		int (*check_acl)(struct inode *, int, unsigned int));
+extern int generic_permission(struct inode *, int, unsigned int);
 
 static inline bool execute_ok(struct inode *inode)
 {
diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h
index 6deef5dc95fb..1a3ca8f80200 100644
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -45,6 +45,7 @@ int reiserfs_permission(struct inode *inode, int mask, unsigned int flags);
 
 #ifdef CONFIG_REISERFS_FS_XATTR
 #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir)
+int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags);
 ssize_t reiserfs_getxattr(struct dentry *dentry, const char *name,
 			  void *buffer, size_t size);
 int reiserfs_setxattr(struct dentry *dentry, const char *name,
@@ -122,6 +123,7 @@ static inline void reiserfs_init_xattr_rwsem(struct inode *inode)
 #define reiserfs_setxattr NULL
 #define reiserfs_listxattr NULL
 #define reiserfs_removexattr NULL
+#define reiserfs_check_acl NULL
 
 static inline void reiserfs_init_xattr_rwsem(struct inode *inode)
 {
-- 
cgit v1.2.3


From 1fc0f78ca9f311c6277e2f1b7655bb4d43ceb311 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 18:59:02 -0400
Subject: ->permission() sanitizing: MAY_NOT_BLOCK

Duplicate the flags argument into mask bitmap.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c            | 7 +++++--
 fs/proc/proc_sysctl.c | 2 +-
 include/linux/fs.h    | 1 +
 3 files changed, 7 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index c5af0f37e67d..723a3fe4bc40 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -318,13 +318,16 @@ static inline int exec_permission(struct inode *inode, unsigned int flags)
 {
 	int ret;
 	struct user_namespace *ns = inode_userns(inode);
+	int mask = MAY_EXEC;
+	if (flags & IPERM_FLAG_RCU)
+		mask |= MAY_NOT_BLOCK;
 
 	if (inode->i_op->permission) {
-		ret = inode->i_op->permission(inode, MAY_EXEC, flags);
+		ret = inode->i_op->permission(inode, mask, flags);
 		if (likely(!ret))
 			goto ok;
 	} else {
-		ret = acl_permission_check(inode, MAY_EXEC, flags);
+		ret = acl_permission_check(inode, mask, flags);
 		if (likely(!ret))
 			goto ok;
 		if (ret != -EACCES)
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index d167de365a8d..349b22f434d8 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -316,7 +316,7 @@ static int proc_sys_permission(struct inode *inode, int mask,unsigned int flags)
 	if (!table) /* global root - r-xr-xr-x */
 		error = mask & MAY_WRITE ? -EACCES : 0;
 	else /* Use the permissions on the sysctl table entry */
-		error = sysctl_perm(head->root, table, mask);
+		error = sysctl_perm(head->root, table, mask & ~MAY_NOT_BLOCK);
 
 	sysctl_head_finish(head);
 	return error;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 0c15d5e459d5..60c1fe65bb2d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -63,6 +63,7 @@ struct inodes_stat_t {
 #define MAY_ACCESS 16
 #define MAY_OPEN 32
 #define MAY_CHDIR 64
+#define MAY_NOT_BLOCK 128	/* called from RCU mode, don't block */
 
 /*
  * flags in file.f_mode.  Note that FMODE_READ and FMODE_WRITE must correspond
-- 
cgit v1.2.3


From 7e40145eb111a5192e6d819f764db9d6828d1abb Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 19:12:17 -0400
Subject: ->permission() sanitizing: don't pass flags to ->check_acl()

not used in the instances anymore.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/Locking |  2 +-
 Documentation/filesystems/vfs.txt |  2 +-
 fs/9p/acl.c                       |  2 +-
 fs/9p/acl.h                       |  2 +-
 fs/btrfs/acl.c                    |  2 +-
 fs/btrfs/ctree.h                  |  2 +-
 fs/ext2/acl.c                     |  2 +-
 fs/ext2/acl.h                     |  2 +-
 fs/ext3/acl.c                     |  2 +-
 fs/ext3/acl.h                     |  2 +-
 fs/ext4/acl.c                     |  2 +-
 fs/ext4/acl.h                     |  2 +-
 fs/generic_acl.c                  |  2 +-
 fs/gfs2/acl.c                     |  2 +-
 fs/gfs2/acl.h                     |  2 +-
 fs/jffs2/acl.c                    |  2 +-
 fs/jffs2/acl.h                    |  2 +-
 fs/jfs/acl.c                      |  2 +-
 fs/jfs/jfs_acl.h                  |  2 +-
 fs/namei.c                        | 10 +++++-----
 fs/ocfs2/acl.c                    |  2 +-
 fs/ocfs2/acl.h                    |  2 +-
 fs/reiserfs/xattr.c               |  2 +-
 fs/xfs/linux-2.6/xfs_acl.c        |  2 +-
 fs/xfs/xfs_acl.h                  |  2 +-
 include/linux/fs.h                |  2 +-
 include/linux/generic_acl.h       |  2 +-
 include/linux/reiserfs_xattr.h    |  2 +-
 28 files changed, 32 insertions(+), 32 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 57d827d6071d..9b6ed7c9f34f 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -52,7 +52,7 @@ ata *);
 	void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
 	int (*permission) (struct inode *, int, unsigned int);
-	int (*check_acl)(struct inode *, int, unsigned int);
+	int (*check_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 88b9f5519af9..8b4c8e04d879 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -334,7 +334,7 @@ struct inode_operations {
         void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
 	int (*permission) (struct inode *, int, unsigned int);
-	int (*check_acl)(struct inode *, int, unsigned int);
+	int (*check_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
diff --git a/fs/9p/acl.c b/fs/9p/acl.c
index 94af68b092af..e98f56d3787d 100644
--- a/fs/9p/acl.c
+++ b/fs/9p/acl.c
@@ -96,7 +96,7 @@ static struct posix_acl *v9fs_get_cached_acl(struct inode *inode, int type)
 	return acl;
 }
 
-int v9fs_check_acl(struct inode *inode, int mask, unsigned int flags)
+int v9fs_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 	struct v9fs_session_info *v9ses;
diff --git a/fs/9p/acl.h b/fs/9p/acl.h
index 7ef3ac9f6d95..59e18c2e8c7e 100644
--- a/fs/9p/acl.h
+++ b/fs/9p/acl.h
@@ -16,7 +16,7 @@
 
 #ifdef CONFIG_9P_FS_POSIX_ACL
 extern int v9fs_get_acl(struct inode *, struct p9_fid *);
-extern int v9fs_check_acl(struct inode *inode, int mask, unsigned int flags);
+extern int v9fs_check_acl(struct inode *inode, int mask);
 extern int v9fs_acl_chmod(struct dentry *);
 extern int v9fs_set_create_acl(struct dentry *,
 			       struct posix_acl *, struct posix_acl *);
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index a25a4a2e0df2..9f62ab2a7282 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -195,7 +195,7 @@ out:
 	return ret;
 }
 
-int btrfs_check_acl(struct inode *inode, int mask, unsigned int flags)
+int btrfs_check_acl(struct inode *inode, int mask)
 {
 	int error = -EAGAIN;
 
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 3b859a3e6a0e..9864cec801ed 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2642,7 +2642,7 @@ do {								\
 
 /* acl.c */
 #ifdef CONFIG_BTRFS_FS_POSIX_ACL
-int btrfs_check_acl(struct inode *inode, int mask, unsigned int flags);
+int btrfs_check_acl(struct inode *inode, int mask);
 #else
 #define btrfs_check_acl NULL
 #endif
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index 6b9442d1be52..bfe651f9ae16 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -232,7 +232,7 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 }
 
 int
-ext2_check_acl(struct inode *inode, int mask, unsigned int flags)
+ext2_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 
diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h
index c939b7b12099..3ff6cbb9ac44 100644
--- a/fs/ext2/acl.h
+++ b/fs/ext2/acl.h
@@ -54,7 +54,7 @@ static inline int ext2_acl_count(size_t size)
 #ifdef CONFIG_EXT2_FS_POSIX_ACL
 
 /* acl.c */
-extern int ext2_check_acl (struct inode *, int, unsigned int);
+extern int ext2_check_acl (struct inode *, int);
 extern int ext2_acl_chmod (struct inode *);
 extern int ext2_init_acl (struct inode *, struct inode *);
 
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 0a6940d6c30c..edfeb293d4cb 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -240,7 +240,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 }
 
 int
-ext3_check_acl(struct inode *inode, int mask, unsigned int flags)
+ext3_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 
diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h
index 5faf8048e906..597334626de9 100644
--- a/fs/ext3/acl.h
+++ b/fs/ext3/acl.h
@@ -54,7 +54,7 @@ static inline int ext3_acl_count(size_t size)
 #ifdef CONFIG_EXT3_FS_POSIX_ACL
 
 /* acl.c */
-extern int ext3_check_acl (struct inode *, int, unsigned int);
+extern int ext3_check_acl (struct inode *, int);
 extern int ext3_acl_chmod (struct inode *);
 extern int ext3_init_acl (handle_t *, struct inode *, struct inode *);
 
diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 4f54252e439e..60d900fcc3db 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -238,7 +238,7 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 }
 
 int
-ext4_check_acl(struct inode *inode, int mask, unsigned int flags)
+ext4_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 
diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h
index dec821168fd4..9d843d5deac4 100644
--- a/fs/ext4/acl.h
+++ b/fs/ext4/acl.h
@@ -54,7 +54,7 @@ static inline int ext4_acl_count(size_t size)
 #ifdef CONFIG_EXT4_FS_POSIX_ACL
 
 /* acl.c */
-extern int ext4_check_acl(struct inode *, int, unsigned int);
+extern int ext4_check_acl(struct inode *, int);
 extern int ext4_acl_chmod(struct inode *);
 extern int ext4_init_acl(handle_t *, struct inode *, struct inode *);
 
diff --git a/fs/generic_acl.c b/fs/generic_acl.c
index 5976bb1fa4ca..70e90b4974ce 100644
--- a/fs/generic_acl.c
+++ b/fs/generic_acl.c
@@ -190,7 +190,7 @@ generic_acl_chmod(struct inode *inode)
 }
 
 int
-generic_check_acl(struct inode *inode, int mask, unsigned int flags)
+generic_check_acl(struct inode *inode, int mask)
 {
 	if (mask & MAY_NOT_BLOCK) {
 		if (!negative_cached_acl(inode, ACL_TYPE_ACCESS))
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index 4d97352d39a1..8ef1079f1665 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -75,7 +75,7 @@ static struct posix_acl *gfs2_acl_get(struct gfs2_inode *ip, int type)
  * Returns: errno
  */
 
-int gfs2_check_acl(struct inode *inode, int mask, unsigned int flags)
+int gfs2_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 	int error;
diff --git a/fs/gfs2/acl.h b/fs/gfs2/acl.h
index a93907c8159b..b522b0cb39ea 100644
--- a/fs/gfs2/acl.h
+++ b/fs/gfs2/acl.h
@@ -16,7 +16,7 @@
 #define GFS2_POSIX_ACL_DEFAULT		"posix_acl_default"
 #define GFS2_ACL_MAX_ENTRIES		25
 
-extern int gfs2_check_acl(struct inode *inode, int mask, unsigned int);
+extern int gfs2_check_acl(struct inode *inode, int mask);
 extern int gfs2_acl_create(struct gfs2_inode *dip, struct inode *inode);
 extern int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr);
 extern const struct xattr_handler gfs2_xattr_system_handler;
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index 952afb59e6f1..3675b3cdee89 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -259,7 +259,7 @@ static int jffs2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	return rc;
 }
 
-int jffs2_check_acl(struct inode *inode, int mask, unsigned int flags)
+int jffs2_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 	int rc;
diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h
index 3119f59253d3..5e42de8d9541 100644
--- a/fs/jffs2/acl.h
+++ b/fs/jffs2/acl.h
@@ -26,7 +26,7 @@ struct jffs2_acl_header {
 
 #ifdef CONFIG_JFFS2_FS_POSIX_ACL
 
-extern int jffs2_check_acl(struct inode *, int, unsigned int);
+extern int jffs2_check_acl(struct inode *, int);
 extern int jffs2_acl_chmod(struct inode *);
 extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *);
 extern int jffs2_init_acl_post(struct inode *);
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 859ae5a92166..8a0a0666d5a6 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -114,7 +114,7 @@ out:
 	return rc;
 }
 
-int jfs_check_acl(struct inode *inode, int mask, unsigned int flags)
+int jfs_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 
diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h
index f9285c4900fa..54e07559878d 100644
--- a/fs/jfs/jfs_acl.h
+++ b/fs/jfs/jfs_acl.h
@@ -20,7 +20,7 @@
 
 #ifdef CONFIG_JFS_POSIX_ACL
 
-int jfs_check_acl(struct inode *, int, unsigned int flags);
+int jfs_check_acl(struct inode *, int);
 int jfs_init_acl(tid_t, struct inode *, struct inode *);
 int jfs_acl_chmod(struct inode *inode);
 
diff --git a/fs/namei.c b/fs/namei.c
index e0624e2f0bba..560fd1dff1d0 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -176,9 +176,9 @@ EXPORT_SYMBOL(putname);
 /*
  * This does basic POSIX ACL permission checking
  */
-static int acl_permission_check(struct inode *inode, int mask, unsigned int flags)
+static int acl_permission_check(struct inode *inode, int mask)
 {
-	int (*check_acl)(struct inode *inode, int mask, unsigned int flags);
+	int (*check_acl)(struct inode *inode, int mask);
 	unsigned int mode = inode->i_mode;
 
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC | MAY_NOT_BLOCK;
@@ -191,7 +191,7 @@ static int acl_permission_check(struct inode *inode, int mask, unsigned int flag
 	else {
 		check_acl = inode->i_op->check_acl;
 		if (IS_POSIXACL(inode) && (mode & S_IRWXG) && check_acl) {
-			int error = check_acl(inode, mask, flags);
+			int error = check_acl(inode, mask);
 			if (error != -EAGAIN)
 				return error;
 		}
@@ -231,7 +231,7 @@ int generic_permission(struct inode *inode, int mask, unsigned int flags)
 	/*
 	 * Do the basic POSIX ACL permission checks.
 	 */
-	ret = acl_permission_check(inode, mask, flags);
+	ret = acl_permission_check(inode, mask);
 	if (ret != -EACCES)
 		return ret;
 
@@ -327,7 +327,7 @@ static inline int exec_permission(struct inode *inode, unsigned int flags)
 		if (likely(!ret))
 			goto ok;
 	} else {
-		ret = acl_permission_check(inode, mask, flags);
+		ret = acl_permission_check(inode, mask);
 		if (likely(!ret))
 			goto ok;
 		if (ret != -EACCES)
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index 4b683ccc4506..1cee970eb55a 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -290,7 +290,7 @@ static int ocfs2_set_acl(handle_t *handle,
 	return ret;
 }
 
-int ocfs2_check_acl(struct inode *inode, int mask, unsigned int flags)
+int ocfs2_check_acl(struct inode *inode, int mask)
 {
 	struct ocfs2_super *osb;
 	struct buffer_head *di_bh = NULL;
diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h
index 4fe7c9cf4bfb..5c5d31f05853 100644
--- a/fs/ocfs2/acl.h
+++ b/fs/ocfs2/acl.h
@@ -26,7 +26,7 @@ struct ocfs2_acl_entry {
 	__le32 e_id;
 };
 
-extern int ocfs2_check_acl(struct inode *, int, unsigned int);
+extern int ocfs2_check_acl(struct inode *, int);
 extern int ocfs2_acl_chmod(struct inode *);
 extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
 			  struct buffer_head *, struct buffer_head *,
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 6747470ec103..6ee3c11aa8d9 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -868,7 +868,7 @@ out:
 	return err;
 }
 
-int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags)
+int reiserfs_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
 	int error = -EAGAIN; /* do regular unix permission checks by default */
diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
index 278e6736135a..a5dcd6a0f1b5 100644
--- a/fs/xfs/linux-2.6/xfs_acl.c
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -219,7 +219,7 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 }
 
 int
-xfs_check_acl(struct inode *inode, int mask, unsigned int flags)
+xfs_check_acl(struct inode *inode, int mask)
 {
 	struct xfs_inode *ip;
 	struct posix_acl *acl;
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 11dd72070cbb..0135e2a669d7 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -42,7 +42,7 @@ struct xfs_acl {
 #define SGI_ACL_DEFAULT_SIZE	(sizeof(SGI_ACL_DEFAULT)-1)
 
 #ifdef CONFIG_XFS_POSIX_ACL
-extern int xfs_check_acl(struct inode *inode, int mask, unsigned int flags);
+extern int xfs_check_acl(struct inode *inode, int mask);
 extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
 extern int xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl);
 extern int xfs_acl_chmod(struct inode *inode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 60c1fe65bb2d..f218b42718aa 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1579,7 +1579,7 @@ struct inode_operations {
 	struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *);
 	void * (*follow_link) (struct dentry *, struct nameidata *);
 	int (*permission) (struct inode *, int, unsigned int);
-	int (*check_acl)(struct inode *, int, unsigned int);
+	int (*check_acl)(struct inode *, int);
 
 	int (*readlink) (struct dentry *, char __user *,int);
 	void (*put_link) (struct dentry *, struct nameidata *, void *);
diff --git a/include/linux/generic_acl.h b/include/linux/generic_acl.h
index 0437e377b555..574bea4013b6 100644
--- a/include/linux/generic_acl.h
+++ b/include/linux/generic_acl.h
@@ -10,6 +10,6 @@ extern const struct xattr_handler generic_acl_default_handler;
 
 int generic_acl_init(struct inode *, struct inode *);
 int generic_acl_chmod(struct inode *);
-int generic_check_acl(struct inode *inode, int mask, unsigned int flags);
+int generic_check_acl(struct inode *inode, int mask);
 
 #endif /* LINUX_GENERIC_ACL_H */
diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h
index 1a3ca8f80200..424ba6416e7c 100644
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -45,7 +45,7 @@ int reiserfs_permission(struct inode *inode, int mask, unsigned int flags);
 
 #ifdef CONFIG_REISERFS_FS_XATTR
 #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir)
-int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags);
+int reiserfs_check_acl(struct inode *inode, int mask);
 ssize_t reiserfs_getxattr(struct dentry *dentry, const char *name,
 			  void *buffer, size_t size);
 int reiserfs_setxattr(struct dentry *dentry, const char *name,
-- 
cgit v1.2.3


From 2830ba7f34ebb27c4e5b8b6ef408cd6d74860890 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 19:16:29 -0400
Subject: ->permission() sanitizing: don't pass flags to generic_permission()

redundant; all callers get it duplicated in mask & MAY_NOT_BLOCK and none of
them removes that bit.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/afs/security.c       | 2 +-
 fs/btrfs/inode.c        | 2 +-
 fs/ceph/inode.c         | 2 +-
 fs/cifs/cifsfs.c        | 2 +-
 fs/fuse/dir.c           | 5 ++---
 fs/gfs2/inode.c         | 2 +-
 fs/hostfs/hostfs_kern.c | 2 +-
 fs/hpfs/namei.c         | 2 +-
 fs/namei.c              | 4 ++--
 fs/nfs/dir.c            | 2 +-
 fs/nilfs2/inode.c       | 2 +-
 fs/ocfs2/file.c         | 2 +-
 fs/proc/base.c          | 2 +-
 fs/reiserfs/xattr.c     | 2 +-
 fs/sysfs/inode.c        | 2 +-
 include/linux/fs.h      | 2 +-
 16 files changed, 18 insertions(+), 19 deletions(-)

(limited to 'include')

diff --git a/fs/afs/security.c b/fs/afs/security.c
index 745ee654165f..ab6b3147f05f 100644
--- a/fs/afs/security.c
+++ b/fs/afs/security.c
@@ -350,7 +350,7 @@ int afs_permission(struct inode *inode, int mask, unsigned int flags)
 	}
 
 	key_put(key);
-	ret = generic_permission(inode, mask, flags);
+	ret = generic_permission(inode, mask);
 	_leave(" = %d", ret);
 	return ret;
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index f0bd87371566..b61b0477a8ac 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7339,7 +7339,7 @@ static int btrfs_permission(struct inode *inode, int mask, unsigned int flags)
 		return -EROFS;
 	if ((BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) && (mask & MAY_WRITE))
 		return -EACCES;
-	return generic_permission(inode, mask, flags);
+	return generic_permission(inode, mask);
 }
 
 static const struct inode_operations btrfs_dir_inode_operations = {
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index beb5d55d6fd2..9c169741b416 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1805,7 +1805,7 @@ int ceph_permission(struct inode *inode, int mask, unsigned int flags)
 	err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
 
 	if (!err)
-		err = generic_permission(inode, mask, flags);
+		err = generic_permission(inode, mask);
 	return err;
 }
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index b79804fa410f..b285b5bcf711 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -239,7 +239,7 @@ static int cifs_permission(struct inode *inode, int mask, unsigned int flags)
 		on the client (above and beyond ACL on servers) for
 		servers which do not support setting and viewing mode bits,
 		so allowing client to check permissions is useful */
-		return generic_permission(inode, mask, flags);
+		return generic_permission(inode, mask);
 }
 
 static struct kmem_cache *cifs_inode_cachep;
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 0a2fcd860ad6..0df56b6c26e2 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -1018,7 +1018,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 	}
 
 	if (fc->flags & FUSE_DEFAULT_PERMISSIONS) {
-		err = generic_permission(inode, mask, flags);
+		err = generic_permission(inode, mask);
 
 		/* If permission is denied, try to refresh file
 		   attributes.  This is also needed, because the root
@@ -1026,8 +1026,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 		if (err == -EACCES && !refreshed) {
 			err = fuse_perm_getattr(inode, flags);
 			if (!err)
-				err = generic_permission(inode, mask,
-							flags);
+				err = generic_permission(inode, mask);
 		}
 
 		/* Note: the opposite of the above test does not
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index d5f0f4ea25dc..b776ec8f9c19 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -1564,7 +1564,7 @@ int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
 	if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode))
 		error = -EACCES;
 	else
-		error = generic_permission(inode, mask, flags);
+		error = generic_permission(inode, mask);
 	if (unlock)
 		gfs2_glock_dq_uninit(&i_gh);
 
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index a98d0d1aef65..b1bc31bde833 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -770,7 +770,7 @@ int hostfs_permission(struct inode *ino, int desired, unsigned int flags)
 		err = access_file(name, r, w, x);
 	__putname(name);
 	if (!err)
-		err = generic_permission(ino, desired, flags);
+		err = generic_permission(ino, desired);
 	return err;
 }
 
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index bd2ce7dd8df3..2df69e2f07cf 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -398,7 +398,7 @@ again:
 			hpfs_unlock(dir->i_sb);
 			return -ENOSPC;
 		}
-		if (generic_permission(inode, MAY_WRITE, 0) ||
+		if (generic_permission(inode, MAY_WRITE) ||
 		    !S_ISREG(inode->i_mode) ||
 		    get_write_access(inode)) {
 			d_rehash(dentry);
diff --git a/fs/namei.c b/fs/namei.c
index 560fd1dff1d0..684e0f30cf4c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -224,7 +224,7 @@ other_perms:
  * request cannot be satisfied (eg. requires blocking or too much complexity).
  * It would then be called again in ref-walk mode.
  */
-int generic_permission(struct inode *inode, int mask, unsigned int flags)
+int generic_permission(struct inode *inode, int mask)
 {
 	int ret;
 
@@ -289,7 +289,7 @@ int inode_permission(struct inode *inode, int mask)
 	if (inode->i_op->permission)
 		retval = inode->i_op->permission(inode, mask, 0);
 	else
-		retval = generic_permission(inode, mask, 0);
+		retval = generic_permission(inode, mask);
 
 	if (retval)
 		return retval;
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 0485dca34fb1..16cf84b4afb9 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -2328,7 +2328,7 @@ out:
 out_notsup:
 	res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
 	if (res == 0)
-		res = generic_permission(inode, mask, flags);
+		res = generic_permission(inode, mask);
 	goto out;
 }
 
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 650aa7755003..0df6de58bbcb 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -806,7 +806,7 @@ int nilfs_permission(struct inode *inode, int mask, unsigned int flags)
 	    root->cno != NILFS_CPTREE_CURRENT_CNO)
 		return -EROFS; /* snapshot is not writable */
 
-	return generic_permission(inode, mask, flags);
+	return generic_permission(inode, mask);
 }
 
 int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh)
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index d058cb7e12d4..ecb52b028964 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1293,7 +1293,7 @@ int ocfs2_permission(struct inode *inode, int mask, unsigned int flags)
 		goto out;
 	}
 
-	ret = generic_permission(inode, mask, flags);
+	ret = generic_permission(inode, mask);
 
 	ocfs2_inode_unlock(inode, 0);
 out:
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 8b8470113576..53a1a961b25f 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2169,7 +2169,7 @@ static const struct file_operations proc_fd_operations = {
  */
 static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags)
 {
-	int rv = generic_permission(inode, mask, flags);
+	int rv = generic_permission(inode, mask);
 	if (rv == 0)
 		return 0;
 	if (task_pid(current) == proc_pid(inode))
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 6ee3c11aa8d9..f17319613a85 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -967,7 +967,7 @@ int reiserfs_permission(struct inode *inode, int mask, unsigned int flags)
 	if (IS_PRIVATE(inode))
 		return 0;
 
-	return generic_permission(inode, mask, flags);
+	return generic_permission(inode, mask);
 }
 
 static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd)
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index a37165c64757..04c81e5ba6f7 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -362,5 +362,5 @@ int sysfs_permission(struct inode *inode, int mask, unsigned int flags)
 	sysfs_refresh_inode(sd, inode);
 	mutex_unlock(&sysfs_mutex);
 
-	return generic_permission(inode, mask, flags);
+	return generic_permission(inode, mask);
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index f218b42718aa..a1689c13eb77 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2188,7 +2188,7 @@ extern sector_t bmap(struct inode *, sector_t);
 #endif
 extern int notify_change(struct dentry *, struct iattr *);
 extern int inode_permission(struct inode *, int);
-extern int generic_permission(struct inode *, int, unsigned int);
+extern int generic_permission(struct inode *, int);
 
 static inline bool execute_ok(struct inode *inode)
 {
-- 
cgit v1.2.3


From 10556cb21a0d0b24d95f00ea6df16f599a3345b2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 19:28:19 -0400
Subject: ->permission() sanitizing: don't pass flags to ->permission()

not used by the instances anymore.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/vfs.txt |  4 ++--
 fs/afs/internal.h                 |  2 +-
 fs/afs/security.c                 |  4 ++--
 fs/bad_inode.c                    |  2 +-
 fs/btrfs/inode.c                  |  2 +-
 fs/ceph/inode.c                   |  4 ++--
 fs/ceph/super.h                   |  2 +-
 fs/cifs/cifsfs.c                  |  2 +-
 fs/coda/coda_linux.h              |  2 +-
 fs/coda/dir.c                     |  4 ++--
 fs/coda/pioctl.c                  |  4 ++--
 fs/ecryptfs/inode.c               |  4 ++--
 fs/fuse/dir.c                     | 14 +++++++-------
 fs/gfs2/file.c                    |  2 +-
 fs/gfs2/inode.c                   | 16 ++++++++--------
 fs/gfs2/inode.h                   |  2 +-
 fs/hostfs/hostfs_kern.c           |  4 ++--
 fs/namei.c                        |  4 ++--
 fs/nfs/dir.c                      |  4 ++--
 fs/nilfs2/inode.c                 |  2 +-
 fs/nilfs2/nilfs.h                 |  2 +-
 fs/ocfs2/file.c                   |  4 ++--
 fs/ocfs2/file.h                   |  2 +-
 fs/proc/base.c                    |  2 +-
 fs/proc/proc_sysctl.c             |  2 +-
 fs/reiserfs/xattr.c               |  2 +-
 fs/sysfs/inode.c                  |  4 ++--
 fs/sysfs/sysfs.h                  |  2 +-
 include/linux/fs.h                |  2 +-
 include/linux/nfs_fs.h            |  2 +-
 include/linux/reiserfs_xattr.h    |  2 +-
 31 files changed, 55 insertions(+), 55 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 8b4c8e04d879..d56151fe77dc 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -333,7 +333,7 @@ struct inode_operations {
         void * (*follow_link) (struct dentry *, struct nameidata *);
         void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
-	int (*permission) (struct inode *, int, unsigned int);
+	int (*permission) (struct inode *, int);
 	int (*check_acl)(struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
@@ -423,7 +423,7 @@ otherwise noted.
   permission: called by the VFS to check for access rights on a POSIX-like
   	filesystem.
 
-	May be called in rcu-walk mode (flags & IPERM_FLAG_RCU). If in rcu-walk
+	May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in rcu-walk
         mode, the filesystem must check the permission without blocking or
 	storing to the inode.
 
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index 5a9b6843bac1..f396d337b817 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -627,7 +627,7 @@ extern void afs_clear_permits(struct afs_vnode *);
 extern void afs_cache_permit(struct afs_vnode *, struct key *, long);
 extern void afs_zap_permits(struct rcu_head *);
 extern struct key *afs_request_key(struct afs_cell *);
-extern int afs_permission(struct inode *, int, unsigned int);
+extern int afs_permission(struct inode *, int);
 
 /*
  * server.c
diff --git a/fs/afs/security.c b/fs/afs/security.c
index ab6b3147f05f..8d010422dc89 100644
--- a/fs/afs/security.c
+++ b/fs/afs/security.c
@@ -285,14 +285,14 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key,
  * - AFS ACLs are attached to directories only, and a file is controlled by its
  *   parent directory's ACL
  */
-int afs_permission(struct inode *inode, int mask, unsigned int flags)
+int afs_permission(struct inode *inode, int mask)
 {
 	struct afs_vnode *vnode = AFS_FS_I(inode);
 	afs_access_t uninitialized_var(access);
 	struct key *key;
 	int ret;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	_enter("{{%x:%u},%lx},%x,",
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index bfcb18feb1df..f024d8aaddef 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -229,7 +229,7 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer,
 	return -EIO;
 }
 
-static int bad_inode_permission(struct inode *inode, int mask, unsigned int flags)
+static int bad_inode_permission(struct inode *inode, int mask)
 {
 	return -EIO;
 }
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b61b0477a8ac..cb170ca51c64 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7331,7 +7331,7 @@ static int btrfs_set_page_dirty(struct page *page)
 	return __set_page_dirty_nobuffers(page);
 }
 
-static int btrfs_permission(struct inode *inode, int mask, unsigned int flags)
+static int btrfs_permission(struct inode *inode, int mask)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 9c169741b416..dfb2831d8d85 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1795,11 +1795,11 @@ int ceph_do_getattr(struct inode *inode, int mask)
  * Check inode permissions.  We verify we have a valid value for
  * the AUTH cap, then call the generic handler.
  */
-int ceph_permission(struct inode *inode, int mask, unsigned int flags)
+int ceph_permission(struct inode *inode, int mask)
 {
 	int err;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index f5cabefa98dc..56c41ef47cad 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -692,7 +692,7 @@ extern void ceph_queue_invalidate(struct inode *inode);
 extern void ceph_queue_writeback(struct inode *inode);
 
 extern int ceph_do_getattr(struct inode *inode, int mask);
-extern int ceph_permission(struct inode *inode, int mask, unsigned int flags);
+extern int ceph_permission(struct inode *inode, int mask);
 extern int ceph_setattr(struct dentry *dentry, struct iattr *attr);
 extern int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry,
 			struct kstat *stat);
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index b285b5bcf711..cbbb55e781d3 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -224,7 +224,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	return 0;
 }
 
-static int cifs_permission(struct inode *inode, int mask, unsigned int flags)
+static int cifs_permission(struct inode *inode, int mask)
 {
 	struct cifs_sb_info *cifs_sb;
 
diff --git a/fs/coda/coda_linux.h b/fs/coda/coda_linux.h
index 9b0c5323890b..44e17e9c21ae 100644
--- a/fs/coda/coda_linux.h
+++ b/fs/coda/coda_linux.h
@@ -39,7 +39,7 @@ extern const struct file_operations coda_ioctl_operations;
 /* operations shared over more than one file */
 int coda_open(struct inode *i, struct file *f);
 int coda_release(struct inode *i, struct file *f);
-int coda_permission(struct inode *inode, int mask, unsigned int flags);
+int coda_permission(struct inode *inode, int mask);
 int coda_revalidate_inode(struct dentry *);
 int coda_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 int coda_setattr(struct dentry *, struct iattr *);
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 2b8dae4d121e..cd5532398c15 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -132,11 +132,11 @@ exit:
 }
 
 
-int coda_permission(struct inode *inode, int mask, unsigned int flags)
+int coda_permission(struct inode *inode, int mask)
 {
 	int error;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c
index cb140ef293e4..ee0981f1375b 100644
--- a/fs/coda/pioctl.c
+++ b/fs/coda/pioctl.c
@@ -24,7 +24,7 @@
 #include "coda_linux.h"
 
 /* pioctl ops */
-static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags);
+static int coda_ioctl_permission(struct inode *inode, int mask);
 static long coda_pioctl(struct file *filp, unsigned int cmd,
 			unsigned long user_data);
 
@@ -41,7 +41,7 @@ const struct file_operations coda_ioctl_operations = {
 };
 
 /* the coda pioctl inode ops */
-static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags)
+static int coda_ioctl_permission(struct inode *inode, int mask)
 {
 	return (mask & MAY_EXEC) ? -EACCES : 0;
 }
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 7349ade17de6..bec75f8e91ac 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -942,9 +942,9 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 }
 
 static int
-ecryptfs_permission(struct inode *inode, int mask, unsigned int flags)
+ecryptfs_permission(struct inode *inode, int mask)
 {
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 	return inode_permission(ecryptfs_inode_to_lower(inode), mask);
 }
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 0df56b6c26e2..e2b14001cea5 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -971,9 +971,9 @@ static int fuse_access(struct inode *inode, int mask)
 	return err;
 }
 
-static int fuse_perm_getattr(struct inode *inode, int flags)
+static int fuse_perm_getattr(struct inode *inode, int mask)
 {
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	return fuse_do_getattr(inode, NULL, NULL);
@@ -992,7 +992,7 @@ static int fuse_perm_getattr(struct inode *inode, int flags)
  * access request is sent.  Execute permission is still checked
  * locally based on file mode.
  */
-static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
+static int fuse_permission(struct inode *inode, int mask)
 {
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	bool refreshed = false;
@@ -1011,7 +1011,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 		if (fi->i_time < get_jiffies_64()) {
 			refreshed = true;
 
-			err = fuse_perm_getattr(inode, flags);
+			err = fuse_perm_getattr(inode, mask);
 			if (err)
 				return err;
 		}
@@ -1024,7 +1024,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 		   attributes.  This is also needed, because the root
 		   node will at first have no permissions */
 		if (err == -EACCES && !refreshed) {
-			err = fuse_perm_getattr(inode, flags);
+			err = fuse_perm_getattr(inode, mask);
 			if (!err)
 				err = generic_permission(inode, mask);
 		}
@@ -1034,7 +1034,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 		   noticed immediately, only after the attribute
 		   timeout has expired */
 	} else if (mask & (MAY_ACCESS | MAY_CHDIR)) {
-		if (flags & IPERM_FLAG_RCU)
+		if (mask & MAY_NOT_BLOCK)
 			return -ECHILD;
 
 		err = fuse_access(inode, mask);
@@ -1043,7 +1043,7 @@ static int fuse_permission(struct inode *inode, int mask, unsigned int flags)
 			if (refreshed)
 				return -EACCES;
 
-			err = fuse_perm_getattr(inode, flags);
+			err = fuse_perm_getattr(inode, mask);
 			if (!err && !(inode->i_mode & S_IXUGO))
 				return -EACCES;
 		}
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index a9f5cbe45cd9..89c39e53760d 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -243,7 +243,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask)
 	    !capable(CAP_LINUX_IMMUTABLE))
 		goto out;
 	if (!IS_IMMUTABLE(inode)) {
-		error = gfs2_permission(inode, MAY_WRITE, 0);
+		error = gfs2_permission(inode, MAY_WRITE);
 		if (error)
 			goto out;
 	}
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index b776ec8f9c19..b1090d66a6fd 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -307,7 +307,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
 	}
 
 	if (!is_root) {
-		error = gfs2_permission(dir, MAY_EXEC, 0);
+		error = gfs2_permission(dir, MAY_EXEC);
 		if (error)
 			goto out;
 	}
@@ -337,7 +337,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name,
 {
 	int error;
 
-	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0);
+	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC);
 	if (error)
 		return error;
 
@@ -857,7 +857,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
 	if (inode->i_nlink == 0)
 		goto out_gunlock;
 
-	error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0);
+	error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC);
 	if (error)
 		goto out_gunlock;
 
@@ -990,7 +990,7 @@ static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
 	if (IS_APPEND(&dip->i_inode))
 		return -EPERM;
 
-	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0);
+	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC);
 	if (error)
 		return error;
 
@@ -1336,7 +1336,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 			}
 		}
 	} else {
-		error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0);
+		error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC);
 		if (error)
 			goto out_gunlock;
 
@@ -1371,7 +1371,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 	/* Check out the dir to be renamed */
 
 	if (dir_rename) {
-		error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0);
+		error = gfs2_permission(odentry->d_inode, MAY_WRITE);
 		if (error)
 			goto out_gunlock;
 	}
@@ -1543,7 +1543,7 @@ static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p)
  * Returns: errno
  */
 
-int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
+int gfs2_permission(struct inode *inode, int mask)
 {
 	struct gfs2_inode *ip;
 	struct gfs2_holder i_gh;
@@ -1553,7 +1553,7 @@ int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
 
 	ip = GFS2_I(inode);
 	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
-		if (flags & IPERM_FLAG_RCU)
+		if (mask & MAY_NOT_BLOCK)
 			return -ECHILD;
 		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
 		if (error)
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 31606076f701..8d90e0c07672 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -108,7 +108,7 @@ extern int gfs2_inode_refresh(struct gfs2_inode *ip);
 
 extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
 				  int is_root);
-extern int gfs2_permission(struct inode *inode, int mask, unsigned int flags);
+extern int gfs2_permission(struct inode *inode, int mask);
 extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr);
 extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name);
 extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf);
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index b1bc31bde833..6e449c599b9d 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -748,12 +748,12 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from,
 	return err;
 }
 
-int hostfs_permission(struct inode *ino, int desired, unsigned int flags)
+int hostfs_permission(struct inode *ino, int desired)
 {
 	char *name;
 	int r = 0, w = 0, x = 0, err;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (desired & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	if (desired & MAY_READ) r = 1;
diff --git a/fs/namei.c b/fs/namei.c
index 684e0f30cf4c..c5c382620a86 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -287,7 +287,7 @@ int inode_permission(struct inode *inode, int mask)
 	}
 
 	if (inode->i_op->permission)
-		retval = inode->i_op->permission(inode, mask, 0);
+		retval = inode->i_op->permission(inode, mask);
 	else
 		retval = generic_permission(inode, mask);
 
@@ -323,7 +323,7 @@ static inline int exec_permission(struct inode *inode, unsigned int flags)
 		mask |= MAY_NOT_BLOCK;
 
 	if (inode->i_op->permission) {
-		ret = inode->i_op->permission(inode, mask, flags);
+		ret = inode->i_op->permission(inode, mask);
 		if (likely(!ret))
 			goto ok;
 	} else {
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 16cf84b4afb9..a86acd6d0cb4 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -2273,12 +2273,12 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags)
 	return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags));
 }
 
-int nfs_permission(struct inode *inode, int mask, unsigned int flags)
+int nfs_permission(struct inode *inode, int mask)
 {
 	struct rpc_cred *cred;
 	int res = 0;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	nfs_inc_stats(inode, NFSIOS_VFSACCESS);
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 0df6de58bbcb..be8664c6a825 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -799,7 +799,7 @@ out_err:
 	return err;
 }
 
-int nilfs_permission(struct inode *inode, int mask, unsigned int flags)
+int nilfs_permission(struct inode *inode, int mask)
 {
 	struct nilfs_root *root = NILFS_I(inode)->i_root;
 	if ((mask & MAY_WRITE) && root &&
diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h
index f02b9ad43a21..6fb7511c0328 100644
--- a/fs/nilfs2/nilfs.h
+++ b/fs/nilfs2/nilfs.h
@@ -264,7 +264,7 @@ extern void nilfs_update_inode(struct inode *, struct buffer_head *);
 extern void nilfs_truncate(struct inode *);
 extern void nilfs_evict_inode(struct inode *);
 extern int nilfs_setattr(struct dentry *, struct iattr *);
-int nilfs_permission(struct inode *inode, int mask, unsigned int flags);
+int nilfs_permission(struct inode *inode, int mask);
 int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh);
 extern int nilfs_inode_dirty(struct inode *);
 int nilfs_set_file_dirty(struct inode *inode, unsigned nr_dirty);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index ecb52b028964..1406c37a5722 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1279,11 +1279,11 @@ bail:
 	return err;
 }
 
-int ocfs2_permission(struct inode *inode, int mask, unsigned int flags)
+int ocfs2_permission(struct inode *inode, int mask)
 {
 	int ret;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	ret = ocfs2_inode_lock(inode, NULL, 0);
diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h
index f5afbbef6703..97bf761c9e7c 100644
--- a/fs/ocfs2/file.h
+++ b/fs/ocfs2/file.h
@@ -61,7 +61,7 @@ int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh,
 int ocfs2_setattr(struct dentry *dentry, struct iattr *attr);
 int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
 		  struct kstat *stat);
-int ocfs2_permission(struct inode *inode, int mask, unsigned int flags);
+int ocfs2_permission(struct inode *inode, int mask);
 
 int ocfs2_should_update_atime(struct inode *inode,
 			      struct vfsmount *vfsmnt);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 53a1a961b25f..be1ff932033b 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2167,7 +2167,7 @@ static const struct file_operations proc_fd_operations = {
  * /proc/pid/fd needs a special permission handler so that a process can still
  * access /proc/self/fd after it has executed a setuid().
  */
-static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags)
+static int proc_fd_permission(struct inode *inode, int mask)
 {
 	int rv = generic_permission(inode, mask);
 	if (rv == 0)
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 349b22f434d8..1a77dbef226f 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -294,7 +294,7 @@ out:
 	return ret;
 }
 
-static int proc_sys_permission(struct inode *inode, int mask,unsigned int flags)
+static int proc_sys_permission(struct inode *inode, int mask)
 {
 	/*
 	 * sysctl entries that are not writeable,
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index f17319613a85..4ea2ab41fdee 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -958,7 +958,7 @@ static int xattr_mount_check(struct super_block *s)
 	return 0;
 }
 
-int reiserfs_permission(struct inode *inode, int mask, unsigned int flags)
+int reiserfs_permission(struct inode *inode, int mask)
 {
 	/*
 	 * We don't do permission checks on the internal objects.
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 04c81e5ba6f7..e3f091a81c72 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -349,11 +349,11 @@ int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const void *ns, const cha
 		return -ENOENT;
 }
 
-int sysfs_permission(struct inode *inode, int mask, unsigned int flags)
+int sysfs_permission(struct inode *inode, int mask)
 {
 	struct sysfs_dirent *sd;
 
-	if (flags & IPERM_FLAG_RCU)
+	if (mask & MAY_NOT_BLOCK)
 		return -ECHILD;
 
 	sd = inode->i_private;
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 2ed2404f3113..845ab3ad229d 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -201,7 +201,7 @@ static inline void __sysfs_put(struct sysfs_dirent *sd)
 struct inode *sysfs_get_inode(struct super_block *sb, struct sysfs_dirent *sd);
 void sysfs_evict_inode(struct inode *inode);
 int sysfs_sd_setattr(struct sysfs_dirent *sd, struct iattr *iattr);
-int sysfs_permission(struct inode *inode, int mask, unsigned int flags);
+int sysfs_permission(struct inode *inode, int mask);
 int sysfs_setattr(struct dentry *dentry, struct iattr *iattr);
 int sysfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat);
 int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index a1689c13eb77..1fdefe3995fd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1578,7 +1578,7 @@ struct file_operations {
 struct inode_operations {
 	struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *);
 	void * (*follow_link) (struct dentry *, struct nameidata *);
-	int (*permission) (struct inode *, int, unsigned int);
+	int (*permission) (struct inode *, int);
 	int (*check_acl)(struct inode *, int);
 
 	int (*readlink) (struct dentry *, char __user *,int);
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 1b93b9c60e55..86014811a0fb 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -360,7 +360,7 @@ extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *);
 extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr);
 extern int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fattr);
 extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
-extern int nfs_permission(struct inode *, int, unsigned int);
+extern int nfs_permission(struct inode *, int);
 extern int nfs_open(struct inode *, struct file *);
 extern int nfs_release(struct inode *, struct file *);
 extern int nfs_attribute_timeout(struct inode *inode);
diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h
index 424ba6416e7c..57958c0e1d38 100644
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -41,7 +41,7 @@ int reiserfs_xattr_init(struct super_block *sb, int mount_flags);
 int reiserfs_lookup_privroot(struct super_block *sb);
 int reiserfs_delete_xattrs(struct inode *inode);
 int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs);
-int reiserfs_permission(struct inode *inode, int mask, unsigned int flags);
+int reiserfs_permission(struct inode *inode, int mask);
 
 #ifdef CONFIG_REISERFS_FS_XATTR
 #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir)
-- 
cgit v1.2.3


From e74f71eb78a4a8b9eaf1bc65f20f761648e85f76 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 19:38:15 -0400
Subject: ->permission() sanitizing: don't pass flags to ->inode_permission()

pass that via mask instead.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/security.h   | 2 +-
 security/capability.c      | 2 +-
 security/security.c        | 7 +++++--
 security/selinux/hooks.c   | 5 +++--
 security/smack/smack_lsm.c | 5 +++--
 5 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'include')

diff --git a/include/linux/security.h b/include/linux/security.h
index 8ce59ef3e5af..ca02f1716736 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1456,7 +1456,7 @@ struct security_operations {
 			     struct inode *new_dir, struct dentry *new_dentry);
 	int (*inode_readlink) (struct dentry *dentry);
 	int (*inode_follow_link) (struct dentry *dentry, struct nameidata *nd);
-	int (*inode_permission) (struct inode *inode, int mask, unsigned flags);
+	int (*inode_permission) (struct inode *inode, int mask);
 	int (*inode_setattr)	(struct dentry *dentry, struct iattr *attr);
 	int (*inode_getattr) (struct vfsmount *mnt, struct dentry *dentry);
 	int (*inode_setxattr) (struct dentry *dentry, const char *name,
diff --git a/security/capability.c b/security/capability.c
index bbb51156261b..2984ea4f776f 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -181,7 +181,7 @@ static int cap_inode_follow_link(struct dentry *dentry,
 	return 0;
 }
 
-static int cap_inode_permission(struct inode *inode, int mask, unsigned flags)
+static int cap_inode_permission(struct inode *inode, int mask)
 {
 	return 0;
 }
diff --git a/security/security.c b/security/security.c
index 4ba6d4cc061f..db3b750da353 100644
--- a/security/security.c
+++ b/security/security.c
@@ -518,14 +518,17 @@ int security_inode_permission(struct inode *inode, int mask)
 {
 	if (unlikely(IS_PRIVATE(inode)))
 		return 0;
-	return security_ops->inode_permission(inode, mask, 0);
+	return security_ops->inode_permission(inode, mask);
 }
 
 int security_inode_exec_permission(struct inode *inode, unsigned int flags)
 {
+	int mask = MAY_EXEC;
 	if (unlikely(IS_PRIVATE(inode)))
 		return 0;
-	return security_ops->inode_permission(inode, MAY_EXEC, flags);
+	if (flags)
+		mask |= MAY_NOT_BLOCK;
+	return security_ops->inode_permission(inode, mask);
 }
 
 int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 20219ef5439a..47a059fff344 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -2659,12 +2659,13 @@ static int selinux_inode_follow_link(struct dentry *dentry, struct nameidata *na
 	return dentry_has_perm(cred, dentry, FILE__READ);
 }
 
-static int selinux_inode_permission(struct inode *inode, int mask, unsigned flags)
+static int selinux_inode_permission(struct inode *inode, int mask)
 {
 	const struct cred *cred = current_cred();
 	struct common_audit_data ad;
 	u32 perms;
 	bool from_access;
+	unsigned __flags = mask & MAY_NOT_BLOCK ? IPERM_FLAG_RCU : 0;
 
 	from_access = mask & MAY_ACCESS;
 	mask &= (MAY_READ|MAY_WRITE|MAY_EXEC|MAY_APPEND);
@@ -2681,7 +2682,7 @@ static int selinux_inode_permission(struct inode *inode, int mask, unsigned flag
 
 	perms = file_mask_to_av(inode->i_mode, mask);
 
-	return inode_has_perm(cred, inode, perms, &ad, flags);
+	return inode_has_perm(cred, inode, perms, &ad, __flags);
 }
 
 static int selinux_inode_setattr(struct dentry *dentry, struct iattr *iattr)
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 9831a39c11f6..f375eb2e1957 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -689,9 +689,10 @@ static int smack_inode_rename(struct inode *old_inode,
  *
  * Returns 0 if access is permitted, -EACCES otherwise
  */
-static int smack_inode_permission(struct inode *inode, int mask, unsigned flags)
+static int smack_inode_permission(struct inode *inode, int mask)
 {
 	struct smk_audit_info ad;
+	int no_block = mask & MAY_NOT_BLOCK;
 
 	mask &= (MAY_READ|MAY_WRITE|MAY_EXEC|MAY_APPEND);
 	/*
@@ -701,7 +702,7 @@ static int smack_inode_permission(struct inode *inode, int mask, unsigned flags)
 		return 0;
 
 	/* May be droppable after audit */
-	if (flags & IPERM_FLAG_RCU)
+	if (no_block)
 		return -ECHILD;
 	smk_ad_init(&ad, __func__, LSM_AUDIT_DATA_INODE);
 	smk_ad_setfield_u_fs_inode(&ad, inode);
-- 
cgit v1.2.3


From eecdd358b467405a084d400d5ec571bbdbfe97a3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 20 Jun 2011 19:48:41 -0400
Subject: ->permission() sanitizing: don't pass flags to exec_permission()

pass mask instead; kill security_inode_exec_permission() since we can use
security_inode_permission() instead.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c               | 17 +++++++----------
 include/linux/security.h |  7 -------
 security/security.c      | 10 ----------
 3 files changed, 7 insertions(+), 27 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index c5c382620a86..21eba95368f2 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -304,7 +304,7 @@ int inode_permission(struct inode *inode, int mask)
 /**
  * exec_permission  -  check for right to do lookups in a given directory
  * @inode:	inode to check permission on
- * @flags:	IPERM_FLAG_ flags.
+ * @mask:	MAY_EXEC and possibly MAY_NOT_BLOCK flags.
  *
  * Short-cut version of inode_permission(), for calling on directories
  * during pathname resolution.  Combines parts of inode_permission()
@@ -314,13 +314,10 @@ int inode_permission(struct inode *inode, int mask)
  * short-cut DAC fails, then call ->permission() to do more
  * complete permission check.
  */
-static inline int exec_permission(struct inode *inode, unsigned int flags)
+static inline int exec_permission(struct inode *inode, int mask)
 {
 	int ret;
 	struct user_namespace *ns = inode_userns(inode);
-	int mask = MAY_EXEC;
-	if (flags & IPERM_FLAG_RCU)
-		mask |= MAY_NOT_BLOCK;
 
 	if (inode->i_op->permission) {
 		ret = inode->i_op->permission(inode, mask);
@@ -338,7 +335,7 @@ static inline int exec_permission(struct inode *inode, unsigned int flags)
 	}
 	return ret;
 ok:
-	return security_inode_exec_permission(inode, flags);
+	return security_inode_permission(inode, mask);
 }
 
 /**
@@ -1214,13 +1211,13 @@ retry:
 static inline int may_lookup(struct nameidata *nd)
 {
 	if (nd->flags & LOOKUP_RCU) {
-		int err = exec_permission(nd->inode, IPERM_FLAG_RCU);
+		int err = exec_permission(nd->inode, MAY_EXEC|MAY_NOT_BLOCK);
 		if (err != -ECHILD)
 			return err;
 		if (unlazy_walk(nd, NULL))
 			return -ECHILD;
 	}
-	return exec_permission(nd->inode, 0);
+	return exec_permission(nd->inode, MAY_EXEC);
 }
 
 static inline int handle_dots(struct nameidata *nd, int type)
@@ -1495,7 +1492,7 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 			if (!S_ISDIR(dentry->d_inode->i_mode))
 				goto fput_fail;
 
-			retval = exec_permission(dentry->d_inode, 0);
+			retval = exec_permission(dentry->d_inode, MAY_EXEC);
 			if (retval)
 				goto fput_fail;
 		}
@@ -1652,7 +1649,7 @@ static struct dentry *__lookup_hash(struct qstr *name,
 	struct dentry *dentry;
 	int err;
 
-	err = exec_permission(inode, 0);
+	err = exec_permission(inode, MAY_EXEC);
 	if (err)
 		return ERR_PTR(err);
 
diff --git a/include/linux/security.h b/include/linux/security.h
index ca02f1716736..ebd2a53a3d07 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1720,7 +1720,6 @@ int security_inode_rename(struct inode *old_dir, struct dentry *old_dentry,
 int security_inode_readlink(struct dentry *dentry);
 int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd);
 int security_inode_permission(struct inode *inode, int mask);
-int security_inode_exec_permission(struct inode *inode, unsigned int flags);
 int security_inode_setattr(struct dentry *dentry, struct iattr *attr);
 int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry);
 int security_inode_setxattr(struct dentry *dentry, const char *name,
@@ -2113,12 +2112,6 @@ static inline int security_inode_permission(struct inode *inode, int mask)
 	return 0;
 }
 
-static inline int security_inode_exec_permission(struct inode *inode,
-						  unsigned int flags)
-{
-	return 0;
-}
-
 static inline int security_inode_setattr(struct dentry *dentry,
 					  struct iattr *attr)
 {
diff --git a/security/security.c b/security/security.c
index db3b750da353..0e4fccfef12c 100644
--- a/security/security.c
+++ b/security/security.c
@@ -521,16 +521,6 @@ int security_inode_permission(struct inode *inode, int mask)
 	return security_ops->inode_permission(inode, mask);
 }
 
-int security_inode_exec_permission(struct inode *inode, unsigned int flags)
-{
-	int mask = MAY_EXEC;
-	if (unlikely(IS_PRIVATE(inode)))
-		return 0;
-	if (flags)
-		mask |= MAY_NOT_BLOCK;
-	return security_ops->inode_permission(inode, mask);
-}
-
 int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	if (unlikely(IS_PRIVATE(dentry->d_inode)))
-- 
cgit v1.2.3


From 729cdb3a1ee03a4363f9c7e66ddd979727e99e1f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 21 Jun 2011 01:01:22 -0400
Subject: kill IPERM_FLAG_RCU

not used anymore

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/fs.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1fdefe3995fd..8494aac189f0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1573,8 +1573,6 @@ struct file_operations {
 			  loff_t len);
 };
 
-#define IPERM_FLAG_RCU	0x0001
-
 struct inode_operations {
 	struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *);
 	void * (*follow_link) (struct dentry *, struct nameidata *);
-- 
cgit v1.2.3


From 3d4ff43d895c50319af45eb4bf04a4618eccdf76 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 22 Jun 2011 18:40:12 -0400
Subject: nfs_open_context doesn't need struct path either

just dentry, please...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/nfs/dir.c           | 14 +++++---------
 fs/nfs/direct.c        |  4 ++--
 fs/nfs/inode.c         | 20 +++++++++++---------
 fs/nfs/nfs4proc.c      | 10 +++++-----
 fs/nfs/pagelist.c      |  4 ++--
 fs/nfs/read.c          |  8 ++++----
 fs/nfs/write.c         | 22 +++++++++++-----------
 include/linux/nfs_fs.h |  4 ++--
 8 files changed, 42 insertions(+), 44 deletions(-)

(limited to 'include')

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index a86acd6d0cb4..a0693f3f11c0 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1345,10 +1345,6 @@ static int is_atomic_open(struct nameidata *nd)
 
 static struct nfs_open_context *nameidata_to_nfs_open_context(struct dentry *dentry, struct nameidata *nd)
 {
-	struct path path = {
-		.mnt = nd->path.mnt,
-		.dentry = dentry,
-	};
 	struct nfs_open_context *ctx;
 	struct rpc_cred *cred;
 	fmode_t fmode = nd->intent.open.flags & (FMODE_READ | FMODE_WRITE | FMODE_EXEC);
@@ -1356,7 +1352,7 @@ static struct nfs_open_context *nameidata_to_nfs_open_context(struct dentry *den
 	cred = rpc_lookup_cred();
 	if (IS_ERR(cred))
 		return ERR_CAST(cred);
-	ctx = alloc_nfs_open_context(&path, cred, fmode);
+	ctx = alloc_nfs_open_context(dentry, cred, fmode);
 	put_rpccred(cred);
 	if (ctx == NULL)
 		return ERR_PTR(-ENOMEM);
@@ -1376,13 +1372,13 @@ static int nfs_intent_set_file(struct nameidata *nd, struct nfs_open_context *ct
 
 	/* If the open_intent is for execute, we have an extra check to make */
 	if (ctx->mode & FMODE_EXEC) {
-		ret = nfs_may_open(ctx->path.dentry->d_inode,
+		ret = nfs_may_open(ctx->dentry->d_inode,
 				ctx->cred,
 				nd->intent.open.flags);
 		if (ret < 0)
 			goto out;
 	}
-	filp = lookup_instantiate_filp(nd, ctx->path.dentry, do_open);
+	filp = lookup_instantiate_filp(nd, ctx->dentry, do_open);
 	if (IS_ERR(filp))
 		ret = PTR_ERR(filp);
 	else
@@ -1463,8 +1459,8 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry
 	res = d_add_unique(dentry, inode);
 	nfs_unblock_sillyrename(dentry->d_parent);
 	if (res != NULL) {
-		dput(ctx->path.dentry);
-		ctx->path.dentry = dget(res);
+		dput(ctx->dentry);
+		ctx->dentry = dget(res);
 		dentry = res;
 	}
 	err = nfs_intent_set_file(nd, ctx);
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 8eea25366717..b35d25b98da6 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -284,7 +284,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 						loff_t pos)
 {
 	struct nfs_open_context *ctx = dreq->ctx;
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode = ctx->dentry->d_inode;
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
 	size_t rsize = NFS_SERVER(inode)->rsize;
@@ -715,7 +715,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 						 loff_t pos, int sync)
 {
 	struct nfs_open_context *ctx = dreq->ctx;
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode = ctx->dentry->d_inode;
 	unsigned long user_addr = (unsigned long)iov->iov_base;
 	size_t count = iov->iov_len;
 	struct rpc_task *task;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 6f4850deb272..fe1203797b2b 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -567,7 +567,7 @@ static struct nfs_lock_context *__nfs_find_lock_context(struct nfs_open_context
 struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx)
 {
 	struct nfs_lock_context *res, *new = NULL;
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode = ctx->dentry->d_inode;
 
 	spin_lock(&inode->i_lock);
 	res = __nfs_find_lock_context(ctx);
@@ -594,7 +594,7 @@ struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx)
 void nfs_put_lock_context(struct nfs_lock_context *l_ctx)
 {
 	struct nfs_open_context *ctx = l_ctx->open_context;
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode = ctx->dentry->d_inode;
 
 	if (!atomic_dec_and_lock(&l_ctx->count, &inode->i_lock))
 		return;
@@ -620,7 +620,7 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
 		return;
 	if (!is_sync)
 		return;
-	inode = ctx->path.dentry->d_inode;
+	inode = ctx->dentry->d_inode;
 	if (!list_empty(&NFS_I(inode)->open_files))
 		return;
 	server = NFS_SERVER(inode);
@@ -629,14 +629,14 @@ void nfs_close_context(struct nfs_open_context *ctx, int is_sync)
 	nfs_revalidate_inode(server, inode);
 }
 
-struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cred *cred, fmode_t f_mode)
+struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred, fmode_t f_mode)
 {
 	struct nfs_open_context *ctx;
 
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
 	if (ctx != NULL) {
-		ctx->path = *path;
-		path_get(&ctx->path);
+		nfs_sb_active(dentry->d_sb);
+		ctx->dentry = dget(dentry);
 		ctx->cred = get_rpccred(cred);
 		ctx->state = NULL;
 		ctx->mode = f_mode;
@@ -658,7 +658,8 @@ struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx)
 
 static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync)
 {
-	struct inode *inode = ctx->path.dentry->d_inode;
+	struct inode *inode = ctx->dentry->d_inode;
+	struct super_block *sb = ctx->dentry->d_sb;
 
 	if (!list_empty(&ctx->list)) {
 		if (!atomic_dec_and_lock(&ctx->lock_context.count, &inode->i_lock))
@@ -671,7 +672,8 @@ static void __put_nfs_open_context(struct nfs_open_context *ctx, int is_sync)
 		NFS_PROTO(inode)->close_context(ctx, is_sync);
 	if (ctx->cred != NULL)
 		put_rpccred(ctx->cred);
-	path_put(&ctx->path);
+	dput(ctx->dentry);
+	nfs_sb_deactive(sb);
 	kfree(ctx);
 }
 
@@ -741,7 +743,7 @@ int nfs_open(struct inode *inode, struct file *filp)
 	cred = rpc_lookup_cred();
 	if (IS_ERR(cred))
 		return PTR_ERR(cred);
-	ctx = alloc_nfs_open_context(&filp->f_path, cred, filp->f_mode);
+	ctx = alloc_nfs_open_context(filp->f_path.dentry, cred, filp->f_mode);
 	put_rpccred(cred);
 	if (ctx == NULL)
 		return -ENOMEM;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index aef22fa0a1c0..26bece8f3083 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1132,7 +1132,7 @@ static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context
 {
 	struct nfs4_opendata *opendata;
 
-	opendata = nfs4_opendata_alloc(ctx->path.dentry, state->owner, 0, 0, NULL, GFP_NOFS);
+	opendata = nfs4_opendata_alloc(ctx->dentry, state->owner, 0, 0, NULL, GFP_NOFS);
 	if (opendata == NULL)
 		return ERR_PTR(-ENOMEM);
 	opendata->state = state;
@@ -1650,7 +1650,7 @@ static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *s
 		return PTR_ERR(opendata);
 	ret = nfs4_open_recover(opendata, state);
 	if (ret == -ESTALE)
-		d_drop(ctx->path.dentry);
+		d_drop(ctx->dentry);
 	nfs4_opendata_put(opendata);
 	return ret;
 }
@@ -2081,7 +2081,7 @@ nfs4_atomic_open(struct inode *dir, struct nfs_open_context *ctx, int open_flags
 	struct nfs4_state *state;
 
 	/* Protect against concurrent sillydeletes */
-	state = nfs4_do_open(dir, ctx->path.dentry, ctx->mode, open_flags, attr, ctx->cred);
+	state = nfs4_do_open(dir, ctx->dentry, ctx->mode, open_flags, attr, ctx->cred);
 	if (IS_ERR(state))
 		return ERR_CAST(state);
 	ctx->state = state;
@@ -2625,7 +2625,7 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr,
 
 	if (ctx != NULL) {
 		cred = ctx->cred;
-		de = ctx->path.dentry;
+		de = ctx->dentry;
 		fmode = ctx->mode;
 	}
 	sattr->ia_mode &= ~current_umask();
@@ -4292,7 +4292,7 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 		memcpy(data->lsp->ls_stateid.data, data->res.stateid.data,
 					sizeof(data->lsp->ls_stateid.data));
 		data->lsp->ls_flags |= NFS_LOCK_INITIALIZED;
-		renew_lease(NFS_SERVER(data->ctx->path.dentry->d_inode), data->timestamp);
+		renew_lease(NFS_SERVER(data->ctx->dentry->d_inode), data->timestamp);
 	}
 out:
 	dprintk("%s: done, ret = %d!\n", __func__, data->rpc_status);
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 009855716286..18449f43c568 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -114,7 +114,7 @@ int nfs_set_page_tag_locked(struct nfs_page *req)
 	if (!nfs_lock_request_dontget(req))
 		return 0;
 	if (test_bit(PG_MAPPED, &req->wb_flags))
-		radix_tree_tag_set(&NFS_I(req->wb_context->path.dentry->d_inode)->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
+		radix_tree_tag_set(&NFS_I(req->wb_context->dentry->d_inode)->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED);
 	return 1;
 }
 
@@ -124,7 +124,7 @@ int nfs_set_page_tag_locked(struct nfs_page *req)
 void nfs_clear_page_tag_locked(struct nfs_page *req)
 {
 	if (test_bit(PG_MAPPED, &req->wb_flags)) {
-		struct inode *inode = req->wb_context->path.dentry->d_inode;
+		struct inode *inode = req->wb_context->dentry->d_inode;
 		struct nfs_inode *nfsi = NFS_I(inode);
 
 		spin_lock(&inode->i_lock);
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 20a7f952e244..a68679f538fc 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -144,7 +144,7 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
 
 static void nfs_readpage_release(struct nfs_page *req)
 {
-	struct inode *d_inode = req->wb_context->path.dentry->d_inode;
+	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);
@@ -152,8 +152,8 @@ static void nfs_readpage_release(struct nfs_page *req)
 	unlock_page(req->wb_page);
 
 	dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
-			req->wb_context->path.dentry->d_inode->i_sb->s_id,
-			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
+			req->wb_context->dentry->d_inode->i_sb->s_id,
+			(long long)NFS_FILEID(req->wb_context->dentry->d_inode),
 			req->wb_bytes,
 			(long long)req_offset(req));
 	nfs_release_request(req);
@@ -207,7 +207,7 @@ static int nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
 		unsigned int count, unsigned int offset,
 		struct pnfs_layout_segment *lseg)
 {
-	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	struct inode *inode = req->wb_context->dentry->d_inode;
 
 	data->req	  = req;
 	data->inode	  = inode;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 727168059684..08579312c57b 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -409,7 +409,7 @@ out:
  */
 static void nfs_inode_remove_request(struct nfs_page *req)
 {
-	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	struct inode *inode = req->wb_context->dentry->d_inode;
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	BUG_ON (!NFS_WBACK_BUSY(req));
@@ -438,7 +438,7 @@ nfs_mark_request_dirty(struct nfs_page *req)
 static void
 nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg)
 {
-	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	struct inode *inode = req->wb_context->dentry->d_inode;
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	spin_lock(&inode->i_lock);
@@ -852,13 +852,13 @@ static int nfs_write_rpcsetup(struct nfs_page *req,
 		struct pnfs_layout_segment *lseg,
 		int how)
 {
-	struct inode *inode = req->wb_context->path.dentry->d_inode;
+	struct inode *inode = req->wb_context->dentry->d_inode;
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
 
 	data->req = req;
-	data->inode = inode = req->wb_context->path.dentry->d_inode;
+	data->inode = inode = req->wb_context->dentry->d_inode;
 	data->cred = req->wb_context->cred;
 	data->lseg = get_lseg(lseg);
 
@@ -1053,9 +1053,9 @@ static void nfs_writeback_done_partial(struct rpc_task *task, void *calldata)
 
 	dprintk("NFS: %5u write(%s/%lld %d@%lld)",
 		task->tk_pid,
-		data->req->wb_context->path.dentry->d_inode->i_sb->s_id,
+		data->req->wb_context->dentry->d_inode->i_sb->s_id,
 		(long long)
-		  NFS_FILEID(data->req->wb_context->path.dentry->d_inode),
+		  NFS_FILEID(data->req->wb_context->dentry->d_inode),
 		data->req->wb_bytes, (long long)req_offset(data->req));
 
 	nfs_writeback_done(task, data);
@@ -1148,8 +1148,8 @@ static void nfs_writeback_release_full(void *calldata)
 
 		dprintk("NFS: %5u write (%s/%lld %d@%lld)",
 			data->task.tk_pid,
-			req->wb_context->path.dentry->d_inode->i_sb->s_id,
-			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
+			req->wb_context->dentry->d_inode->i_sb->s_id,
+			(long long)NFS_FILEID(req->wb_context->dentry->d_inode),
 			req->wb_bytes,
 			(long long)req_offset(req));
 
@@ -1347,7 +1347,7 @@ void nfs_init_commit(struct nfs_write_data *data,
 			    struct pnfs_layout_segment *lseg)
 {
 	struct nfs_page *first = nfs_list_entry(head->next);
-	struct inode *inode = first->wb_context->path.dentry->d_inode;
+	struct inode *inode = first->wb_context->dentry->d_inode;
 
 	/* Set up the RPC argument and reply structs
 	 * NB: take care not to mess about with data->commit et al. */
@@ -1435,8 +1435,8 @@ void nfs_commit_release_pages(struct nfs_write_data *data)
 		nfs_clear_request_commit(req);
 
 		dprintk("NFS:       commit (%s/%lld %d@%lld)",
-			req->wb_context->path.dentry->d_inode->i_sb->s_id,
-			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
+			req->wb_context->dentry->d_sb->s_id,
+			(long long)NFS_FILEID(req->wb_context->dentry->d_inode),
 			req->wb_bytes,
 			(long long)req_offset(req));
 		if (status < 0) {
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 86014811a0fb..8b579beb6358 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -85,7 +85,7 @@ struct nfs_lock_context {
 struct nfs4_state;
 struct nfs_open_context {
 	struct nfs_lock_context lock_context;
-	struct path path;
+	struct dentry *dentry;
 	struct rpc_cred *cred;
 	struct nfs4_state *state;
 	fmode_t mode;
@@ -372,7 +372,7 @@ extern void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr);
 extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx);
 extern void put_nfs_open_context(struct nfs_open_context *ctx);
 extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, fmode_t mode);
-extern struct nfs_open_context *alloc_nfs_open_context(struct path *path, struct rpc_cred *cred, fmode_t f_mode);
+extern struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred, fmode_t f_mode);
 extern void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx);
 extern struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx);
 extern void nfs_put_lock_context(struct nfs_lock_context *l_ctx);
-- 
cgit v1.2.3


From 49084c3bb2055c401f3493c13edae14d49128ca0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 25 Jun 2011 21:59:52 -0400
Subject: kill LOOKUP_CONTINUE

LOOKUP_PARENT is equivalent to it now

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c            | 11 +++--------
 include/linux/namei.h |  1 -
 2 files changed, 3 insertions(+), 9 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index 5e65f67ee926..f49d6abfa799 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -664,7 +664,7 @@ static int follow_automount(struct path *path, unsigned flags,
 	/* We don't want to mount if someone supplied AT_NO_AUTOMOUNT
 	 * and this is the terminal part of the path.
 	 */
-	if ((flags & LOOKUP_NO_AUTOMOUNT) && !(flags & LOOKUP_CONTINUE))
+	if ((flags & LOOKUP_NO_AUTOMOUNT) && !(flags & LOOKUP_PARENT))
 		return -EISDIR; /* we actually want to stop here */
 
 	/* We want to mount if someone is trying to open/create a file of any
@@ -676,7 +676,7 @@ static int follow_automount(struct path *path, unsigned flags,
 	 * appended a '/' to the name.
 	 */
 	if (!(flags & LOOKUP_FOLLOW) &&
-	    !(flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY |
+	    !(flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY |
 		       LOOKUP_OPEN | LOOKUP_CREATE)))
 		return -EISDIR;
 
@@ -695,7 +695,7 @@ static int follow_automount(struct path *path, unsigned flags,
 		 * the path being looked up; if it wasn't then the remainder of
 		 * the path is inaccessible and we should say so.
 		 */
-		if (PTR_ERR(mnt) == -EISDIR && (flags & LOOKUP_CONTINUE))
+		if (PTR_ERR(mnt) == -EISDIR && (flags & LOOKUP_PARENT))
 			return -EREMOTE;
 		return PTR_ERR(mnt);
 	}
@@ -1281,7 +1281,6 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 {
 	struct path next;
 	int err;
-	unsigned int lookup_flags = nd->flags;
 	
 	while (*name=='/')
 		name++;
@@ -1295,8 +1294,6 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 		unsigned int c;
 		int type;
 
-		nd->flags |= LOOKUP_CONTINUE;
-
 		err = may_lookup(nd);
  		if (err)
 			break;
@@ -1358,8 +1355,6 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 		/* here ends the main loop */
 
 last_component:
-		/* Clear LOOKUP_CONTINUE iff it was previously unset */
-		nd->flags &= lookup_flags | ~LOOKUP_CONTINUE;
 		nd->last = this;
 		nd->last_type = type;
 		return 0;
diff --git a/include/linux/namei.h b/include/linux/namei.h
index eba45ea10298..3439ab862e1d 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -48,7 +48,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
  */
 #define LOOKUP_FOLLOW		0x0001
 #define LOOKUP_DIRECTORY	0x0002
-#define LOOKUP_CONTINUE		0x0004
 
 #define LOOKUP_PARENT		0x0010
 #define LOOKUP_REVAL		0x0020
-- 
cgit v1.2.3


From dae6ad8f37529963ae7df52baaccf056b38f210e Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 26 Jun 2011 11:50:15 -0400
Subject: new helpers: kern_path_create/user_path_create

combination of kern_path_parent() and lookup_create().  Does *not*
expose struct nameidata to caller.  Syscalls converted to that...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c              | 154 ++++++++++++++++++++++++------------------------
 fs/ocfs2/refcounttree.c |  49 ++++-----------
 include/linux/namei.h   |   2 +
 net/unix/af_unix.c      |  38 ++++++------
 4 files changed, 106 insertions(+), 137 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index f49d6abfa799..b292eb03d9d2 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2311,6 +2311,35 @@ fail:
 }
 EXPORT_SYMBOL_GPL(lookup_create);
 
+struct dentry *kern_path_create(int dfd, const char *pathname, struct path *path, int is_dir)
+{
+	struct nameidata nd;
+	struct dentry *res;
+	int error = do_path_lookup(dfd, pathname, LOOKUP_PARENT, &nd);
+	if (error)
+		return ERR_PTR(error);
+	res = lookup_create(&nd, is_dir);
+	if (IS_ERR(res)) {
+		mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+		path_put(&nd.path);
+	}
+	*path = nd.path;
+	return res;
+}
+EXPORT_SYMBOL(kern_path_create);
+
+struct dentry *user_path_create(int dfd, const char __user *pathname, struct path *path, int is_dir)
+{
+	char *tmp = getname(pathname);
+	struct dentry *res;
+	if (IS_ERR(tmp))
+		return ERR_CAST(tmp);
+	res = kern_path_create(dfd, tmp, path, is_dir);
+	putname(tmp);
+	return res;
+}
+EXPORT_SYMBOL(user_path_create);
+
 int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
 {
 	int error = may_create(dir, dentry);
@@ -2359,54 +2388,46 @@ static int may_mknod(mode_t mode)
 SYSCALL_DEFINE4(mknodat, int, dfd, const char __user *, filename, int, mode,
 		unsigned, dev)
 {
-	int error;
-	char *tmp;
 	struct dentry *dentry;
-	struct nameidata nd;
+	struct path path;
+	int error;
 
 	if (S_ISDIR(mode))
 		return -EPERM;
 
-	error = user_path_parent(dfd, filename, &nd, &tmp);
-	if (error)
-		return error;
+	dentry = user_path_create(dfd, filename, &path, 0);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
 
-	dentry = lookup_create(&nd, 0);
-	if (IS_ERR(dentry)) {
-		error = PTR_ERR(dentry);
-		goto out_unlock;
-	}
-	if (!IS_POSIXACL(nd.path.dentry->d_inode))
+	if (!IS_POSIXACL(path.dentry->d_inode))
 		mode &= ~current_umask();
 	error = may_mknod(mode);
 	if (error)
 		goto out_dput;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_dput;
-	error = security_path_mknod(&nd.path, dentry, mode, dev);
+	error = security_path_mknod(&path, dentry, mode, dev);
 	if (error)
 		goto out_drop_write;
 	switch (mode & S_IFMT) {
 		case 0: case S_IFREG:
-			error = vfs_create(nd.path.dentry->d_inode,dentry,mode,NULL);
+			error = vfs_create(path.dentry->d_inode,dentry,mode,NULL);
 			break;
 		case S_IFCHR: case S_IFBLK:
-			error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,
+			error = vfs_mknod(path.dentry->d_inode,dentry,mode,
 					new_decode_dev(dev));
 			break;
 		case S_IFIFO: case S_IFSOCK:
-			error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,0);
+			error = vfs_mknod(path.dentry->d_inode,dentry,mode,0);
 			break;
 	}
 out_drop_write:
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(path.mnt);
 out_dput:
 	dput(dentry);
-out_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-	path_put(&nd.path);
-	putname(tmp);
+	mutex_unlock(&path.dentry->d_inode->i_mutex);
+	path_put(&path);
 
 	return error;
 }
@@ -2439,38 +2460,29 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 
 SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, int, mode)
 {
-	int error = 0;
-	char * tmp;
 	struct dentry *dentry;
-	struct nameidata nd;
-
-	error = user_path_parent(dfd, pathname, &nd, &tmp);
-	if (error)
-		goto out_err;
+	struct path path;
+	int error;
 
-	dentry = lookup_create(&nd, 1);
-	error = PTR_ERR(dentry);
+	dentry = user_path_create(dfd, pathname, &path, 1);
 	if (IS_ERR(dentry))
-		goto out_unlock;
+		return PTR_ERR(dentry);
 
-	if (!IS_POSIXACL(nd.path.dentry->d_inode))
+	if (!IS_POSIXACL(path.dentry->d_inode))
 		mode &= ~current_umask();
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_dput;
-	error = security_path_mkdir(&nd.path, dentry, mode);
+	error = security_path_mkdir(&path, dentry, mode);
 	if (error)
 		goto out_drop_write;
-	error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode);
+	error = vfs_mkdir(path.dentry->d_inode, dentry, mode);
 out_drop_write:
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(path.mnt);
 out_dput:
 	dput(dentry);
-out_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-	path_put(&nd.path);
-	putname(tmp);
-out_err:
+	mutex_unlock(&path.dentry->d_inode->i_mutex);
+	path_put(&path);
 	return error;
 }
 
@@ -2730,38 +2742,31 @@ SYSCALL_DEFINE3(symlinkat, const char __user *, oldname,
 {
 	int error;
 	char *from;
-	char *to;
 	struct dentry *dentry;
-	struct nameidata nd;
+	struct path path;
 
 	from = getname(oldname);
 	if (IS_ERR(from))
 		return PTR_ERR(from);
 
-	error = user_path_parent(newdfd, newname, &nd, &to);
-	if (error)
-		goto out_putname;
-
-	dentry = lookup_create(&nd, 0);
+	dentry = user_path_create(newdfd, newname, &path, 0);
 	error = PTR_ERR(dentry);
 	if (IS_ERR(dentry))
-		goto out_unlock;
+		goto out_putname;
 
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_dput;
-	error = security_path_symlink(&nd.path, dentry, from);
+	error = security_path_symlink(&path, dentry, from);
 	if (error)
 		goto out_drop_write;
-	error = vfs_symlink(nd.path.dentry->d_inode, dentry, from);
+	error = vfs_symlink(path.dentry->d_inode, dentry, from);
 out_drop_write:
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(path.mnt);
 out_dput:
 	dput(dentry);
-out_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-	path_put(&nd.path);
-	putname(to);
+	mutex_unlock(&path.dentry->d_inode->i_mutex);
+	path_put(&path);
 out_putname:
 	putname(from);
 	return error;
@@ -2826,11 +2831,9 @@ SYSCALL_DEFINE5(linkat, int, olddfd, const char __user *, oldname,
 		int, newdfd, const char __user *, newname, int, flags)
 {
 	struct dentry *new_dentry;
-	struct nameidata nd;
-	struct path old_path;
+	struct path old_path, new_path;
 	int how = 0;
 	int error;
-	char *to;
 
 	if ((flags & ~(AT_SYMLINK_FOLLOW | AT_EMPTY_PATH)) != 0)
 		return -EINVAL;
@@ -2852,32 +2855,27 @@ SYSCALL_DEFINE5(linkat, int, olddfd, const char __user *, oldname,
 	if (error)
 		return error;
 
-	error = user_path_parent(newdfd, newname, &nd, &to);
-	if (error)
-		goto out;
-	error = -EXDEV;
-	if (old_path.mnt != nd.path.mnt)
-		goto out_release;
-	new_dentry = lookup_create(&nd, 0);
+	new_dentry = user_path_create(newdfd, newname, &new_path, 0);
 	error = PTR_ERR(new_dentry);
 	if (IS_ERR(new_dentry))
-		goto out_unlock;
-	error = mnt_want_write(nd.path.mnt);
+		goto out;
+
+	error = -EXDEV;
+	if (old_path.mnt != new_path.mnt)
+		goto out_dput;
+	error = mnt_want_write(new_path.mnt);
 	if (error)
 		goto out_dput;
-	error = security_path_link(old_path.dentry, &nd.path, new_dentry);
+	error = security_path_link(old_path.dentry, &new_path, new_dentry);
 	if (error)
 		goto out_drop_write;
-	error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry);
+	error = vfs_link(old_path.dentry, new_path.dentry->d_inode, new_dentry);
 out_drop_write:
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(new_path.mnt);
 out_dput:
 	dput(new_dentry);
-out_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-out_release:
-	path_put(&nd.path);
-	putname(to);
+	mutex_unlock(&new_path.dentry->d_inode->i_mutex);
+	path_put(&new_path);
 out:
 	path_put(&old_path);
 
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index ebfd3825f12a..cf7823382664 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -4368,25 +4368,6 @@ static inline int ocfs2_may_create(struct inode *dir, struct dentry *child)
 	return inode_permission(dir, MAY_WRITE | MAY_EXEC);
 }
 
-/* copied from user_path_parent. */
-static int ocfs2_user_path_parent(const char __user *path,
-				  struct nameidata *nd, char **name)
-{
-	char *s = getname(path);
-	int error;
-
-	if (IS_ERR(s))
-		return PTR_ERR(s);
-
-	error = kern_path_parent(s, nd);
-	if (error)
-		putname(s);
-	else
-		*name = s;
-
-	return error;
-}
-
 /**
  * ocfs2_vfs_reflink - Create a reference-counted link
  *
@@ -4460,10 +4441,8 @@ int ocfs2_reflink_ioctl(struct inode *inode,
 			bool preserve)
 {
 	struct dentry *new_dentry;
-	struct nameidata nd;
-	struct path old_path;
+	struct path old_path, new_path;
 	int error;
-	char *to = NULL;
 
 	if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb)))
 		return -EOPNOTSUPP;
@@ -4474,39 +4453,33 @@ int ocfs2_reflink_ioctl(struct inode *inode,
 		return error;
 	}
 
-	error = ocfs2_user_path_parent(newname, &nd, &to);
-	if (error) {
+	new_dentry = user_path_create(AT_FDCWD, newname, &new_path, 0);
+	error = PTR_ERR(new_dentry);
+	if (IS_ERR(new_dentry)) {
 		mlog_errno(error);
 		goto out;
 	}
 
 	error = -EXDEV;
-	if (old_path.mnt != nd.path.mnt)
-		goto out_release;
-	new_dentry = lookup_create(&nd, 0);
-	error = PTR_ERR(new_dentry);
-	if (IS_ERR(new_dentry)) {
+	if (old_path.mnt != new_path.mnt) {
 		mlog_errno(error);
-		goto out_unlock;
+		goto out_dput;
 	}
 
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(new_path.mnt);
 	if (error) {
 		mlog_errno(error);
 		goto out_dput;
 	}
 
 	error = ocfs2_vfs_reflink(old_path.dentry,
-				  nd.path.dentry->d_inode,
+				  new_path.dentry->d_inode,
 				  new_dentry, preserve);
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(new_path.mnt);
 out_dput:
 	dput(new_dentry);
-out_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-out_release:
-	path_put(&nd.path);
-	putname(to);
+	mutex_unlock(&new_path.dentry->d_inode->i_mutex);
+	path_put(&new_path);
 out:
 	path_put(&old_path);
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 3439ab862e1d..b8cea804d31a 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -74,6 +74,8 @@ extern int user_path_at(int, const char __user *, unsigned, struct path *);
 
 extern int kern_path(const char *, unsigned, struct path *);
 
+extern struct dentry *kern_path_create(int, const char *, struct path *, int);
+extern struct dentry *user_path_create(int, const char __user *, struct path *, int);
 extern int kern_path_parent(const char *, struct nameidata *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct nameidata *);
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 0722a25a3a33..ec68e1c05b85 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -808,8 +808,9 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	struct net *net = sock_net(sk);
 	struct unix_sock *u = unix_sk(sk);
 	struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr;
+	char *sun_path = sunaddr->sun_path;
 	struct dentry *dentry = NULL;
-	struct nameidata nd;
+	struct path path;
 	int err;
 	unsigned hash;
 	struct unix_address *addr;
@@ -845,48 +846,44 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 	addr->hash = hash ^ sk->sk_type;
 	atomic_set(&addr->refcnt, 1);
 
-	if (sunaddr->sun_path[0]) {
+	if (sun_path[0]) {
 		unsigned int mode;
 		err = 0;
 		/*
 		 * Get the parent directory, calculate the hash for last
 		 * component.
 		 */
-		err = kern_path_parent(sunaddr->sun_path, &nd);
-		if (err)
-			goto out_mknod_parent;
-
-		dentry = lookup_create(&nd, 0);
+		dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0);
 		err = PTR_ERR(dentry);
 		if (IS_ERR(dentry))
-			goto out_mknod_unlock;
+			goto out_mknod_parent;
 
 		/*
 		 * All right, let's create it.
 		 */
 		mode = S_IFSOCK |
 		       (SOCK_INODE(sock)->i_mode & ~current_umask());
-		err = mnt_want_write(nd.path.mnt);
+		err = mnt_want_write(path.mnt);
 		if (err)
 			goto out_mknod_dput;
-		err = security_path_mknod(&nd.path, dentry, mode, 0);
+		err = security_path_mknod(&path, dentry, mode, 0);
 		if (err)
 			goto out_mknod_drop_write;
-		err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0);
+		err = vfs_mknod(path.dentry->d_inode, dentry, mode, 0);
 out_mknod_drop_write:
-		mnt_drop_write(nd.path.mnt);
+		mnt_drop_write(path.mnt);
 		if (err)
 			goto out_mknod_dput;
-		mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-		dput(nd.path.dentry);
-		nd.path.dentry = dentry;
+		mutex_unlock(&path.dentry->d_inode->i_mutex);
+		dput(path.dentry);
+		path.dentry = dentry;
 
 		addr->hash = UNIX_HASH_SIZE;
 	}
 
 	spin_lock(&unix_table_lock);
 
-	if (!sunaddr->sun_path[0]) {
+	if (!sun_path[0]) {
 		err = -EADDRINUSE;
 		if (__unix_find_socket_byname(net, sunaddr, addr_len,
 					      sk->sk_type, hash)) {
@@ -897,8 +894,8 @@ out_mknod_drop_write:
 		list = &unix_socket_table[addr->hash];
 	} else {
 		list = &unix_socket_table[dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1)];
-		u->dentry = nd.path.dentry;
-		u->mnt    = nd.path.mnt;
+		u->dentry = path.dentry;
+		u->mnt    = path.mnt;
 	}
 
 	err = 0;
@@ -915,9 +912,8 @@ out:
 
 out_mknod_dput:
 	dput(dentry);
-out_mknod_unlock:
-	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-	path_put(&nd.path);
+	mutex_unlock(&path.dentry->d_inode->i_mutex);
+	path_put(&path);
 out_mknod_parent:
 	if (err == -EEXIST)
 		err = -EADDRINUSE;
-- 
cgit v1.2.3


From 6657719390cd05be45f4e3b501d8bb46889c0a19 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 28 Jun 2011 15:41:10 -0400
Subject: make sure that nsproxy_cache is initialized early enough

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/nsproxy.h | 1 +
 kernel/fork.c           | 1 +
 kernel/nsproxy.c        | 4 +---
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include')

diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h
index 50d20aba57d3..cc37a55ad004 100644
--- a/include/linux/nsproxy.h
+++ b/include/linux/nsproxy.h
@@ -68,6 +68,7 @@ void switch_task_namespaces(struct task_struct *tsk, struct nsproxy *new);
 void free_nsproxy(struct nsproxy *ns);
 int unshare_nsproxy_namespaces(unsigned long, struct nsproxy **,
 	struct fs_struct *);
+int __init nsproxy_cache_init(void);
 
 static inline void put_nsproxy(struct nsproxy *ns)
 {
diff --git a/kernel/fork.c b/kernel/fork.c
index 0276c30401a0..31fa13e63b70 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1574,6 +1574,7 @@ void __init proc_caches_init(void)
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK, NULL);
 	vm_area_cachep = KMEM_CACHE(vm_area_struct, SLAB_PANIC);
 	mmap_init();
+	nsproxy_cache_init();
 }
 
 /*
diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c
index d6a00f3de15d..9aeab4b98c64 100644
--- a/kernel/nsproxy.c
+++ b/kernel/nsproxy.c
@@ -271,10 +271,8 @@ out:
 	return err;
 }
 
-static int __init nsproxy_cache_init(void)
+int __init nsproxy_cache_init(void)
 {
 	nsproxy_cachep = KMEM_CACHE(nsproxy, SLAB_PANIC);
 	return 0;
 }
-
-module_init(nsproxy_cache_init);
-- 
cgit v1.2.3


From ed75e95de574c99575e5f3e1d9ca59ea8c12a9cb Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 27 Jun 2011 16:53:43 -0400
Subject: kill lookup_create()

folded into the only caller (kern_path_create())

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c             | 54 +++++++++++++++++---------------------------------
 include/linux/dcache.h |  1 -
 2 files changed, 18 insertions(+), 37 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index b292eb03d9d2..b45a039216c7 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2258,35 +2258,29 @@ struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt,
 	return file;
 }
 
-/**
- * lookup_create - lookup a dentry, creating it if it doesn't exist
- * @nd: nameidata info
- * @is_dir: directory flag
- *
- * Simple function to lookup and return a dentry and create it
- * if it doesn't exist.  Is SMP-safe.
- *
- * Returns with nd->path.dentry->d_inode->i_mutex locked.
- */
-struct dentry *lookup_create(struct nameidata *nd, int is_dir)
+struct dentry *kern_path_create(int dfd, const char *pathname, struct path *path, int is_dir)
 {
 	struct dentry *dentry = ERR_PTR(-EEXIST);
+	struct nameidata nd;
+	int error = do_path_lookup(dfd, pathname, LOOKUP_PARENT, &nd);
+	if (error)
+		return ERR_PTR(error);
 
-	mutex_lock_nested(&nd->path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
 	/*
 	 * Yucky last component or no last component at all?
 	 * (foo/., foo/.., /////)
 	 */
-	if (nd->last_type != LAST_NORM)
-		goto fail;
-	nd->flags &= ~LOOKUP_PARENT;
-	nd->flags |= LOOKUP_CREATE | LOOKUP_EXCL;
-	nd->intent.open.flags = O_EXCL;
+	if (nd.last_type != LAST_NORM)
+		goto out;
+	nd.flags &= ~LOOKUP_PARENT;
+	nd.flags |= LOOKUP_CREATE | LOOKUP_EXCL;
+	nd.intent.open.flags = O_EXCL;
 
 	/*
 	 * Do the final lookup.
 	 */
-	dentry = lookup_hash(nd);
+	mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
+	dentry = lookup_hash(&nd);
 	if (IS_ERR(dentry))
 		goto fail;
 
@@ -2298,34 +2292,22 @@ struct dentry *lookup_create(struct nameidata *nd, int is_dir)
 	 * all is fine. Let's be bastards - you had / on the end, you've
 	 * been asking for (non-existent) directory. -ENOENT for you.
 	 */
-	if (unlikely(!is_dir && nd->last.name[nd->last.len])) {
+	if (unlikely(!is_dir && nd.last.name[nd.last.len])) {
 		dput(dentry);
 		dentry = ERR_PTR(-ENOENT);
+		goto fail;
 	}
+	*path = nd.path;
 	return dentry;
 eexist:
 	dput(dentry);
 	dentry = ERR_PTR(-EEXIST);
 fail:
+	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+out:
+	path_put(&nd.path);
 	return dentry;
 }
-EXPORT_SYMBOL_GPL(lookup_create);
-
-struct dentry *kern_path_create(int dfd, const char *pathname, struct path *path, int is_dir)
-{
-	struct nameidata nd;
-	struct dentry *res;
-	int error = do_path_lookup(dfd, pathname, LOOKUP_PARENT, &nd);
-	if (error)
-		return ERR_PTR(error);
-	res = lookup_create(&nd, is_dir);
-	if (IS_ERR(res)) {
-		mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
-		path_put(&nd.path);
-	}
-	*path = nd.path;
-	return res;
-}
 EXPORT_SYMBOL(kern_path_create);
 
 struct dentry *user_path_create(int dfd, const char __user *pathname, struct path *path, int is_dir)
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 5fa5bd33b979..3f22d8d6d8a3 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -423,7 +423,6 @@ static inline bool d_need_lookup(struct dentry *dentry)
 }
 
 extern void d_clear_need_lookup(struct dentry *dentry);
-extern struct dentry *lookup_create(struct nameidata *nd, int is_dir);
 
 extern int sysctl_vfs_cache_pressure;
 
-- 
cgit v1.2.3


From e0a0124936171af6156b80fe8ac8799f039e767f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 27 Jun 2011 17:00:37 -0400
Subject: switch vfs_path_lookup() to struct path

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c            | 16 +++++++++++-----
 fs/nfs/cache_lib.c    |  9 ++++-----
 fs/nfs/super.c        | 16 +++++-----------
 include/linux/namei.h |  2 +-
 net/sunrpc/clnt.c     | 11 +++++------
 5 files changed, 26 insertions(+), 28 deletions(-)

(limited to 'include')

diff --git a/fs/namei.c b/fs/namei.c
index b45a039216c7..7e6ba8c80e77 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1575,16 +1575,22 @@ int kern_path(const char *name, unsigned int flags, struct path *path)
  * @mnt: pointer to vfs mount of the base directory
  * @name: pointer to file name
  * @flags: lookup flags
- * @nd: pointer to nameidata
+ * @path: pointer to struct path to fill
  */
 int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt,
 		    const char *name, unsigned int flags,
-		    struct nameidata *nd)
+		    struct path *path)
 {
-	nd->root.dentry = dentry;
-	nd->root.mnt = mnt;
+	struct nameidata nd;
+	int err;
+	nd.root.dentry = dentry;
+	nd.root.mnt = mnt;
+	BUG_ON(flags & LOOKUP_PARENT);
 	/* the first argument of do_path_lookup() is ignored with LOOKUP_ROOT */
-	return do_path_lookup(AT_FDCWD, name, flags | LOOKUP_ROOT, nd);
+	err = do_path_lookup(AT_FDCWD, name, flags | LOOKUP_ROOT, &nd);
+	if (!err)
+		*path = nd.path;
+	return err;
 }
 
 static struct dentry *__lookup_hash(struct qstr *name,
diff --git a/fs/nfs/cache_lib.c b/fs/nfs/cache_lib.c
index 84690319e625..c98b439332fc 100644
--- a/fs/nfs/cache_lib.c
+++ b/fs/nfs/cache_lib.c
@@ -113,19 +113,18 @@ int nfs_cache_wait_for_upcall(struct nfs_cache_defer_req *dreq)
 
 int nfs_cache_register(struct cache_detail *cd)
 {
-	struct nameidata nd;
 	struct vfsmount *mnt;
+	struct path path;
 	int ret;
 
 	mnt = rpc_get_mount();
 	if (IS_ERR(mnt))
 		return PTR_ERR(mnt);
-	ret = vfs_path_lookup(mnt->mnt_root, mnt, "/cache", 0, &nd);
+	ret = vfs_path_lookup(mnt->mnt_root, mnt, "/cache", 0, &path);
 	if (ret)
 		goto err;
-	ret = sunrpc_cache_register_pipefs(nd.path.dentry,
-			cd->name, 0600, cd);
-	path_put(&nd.path);
+	ret = sunrpc_cache_register_pipefs(path.dentry, cd->name, 0600, cd);
+	path_put(&path);
 	if (!ret)
 		return ret;
 err:
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index ce40e5c568ba..b961ceac66b4 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2773,16 +2773,12 @@ static void nfs_referral_loop_unprotect(void)
 static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
 		const char *export_path)
 {
-	struct nameidata *nd = NULL;
 	struct mnt_namespace *ns_private;
 	struct super_block *s;
 	struct dentry *dentry;
+	struct path path;
 	int ret;
 
-	nd = kmalloc(sizeof(*nd), GFP_KERNEL);
-	if (nd == NULL)
-		return ERR_PTR(-ENOMEM);
-
 	ns_private = create_mnt_ns(root_mnt);
 	ret = PTR_ERR(ns_private);
 	if (IS_ERR(ns_private))
@@ -2793,7 +2789,7 @@ static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
 		goto out_put_mnt_ns;
 
 	ret = vfs_path_lookup(root_mnt->mnt_root, root_mnt,
-			export_path, LOOKUP_FOLLOW, nd);
+			export_path, LOOKUP_FOLLOW, &path);
 
 	nfs_referral_loop_unprotect();
 	put_mnt_ns(ns_private);
@@ -2801,12 +2797,11 @@ static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
 	if (ret != 0)
 		goto out_err;
 
-	s = nd->path.mnt->mnt_sb;
+	s = path.mnt->mnt_sb;
 	atomic_inc(&s->s_active);
-	dentry = dget(nd->path.dentry);
+	dentry = dget(path.dentry);
 
-	path_put(&nd->path);
-	kfree(nd);
+	path_put(&path);
 	down_write(&s->s_umount);
 	return dentry;
 out_put_mnt_ns:
@@ -2814,7 +2809,6 @@ out_put_mnt_ns:
 out_mntput:
 	mntput(root_mnt);
 out_err:
-	kfree(nd);
 	return ERR_PTR(ret);
 }
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index b8cea804d31a..76fe2c62ae71 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -78,7 +78,7 @@ extern struct dentry *kern_path_create(int, const char *, struct path *, int);
 extern struct dentry *user_path_create(int, const char __user *, struct path *, int);
 extern int kern_path_parent(const char *, struct nameidata *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
-			   const char *, unsigned int, struct nameidata *);
+			   const char *, unsigned int, struct path *);
 
 extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry,
 		int (*open)(struct inode *, struct file *));
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 8c9141583d6f..304f403a0411 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -97,8 +97,7 @@ static int
 rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
 {
 	static uint32_t clntid;
-	struct nameidata nd;
-	struct path path;
+	struct path path, dir;
 	char name[15];
 	struct qstr q = {
 		.name = name,
@@ -113,7 +112,7 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
 	path.mnt = rpc_get_mount();
 	if (IS_ERR(path.mnt))
 		return PTR_ERR(path.mnt);
-	error = vfs_path_lookup(path.mnt->mnt_root, path.mnt, dir_name, 0, &nd);
+	error = vfs_path_lookup(path.mnt->mnt_root, path.mnt, dir_name, 0, &dir);
 	if (error)
 		goto err;
 
@@ -121,7 +120,7 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
 		q.len = snprintf(name, sizeof(name), "clnt%x", (unsigned int)clntid++);
 		name[sizeof(name) - 1] = '\0';
 		q.hash = full_name_hash(q.name, q.len);
-		path.dentry = rpc_create_client_dir(nd.path.dentry, &q, clnt);
+		path.dentry = rpc_create_client_dir(dir.dentry, &q, clnt);
 		if (!IS_ERR(path.dentry))
 			break;
 		error = PTR_ERR(path.dentry);
@@ -132,11 +131,11 @@ rpc_setup_pipedir(struct rpc_clnt *clnt, char *dir_name)
 			goto err_path_put;
 		}
 	}
-	path_put(&nd.path);
+	path_put(&dir);
 	clnt->cl_path = path;
 	return 0;
 err_path_put:
-	path_put(&nd.path);
+	path_put(&dir);
 err:
 	rpc_put_mount();
 	return error;
-- 
cgit v1.2.3


From 0ee5dc676a5f8fadede608c7281dfedb1ae714ea Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 7 Jul 2011 15:44:25 -0400
Subject: btrfs: kill magical embedded struct superblock

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/btrfs/ctree.h   |  2 +-
 fs/btrfs/disk-io.c | 15 ++++-----------
 fs/btrfs/inode.c   |  2 +-
 fs/super.c         | 32 +++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 5 files changed, 31 insertions(+), 22 deletions(-)

(limited to 'include')

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 9864cec801ed..9552afc24ef7 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1219,7 +1219,7 @@ struct btrfs_root {
 	 * right now this just gets used so that a root has its own devid
 	 * for stat.  It may be used for more later
 	 */
-	struct super_block anon_super;
+	dev_t anon_dev;
 };
 
 struct btrfs_ioctl_defrag_range_args {
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 1ac8db5dc0a3..b231ae13b269 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1077,12 +1077,7 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	init_completion(&root->kobj_unregister);
 	root->defrag_running = 0;
 	root->root_key.objectid = objectid;
-	root->anon_super.s_root = NULL;
-	root->anon_super.s_dev = 0;
-	INIT_LIST_HEAD(&root->anon_super.s_list);
-	INIT_LIST_HEAD(&root->anon_super.s_instances);
-	init_rwsem(&root->anon_super.s_umount);
-
+	root->anon_dev = 0;
 	return 0;
 }
 
@@ -1311,7 +1306,7 @@ again:
 	spin_lock_init(&root->cache_lock);
 	init_waitqueue_head(&root->cache_wait);
 
-	ret = set_anon_super(&root->anon_super, NULL);
+	ret = get_anon_bdev(&root->anon_dev);
 	if (ret)
 		goto fail;
 
@@ -2393,10 +2388,8 @@ static void free_fs_root(struct btrfs_root *root)
 {
 	iput(root->cache_inode);
 	WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
-	if (root->anon_super.s_dev) {
-		down_write(&root->anon_super.s_umount);
-		kill_anon_super(&root->anon_super);
-	}
+	if (root->anon_dev)
+		free_anon_bdev(root->anon_dev);
 	free_extent_buffer(root->node);
 	free_extent_buffer(root->commit_root);
 	kfree(root->free_ino_ctl);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index cb170ca51c64..ecf0fac712d6 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6900,7 +6900,7 @@ static int btrfs_getattr(struct vfsmount *mnt,
 {
 	struct inode *inode = dentry->d_inode;
 	generic_fillattr(inode, stat);
-	stat->dev = BTRFS_I(inode)->root->anon_super.s_dev;
+	stat->dev = BTRFS_I(inode)->root->anon_dev;
 	stat->blksize = PAGE_CACHE_SIZE;
 	stat->blocks = (inode_get_bytes(inode) +
 			BTRFS_I(inode)->delalloc_bytes) >> 9;
diff --git a/fs/super.c b/fs/super.c
index 444da9579068..263edeb9f0e9 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -693,7 +693,7 @@ static DEFINE_IDA(unnamed_dev_ida);
 static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */
 static int unnamed_dev_start = 0; /* don't bother trying below it */
 
-int set_anon_super(struct super_block *s, void *data)
+int get_anon_bdev(dev_t *p)
 {
 	int dev;
 	int error;
@@ -720,24 +720,38 @@ int set_anon_super(struct super_block *s, void *data)
 		spin_unlock(&unnamed_dev_lock);
 		return -EMFILE;
 	}
-	s->s_dev = MKDEV(0, dev & MINORMASK);
-	s->s_bdi = &noop_backing_dev_info;
+	*p = MKDEV(0, dev & MINORMASK);
 	return 0;
 }
+EXPORT_SYMBOL(get_anon_bdev);
 
-EXPORT_SYMBOL(set_anon_super);
-
-void kill_anon_super(struct super_block *sb)
+void free_anon_bdev(dev_t dev)
 {
-	int slot = MINOR(sb->s_dev);
-
-	generic_shutdown_super(sb);
+	int slot = MINOR(dev);
 	spin_lock(&unnamed_dev_lock);
 	ida_remove(&unnamed_dev_ida, slot);
 	if (slot < unnamed_dev_start)
 		unnamed_dev_start = slot;
 	spin_unlock(&unnamed_dev_lock);
 }
+EXPORT_SYMBOL(free_anon_bdev);
+
+int set_anon_super(struct super_block *s, void *data)
+{
+	int error = get_anon_bdev(&s->s_dev);
+	if (!error)
+		s->s_bdi = &noop_backing_dev_info;
+	return error;
+}
+
+EXPORT_SYMBOL(set_anon_super);
+
+void kill_anon_super(struct super_block *sb)
+{
+	dev_t dev = sb->s_dev;
+	generic_shutdown_super(sb);
+	free_anon_bdev(dev);
+}
 
 EXPORT_SYMBOL(kill_anon_super);
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 8494aac189f0..a0011aef4338 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1835,6 +1835,8 @@ void kill_litter_super(struct super_block *sb);
 void deactivate_super(struct super_block *sb);
 void deactivate_locked_super(struct super_block *sb);
 int set_anon_super(struct super_block *s, void *data);
+int get_anon_bdev(dev_t *);
+void free_anon_bdev(dev_t);
 struct super_block *sget(struct file_system_type *type,
 			int (*test)(struct super_block *,void *),
 			int (*set)(struct super_block *,void *),
-- 
cgit v1.2.3


From 095760730c1047c69159ce88021a7fa3833502c8 Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:34 +1000
Subject: vmscan: add shrink_slab tracepoints

It is impossible to understand what the shrinkers are actually doing
without instrumenting the code, so add a some tracepoints to allow
insight to be gained.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/trace/events/vmscan.h | 77 +++++++++++++++++++++++++++++++++++++++++++
 mm/vmscan.c                   |  8 ++++-
 2 files changed, 84 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index b2c33bd955fa..36851f7f13da 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -179,6 +179,83 @@ DEFINE_EVENT(mm_vmscan_direct_reclaim_end_template, mm_vmscan_memcg_softlimit_re
 	TP_ARGS(nr_reclaimed)
 );
 
+TRACE_EVENT(mm_shrink_slab_start,
+	TP_PROTO(struct shrinker *shr, struct shrink_control *sc,
+		long nr_objects_to_shrink, unsigned long pgs_scanned,
+		unsigned long lru_pgs, unsigned long cache_items,
+		unsigned long long delta, unsigned long total_scan),
+
+	TP_ARGS(shr, sc, nr_objects_to_shrink, pgs_scanned, lru_pgs,
+		cache_items, delta, total_scan),
+
+	TP_STRUCT__entry(
+		__field(struct shrinker *, shr)
+		__field(void *, shrink)
+		__field(long, nr_objects_to_shrink)
+		__field(gfp_t, gfp_flags)
+		__field(unsigned long, pgs_scanned)
+		__field(unsigned long, lru_pgs)
+		__field(unsigned long, cache_items)
+		__field(unsigned long long, delta)
+		__field(unsigned long, total_scan)
+	),
+
+	TP_fast_assign(
+		__entry->shr = shr;
+		__entry->shrink = shr->shrink;
+		__entry->nr_objects_to_shrink = nr_objects_to_shrink;
+		__entry->gfp_flags = sc->gfp_mask;
+		__entry->pgs_scanned = pgs_scanned;
+		__entry->lru_pgs = lru_pgs;
+		__entry->cache_items = cache_items;
+		__entry->delta = delta;
+		__entry->total_scan = total_scan;
+	),
+
+	TP_printk("%pF %p: objects to shrink %ld gfp_flags %s pgs_scanned %ld lru_pgs %ld cache items %ld delta %lld total_scan %ld",
+		__entry->shrink,
+		__entry->shr,
+		__entry->nr_objects_to_shrink,
+		show_gfp_flags(__entry->gfp_flags),
+		__entry->pgs_scanned,
+		__entry->lru_pgs,
+		__entry->cache_items,
+		__entry->delta,
+		__entry->total_scan)
+);
+
+TRACE_EVENT(mm_shrink_slab_end,
+	TP_PROTO(struct shrinker *shr, int shrinker_retval,
+		long unused_scan_cnt, long new_scan_cnt),
+
+	TP_ARGS(shr, shrinker_retval, unused_scan_cnt, new_scan_cnt),
+
+	TP_STRUCT__entry(
+		__field(struct shrinker *, shr)
+		__field(void *, shrink)
+		__field(long, unused_scan)
+		__field(long, new_scan)
+		__field(int, retval)
+		__field(long, total_scan)
+	),
+
+	TP_fast_assign(
+		__entry->shr = shr;
+		__entry->shrink = shr->shrink;
+		__entry->unused_scan = unused_scan_cnt;
+		__entry->new_scan = new_scan_cnt;
+		__entry->retval = shrinker_retval;
+		__entry->total_scan = new_scan_cnt - unused_scan_cnt;
+	),
+
+	TP_printk("%pF %p: unused scan count %ld new scan count %ld total_scan %ld last shrinker return val %d",
+		__entry->shrink,
+		__entry->shr,
+		__entry->unused_scan,
+		__entry->new_scan,
+		__entry->total_scan,
+		__entry->retval)
+);
 
 DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index d036e59d302b..255226554a3d 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -250,6 +250,7 @@ unsigned long shrink_slab(struct shrink_control *shrink,
 		unsigned long long delta;
 		unsigned long total_scan;
 		unsigned long max_pass;
+		int shrink_ret = 0;
 
 		max_pass = do_shrinker_shrink(shrinker, shrink, 0);
 		delta = (4 * nr_pages_scanned) / shrinker->seeks;
@@ -274,9 +275,12 @@ unsigned long shrink_slab(struct shrink_control *shrink,
 		total_scan = shrinker->nr;
 		shrinker->nr = 0;
 
+		trace_mm_shrink_slab_start(shrinker, shrink, total_scan,
+					nr_pages_scanned, lru_pages,
+					max_pass, delta, total_scan);
+
 		while (total_scan >= SHRINK_BATCH) {
 			long this_scan = SHRINK_BATCH;
-			int shrink_ret;
 			int nr_before;
 
 			nr_before = do_shrinker_shrink(shrinker, shrink, 0);
@@ -293,6 +297,8 @@ unsigned long shrink_slab(struct shrink_control *shrink,
 		}
 
 		shrinker->nr += total_scan;
+		trace_mm_shrink_slab_end(shrinker, shrink_ret, total_scan,
+					 shrinker->nr);
 	}
 	up_read(&shrinker_rwsem);
 out:
-- 
cgit v1.2.3


From e9299f5058595a655c3b207cda9635e28b9197e6 Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:37 +1000
Subject: vmscan: add customisable shrinker batch size

For shrinkers that have their own cond_resched* calls, having
shrink_slab break the work down into small batches is not
paticularly efficient. Add a custom batchsize field to the struct
shrinker so that shrinkers can use a larger batch size if they
desire.

A value of zero (uninitialised) means "use the default", so
behaviour is unchanged by this patch.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/mm.h |  1 +
 mm/vmscan.c        | 11 ++++++-----
 2 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 9670f71d7be9..9b9777ac726d 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1150,6 +1150,7 @@ struct shrink_control {
 struct shrinker {
 	int (*shrink)(struct shrinker *, struct shrink_control *sc);
 	int seeks;	/* seeks to recreate an obj */
+	long batch;	/* reclaim batch size, 0 = default */
 
 	/* These are for internal use */
 	struct list_head list;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 387422470c95..febbc044e792 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -253,6 +253,8 @@ unsigned long shrink_slab(struct shrink_control *shrink,
 		int shrink_ret = 0;
 		long nr;
 		long new_nr;
+		long batch_size = shrinker->batch ? shrinker->batch
+						  : SHRINK_BATCH;
 
 		/*
 		 * copy the current shrinker scan count into a local variable
@@ -303,19 +305,18 @@ unsigned long shrink_slab(struct shrink_control *shrink,
 					nr_pages_scanned, lru_pages,
 					max_pass, delta, total_scan);
 
-		while (total_scan >= SHRINK_BATCH) {
-			long this_scan = SHRINK_BATCH;
+		while (total_scan >= batch_size) {
 			int nr_before;
 
 			nr_before = do_shrinker_shrink(shrinker, shrink, 0);
 			shrink_ret = do_shrinker_shrink(shrinker, shrink,
-							this_scan);
+							batch_size);
 			if (shrink_ret == -1)
 				break;
 			if (shrink_ret < nr_before)
 				ret += nr_before - shrink_ret;
-			count_vm_events(SLABS_SCANNED, this_scan);
-			total_scan -= this_scan;
+			count_vm_events(SLABS_SCANNED, batch_size);
+			total_scan -= batch_size;
 
 			cond_resched();
 		}
-- 
cgit v1.2.3


From 98b745c647a5a90c3c21ea43cbfad9a47b0dfad7 Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:39 +1000
Subject: inode: Make unused inode LRU per superblock

The inode unused list is currently a global LRU. This does not match
the other global filesystem cache - the dentry cache - which uses
per-superblock LRU lists. Hence we have related filesystem object
types using different LRU reclaimation schemes.

To enable a per-superblock filesystem cache shrinker, both of these
caches need to have per-sb unused object LRU lists. Hence this patch
converts the global inode LRU to per-sb LRUs.

The patch only does rudimentary per-sb propotioning in the shrinker
infrastructure, as this gets removed when the per-sb shrinker
callouts are introduced later on.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/inode.c         | 91 +++++++++++++++++++++++++++++++++++++++++++++++-------
 fs/super.c         |  1 +
 include/linux/fs.h |  4 +++
 3 files changed, 85 insertions(+), 11 deletions(-)

(limited to 'include')

diff --git a/fs/inode.c b/fs/inode.c
index 9a0361121712..8c3491302e0c 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -34,7 +34,7 @@
  * inode->i_lock protects:
  *   inode->i_state, inode->i_hash, __iget()
  * inode_lru_lock protects:
- *   inode_lru, inode->i_lru
+ *   inode->i_sb->s_inode_lru, inode->i_lru
  * inode_sb_list_lock protects:
  *   sb->s_inodes, inode->i_sb_list
  * inode_wb_list_lock protects:
@@ -64,7 +64,6 @@ static unsigned int i_hash_shift __read_mostly;
 static struct hlist_head *inode_hashtable __read_mostly;
 static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock);
 
-static LIST_HEAD(inode_lru);
 static DEFINE_SPINLOCK(inode_lru_lock);
 
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_sb_list_lock);
@@ -345,7 +344,8 @@ static void inode_lru_list_add(struct inode *inode)
 {
 	spin_lock(&inode_lru_lock);
 	if (list_empty(&inode->i_lru)) {
-		list_add(&inode->i_lru, &inode_lru);
+		list_add(&inode->i_lru, &inode->i_sb->s_inode_lru);
+		inode->i_sb->s_nr_inodes_unused++;
 		this_cpu_inc(nr_unused);
 	}
 	spin_unlock(&inode_lru_lock);
@@ -356,6 +356,7 @@ static void inode_lru_list_del(struct inode *inode)
 	spin_lock(&inode_lru_lock);
 	if (!list_empty(&inode->i_lru)) {
 		list_del_init(&inode->i_lru);
+		inode->i_sb->s_nr_inodes_unused--;
 		this_cpu_dec(nr_unused);
 	}
 	spin_unlock(&inode_lru_lock);
@@ -628,21 +629,20 @@ static int can_unuse(struct inode *inode)
  * LRU does not have strict ordering. Hence we don't want to reclaim inodes
  * with this flag set because they are the inodes that are out of order.
  */
-static void prune_icache(int nr_to_scan)
+static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 {
 	LIST_HEAD(freeable);
 	int nr_scanned;
 	unsigned long reap = 0;
 
-	down_read(&iprune_sem);
 	spin_lock(&inode_lru_lock);
-	for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
+	for (nr_scanned = *nr_to_scan; nr_scanned >= 0; nr_scanned--) {
 		struct inode *inode;
 
-		if (list_empty(&inode_lru))
+		if (list_empty(&sb->s_inode_lru))
 			break;
 
-		inode = list_entry(inode_lru.prev, struct inode, i_lru);
+		inode = list_entry(sb->s_inode_lru.prev, struct inode, i_lru);
 
 		/*
 		 * we are inverting the inode_lru_lock/inode->i_lock here,
@@ -650,7 +650,7 @@ static void prune_icache(int nr_to_scan)
 		 * inode to the back of the list so we don't spin on it.
 		 */
 		if (!spin_trylock(&inode->i_lock)) {
-			list_move(&inode->i_lru, &inode_lru);
+			list_move(&inode->i_lru, &sb->s_inode_lru);
 			continue;
 		}
 
@@ -662,6 +662,7 @@ static void prune_icache(int nr_to_scan)
 		    (inode->i_state & ~I_REFERENCED)) {
 			list_del_init(&inode->i_lru);
 			spin_unlock(&inode->i_lock);
+			sb->s_nr_inodes_unused--;
 			this_cpu_dec(nr_unused);
 			continue;
 		}
@@ -669,7 +670,7 @@ static void prune_icache(int nr_to_scan)
 		/* recently referenced inodes get one more pass */
 		if (inode->i_state & I_REFERENCED) {
 			inode->i_state &= ~I_REFERENCED;
-			list_move(&inode->i_lru, &inode_lru);
+			list_move(&inode->i_lru, &sb->s_inode_lru);
 			spin_unlock(&inode->i_lock);
 			continue;
 		}
@@ -683,7 +684,7 @@ static void prune_icache(int nr_to_scan)
 			iput(inode);
 			spin_lock(&inode_lru_lock);
 
-			if (inode != list_entry(inode_lru.next,
+			if (inode != list_entry(sb->s_inode_lru.next,
 						struct inode, i_lru))
 				continue;	/* wrong inode or list_empty */
 			/* avoid lock inversions with trylock */
@@ -699,6 +700,7 @@ static void prune_icache(int nr_to_scan)
 		spin_unlock(&inode->i_lock);
 
 		list_move(&inode->i_lru, &freeable);
+		sb->s_nr_inodes_unused--;
 		this_cpu_dec(nr_unused);
 	}
 	if (current_is_kswapd())
@@ -706,8 +708,75 @@ static void prune_icache(int nr_to_scan)
 	else
 		__count_vm_events(PGINODESTEAL, reap);
 	spin_unlock(&inode_lru_lock);
+	*nr_to_scan = nr_scanned;
 
 	dispose_list(&freeable);
+}
+
+static void prune_icache(int count)
+{
+	struct super_block *sb, *p = NULL;
+	int w_count;
+	int unused = inodes_stat.nr_unused;
+	int prune_ratio;
+	int pruned;
+
+	if (unused == 0 || count == 0)
+		return;
+	down_read(&iprune_sem);
+	if (count >= unused)
+		prune_ratio = 1;
+	else
+		prune_ratio = unused / count;
+	spin_lock(&sb_lock);
+	list_for_each_entry(sb, &super_blocks, s_list) {
+		if (list_empty(&sb->s_instances))
+			continue;
+		if (sb->s_nr_inodes_unused == 0)
+			continue;
+		sb->s_count++;
+		/* Now, we reclaim unused dentrins with fairness.
+		 * We reclaim them same percentage from each superblock.
+		 * We calculate number of dentries to scan on this sb
+		 * as follows, but the implementation is arranged to avoid
+		 * overflows:
+		 * number of dentries to scan on this sb =
+		 * count * (number of dentries on this sb /
+		 * number of dentries in the machine)
+		 */
+		spin_unlock(&sb_lock);
+		if (prune_ratio != 1)
+			w_count = (sb->s_nr_inodes_unused / prune_ratio) + 1;
+		else
+			w_count = sb->s_nr_inodes_unused;
+		pruned = w_count;
+		/*
+		 * We need to be sure this filesystem isn't being unmounted,
+		 * otherwise we could race with generic_shutdown_super(), and
+		 * end up holding a reference to an inode while the filesystem
+		 * is unmounted.  So we try to get s_umount, and make sure
+		 * s_root isn't NULL.
+		 */
+		if (down_read_trylock(&sb->s_umount)) {
+			if ((sb->s_root != NULL) &&
+			    (!list_empty(&sb->s_dentry_lru))) {
+				shrink_icache_sb(sb, &w_count);
+				pruned -= w_count;
+			}
+			up_read(&sb->s_umount);
+		}
+		spin_lock(&sb_lock);
+		if (p)
+			__put_super(p);
+		count -= pruned;
+		p = sb;
+		/* more work left to do? */
+		if (count <= 0)
+			break;
+	}
+	if (p)
+		__put_super(p);
+	spin_unlock(&sb_lock);
 	up_read(&iprune_sem);
 }
 
diff --git a/fs/super.c b/fs/super.c
index 263edeb9f0e9..e8e6dbfefe8c 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -77,6 +77,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		INIT_HLIST_BL_HEAD(&s->s_anon);
 		INIT_LIST_HEAD(&s->s_inodes);
 		INIT_LIST_HEAD(&s->s_dentry_lru);
+		INIT_LIST_HEAD(&s->s_inode_lru);
 		init_rwsem(&s->s_umount);
 		mutex_init(&s->s_lock);
 		lockdep_set_class(&s->s_umount, &type->s_umount_key);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index a0011aef4338..9724f0a48742 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1397,6 +1397,10 @@ struct super_block {
 	struct list_head	s_dentry_lru;	/* unused dentry lru */
 	int			s_nr_dentry_unused;	/* # of dentry on lru */
 
+	/* inode_lru_lock protects s_inode_lru and s_nr_inodes_unused */
+	struct list_head	s_inode_lru;		/* unused inode lru */
+	int			s_nr_inodes_unused;	/* # of inodes on lru */
+
 	struct block_device	*s_bdev;
 	struct backing_dev_info *s_bdi;
 	struct mtd_info		*s_mtd;
-- 
cgit v1.2.3


From 09cc9fc7a7c3d872065426d7fb0f0ad6d3eb90fc Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:40 +1000
Subject: inode: move to per-sb LRU locks

With the inode LRUs moving to per-sb structures, there is no longer
a need for a global inode_lru_lock. The locking can be made more
fine-grained by moving to a per-sb LRU lock, isolating the LRU
operations of different filesytsems completely from each other.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/inode.c         | 27 +++++++++++++--------------
 fs/super.c         |  1 +
 include/linux/fs.h |  3 ++-
 3 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/fs/inode.c b/fs/inode.c
index 8c3491302e0c..0450e25aeda0 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -33,7 +33,7 @@
  *
  * inode->i_lock protects:
  *   inode->i_state, inode->i_hash, __iget()
- * inode_lru_lock protects:
+ * inode->i_sb->s_inode_lru_lock protects:
  *   inode->i_sb->s_inode_lru, inode->i_lru
  * inode_sb_list_lock protects:
  *   sb->s_inodes, inode->i_sb_list
@@ -46,7 +46,7 @@
  *
  * inode_sb_list_lock
  *   inode->i_lock
- *     inode_lru_lock
+ *     inode->i_sb->s_inode_lru_lock
  *
  * inode_wb_list_lock
  *   inode->i_lock
@@ -64,8 +64,6 @@ static unsigned int i_hash_shift __read_mostly;
 static struct hlist_head *inode_hashtable __read_mostly;
 static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock);
 
-static DEFINE_SPINLOCK(inode_lru_lock);
-
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_sb_list_lock);
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_wb_list_lock);
 
@@ -342,24 +340,24 @@ EXPORT_SYMBOL(ihold);
 
 static void inode_lru_list_add(struct inode *inode)
 {
-	spin_lock(&inode_lru_lock);
+	spin_lock(&inode->i_sb->s_inode_lru_lock);
 	if (list_empty(&inode->i_lru)) {
 		list_add(&inode->i_lru, &inode->i_sb->s_inode_lru);
 		inode->i_sb->s_nr_inodes_unused++;
 		this_cpu_inc(nr_unused);
 	}
-	spin_unlock(&inode_lru_lock);
+	spin_unlock(&inode->i_sb->s_inode_lru_lock);
 }
 
 static void inode_lru_list_del(struct inode *inode)
 {
-	spin_lock(&inode_lru_lock);
+	spin_lock(&inode->i_sb->s_inode_lru_lock);
 	if (!list_empty(&inode->i_lru)) {
 		list_del_init(&inode->i_lru);
 		inode->i_sb->s_nr_inodes_unused--;
 		this_cpu_dec(nr_unused);
 	}
-	spin_unlock(&inode_lru_lock);
+	spin_unlock(&inode->i_sb->s_inode_lru_lock);
 }
 
 /**
@@ -615,7 +613,8 @@ static int can_unuse(struct inode *inode)
 
 /*
  * Scan `goal' inodes on the unused list for freeable ones. They are moved to a
- * temporary list and then are freed outside inode_lru_lock by dispose_list().
+ * temporary list and then are freed outside sb->s_inode_lru_lock by
+ * dispose_list().
  *
  * Any inodes which are pinned purely because of attached pagecache have their
  * pagecache removed.  If the inode has metadata buffers attached to
@@ -635,7 +634,7 @@ static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 	int nr_scanned;
 	unsigned long reap = 0;
 
-	spin_lock(&inode_lru_lock);
+	spin_lock(&sb->s_inode_lru_lock);
 	for (nr_scanned = *nr_to_scan; nr_scanned >= 0; nr_scanned--) {
 		struct inode *inode;
 
@@ -645,7 +644,7 @@ static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 		inode = list_entry(sb->s_inode_lru.prev, struct inode, i_lru);
 
 		/*
-		 * we are inverting the inode_lru_lock/inode->i_lock here,
+		 * we are inverting the sb->s_inode_lru_lock/inode->i_lock here,
 		 * so use a trylock. If we fail to get the lock, just move the
 		 * inode to the back of the list so we don't spin on it.
 		 */
@@ -677,12 +676,12 @@ static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 		if (inode_has_buffers(inode) || inode->i_data.nrpages) {
 			__iget(inode);
 			spin_unlock(&inode->i_lock);
-			spin_unlock(&inode_lru_lock);
+			spin_unlock(&sb->s_inode_lru_lock);
 			if (remove_inode_buffers(inode))
 				reap += invalidate_mapping_pages(&inode->i_data,
 								0, -1);
 			iput(inode);
-			spin_lock(&inode_lru_lock);
+			spin_lock(&sb->s_inode_lru_lock);
 
 			if (inode != list_entry(sb->s_inode_lru.next,
 						struct inode, i_lru))
@@ -707,7 +706,7 @@ static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 		__count_vm_events(KSWAPD_INODESTEAL, reap);
 	else
 		__count_vm_events(PGINODESTEAL, reap);
-	spin_unlock(&inode_lru_lock);
+	spin_unlock(&sb->s_inode_lru_lock);
 	*nr_to_scan = nr_scanned;
 
 	dispose_list(&freeable);
diff --git a/fs/super.c b/fs/super.c
index e8e6dbfefe8c..73ab9f9b3571 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -78,6 +78,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		INIT_LIST_HEAD(&s->s_inodes);
 		INIT_LIST_HEAD(&s->s_dentry_lru);
 		INIT_LIST_HEAD(&s->s_inode_lru);
+		spin_lock_init(&s->s_inode_lru_lock);
 		init_rwsem(&s->s_umount);
 		mutex_init(&s->s_lock);
 		lockdep_set_class(&s->s_umount, &type->s_umount_key);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9724f0a48742..460d2cc21ec6 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1397,7 +1397,8 @@ struct super_block {
 	struct list_head	s_dentry_lru;	/* unused dentry lru */
 	int			s_nr_dentry_unused;	/* # of dentry on lru */
 
-	/* inode_lru_lock protects s_inode_lru and s_nr_inodes_unused */
+	/* s_inode_lru_lock protects s_inode_lru and s_nr_inodes_unused */
+	spinlock_t		s_inode_lru_lock ____cacheline_aligned_in_smp;
 	struct list_head	s_inode_lru;		/* unused inode lru */
 	int			s_nr_inodes_unused;	/* # of inodes on lru */
 
-- 
cgit v1.2.3


From b0d40c92adafde7c2d81203ce7c1c69275f41140 Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:42 +1000
Subject: superblock: introduce per-sb cache shrinker infrastructure

With context based shrinkers, we can implement a per-superblock
shrinker that shrinks the caches attached to the superblock. We
currently have global shrinkers for the inode and dentry caches that
split up into per-superblock operations via a coarse proportioning
method that does not batch very well.  The global shrinkers also
have a dependency - dentries pin inodes - so we have to be very
careful about how we register the global shrinkers so that the
implicit call order is always correct.

With a per-sb shrinker callout, we can encode this dependency
directly into the per-sb shrinker, hence avoiding the need for
strictly ordering shrinker registrations. We also have no need for
any proportioning code for the shrinker subsystem already provides
this functionality across all shrinkers. Allowing the shrinker to
operate on a single superblock at a time means that we do less
superblock list traversals and locking and reclaim should batch more
effectively. This should result in less CPU overhead for reclaim and
potentially faster reclaim of items from each filesystem.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/dcache.c              | 121 +++++------------------------------------------
 fs/inode.c               | 117 ++++-----------------------------------------
 fs/super.c               |  51 +++++++++++++++++++-
 include/linux/fs.h       |   7 +++
 include/linux/mm.h       |  40 +---------------
 include/linux/shrinker.h |  42 ++++++++++++++++
 6 files changed, 121 insertions(+), 257 deletions(-)
 create mode 100644 include/linux/shrinker.h

(limited to 'include')

diff --git a/fs/dcache.c b/fs/dcache.c
index 41e2085d430b..2762804a140d 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -743,13 +743,11 @@ static void shrink_dentry_list(struct list_head *list)
  *
  * If flags contains DCACHE_REFERENCED reference dentries will not be pruned.
  */
-static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags)
+static void __shrink_dcache_sb(struct super_block *sb, int count, int flags)
 {
-	/* called from prune_dcache() and shrink_dcache_parent() */
 	struct dentry *dentry;
 	LIST_HEAD(referenced);
 	LIST_HEAD(tmp);
-	int cnt = *count;
 
 relock:
 	spin_lock(&dcache_lru_lock);
@@ -777,7 +775,7 @@ relock:
 		} else {
 			list_move_tail(&dentry->d_lru, &tmp);
 			spin_unlock(&dentry->d_lock);
-			if (!--cnt)
+			if (!--count)
 				break;
 		}
 		cond_resched_lock(&dcache_lru_lock);
@@ -787,83 +785,22 @@ relock:
 	spin_unlock(&dcache_lru_lock);
 
 	shrink_dentry_list(&tmp);
-
-	*count = cnt;
 }
 
 /**
- * prune_dcache - shrink the dcache
- * @count: number of entries to try to free
+ * prune_dcache_sb - shrink the dcache
+ * @nr_to_scan: number of entries to try to free
  *
- * Shrink the dcache. This is done when we need more memory, or simply when we
- * need to unmount something (at which point we need to unuse all dentries).
+ * Attempt to shrink the superblock dcache LRU by @nr_to_scan entries. This is
+ * done when we need more memory an called from the superblock shrinker
+ * function.
  *
- * This function may fail to free any resources if all the dentries are in use.
+ * This function may fail to free any resources if all the dentries are in
+ * use.
  */
-static void prune_dcache(int count)
+void prune_dcache_sb(struct super_block *sb, int nr_to_scan)
 {
-	struct super_block *sb, *p = NULL;
-	int w_count;
-	int unused = dentry_stat.nr_unused;
-	int prune_ratio;
-	int pruned;
-
-	if (unused == 0 || count == 0)
-		return;
-	if (count >= unused)
-		prune_ratio = 1;
-	else
-		prune_ratio = unused / count;
-	spin_lock(&sb_lock);
-	list_for_each_entry(sb, &super_blocks, s_list) {
-		if (list_empty(&sb->s_instances))
-			continue;
-		if (sb->s_nr_dentry_unused == 0)
-			continue;
-		sb->s_count++;
-		/* Now, we reclaim unused dentrins with fairness.
-		 * We reclaim them same percentage from each superblock.
-		 * We calculate number of dentries to scan on this sb
-		 * as follows, but the implementation is arranged to avoid
-		 * overflows:
-		 * number of dentries to scan on this sb =
-		 * count * (number of dentries on this sb /
-		 * number of dentries in the machine)
-		 */
-		spin_unlock(&sb_lock);
-		if (prune_ratio != 1)
-			w_count = (sb->s_nr_dentry_unused / prune_ratio) + 1;
-		else
-			w_count = sb->s_nr_dentry_unused;
-		pruned = w_count;
-		/*
-		 * We need to be sure this filesystem isn't being unmounted,
-		 * otherwise we could race with generic_shutdown_super(), and
-		 * end up holding a reference to an inode while the filesystem
-		 * is unmounted.  So we try to get s_umount, and make sure
-		 * s_root isn't NULL.
-		 */
-		if (down_read_trylock(&sb->s_umount)) {
-			if ((sb->s_root != NULL) &&
-			    (!list_empty(&sb->s_dentry_lru))) {
-				__shrink_dcache_sb(sb, &w_count,
-						DCACHE_REFERENCED);
-				pruned -= w_count;
-			}
-			up_read(&sb->s_umount);
-		}
-		spin_lock(&sb_lock);
-		if (p)
-			__put_super(p);
-		count -= pruned;
-		p = sb;
-		/* more work left to do? */
-		if (count <= 0)
-			break;
-	}
-	if (p)
-		__put_super(p);
-	spin_unlock(&sb_lock);
+	__shrink_dcache_sb(sb, nr_to_scan, DCACHE_REFERENCED);
 }
 
 /**
@@ -1238,42 +1175,10 @@ void shrink_dcache_parent(struct dentry * parent)
 	int found;
 
 	while ((found = select_parent(parent)) != 0)
-		__shrink_dcache_sb(sb, &found, 0);
+		__shrink_dcache_sb(sb, found, 0);
 }
 EXPORT_SYMBOL(shrink_dcache_parent);
 
-/*
- * Scan `sc->nr_slab_to_reclaim' dentries and return the number which remain.
- *
- * We need to avoid reentering the filesystem if the caller is performing a
- * GFP_NOFS allocation attempt.  One example deadlock is:
- *
- * ext2_new_block->getblk->GFP->shrink_dcache_memory->prune_dcache->
- * prune_one_dentry->dput->dentry_iput->iput->inode->i_sb->s_op->put_inode->
- * ext2_discard_prealloc->ext2_free_blocks->lock_super->DEADLOCK.
- *
- * In this case we return -1 to tell the caller that we baled.
- */
-static int shrink_dcache_memory(struct shrinker *shrink,
-				struct shrink_control *sc)
-{
-	int nr = sc->nr_to_scan;
-	gfp_t gfp_mask = sc->gfp_mask;
-
-	if (nr) {
-		if (!(gfp_mask & __GFP_FS))
-			return -1;
-		prune_dcache(nr);
-	}
-
-	return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
-}
-
-static struct shrinker dcache_shrinker = {
-	.shrink = shrink_dcache_memory,
-	.seeks = DEFAULT_SEEKS,
-};
-
 /**
  * __d_alloc	-	allocate a dcache entry
  * @sb: filesystem it will belong to
@@ -3083,8 +2988,6 @@ static void __init dcache_init(void)
 	 */
 	dentry_cache = KMEM_CACHE(dentry,
 		SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|SLAB_MEM_SPREAD);
-	
-	register_shrinker(&dcache_shrinker);
 
 	/* Hash may have been set up in dcache_init_early */
 	if (!hashdist)
diff --git a/fs/inode.c b/fs/inode.c
index 0450e25aeda0..1fdbb64a952f 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -73,7 +73,7 @@ __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_wb_list_lock);
  *
  * We don't actually need it to protect anything in the umount path,
  * but only need to cycle through it to make sure any inode that
- * prune_icache took off the LRU list has been fully torn down by the
+ * prune_icache_sb took off the LRU list has been fully torn down by the
  * time we are past evict_inodes.
  */
 static DECLARE_RWSEM(iprune_sem);
@@ -544,7 +544,7 @@ void evict_inodes(struct super_block *sb)
 	dispose_list(&dispose);
 
 	/*
-	 * Cycle through iprune_sem to make sure any inode that prune_icache
+	 * Cycle through iprune_sem to make sure any inode that prune_icache_sb
 	 * moved off the list before we took the lock has been fully torn
 	 * down.
 	 */
@@ -612,9 +612,10 @@ static int can_unuse(struct inode *inode)
 }
 
 /*
- * Scan `goal' inodes on the unused list for freeable ones. They are moved to a
- * temporary list and then are freed outside sb->s_inode_lru_lock by
- * dispose_list().
+ * Walk the superblock inode LRU for freeable inodes and attempt to free them.
+ * This is called from the superblock shrinker function with a number of inodes
+ * to trim from the LRU. Inodes to be freed are moved to a temporary list and
+ * then are freed outside inode_lock by dispose_list().
  *
  * Any inodes which are pinned purely because of attached pagecache have their
  * pagecache removed.  If the inode has metadata buffers attached to
@@ -628,14 +629,15 @@ static int can_unuse(struct inode *inode)
  * LRU does not have strict ordering. Hence we don't want to reclaim inodes
  * with this flag set because they are the inodes that are out of order.
  */
-static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
+void prune_icache_sb(struct super_block *sb, int nr_to_scan)
 {
 	LIST_HEAD(freeable);
 	int nr_scanned;
 	unsigned long reap = 0;
 
+	down_read(&iprune_sem);
 	spin_lock(&sb->s_inode_lru_lock);
-	for (nr_scanned = *nr_to_scan; nr_scanned >= 0; nr_scanned--) {
+	for (nr_scanned = nr_to_scan; nr_scanned >= 0; nr_scanned--) {
 		struct inode *inode;
 
 		if (list_empty(&sb->s_inode_lru))
@@ -707,111 +709,11 @@ static void shrink_icache_sb(struct super_block *sb, int *nr_to_scan)
 	else
 		__count_vm_events(PGINODESTEAL, reap);
 	spin_unlock(&sb->s_inode_lru_lock);
-	*nr_to_scan = nr_scanned;
 
 	dispose_list(&freeable);
-}
-
-static void prune_icache(int count)
-{
-	struct super_block *sb, *p = NULL;
-	int w_count;
-	int unused = inodes_stat.nr_unused;
-	int prune_ratio;
-	int pruned;
-
-	if (unused == 0 || count == 0)
-		return;
-	down_read(&iprune_sem);
-	if (count >= unused)
-		prune_ratio = 1;
-	else
-		prune_ratio = unused / count;
-	spin_lock(&sb_lock);
-	list_for_each_entry(sb, &super_blocks, s_list) {
-		if (list_empty(&sb->s_instances))
-			continue;
-		if (sb->s_nr_inodes_unused == 0)
-			continue;
-		sb->s_count++;
-		/* Now, we reclaim unused dentrins with fairness.
-		 * We reclaim them same percentage from each superblock.
-		 * We calculate number of dentries to scan on this sb
-		 * as follows, but the implementation is arranged to avoid
-		 * overflows:
-		 * number of dentries to scan on this sb =
-		 * count * (number of dentries on this sb /
-		 * number of dentries in the machine)
-		 */
-		spin_unlock(&sb_lock);
-		if (prune_ratio != 1)
-			w_count = (sb->s_nr_inodes_unused / prune_ratio) + 1;
-		else
-			w_count = sb->s_nr_inodes_unused;
-		pruned = w_count;
-		/*
-		 * We need to be sure this filesystem isn't being unmounted,
-		 * otherwise we could race with generic_shutdown_super(), and
-		 * end up holding a reference to an inode while the filesystem
-		 * is unmounted.  So we try to get s_umount, and make sure
-		 * s_root isn't NULL.
-		 */
-		if (down_read_trylock(&sb->s_umount)) {
-			if ((sb->s_root != NULL) &&
-			    (!list_empty(&sb->s_dentry_lru))) {
-				shrink_icache_sb(sb, &w_count);
-				pruned -= w_count;
-			}
-			up_read(&sb->s_umount);
-		}
-		spin_lock(&sb_lock);
-		if (p)
-			__put_super(p);
-		count -= pruned;
-		p = sb;
-		/* more work left to do? */
-		if (count <= 0)
-			break;
-	}
-	if (p)
-		__put_super(p);
-	spin_unlock(&sb_lock);
 	up_read(&iprune_sem);
 }
 
-/*
- * shrink_icache_memory() will attempt to reclaim some unused inodes.  Here,
- * "unused" means that no dentries are referring to the inodes: the files are
- * not open and the dcache references to those inodes have already been
- * reclaimed.
- *
- * This function is passed the number of inodes to scan, and it returns the
- * total number of remaining possibly-reclaimable inodes.
- */
-static int shrink_icache_memory(struct shrinker *shrink,
-				struct shrink_control *sc)
-{
-	int nr = sc->nr_to_scan;
-	gfp_t gfp_mask = sc->gfp_mask;
-
-	if (nr) {
-		/*
-		 * Nasty deadlock avoidance.  We may hold various FS locks,
-		 * and we don't want to recurse into the FS that called us
-		 * in clear_inode() and friends..
-		 */
-		if (!(gfp_mask & __GFP_FS))
-			return -1;
-		prune_icache(nr);
-	}
-	return (get_nr_inodes_unused() / 100) * sysctl_vfs_cache_pressure;
-}
-
-static struct shrinker icache_shrinker = {
-	.shrink = shrink_icache_memory,
-	.seeks = DEFAULT_SEEKS,
-};
-
 static void __wait_on_freeing_inode(struct inode *inode);
 /*
  * Called with the inode lock held.
@@ -1691,7 +1593,6 @@ void __init inode_init(void)
 					 (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|
 					 SLAB_MEM_SPREAD),
 					 init_once);
-	register_shrinker(&icache_shrinker);
 
 	/* Hash may have been set up in inode_init_early */
 	if (!hashdist)
diff --git a/fs/super.c b/fs/super.c
index e63c754447ce..37a75410079e 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -38,6 +38,48 @@
 LIST_HEAD(super_blocks);
 DEFINE_SPINLOCK(sb_lock);
 
+/*
+ * One thing we have to be careful of with a per-sb shrinker is that we don't
+ * drop the last active reference to the superblock from within the shrinker.
+ * If that happens we could trigger unregistering the shrinker from within the
+ * shrinker path and that leads to deadlock on the shrinker_rwsem. Hence we
+ * take a passive reference to the superblock to avoid this from occurring.
+ */
+static int prune_super(struct shrinker *shrink, struct shrink_control *sc)
+{
+	struct super_block *sb;
+	int count;
+
+	sb = container_of(shrink, struct super_block, s_shrink);
+
+	/*
+	 * Deadlock avoidance.  We may hold various FS locks, and we don't want
+	 * to recurse into the FS that called us in clear_inode() and friends..
+	 */
+	if (sc->nr_to_scan && !(sc->gfp_mask & __GFP_FS))
+		return -1;
+
+	if (!grab_super_passive(sb))
+		return -1;
+
+	if (sc->nr_to_scan) {
+		/* proportion the scan between the two caches */
+		int total;
+
+		total = sb->s_nr_dentry_unused + sb->s_nr_inodes_unused + 1;
+		count = (sc->nr_to_scan * sb->s_nr_dentry_unused) / total;
+
+		/* prune dcache first as icache is pinned by it */
+		prune_dcache_sb(sb, count);
+		prune_icache_sb(sb, sc->nr_to_scan - count);
+	}
+
+	count = ((sb->s_nr_dentry_unused + sb->s_nr_inodes_unused) / 100)
+						* sysctl_vfs_cache_pressure;
+	drop_super(sb);
+	return count;
+}
+
 /**
  *	alloc_super	-	create new superblock
  *	@type:	filesystem type superblock should belong to
@@ -116,6 +158,9 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		s->s_op = &default_op;
 		s->s_time_gran = 1000000000;
 		s->cleancache_poolid = -1;
+
+		s->s_shrink.seeks = DEFAULT_SEEKS;
+		s->s_shrink.shrink = prune_super;
 	}
 out:
 	return s;
@@ -183,6 +228,10 @@ void deactivate_locked_super(struct super_block *s)
 	if (atomic_dec_and_test(&s->s_active)) {
 		cleancache_flush_fs(s);
 		fs->kill_sb(s);
+
+		/* caches are now gone, we can safely kill the shrinker now */
+		unregister_shrinker(&s->s_shrink);
+
 		/*
 		 * We need to call rcu_barrier so all the delayed rcu free
 		 * inodes are flushed before we release the fs module.
@@ -311,7 +360,6 @@ void generic_shutdown_super(struct super_block *sb)
 {
 	const struct super_operations *sop = sb->s_op;
 
-
 	if (sb->s_root) {
 		shrink_dcache_for_umount(sb);
 		sync_filesystem(sb);
@@ -399,6 +447,7 @@ retry:
 	list_add(&s->s_instances, &type->fs_supers);
 	spin_unlock(&sb_lock);
 	get_filesystem(type);
+	register_shrinker(&s->s_shrink);
 	return s;
 }
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 460d2cc21ec6..d7f35e90b84a 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -393,6 +393,7 @@ struct inodes_stat_t {
 #include <linux/semaphore.h>
 #include <linux/fiemap.h>
 #include <linux/rculist_bl.h>
+#include <linux/shrinker.h>
 #include <linux/atomic.h>
 
 #include <asm/byteorder.h>
@@ -1444,8 +1445,14 @@ struct super_block {
 	 * Saved pool identifier for cleancache (-1 means none)
 	 */
 	int cleancache_poolid;
+
+	struct shrinker s_shrink;	/* per-sb shrinker handle */
 };
 
+/* superblock cache pruning functions */
+extern void prune_icache_sb(struct super_block *sb, int nr_to_scan);
+extern void prune_dcache_sb(struct super_block *sb, int nr_to_scan);
+
 extern struct timespec current_fs_time(struct super_block *sb);
 
 /*
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 9b9777ac726d..e3a1a9eec0b1 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -15,6 +15,7 @@
 #include <linux/range.h>
 #include <linux/pfn.h>
 #include <linux/bit_spinlock.h>
+#include <linux/shrinker.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -1121,45 +1122,6 @@ static inline void sync_mm_rss(struct task_struct *task, struct mm_struct *mm)
 }
 #endif
 
-/*
- * This struct is used to pass information from page reclaim to the shrinkers.
- * We consolidate the values for easier extention later.
- */
-struct shrink_control {
-	gfp_t gfp_mask;
-
-	/* How many slab objects shrinker() should scan and try to reclaim */
-	unsigned long nr_to_scan;
-};
-
-/*
- * A callback you can register to apply pressure to ageable caches.
- *
- * 'sc' is passed shrink_control which includes a count 'nr_to_scan'
- * and a 'gfpmask'.  It should look through the least-recently-used
- * 'nr_to_scan' entries and attempt to free them up.  It should return
- * the number of objects which remain in the cache.  If it returns -1, it means
- * it cannot do any scanning at this time (eg. there is a risk of deadlock).
- *
- * The 'gfpmask' refers to the allocation we are currently trying to
- * fulfil.
- *
- * Note that 'shrink' will be passed nr_to_scan == 0 when the VM is
- * querying the cache size, so a fastpath for that case is appropriate.
- */
-struct shrinker {
-	int (*shrink)(struct shrinker *, struct shrink_control *sc);
-	int seeks;	/* seeks to recreate an obj */
-	long batch;	/* reclaim batch size, 0 = default */
-
-	/* These are for internal use */
-	struct list_head list;
-	long nr;	/* objs pending delete */
-};
-#define DEFAULT_SEEKS 2 /* A good number if you don't know better. */
-extern void register_shrinker(struct shrinker *);
-extern void unregister_shrinker(struct shrinker *);
-
 int vma_wants_writenotify(struct vm_area_struct *vma);
 
 extern pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr,
diff --git a/include/linux/shrinker.h b/include/linux/shrinker.h
new file mode 100644
index 000000000000..790651b4e5ba
--- /dev/null
+++ b/include/linux/shrinker.h
@@ -0,0 +1,42 @@
+#ifndef _LINUX_SHRINKER_H
+#define _LINUX_SHRINKER_H
+
+/*
+ * This struct is used to pass information from page reclaim to the shrinkers.
+ * We consolidate the values for easier extention later.
+ */
+struct shrink_control {
+	gfp_t gfp_mask;
+
+	/* How many slab objects shrinker() should scan and try to reclaim */
+	unsigned long nr_to_scan;
+};
+
+/*
+ * A callback you can register to apply pressure to ageable caches.
+ *
+ * 'sc' is passed shrink_control which includes a count 'nr_to_scan'
+ * and a 'gfpmask'.  It should look through the least-recently-used
+ * 'nr_to_scan' entries and attempt to free them up.  It should return
+ * the number of objects which remain in the cache.  If it returns -1, it means
+ * it cannot do any scanning at this time (eg. there is a risk of deadlock).
+ *
+ * The 'gfpmask' refers to the allocation we are currently trying to
+ * fulfil.
+ *
+ * Note that 'shrink' will be passed nr_to_scan == 0 when the VM is
+ * querying the cache size, so a fastpath for that case is appropriate.
+ */
+struct shrinker {
+	int (*shrink)(struct shrinker *, struct shrink_control *sc);
+	int seeks;	/* seeks to recreate an obj */
+	long batch;	/* reclaim batch size, 0 = default */
+
+	/* These are for internal use */
+	struct list_head list;
+	long nr;	/* objs pending delete */
+};
+#define DEFAULT_SEEKS 2 /* A good number if you don't know better. */
+extern void register_shrinker(struct shrinker *);
+extern void unregister_shrinker(struct shrinker *);
+#endif
-- 
cgit v1.2.3


From 0e1fdafd93980eac62e778798549ce0f6073905c Mon Sep 17 00:00:00 2001
From: Dave Chinner <dchinner@redhat.com>
Date: Fri, 8 Jul 2011 14:14:44 +1000
Subject: superblock: add filesystem shrinker operations

Now we have a per-superblock shrinker implementation, we can add a
filesystem specific callout to it to allow filesystem internal
caches to be shrunk by the superblock shrinker.

Rather than perpetuate the multipurpose shrinker callback API (i.e.
nr_to_scan == 0 meaning "tell me how many objects freeable in the
cache), two operations will be added. The first will return the
number of objects that are freeable, the second is the actual
shrinker call.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/vfs.txt | 16 ++++++++++++++
 fs/super.c                        | 45 ++++++++++++++++++++++++++++-----------
 include/linux/fs.h                |  2 ++
 3 files changed, 51 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index d56151fe77dc..fd24f34f120f 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -229,6 +229,8 @@ struct super_operations {
 
         ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
         ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
+	int (*nr_cached_objects)(struct super_block *);
+	void (*free_cached_objects)(struct super_block *, int);
 };
 
 All methods are called without any locks being held, unless otherwise
@@ -301,6 +303,20 @@ or bottom half).
 
   quota_write: called by the VFS to write to filesystem quota file.
 
+  nr_cached_objects: called by the sb cache shrinking function for the
+	filesystem to return the number of freeable cached objects it contains.
+	Optional.
+
+  free_cache_objects: called by the sb cache shrinking function for the
+	filesystem to scan the number of objects indicated to try to free them.
+	Optional, but any filesystem implementing this method needs to also
+	implement ->nr_cached_objects for it to be called correctly.
+
+	We can't do anything with any errors that the filesystem might
+	encountered, hence the void return type. This will never be called if
+	the VM is trying to reclaim under GFP_NOFS conditions, hence this
+	method does not need to handle that situation itself.
+
 Whoever sets up the inode is responsible for filling in the "i_op" field. This
 is a pointer to a "struct inode_operations" which describes the methods that
 can be performed on individual inodes.
diff --git a/fs/super.c b/fs/super.c
index 37a75410079e..5101f0544960 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -48,7 +48,8 @@ DEFINE_SPINLOCK(sb_lock);
 static int prune_super(struct shrinker *shrink, struct shrink_control *sc)
 {
 	struct super_block *sb;
-	int count;
+	int	fs_objects = 0;
+	int	total_objects;
 
 	sb = container_of(shrink, struct super_block, s_shrink);
 
@@ -62,22 +63,42 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc)
 	if (!grab_super_passive(sb))
 		return -1;
 
-	if (sc->nr_to_scan) {
-		/* proportion the scan between the two caches */
-		int total;
+	if (sb->s_op && sb->s_op->nr_cached_objects)
+		fs_objects = sb->s_op->nr_cached_objects(sb);
+
+	total_objects = sb->s_nr_dentry_unused +
+			sb->s_nr_inodes_unused + fs_objects + 1;
 
-		total = sb->s_nr_dentry_unused + sb->s_nr_inodes_unused + 1;
-		count = (sc->nr_to_scan * sb->s_nr_dentry_unused) / total;
+	if (sc->nr_to_scan) {
+		int	dentries;
+		int	inodes;
+
+		/* proportion the scan between the caches */
+		dentries = (sc->nr_to_scan * sb->s_nr_dentry_unused) /
+							total_objects;
+		inodes = (sc->nr_to_scan * sb->s_nr_inodes_unused) /
+							total_objects;
+		if (fs_objects)
+			fs_objects = (sc->nr_to_scan * fs_objects) /
+							total_objects;
+		/*
+		 * prune the dcache first as the icache is pinned by it, then
+		 * prune the icache, followed by the filesystem specific caches
+		 */
+		prune_dcache_sb(sb, dentries);
+		prune_icache_sb(sb, inodes);
 
-		/* prune dcache first as icache is pinned by it */
-		prune_dcache_sb(sb, count);
-		prune_icache_sb(sb, sc->nr_to_scan - count);
+		if (fs_objects && sb->s_op->free_cached_objects) {
+			sb->s_op->free_cached_objects(sb, fs_objects);
+			fs_objects = sb->s_op->nr_cached_objects(sb);
+		}
+		total_objects = sb->s_nr_dentry_unused +
+				sb->s_nr_inodes_unused + fs_objects;
 	}
 
-	count = ((sb->s_nr_dentry_unused + sb->s_nr_inodes_unused) / 100)
-						* sysctl_vfs_cache_pressure;
+	total_objects = (total_objects / 100) * sysctl_vfs_cache_pressure;
 	drop_super(sb);
-	return count;
+	return total_objects;
 }
 
 /**
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d7f35e90b84a..1393742bba9b 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1655,6 +1655,8 @@ struct super_operations {
 	ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
 #endif
 	int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t);
+	int (*nr_cached_objects)(struct super_block *);
+	void (*free_cached_objects)(struct super_block *, int);
 };
 
 /*
-- 
cgit v1.2.3


From e46ebd27823b27273da780376d54c080250ff1a2 Mon Sep 17 00:00:00 2001
From: Tomasz Stanislawski <t.stanislaws@samsung.com>
Date: Tue, 12 Jul 2011 11:27:20 +0200
Subject: anonfd: fix missing declaration

The forward declaration of struct file_operations is
added to avoid compilation warnings.

Signed-off-by: Tomasz Stanislawski <t.stanislaws@samsung.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/anon_inodes.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include')

diff --git a/include/linux/anon_inodes.h b/include/linux/anon_inodes.h
index 69a21e0ebd33..8013a45242fe 100644
--- a/include/linux/anon_inodes.h
+++ b/include/linux/anon_inodes.h
@@ -8,6 +8,8 @@
 #ifndef _LINUX_ANON_INODES_H
 #define _LINUX_ANON_INODES_H
 
+struct file_operations;
+
 struct file *anon_inode_getfile(const char *name,
 				const struct file_operations *fops,
 				void *priv, int flags);
-- 
cgit v1.2.3


From bd5fe6c5eb9c548d7f07fe8f89a150bb6705e8e3 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Fri, 24 Jun 2011 14:29:43 -0400
Subject: fs: kill i_alloc_sem

i_alloc_sem is a rather special rw_semaphore.  It's the last one that may
be released by a non-owner, and it's write side is always mirrored by
real exclusion.  It's intended use it to wait for all pending direct I/O
requests to finish before starting a truncate.

Replace it with a hand-grown construct:

 - exclusion for truncates is already guaranteed by i_mutex, so it can
   simply fall way
 - the reader side is replaced by an i_dio_count member in struct inode
   that counts the number of pending direct I/O requests.  Truncate can't
   proceed as long as it's non-zero
 - when i_dio_count reaches non-zero we wake up a pending truncate using
   wake_up_bit on a new bit in i_flags
 - new references to i_dio_count can't appear while we are waiting for
   it to read zero because the direct I/O count always needs i_mutex
   (or an equivalent like XFS's i_iolock) for starting a new operation.

This scheme is much simpler, and saves the space of a spinlock_t and a
struct list_head in struct inode (typically 160 bits on a non-debug 64-bit
system).

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/attr.c           |  5 +----
 fs/direct-io.c      | 65 +++++++++++++++++++++++++++++++++++++++++------------
 fs/inode.c          |  3 +--
 fs/ntfs/file.c      |  3 +--
 fs/ntfs/inode.c     | 10 ++-------
 fs/ocfs2/aops.c     |  7 +++---
 fs/ocfs2/file.c     | 15 ++++++-------
 fs/reiserfs/xattr.c |  3 +--
 include/linux/fs.h  | 11 +++++++--
 mm/filemap.c        |  3 ---
 mm/madvise.c        |  2 +-
 mm/rmap.c           |  1 -
 mm/truncate.c       |  3 +--
 13 files changed, 78 insertions(+), 53 deletions(-)

(limited to 'include')

diff --git a/fs/attr.c b/fs/attr.c
index caf2aa521e2b..f177ac86fa48 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -233,16 +233,13 @@ int notify_change(struct dentry * dentry, struct iattr * attr)
 		return error;
 
 	if (ia_valid & ATTR_SIZE)
-		down_write(&dentry->d_inode->i_alloc_sem);
+		inode_dio_wait(inode);
 
 	if (inode->i_op->setattr)
 		error = inode->i_op->setattr(dentry, attr);
 	else
 		error = simple_setattr(dentry, attr);
 
-	if (ia_valid & ATTR_SIZE)
-		up_write(&dentry->d_inode->i_alloc_sem);
-
 	if (!error)
 		fsnotify_change(dentry, ia_valid);
 
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 98ce3ac0d94b..354cbdbc14bd 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -135,6 +135,50 @@ struct dio {
 	struct page *pages[DIO_PAGES];	/* page buffer */
 };
 
+static void __inode_dio_wait(struct inode *inode)
+{
+	wait_queue_head_t *wq = bit_waitqueue(&inode->i_state, __I_DIO_WAKEUP);
+	DEFINE_WAIT_BIT(q, &inode->i_state, __I_DIO_WAKEUP);
+
+	do {
+		prepare_to_wait(wq, &q.wait, TASK_UNINTERRUPTIBLE);
+		if (atomic_read(&inode->i_dio_count))
+			schedule();
+	} while (atomic_read(&inode->i_dio_count));
+	finish_wait(wq, &q.wait);
+}
+
+/**
+ * inode_dio_wait - wait for outstanding DIO requests to finish
+ * @inode: inode to wait for
+ *
+ * Waits for all pending direct I/O requests to finish so that we can
+ * proceed with a truncate or equivalent operation.
+ *
+ * Must be called under a lock that serializes taking new references
+ * to i_dio_count, usually by inode->i_mutex.
+ */
+void inode_dio_wait(struct inode *inode)
+{
+	if (atomic_read(&inode->i_dio_count))
+		__inode_dio_wait(inode);
+}
+EXPORT_SYMBOL_GPL(inode_dio_wait);
+
+/*
+ * inode_dio_done - signal finish of a direct I/O requests
+ * @inode: inode the direct I/O happens on
+ *
+ * This is called once we've finished processing a direct I/O request,
+ * and is used to wake up callers waiting for direct I/O to be quiesced.
+ */
+void inode_dio_done(struct inode *inode)
+{
+	if (atomic_dec_and_test(&inode->i_dio_count))
+		wake_up_bit(&inode->i_state, __I_DIO_WAKEUP);
+}
+EXPORT_SYMBOL_GPL(inode_dio_done);
+
 /*
  * How many pages are in the queue?
  */
@@ -254,9 +298,7 @@ static ssize_t dio_complete(struct dio *dio, loff_t offset, ssize_t ret, bool is
 	}
 
 	if (dio->flags & DIO_LOCKING)
-		/* lockdep: non-owner release */
-		up_read_non_owner(&dio->inode->i_alloc_sem);
-
+		inode_dio_done(dio->inode);
 	return ret;
 }
 
@@ -980,9 +1022,6 @@ out:
 	return ret;
 }
 
-/*
- * Releases both i_mutex and i_alloc_sem
- */
 static ssize_t
 direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode, 
 	const struct iovec *iov, loff_t offset, unsigned long nr_segs, 
@@ -1146,15 +1185,14 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
  *    For writes this function is called under i_mutex and returns with
  *    i_mutex held, for reads, i_mutex is not held on entry, but it is
  *    taken and dropped again before returning.
- *    For reads and writes i_alloc_sem is taken in shared mode and released
- *    on I/O completion (which may happen asynchronously after returning to
- *    the caller).
+ *    The i_dio_count counter keeps track of the number of outstanding
+ *    direct I/O requests, and truncate waits for it to reach zero.
+ *    New references to i_dio_count must only be grabbed with i_mutex
+ *    held.
  *
  *  - if the flags value does NOT contain DIO_LOCKING we don't use any
  *    internal locking but rather rely on the filesystem to synchronize
  *    direct I/O reads/writes versus each other and truncate.
- *    For reads and writes both i_mutex and i_alloc_sem are not held on
- *    entry and are never taken.
  */
 ssize_t
 __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
@@ -1234,10 +1272,9 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 		}
 
 		/*
-		 * Will be released at I/O completion, possibly in a
-		 * different thread.
+		 * Will be decremented at I/O completion time.
 		 */
-		down_read_non_owner(&inode->i_alloc_sem);
+		atomic_inc(&inode->i_dio_count);
 	}
 
 	/*
diff --git a/fs/inode.c b/fs/inode.c
index cf81baf1898a..96c77b81167c 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -168,8 +168,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode)
 	mutex_init(&inode->i_mutex);
 	lockdep_set_class(&inode->i_mutex, &sb->s_type->i_mutex_key);
 
-	init_rwsem(&inode->i_alloc_sem);
-	lockdep_set_class(&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key);
+	atomic_set(&inode->i_dio_count, 0);
 
 	mapping->a_ops = &empty_aops;
 	mapping->host = inode;
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index f4b1057abdd2..b59f5ac26bef 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -1832,9 +1832,8 @@ static ssize_t ntfs_file_buffered_write(struct kiocb *iocb,
 	 * fails again.
 	 */
 	if (unlikely(NInoTruncateFailed(ni))) {
-		down_write(&vi->i_alloc_sem);
+		inode_dio_wait(vi);
 		err = ntfs_truncate(vi);
-		up_write(&vi->i_alloc_sem);
 		if (err || NInoTruncateFailed(ni)) {
 			if (!err)
 				err = -EIO;
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index c05d6dcf77a4..1371487da955 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -2357,12 +2357,7 @@ static const char *es = "  Leaving inconsistent metadata.  Unmount and run "
  *
  * Returns 0 on success or -errno on error.
  *
- * Called with ->i_mutex held.  In all but one case ->i_alloc_sem is held for
- * writing.  The only case in the kernel where ->i_alloc_sem is not held is
- * mm/filemap.c::generic_file_buffered_write() where vmtruncate() is called
- * with the current i_size as the offset.  The analogous place in NTFS is in
- * fs/ntfs/file.c::ntfs_file_buffered_write() where we call vmtruncate() again
- * without holding ->i_alloc_sem.
+ * Called with ->i_mutex held.
  */
 int ntfs_truncate(struct inode *vi)
 {
@@ -2887,8 +2882,7 @@ void ntfs_truncate_vfs(struct inode *vi) {
  * We also abort all changes of user, group, and mode as we do not implement
  * the NTFS ACLs yet.
  *
- * Called with ->i_mutex held.  For the ATTR_SIZE (i.e. ->truncate) case, also
- * called with ->i_alloc_sem held for writing.
+ * Called with ->i_mutex held.
  */
 int ntfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index ac97bca282d2..de1d3953599d 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -551,9 +551,8 @@ bail:
 
 /*
  * ocfs2_dio_end_io is called by the dio core when a dio is finished.  We're
- * particularly interested in the aio/dio case.  Like the core uses
- * i_alloc_sem, we use the rw_lock DLM lock to protect io on one node from
- * truncation on another.
+ * particularly interested in the aio/dio case.  We use the rw_lock DLM lock
+ * to protect io on one node from truncation on another.
  */
 static void ocfs2_dio_end_io(struct kiocb *iocb,
 			     loff_t offset,
@@ -569,7 +568,7 @@ static void ocfs2_dio_end_io(struct kiocb *iocb,
 	BUG_ON(!ocfs2_iocb_is_rw_locked(iocb));
 
 	if (ocfs2_iocb_is_sem_locked(iocb)) {
-		up_read(&inode->i_alloc_sem);
+		inode_dio_done(inode);
 		ocfs2_iocb_clear_sem_locked(iocb);
 	}
 
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 1406c37a5722..2c3a465514a2 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2236,9 +2236,9 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 	ocfs2_iocb_clear_sem_locked(iocb);
 
 relock:
-	/* to match setattr's i_mutex -> i_alloc_sem -> rw_lock ordering */
+	/* to match setattr's i_mutex -> rw_lock ordering */
 	if (direct_io) {
-		down_read(&inode->i_alloc_sem);
+		atomic_inc(&inode->i_dio_count);
 		have_alloc_sem = 1;
 		/* communicate with ocfs2_dio_end_io */
 		ocfs2_iocb_set_sem_locked(iocb);
@@ -2290,7 +2290,7 @@ relock:
 	 */
 	if (direct_io && !can_do_direct) {
 		ocfs2_rw_unlock(inode, rw_level);
-		up_read(&inode->i_alloc_sem);
+		inode_dio_done(inode);
 
 		have_alloc_sem = 0;
 		rw_level = -1;
@@ -2361,8 +2361,7 @@ out_dio:
 	/*
 	 * deep in g_f_a_w_n()->ocfs2_direct_IO we pass in a ocfs2_dio_end_io
 	 * function pointer which is called when o_direct io completes so that
-	 * it can unlock our rw lock.  (it's the clustered equivalent of
-	 * i_alloc_sem; protects truncate from racing with pending ios).
+	 * it can unlock our rw lock.
 	 * Unfortunately there are error cases which call end_io and others
 	 * that don't.  so we don't have to unlock the rw_lock if either an
 	 * async dio is going to do it in the future or an end_io after an
@@ -2379,7 +2378,7 @@ out:
 
 out_sems:
 	if (have_alloc_sem) {
-		up_read(&inode->i_alloc_sem);
+		inode_dio_done(inode);
 		ocfs2_iocb_clear_sem_locked(iocb);
 	}
 
@@ -2531,8 +2530,8 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
 	 * need locks to protect pending reads from racing with truncate.
 	 */
 	if (filp->f_flags & O_DIRECT) {
-		down_read(&inode->i_alloc_sem);
 		have_alloc_sem = 1;
+		atomic_inc(&inode->i_dio_count);
 		ocfs2_iocb_set_sem_locked(iocb);
 
 		ret = ocfs2_rw_lock(inode, 0);
@@ -2575,7 +2574,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
 
 bail:
 	if (have_alloc_sem) {
-		up_read(&inode->i_alloc_sem);
+		inode_dio_done(inode);
 		ocfs2_iocb_clear_sem_locked(iocb);
 	}
 	if (rw_level != -1)
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 4ea2ab41fdee..6938d8c68d6e 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -555,11 +555,10 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
 
 		reiserfs_write_unlock(inode->i_sb);
 		mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_XATTR);
-		down_write(&dentry->d_inode->i_alloc_sem);
+		inode_dio_wait(dentry->d_inode);
 		reiserfs_write_lock(inode->i_sb);
 
 		err = reiserfs_setattr(dentry, &newattrs);
-		up_write(&dentry->d_inode->i_alloc_sem);
 		mutex_unlock(&dentry->d_inode->i_mutex);
 	} else
 		update_ctime(inode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1393742bba9b..2fe920774abf 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -779,7 +779,7 @@ struct inode {
 	struct timespec		i_ctime;
 	blkcnt_t		i_blocks;
 	unsigned short          i_bytes;
-	struct rw_semaphore	i_alloc_sem;
+	atomic_t		i_dio_count;
 	const struct file_operations	*i_fop;	/* former ->i_op->default_file_ops */
 	struct file_lock	*i_flock;
 	struct address_space	*i_mapping;
@@ -1705,6 +1705,10 @@ struct super_operations {
  *			set during data writeback, and cleared with a wakeup
  *			on the bit address once it is done.
  *
+ * I_REFERENCED		Marks the inode as recently references on the LRU list.
+ *
+ * I_DIO_WAKEUP		Never set.  Only used as a key for wait_on_bit().
+ *
  * Q: What is the difference between I_WILL_FREE and I_FREEING?
  */
 #define I_DIRTY_SYNC		(1 << 0)
@@ -1718,6 +1722,8 @@ struct super_operations {
 #define __I_SYNC		7
 #define I_SYNC			(1 << __I_SYNC)
 #define I_REFERENCED		(1 << 8)
+#define __I_DIO_WAKEUP		9
+#define I_DIO_WAKEUP		(1 << I_DIO_WAKEUP)
 
 #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES)
 
@@ -1828,7 +1834,6 @@ struct file_system_type {
 	struct lock_class_key i_lock_key;
 	struct lock_class_key i_mutex_key;
 	struct lock_class_key i_mutex_dir_key;
-	struct lock_class_key i_alloc_sem_key;
 };
 
 extern struct dentry *mount_ns(struct file_system_type *fs_type, int flags,
@@ -2404,6 +2409,8 @@ enum {
 };
 
 void dio_end_io(struct bio *bio, int error);
+void inode_dio_wait(struct inode *inode);
+void inode_dio_done(struct inode *inode);
 
 ssize_t __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	struct block_device *bdev, const struct iovec *iov, loff_t offset,
diff --git a/mm/filemap.c b/mm/filemap.c
index a8251a8d3457..f820e600f1ad 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -78,9 +78,6 @@
  *  ->i_mutex			(generic_file_buffered_write)
  *    ->mmap_sem		(fault_in_pages_readable->do_page_fault)
  *
- *  ->i_mutex
- *    ->i_alloc_sem             (various)
- *
  *  inode_wb_list_lock
  *    sb_lock			(fs/fs-writeback.c)
  *    ->mapping->tree_lock	(__sync_single_inode)
diff --git a/mm/madvise.c b/mm/madvise.c
index 2221491ed503..74bf193eff04 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -218,7 +218,7 @@ static long madvise_remove(struct vm_area_struct *vma,
 	endoff = (loff_t)(end - vma->vm_start - 1)
 			+ ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
 
-	/* vmtruncate_range needs to take i_mutex and i_alloc_sem */
+	/* vmtruncate_range needs to take i_mutex */
 	up_read(&current->mm->mmap_sem);
 	error = vmtruncate_range(mapping->host, offset, endoff);
 	down_read(&current->mm->mmap_sem);
diff --git a/mm/rmap.c b/mm/rmap.c
index 23295f65ae43..2540a39eea4a 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -21,7 +21,6 @@
  * Lock ordering in mm:
  *
  * inode->i_mutex	(while writing or truncating, not reading or faulting)
- *   inode->i_alloc_sem (vmtruncate_range)
  *   mm->mmap_sem
  *     page->flags PG_locked (lock_page)
  *       mapping->i_mmap_mutex
diff --git a/mm/truncate.c b/mm/truncate.c
index e13f22efaad7..003c6c685fc8 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -622,12 +622,11 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end)
 		return -ENOSYS;
 
 	mutex_lock(&inode->i_mutex);
-	down_write(&inode->i_alloc_sem);
+	inode_dio_wait(inode);
 	unmap_mapping_range(mapping, offset, (end - offset), 1);
 	inode->i_op->truncate_range(inode, offset, end);
 	/* unmap again to remove racily COWed private pages */
 	unmap_mapping_range(mapping, offset, (end - offset), 1);
-	up_write(&inode->i_alloc_sem);
 	mutex_unlock(&inode->i_mutex);
 
 	return 0;
-- 
cgit v1.2.3


From 11b80f459adaf91a712f95e7734a17655a36bf30 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Fri, 24 Jun 2011 14:29:44 -0400
Subject: rw_semaphore: remove up/down_read_non_owner

Now that the last users is gone these can be removed.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/rwsem.h | 10 ----------
 kernel/rwsem.c        | 16 ----------------
 2 files changed, 26 deletions(-)

(limited to 'include')

diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h
index a8afe9cd000c..77950dfa0a9e 100644
--- a/include/linux/rwsem.h
+++ b/include/linux/rwsem.h
@@ -124,19 +124,9 @@ extern void downgrade_write(struct rw_semaphore *sem);
  */
 extern void down_read_nested(struct rw_semaphore *sem, int subclass);
 extern void down_write_nested(struct rw_semaphore *sem, int subclass);
-/*
- * Take/release a lock when not the owner will release it.
- *
- * [ This API should be avoided as much as possible - the
- *   proper abstraction for this case is completions. ]
- */
-extern void down_read_non_owner(struct rw_semaphore *sem);
-extern void up_read_non_owner(struct rw_semaphore *sem);
 #else
 # define down_read_nested(sem, subclass)		down_read(sem)
 # define down_write_nested(sem, subclass)	down_write(sem)
-# define down_read_non_owner(sem)		down_read(sem)
-# define up_read_non_owner(sem)			up_read(sem)
 #endif
 
 #endif /* _LINUX_RWSEM_H */
diff --git a/kernel/rwsem.c b/kernel/rwsem.c
index cae050b05f5e..176e5e56ffab 100644
--- a/kernel/rwsem.c
+++ b/kernel/rwsem.c
@@ -117,15 +117,6 @@ void down_read_nested(struct rw_semaphore *sem, int subclass)
 
 EXPORT_SYMBOL(down_read_nested);
 
-void down_read_non_owner(struct rw_semaphore *sem)
-{
-	might_sleep();
-
-	__down_read(sem);
-}
-
-EXPORT_SYMBOL(down_read_non_owner);
-
 void down_write_nested(struct rw_semaphore *sem, int subclass)
 {
 	might_sleep();
@@ -136,13 +127,6 @@ void down_write_nested(struct rw_semaphore *sem, int subclass)
 
 EXPORT_SYMBOL(down_write_nested);
 
-void up_read_non_owner(struct rw_semaphore *sem)
-{
-	__up_read(sem);
-}
-
-EXPORT_SYMBOL(up_read_non_owner);
-
 #endif
 
 
-- 
cgit v1.2.3


From aacfc19c626ebd3daa675652457d71019a1f583f Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Fri, 24 Jun 2011 14:29:47 -0400
Subject: fs: simplify the blockdev_direct_IO prototype

Simple filesystems always pass inode->i_sb_bdev as the block device
argument, and never need a end_io handler.  Let's simply things for
them and for my grepping activity by dropping these arguments.  The
only thing not falling into that scheme is ext4, which passes and
end_io handler without needing special flags (yet), but given how
messy the direct I/O code there is use of __blockdev_direct_IO
in one instead of two out of three cases isn't going to make a large
difference anyway.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ext2/inode.c     |  4 ++--
 fs/ext3/inode.c     |  5 ++---
 fs/ext4/inode.c     | 12 ++++++------
 fs/fat/inode.c      |  4 ++--
 fs/hfs/inode.c      |  4 ++--
 fs/hfsplus/inode.c  |  4 ++--
 fs/jfs/inode.c      |  4 ++--
 fs/nilfs2/inode.c   |  4 ++--
 fs/reiserfs/inode.c |  5 ++---
 include/linux/fs.h  |  9 ++++-----
 10 files changed, 26 insertions(+), 29 deletions(-)

(limited to 'include')

diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 06e7c767ab35..a8a58f63f07c 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -843,8 +843,8 @@ ext2_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
 	struct inode *inode = mapping->host;
 	ssize_t ret;
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev,
-				iov, offset, nr_segs, ext2_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 ext2_get_block);
 	if (ret < 0 && (rw & WRITE))
 		ext2_write_failed(mapping, offset + iov_length(iov, nr_segs));
 	return ret;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 99c28b246b89..2978a2a17a59 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -1816,9 +1816,8 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb,
 	}
 
 retry:
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				 offset, nr_segs,
-				 ext3_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 ext3_get_block);
 	/*
 	 * In case of error extending write may have instantiated a few
 	 * blocks outside i_size. Trim these off again.
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 9ec0a2ba2502..1f35573a34e1 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3501,10 +3501,8 @@ retry:
 				 offset, nr_segs,
 				 ext4_get_block, NULL, NULL, 0);
 	else {
-		ret = blockdev_direct_IO(rw, iocb, inode,
-				 inode->i_sb->s_bdev, iov,
-				 offset, nr_segs,
-				 ext4_get_block, NULL);
+		ret = blockdev_direct_IO(rw, iocb, inode, iov,
+				 offset, nr_segs, ext4_get_block);
 
 		if (unlikely((rw & WRITE) && ret < 0)) {
 			loff_t isize = i_size_read(inode);
@@ -3748,11 +3746,13 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
 			EXT4_I(inode)->cur_aio_dio = iocb->private;
 		}
 
-		ret = blockdev_direct_IO(rw, iocb, inode,
+		ret = __blockdev_direct_IO(rw, iocb, inode,
 					 inode->i_sb->s_bdev, iov,
 					 offset, nr_segs,
 					 ext4_get_block_write,
-					 ext4_end_io_dio);
+					 ext4_end_io_dio,
+					 NULL,
+					 DIO_LOCKING | DIO_SKIP_HOLES);
 		if (iocb->private)
 			EXT4_I(inode)->cur_aio_dio = NULL;
 		/*
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 3decce46c38f..5942fec22c65 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -211,8 +211,8 @@ static ssize_t fat_direct_IO(int rw, struct kiocb *iocb,
 	 * FAT need to use the DIO_LOCKING for avoiding the race
 	 * condition of fat_get_block() and ->truncate().
 	 */
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev,
-				 iov, offset, nr_segs, fat_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 fat_get_block);
 	if (ret < 0 && (rw & WRITE))
 		fat_write_failed(mapping, offset + iov_length(iov, nr_segs));
 
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 48d567cc203e..5e7c3f309617 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -123,8 +123,8 @@ static ssize_t hfs_direct_IO(int rw, struct kiocb *iocb,
 	struct inode *inode = file->f_path.dentry->d_inode->i_mapping->host;
 	ssize_t ret;
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				  offset, nr_segs, hfs_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 hfs_get_block);
 
 	/*
 	 * In case of error extending write may have instantiated a few
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index b0a0a4b621eb..5b1cb98741cc 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -119,8 +119,8 @@ static ssize_t hfsplus_direct_IO(int rw, struct kiocb *iocb,
 	struct inode *inode = file->f_path.dentry->d_inode->i_mapping->host;
 	ssize_t ret;
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				  offset, nr_segs, hfsplus_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 hfsplus_get_block);
 
 	/*
 	 * In case of error extending write may have instantiated a few
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 109655904bbc..77b69b27f825 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -329,8 +329,8 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb,
 	struct inode *inode = file->f_mapping->host;
 	ssize_t ret;
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				offset, nr_segs, jfs_get_block, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				 jfs_get_block);
 
 	/*
 	 * In case of error extending write may have instantiated a few
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 13f113154a9f..666628b395f1 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -259,8 +259,8 @@ nilfs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
 		return 0;
 
 	/* Needs synchronization with the cleaner */
-	size = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				  offset, nr_segs, nilfs_get_block, NULL);
+	size = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				  nilfs_get_block);
 
 	/*
 	 * In case of error extending write may have instantiated a few
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index dcf543d8caf1..2922b90ceac1 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -3068,9 +3068,8 @@ static ssize_t reiserfs_direct_IO(int rw, struct kiocb *iocb,
 	struct inode *inode = file->f_mapping->host;
 	ssize_t ret;
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				  offset, nr_segs,
-				  reiserfs_get_blocks_direct_io, NULL);
+	ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs,
+				  reiserfs_get_blocks_direct_io);
 
 	/*
 	 * In case of error extending write may have instantiated a few
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 2fe920774abf..824453be9fee 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2418,12 +2418,11 @@ ssize_t __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	dio_submit_t submit_io,	int flags);
 
 static inline ssize_t blockdev_direct_IO(int rw, struct kiocb *iocb,
-	struct inode *inode, struct block_device *bdev, const struct iovec *iov,
-	loff_t offset, unsigned long nr_segs, get_block_t get_block,
-	dio_iodone_t end_io)
+		struct inode *inode, const struct iovec *iov, loff_t offset,
+		unsigned long nr_segs, get_block_t get_block)
 {
-	return __blockdev_direct_IO(rw, iocb, inode, bdev, iov, offset,
-				    nr_segs, get_block, end_io, NULL,
+	return __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
+				    offset, nr_segs, get_block, NULL, NULL,
 				    DIO_LOCKING | DIO_SKIP_HOLES);
 }
 #endif
-- 
cgit v1.2.3


From f15146380d28b746df3c8b81b392812eb982382a Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Tue, 12 Jul 2011 20:48:39 +0200
Subject: fs: seq_file - add event counter to simplify poll() support

Moving the event counter into the dynamically allocated 'struc seq_file'
allows poll() support without the need to allocate its own tracking
structure.

All current users are switched over to use the new counter.

Requested-by: Andrew Morton akpm@linux-foundation.org
Acked-by: NeilBrown <neilb@suse.de>
Tested-by: Lucas De Marchi lucas.demarchi@profusion.mobi
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 drivers/md/md.c               | 26 ++++++++------------------
 fs/namespace.c                |  4 ++--
 fs/proc/base.c                |  2 +-
 include/linux/mnt_namespace.h |  1 -
 include/linux/seq_file.h      |  1 +
 mm/swapfile.c                 | 29 ++++++++---------------------
 6 files changed, 20 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 91e31e260b4a..dfc9425db70b 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -6394,16 +6394,11 @@ static void md_seq_stop(struct seq_file *seq, void *v)
 		mddev_put(mddev);
 }
 
-struct mdstat_info {
-	int event;
-};
-
 static int md_seq_show(struct seq_file *seq, void *v)
 {
 	mddev_t *mddev = v;
 	sector_t sectors;
 	mdk_rdev_t *rdev;
-	struct mdstat_info *mi = seq->private;
 	struct bitmap *bitmap;
 
 	if (v == (void*)1) {
@@ -6415,7 +6410,7 @@ static int md_seq_show(struct seq_file *seq, void *v)
 
 		spin_unlock(&pers_lock);
 		seq_printf(seq, "\n");
-		mi->event = atomic_read(&md_event_count);
+		seq->poll_event = atomic_read(&md_event_count);
 		return 0;
 	}
 	if (v == (void*)2) {
@@ -6527,26 +6522,21 @@ static const struct seq_operations md_seq_ops = {
 
 static int md_seq_open(struct inode *inode, struct file *file)
 {
+	struct seq_file *seq;
 	int error;
-	struct mdstat_info *mi = kmalloc(sizeof(*mi), GFP_KERNEL);
-	if (mi == NULL)
-		return -ENOMEM;
 
 	error = seq_open(file, &md_seq_ops);
 	if (error)
-		kfree(mi);
-	else {
-		struct seq_file *p = file->private_data;
-		p->private = mi;
-		mi->event = atomic_read(&md_event_count);
-	}
+		return error;
+
+	seq = file->private_data;
+	seq->poll_event = atomic_read(&md_event_count);
 	return error;
 }
 
 static unsigned int mdstat_poll(struct file *filp, poll_table *wait)
 {
-	struct seq_file *m = filp->private_data;
-	struct mdstat_info *mi = m->private;
+	struct seq_file *seq = filp->private_data;
 	int mask;
 
 	poll_wait(filp, &md_event_waiters, wait);
@@ -6554,7 +6544,7 @@ static unsigned int mdstat_poll(struct file *filp, poll_table *wait)
 	/* always allow read */
 	mask = POLLIN | POLLRDNORM;
 
-	if (mi->event != atomic_read(&md_event_count))
+	if (seq->poll_event != atomic_read(&md_event_count))
 		mask |= POLLERR | POLLPRI;
 	return mask;
 }
diff --git a/fs/namespace.c b/fs/namespace.c
index fe59bd145d21..cda50fe9250a 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -934,8 +934,8 @@ int mnt_had_events(struct proc_mounts *p)
 	int res = 0;
 
 	br_read_lock(vfsmount_lock);
-	if (p->event != ns->event) {
-		p->event = ns->event;
+	if (p->m.poll_event != ns->event) {
+		p->m.poll_event = ns->event;
 		res = 1;
 	}
 	br_read_unlock(vfsmount_lock);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index be1ff932033b..3dc5e2a5cc38 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -673,7 +673,7 @@ static int mounts_open_common(struct inode *inode, struct file *file,
 	p->m.private = p;
 	p->ns = ns;
 	p->root = root;
-	p->event = ns->event;
+	p->m.poll_event = ns->event;
 
 	return 0;
 
diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
index 0b89efc6f215..29304855652d 100644
--- a/include/linux/mnt_namespace.h
+++ b/include/linux/mnt_namespace.h
@@ -18,7 +18,6 @@ struct proc_mounts {
 	struct seq_file m; /* must be the first element */
 	struct mnt_namespace *ns;
 	struct path root;
-	int event;
 };
 
 struct fs_struct;
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 03c0232b4169..be720cd2038d 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -23,6 +23,7 @@ struct seq_file {
 	u64 version;
 	struct mutex lock;
 	const struct seq_operations *op;
+	int poll_event;
 	void *private;
 };
 
diff --git a/mm/swapfile.c b/mm/swapfile.c
index ff8dc1a18cb4..1b8c33907242 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1681,19 +1681,14 @@ out:
 }
 
 #ifdef CONFIG_PROC_FS
-struct proc_swaps {
-	struct seq_file seq;
-	int event;
-};
-
 static unsigned swaps_poll(struct file *file, poll_table *wait)
 {
-	struct proc_swaps *s = file->private_data;
+	struct seq_file *seq = file->private_data;
 
 	poll_wait(file, &proc_poll_wait, wait);
 
-	if (s->event != atomic_read(&proc_poll_event)) {
-		s->event = atomic_read(&proc_poll_event);
+	if (seq->poll_event != atomic_read(&proc_poll_event)) {
+		seq->poll_event = atomic_read(&proc_poll_event);
 		return POLLIN | POLLRDNORM | POLLERR | POLLPRI;
 	}
 
@@ -1783,24 +1778,16 @@ static const struct seq_operations swaps_op = {
 
 static int swaps_open(struct inode *inode, struct file *file)
 {
-	struct proc_swaps *s;
+	struct seq_file *seq;
 	int ret;
 
-	s = kmalloc(sizeof(struct proc_swaps), GFP_KERNEL);
-	if (!s)
-		return -ENOMEM;
-
-	file->private_data = s;
-
 	ret = seq_open(file, &swaps_op);
-	if (ret) {
-		kfree(s);
+	if (ret)
 		return ret;
-	}
 
-	s->seq.private = s;
-	s->event = atomic_read(&proc_poll_event);
-	return ret;
+	seq = file->private_data;
+	seq->poll_event = atomic_read(&proc_poll_event);
+	return 0;
 }
 
 static const struct file_operations proc_swaps_operations = {
-- 
cgit v1.2.3


From 982d816581eeeacfe5b2b7c6d47d13a157616eff Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Mon, 18 Jul 2011 13:21:35 -0400
Subject: fs: add SEEK_HOLE and SEEK_DATA flags

This just gets us ready to support the SEEK_HOLE and SEEK_DATA flags.  Turns out
using fiemap in things like cp cause more problems than it solves, so lets try
and give userspace an interface that doesn't suck.  We need to match solaris
here, and the definitions are

*o* If /whence/ is SEEK_HOLE, the offset of the start of the
next hole greater than or equal to the supplied offset
is returned. The definition of a hole is provided near
the end of the DESCRIPTION.

*o* If /whence/ is SEEK_DATA, the file pointer is set to the
start of the next non-hole file region greater than or
equal to the supplied offset.

So in the generic case the entire file is data and there is a virtual hole at
the end.  That means we will just return i_size for SEEK_HOLE and will return
the same offset for SEEK_DATA.  This is how Solaris does it so we have to do it
the same way.

Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/porting | 10 +++++++++
 fs/read_write.c                   | 44 ++++++++++++++++++++++++++++++++++++---
 include/linux/fs.h                |  4 +++-
 3 files changed, 54 insertions(+), 4 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting
index 0eeb3954dea3..6b96773e27cb 100644
--- a/Documentation/filesystems/porting
+++ b/Documentation/filesystems/porting
@@ -411,3 +411,13 @@ to some pointer to returning that pointer.  On errors return ERR_PTR(...).
 argument; instead of passing IPERM_FLAG_RCU we add MAY_NOT_BLOCK into mask.
 	generic_permission() has also lost the check_acl argument; if you want
 non-NULL to be used for that inode, put it into ->i_op->check_acl.
+
+--
+[mandatory]
+	If you implement your own ->llseek() you must handle SEEK_HOLE and
+SEEK_DATA.  You can hanle this by returning -EINVAL, but it would be nicer to
+support it in some way.  The generic handler assumes that the entire file is
+data and there is a virtual hole at the end of the file.  So if the provided
+offset is less than i_size and SEEK_DATA is specified, return the same offset.
+If the above is true for the offset and you are given SEEK_HOLE, return the end
+of the file.  If the offset is i_size or greater return -ENXIO in either case.
diff --git a/fs/read_write.c b/fs/read_write.c
index 5520f8ad5504..5907b49e4d7e 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -64,6 +64,23 @@ generic_file_llseek_unlocked(struct file *file, loff_t offset, int origin)
 			return file->f_pos;
 		offset += file->f_pos;
 		break;
+	case SEEK_DATA:
+		/*
+		 * In the generic case the entire file is data, so as long as
+		 * offset isn't at the end of the file then the offset is data.
+		 */
+		if (offset >= inode->i_size)
+			return -ENXIO;
+		break;
+	case SEEK_HOLE:
+		/*
+		 * There is a virtual hole at the end of the file, so as long as
+		 * offset isn't i_size or larger, return i_size.
+		 */
+		if (offset >= inode->i_size)
+			return -ENXIO;
+		offset = inode->i_size;
+		break;
 	}
 
 	if (offset < 0 && !unsigned_offsets(file))
@@ -128,12 +145,13 @@ EXPORT_SYMBOL(no_llseek);
 
 loff_t default_llseek(struct file *file, loff_t offset, int origin)
 {
+	struct inode *inode = file->f_path.dentry->d_inode;
 	loff_t retval;
 
-	mutex_lock(&file->f_dentry->d_inode->i_mutex);
+	mutex_lock(&inode->i_mutex);
 	switch (origin) {
 		case SEEK_END:
-			offset += i_size_read(file->f_path.dentry->d_inode);
+			offset += i_size_read(inode);
 			break;
 		case SEEK_CUR:
 			if (offset == 0) {
@@ -141,6 +159,26 @@ loff_t default_llseek(struct file *file, loff_t offset, int origin)
 				goto out;
 			}
 			offset += file->f_pos;
+			break;
+		case SEEK_DATA:
+			/*
+			 * In the generic case the entire file is data, so as
+			 * long as offset isn't at the end of the file then the
+			 * offset is data.
+			 */
+			if (offset >= inode->i_size)
+				return -ENXIO;
+			break;
+		case SEEK_HOLE:
+			/*
+			 * There is a virtual hole at the end of the file, so
+			 * as long as offset isn't i_size or larger, return
+			 * i_size.
+			 */
+			if (offset >= inode->i_size)
+				return -ENXIO;
+			offset = inode->i_size;
+			break;
 	}
 	retval = -EINVAL;
 	if (offset >= 0 || unsigned_offsets(file)) {
@@ -151,7 +189,7 @@ loff_t default_llseek(struct file *file, loff_t offset, int origin)
 		retval = offset;
 	}
 out:
-	mutex_unlock(&file->f_dentry->d_inode->i_mutex);
+	mutex_unlock(&inode->i_mutex);
 	return retval;
 }
 EXPORT_SYMBOL(default_llseek);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 824453be9fee..4a61f98823a6 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -32,7 +32,9 @@
 #define SEEK_SET	0	/* seek relative to beginning of file */
 #define SEEK_CUR	1	/* seek relative to current file position */
 #define SEEK_END	2	/* seek relative to end of file */
-#define SEEK_MAX	SEEK_END
+#define SEEK_DATA	3	/* seek to the next data */
+#define SEEK_HOLE	4	/* seek to the next hole */
+#define SEEK_MAX	SEEK_HOLE
 
 struct fstrim_range {
 	__u64 start;
-- 
cgit v1.2.3


From 02c24a82187d5a628c68edfe71ae60dc135cd178 Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Sat, 16 Jul 2011 20:44:56 -0400
Subject: fs: push i_mutex and filemap_write_and_wait down into ->fsync()
 handlers

Btrfs needs to be able to control how filemap_write_and_wait_range() is called
in fsync to make it less of a painful operation, so push down taking i_mutex and
the calling of filemap_write_and_wait() down into the ->fsync() handlers.  Some
file systems can drop taking the i_mutex altogether it seems, like ext3 and
ocfs2.  For correctness sake I just pushed everything down in all cases to make
sure that we keep the current behavior the same for everybody, and then each
individual fs maintainer can make up their mind about what to do from there.
Thanks,

Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/Locking        |  6 ++---
 Documentation/filesystems/porting        |  7 ++++++
 Documentation/filesystems/vfs.txt        |  2 +-
 arch/powerpc/platforms/cell/spufs/file.c | 11 +++++++--
 drivers/char/ps3flash.c                  |  9 ++++++--
 drivers/mtd/ubi/cdev.c                   | 10 ++++++---
 drivers/staging/pohmelfs/inode.c         | 11 ++++++---
 drivers/usb/gadget/printer.c             |  5 ++++-
 drivers/video/fb_defio.c                 | 11 +++++++--
 fs/9p/v9fs_vfs.h                         |  3 ++-
 fs/9p/vfs_file.c                         | 22 ++++++++++++++++--
 fs/affs/affs.h                           |  2 +-
 fs/affs/file.c                           |  8 ++++++-
 fs/afs/internal.h                        |  2 +-
 fs/afs/write.c                           | 18 +++++++++++----
 fs/bad_inode.c                           |  3 ++-
 fs/block_dev.c                           |  6 +----
 fs/btrfs/ctree.h                         |  2 +-
 fs/btrfs/file.c                          | 21 +++++++++++++-----
 fs/ceph/caps.c                           |  6 +++--
 fs/ceph/dir.c                            | 10 ++++++++-
 fs/ceph/super.h                          |  3 ++-
 fs/cifs/cifsfs.h                         |  4 ++--
 fs/cifs/file.c                           | 18 +++++++++++++--
 fs/coda/coda_int.h                       |  2 +-
 fs/coda/file.c                           |  8 ++++++-
 fs/ecryptfs/file.c                       |  7 +++---
 fs/exofs/file.c                          | 10 ++++++++-
 fs/ext2/ext2.h                           |  3 ++-
 fs/ext2/file.c                           |  4 ++--
 fs/ext3/fsync.c                          | 18 +++++++++++++--
 fs/ext4/ext4.h                           |  2 +-
 fs/ext4/fsync.c                          | 38 +++++++++++++++++++++++++++++---
 fs/fat/fat.h                             |  3 ++-
 fs/fat/file.c                            |  4 ++--
 fs/fuse/dir.c                            |  5 +++--
 fs/fuse/file.c                           | 24 +++++++++++++++-----
 fs/fuse/fuse_i.h                         |  3 ++-
 fs/gfs2/file.c                           | 17 +++++++++++---
 fs/hfs/inode.c                           |  9 +++++++-
 fs/hfsplus/hfsplus_fs.h                  |  3 ++-
 fs/hfsplus/inode.c                       | 10 ++++++++-
 fs/hostfs/hostfs_kern.c                  | 15 +++++++++++--
 fs/hpfs/file.c                           |  7 +++++-
 fs/hpfs/hpfs_fn.h                        |  2 +-
 fs/hppfs/hppfs.c                         |  5 +++--
 fs/jffs2/file.c                          |  9 +++++++-
 fs/jffs2/os-linux.h                      |  2 +-
 fs/jfs/file.c                            |  9 +++++++-
 fs/jfs/jfs_inode.h                       |  2 +-
 fs/libfs.c                               | 16 ++++++++++----
 fs/logfs/file.c                          | 11 ++++++++-
 fs/logfs/logfs.h                         |  2 +-
 fs/ncpfs/file.c                          |  4 ++--
 fs/nfs/dir.c                             |  8 +++++--
 fs/nfs/file.c                            | 11 ++++++---
 fs/nilfs2/file.c                         | 12 ++++++++--
 fs/nilfs2/nilfs.h                        |  2 +-
 fs/ntfs/dir.c                            | 10 ++++++++-
 fs/ntfs/file.c                           | 10 ++++++++-
 fs/ocfs2/file.c                          | 14 +++++++++++-
 fs/reiserfs/dir.c                        | 13 +++++++++--
 fs/reiserfs/file.c                       |  9 +++++++-
 fs/sync.c                                | 25 +++------------------
 fs/ubifs/file.c                          | 21 +++++++++---------
 fs/ubifs/ubifs.h                         |  2 +-
 fs/xfs/linux-2.6/xfs_file.c              | 17 +++++++-------
 include/linux/ext3_fs.h                  |  2 +-
 include/linux/fb.h                       |  3 ++-
 include/linux/fs.h                       |  9 ++++----
 ipc/shm.c                                |  4 ++--
 71 files changed, 462 insertions(+), 164 deletions(-)

(limited to 'include')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 9b6ed7c9f34f..ca7e25292542 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -412,7 +412,7 @@ prototypes:
 	int (*open) (struct inode *, struct file *);
 	int (*flush) (struct file *);
 	int (*release) (struct inode *, struct file *);
-	int (*fsync) (struct file *, int datasync);
+	int (*fsync) (struct file *, loff_t start, loff_t end, int datasync);
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);
@@ -438,9 +438,7 @@ prototypes:
 
 locking rules:
 	All may block except for ->setlease.
-	No VFS locks held on entry except for ->fsync and ->setlease.
-
-->fsync() has i_mutex on inode.
+	No VFS locks held on entry except for ->setlease.
 
 ->setlease has the file_list_lock held and must not sleep.
 
diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting
index 6b96773e27cb..7f8861d341ea 100644
--- a/Documentation/filesystems/porting
+++ b/Documentation/filesystems/porting
@@ -421,3 +421,10 @@ data and there is a virtual hole at the end of the file.  So if the provided
 offset is less than i_size and SEEK_DATA is specified, return the same offset.
 If the above is true for the offset and you are given SEEK_HOLE, return the end
 of the file.  If the offset is i_size or greater return -ENXIO in either case.
+
+[mandatory]
+	If you have your own ->fsync() you must make sure to call
+filemap_write_and_wait_range() so that all dirty pages are synced out properly.
+You must also keep in mind that ->fsync() is not called with i_mutex held
+anymore, so if you require i_mutex locking you must make sure to take it and
+release it yourself.
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 6bf85b78cfea..eff6617c9a0f 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -777,7 +777,7 @@ struct file_operations {
 	int (*open) (struct inode *, struct file *);
 	int (*flush) (struct file *);
 	int (*release) (struct inode *, struct file *);
-	int (*fsync) (struct file *, int datasync);
+	int (*fsync) (struct file *, loff_t, loff_t, int datasync);
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);
diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c
index 3c7c3f82d842..fb59c46e9e9e 100644
--- a/arch/powerpc/platforms/cell/spufs/file.c
+++ b/arch/powerpc/platforms/cell/spufs/file.c
@@ -1850,9 +1850,16 @@ out:
 	return ret;
 }
 
-static int spufs_mfc_fsync(struct file *file, int datasync)
+static int spufs_mfc_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	return spufs_mfc_flush(file, NULL);
+	struct inode *inode = file->f_path.dentry->d_inode;
+	int err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (!err) {
+		mutex_lock(&inode->i_mutex);
+		err = spufs_mfc_flush(file, NULL);
+		mutex_unlock(&inode->i_mutex);
+	}
+	return err;
 }
 
 static int spufs_mfc_fasync(int fd, struct file *file, int on)
diff --git a/drivers/char/ps3flash.c b/drivers/char/ps3flash.c
index 5a06787e5be3..d0c57c2e2909 100644
--- a/drivers/char/ps3flash.c
+++ b/drivers/char/ps3flash.c
@@ -309,9 +309,14 @@ static int ps3flash_flush(struct file *file, fl_owner_t id)
 	return ps3flash_writeback(ps3flash_dev);
 }
 
-static int ps3flash_fsync(struct file *file, int datasync)
+static int ps3flash_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	return ps3flash_writeback(ps3flash_dev);
+	struct inode *inode = file->f_path.dentry->d_inode;
+	int err;
+	mutex_lock(&inode->i_mutex);
+	err = ps3flash_writeback(ps3flash_dev);
+	mutex_unlock(&inode->i_mutex);
+	return err;
 }
 
 static irqreturn_t ps3flash_interrupt(int irq, void *data)
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 191f3bb3c41a..3320a50ba4f0 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -189,12 +189,16 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin)
 	return new_offset;
 }
 
-static int vol_cdev_fsync(struct file *file, int datasync)
+static int vol_cdev_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct ubi_volume_desc *desc = file->private_data;
 	struct ubi_device *ubi = desc->vol->ubi;
-
-	return ubi_sync(ubi->ubi_num);
+	struct inode *inode = file->f_path.dentry->d_inode;
+	int err;
+	mutex_lock(&inode->i_mutex);
+	err = ubi_sync(ubi->ubi_num);
+	mutex_unlock(&inode->i_mutex);
+	return err;
 }
 
 
diff --git a/drivers/staging/pohmelfs/inode.c b/drivers/staging/pohmelfs/inode.c
index c0f0ac7c1cdb..f3c6060c96b8 100644
--- a/drivers/staging/pohmelfs/inode.c
+++ b/drivers/staging/pohmelfs/inode.c
@@ -887,11 +887,16 @@ static struct inode *pohmelfs_alloc_inode(struct super_block *sb)
 /*
  * We want fsync() to work on POHMELFS.
  */
-static int pohmelfs_fsync(struct file *file, int datasync)
+static int pohmelfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
-
-	return sync_inode_metadata(inode, 1);
+	int err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (!err) {
+		mutex_lock(&inode->i_mutex);
+		err = sync_inode_metadata(inode, 1);
+		mutex_unlock(&inode->i_mutex);
+	}
+	return err;
 }
 
 ssize_t pohmelfs_write(struct file *file, const char __user *buf,
diff --git a/drivers/usb/gadget/printer.c b/drivers/usb/gadget/printer.c
index 271ef94668e7..978e6a101bf2 100644
--- a/drivers/usb/gadget/printer.c
+++ b/drivers/usb/gadget/printer.c
@@ -795,12 +795,14 @@ printer_write(struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 }
 
 static int
-printer_fsync(struct file *fd, int datasync)
+printer_fsync(struct file *fd, loff_t start, loff_t end, int datasync)
 {
 	struct printer_dev	*dev = fd->private_data;
+	struct inode *inode = fd->f_path.dentry->d_inode;
 	unsigned long		flags;
 	int			tx_list_empty;
 
+	mutex_lock(&inode->i_mutex);
 	spin_lock_irqsave(&dev->lock, flags);
 	tx_list_empty = (likely(list_empty(&dev->tx_reqs)));
 	spin_unlock_irqrestore(&dev->lock, flags);
@@ -810,6 +812,7 @@ printer_fsync(struct file *fd, int datasync)
 		wait_event_interruptible(dev->tx_flush_wait,
 				(likely(list_empty(&dev->tx_reqs_active))));
 	}
+	mutex_unlock(&inode->i_mutex);
 
 	return 0;
 }
diff --git a/drivers/video/fb_defio.c b/drivers/video/fb_defio.c
index 804000183c5e..32814e8800e0 100644
--- a/drivers/video/fb_defio.c
+++ b/drivers/video/fb_defio.c
@@ -66,19 +66,26 @@ static int fb_deferred_io_fault(struct vm_area_struct *vma,
 	return 0;
 }
 
-int fb_deferred_io_fsync(struct file *file, int datasync)
+int fb_deferred_io_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct fb_info *info = file->private_data;
+	struct inode *inode = file->f_path.dentry->d_inode;
+	int err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
 
 	/* Skip if deferred io is compiled-in but disabled on this fbdev */
 	if (!info->fbdefio)
 		return 0;
 
+	mutex_lock(&inode->i_mutex);
 	/* Kill off the delayed work */
 	cancel_delayed_work_sync(&info->deferred_work);
 
 	/* Run it immediately */
-	return schedule_delayed_work(&info->deferred_work, 0);
+	err = schedule_delayed_work(&info->deferred_work, 0);
+	mutex_unlock(&inode->i_mutex);
+	return err;
 }
 EXPORT_SYMBOL_GPL(fb_deferred_io_fsync);
 
diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h
index 4014160903a9..46ce357ca1ab 100644
--- a/fs/9p/v9fs_vfs.h
+++ b/fs/9p/v9fs_vfs.h
@@ -70,7 +70,8 @@ ssize_t v9fs_file_readn(struct file *, char *, char __user *, u32, u64);
 ssize_t v9fs_fid_readn(struct p9_fid *, char *, char __user *, u32, u64);
 void v9fs_blank_wstat(struct p9_wstat *wstat);
 int v9fs_vfs_setattr_dotl(struct dentry *, struct iattr *);
-int v9fs_file_fsync_dotl(struct file *filp, int datasync);
+int v9fs_file_fsync_dotl(struct file *filp, loff_t start, loff_t end,
+			 int datasync);
 ssize_t v9fs_file_write_internal(struct inode *, struct p9_fid *,
 				 const char __user *, size_t, loff_t *, int);
 int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode);
diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index ffed55817f0c..3c173fcc2c5a 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -519,32 +519,50 @@ out:
 }
 
 
-static int v9fs_file_fsync(struct file *filp, int datasync)
+static int v9fs_file_fsync(struct file *filp, loff_t start, loff_t end,
+			   int datasync)
 {
 	struct p9_fid *fid;
+	struct inode *inode = filp->f_mapping->host;
 	struct p9_wstat wstat;
 	int retval;
 
+	retval = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (retval)
+		return retval;
+
+	mutex_lock(&inode->i_mutex);
 	P9_DPRINTK(P9_DEBUG_VFS, "filp %p datasync %x\n", filp, datasync);
 
 	fid = filp->private_data;
 	v9fs_blank_wstat(&wstat);
 
 	retval = p9_client_wstat(fid, &wstat);
+	mutex_unlock(&inode->i_mutex);
+
 	return retval;
 }
 
-int v9fs_file_fsync_dotl(struct file *filp, int datasync)
+int v9fs_file_fsync_dotl(struct file *filp, loff_t start, loff_t end,
+			 int datasync)
 {
 	struct p9_fid *fid;
+	struct inode *inode = filp->f_mapping->host;
 	int retval;
 
+	retval = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (retval)
+		return retval;
+
+	mutex_lock(&inode->i_mutex);
 	P9_DPRINTK(P9_DEBUG_VFS, "v9fs_file_fsync_dotl: filp %p datasync %x\n",
 			filp, datasync);
 
 	fid = filp->private_data;
 
 	retval = p9_client_fsync(fid, datasync);
+	mutex_unlock(&inode->i_mutex);
+
 	return retval;
 }
 
diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index 0e95f73a7023..c2b9c79eb64e 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -182,7 +182,7 @@ extern int			 affs_add_entry(struct inode *dir, struct inode *inode, struct dent
 
 void		affs_free_prealloc(struct inode *inode);
 extern void	affs_truncate(struct inode *);
-int		affs_file_fsync(struct file *, int);
+int		affs_file_fsync(struct file *, loff_t, loff_t, int);
 
 /* dir.c */
 
diff --git a/fs/affs/file.c b/fs/affs/file.c
index acf321b70fcd..2f4c935cb327 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -923,14 +923,20 @@ affs_truncate(struct inode *inode)
 	affs_free_prealloc(inode);
 }
 
-int affs_file_fsync(struct file *filp, int datasync)
+int affs_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	int ret, err;
 
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
+	mutex_lock(&inode->i_mutex);
 	ret = write_inode_now(inode, 0);
 	err = sync_blockdev(inode->i_sb->s_bdev);
 	if (!ret)
 		ret = err;
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index f396d337b817..d2b0888126d4 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -750,7 +750,7 @@ extern void afs_pages_written_back(struct afs_vnode *, struct afs_call *);
 extern ssize_t afs_file_write(struct kiocb *, const struct iovec *,
 			      unsigned long, loff_t);
 extern int afs_writeback_all(struct afs_vnode *);
-extern int afs_fsync(struct file *, int);
+extern int afs_fsync(struct file *, loff_t, loff_t, int);
 
 
 /*****************************************************************************/
diff --git a/fs/afs/write.c b/fs/afs/write.c
index b806285ff853..9aa52d93c73c 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -681,9 +681,10 @@ int afs_writeback_all(struct afs_vnode *vnode)
  * - the return status from this call provides a reliable indication of
  *   whether any write errors occurred for this process.
  */
-int afs_fsync(struct file *file, int datasync)
+int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct dentry *dentry = file->f_path.dentry;
+	struct inode *inode = file->f_mapping->host;
 	struct afs_writeback *wb, *xwb;
 	struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode);
 	int ret;
@@ -692,12 +693,19 @@ int afs_fsync(struct file *file, int datasync)
 	       vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name,
 	       datasync);
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	/* use a writeback record as a marker in the queue - when this reaches
 	 * the front of the queue, all the outstanding writes are either
 	 * completed or rejected */
 	wb = kzalloc(sizeof(*wb), GFP_KERNEL);
-	if (!wb)
-		return -ENOMEM;
+	if (!wb) {
+		ret = -ENOMEM;
+		goto out;
+	}
 	wb->vnode = vnode;
 	wb->first = 0;
 	wb->last = -1;
@@ -720,7 +728,7 @@ int afs_fsync(struct file *file, int datasync)
 	if (ret < 0) {
 		afs_put_writeback(wb);
 		_leave(" = %d [wb]", ret);
-		return ret;
+		goto out;
 	}
 
 	/* wait for the preceding writes to actually complete */
@@ -729,6 +737,8 @@ int afs_fsync(struct file *file, int datasync)
 				       vnode->writebacks.next == &wb->link);
 	afs_put_writeback(wb);
 	_leave(" = %d", ret);
+out:
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index f024d8aaddef..9205cf25f1c6 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -87,7 +87,8 @@ static int bad_file_release(struct inode *inode, struct file *filp)
 	return -EIO;
 }
 
-static int bad_file_fsync(struct file *file, int datasync)
+static int bad_file_fsync(struct file *file, loff_t start, loff_t end,
+			  int datasync)
 {
 	return -EIO;
 }
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 966617a422d9..9fb0b15331d3 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -378,7 +378,7 @@ out:
 	return retval;
 }
 	
-int blkdev_fsync(struct file *filp, int datasync)
+int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 {
 	struct inode *bd_inode = filp->f_mapping->host;
 	struct block_device *bdev = I_BDEV(bd_inode);
@@ -389,14 +389,10 @@ int blkdev_fsync(struct file *filp, int datasync)
 	 * i_mutex and doing so causes performance issues with concurrent
 	 * O_SYNC writers to a block device.
 	 */
-	mutex_unlock(&bd_inode->i_mutex);
-
 	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL);
 	if (error == -EOPNOTSUPP)
 		error = 0;
 
-	mutex_lock(&bd_inode->i_mutex);
-
 	return error;
 }
 EXPORT_SYMBOL(blkdev_fsync);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index f1ff62bff1b3..82be74efbb26 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2605,7 +2605,7 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
 int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans,
 			   struct inode *inode);
 int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info);
-int btrfs_sync_file(struct file *file, int datasync);
+int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync);
 int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
 			    int skip_pinned);
 extern const struct file_operations btrfs_file_operations;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index bd4d061c6e4d..59cbdb120ad0 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1452,7 +1452,7 @@ int btrfs_release_file(struct inode *inode, struct file *filp)
  * important optimization for directories because holding the mutex prevents
  * new operations on the dir while we write to disk.
  */
-int btrfs_sync_file(struct file *file, int datasync)
+int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct dentry *dentry = file->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
@@ -1462,9 +1462,13 @@ int btrfs_sync_file(struct file *file, int datasync)
 
 	trace_btrfs_sync_file(file, datasync);
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	/* we wait first, since the writeback may change the inode */
 	root->log_batch++;
-	/* the VFS called filemap_fdatawrite for us */
 	btrfs_wait_ordered_range(inode, 0, (u64)-1);
 	root->log_batch++;
 
@@ -1472,8 +1476,10 @@ int btrfs_sync_file(struct file *file, int datasync)
 	 * check the transaction that last modified this inode
 	 * and see if its already been committed
 	 */
-	if (!BTRFS_I(inode)->last_trans)
+	if (!BTRFS_I(inode)->last_trans) {
+		mutex_unlock(&inode->i_mutex);
 		goto out;
+	}
 
 	/*
 	 * if the last transaction that changed this file was before
@@ -1484,6 +1490,7 @@ int btrfs_sync_file(struct file *file, int datasync)
 	if (BTRFS_I(inode)->last_trans <=
 	    root->fs_info->last_trans_committed) {
 		BTRFS_I(inode)->last_trans = 0;
+		mutex_unlock(&inode->i_mutex);
 		goto out;
 	}
 
@@ -1496,12 +1503,15 @@ int btrfs_sync_file(struct file *file, int datasync)
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
+		mutex_unlock(&inode->i_mutex);
 		goto out;
 	}
 
 	ret = btrfs_log_dentry_safe(trans, root, dentry);
-	if (ret < 0)
+	if (ret < 0) {
+		mutex_unlock(&inode->i_mutex);
 		goto out;
+	}
 
 	/* we've logged all the items and now have a consistent
 	 * version of the file in the log.  It is possible that
@@ -1513,7 +1523,7 @@ int btrfs_sync_file(struct file *file, int datasync)
 	 * file again, but that will end up using the synchronization
 	 * inside btrfs_sync_log to keep things safe.
 	 */
-	mutex_unlock(&dentry->d_inode->i_mutex);
+	mutex_unlock(&inode->i_mutex);
 
 	if (ret != BTRFS_NO_LOG_SYNC) {
 		if (ret > 0) {
@@ -1528,7 +1538,6 @@ int btrfs_sync_file(struct file *file, int datasync)
 	} else {
 		ret = btrfs_end_transaction(trans, root);
 	}
-	mutex_lock(&dentry->d_inode->i_mutex);
 out:
 	return ret > 0 ? -EIO : ret;
 }
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index f605753c8fe9..8d74ad7ba556 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1811,7 +1811,7 @@ out:
 	spin_unlock(&ci->i_unsafe_lock);
 }
 
-int ceph_fsync(struct file *file, int datasync)
+int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1822,9 +1822,10 @@ int ceph_fsync(struct file *file, int datasync)
 	dout("fsync %p%s\n", inode, datasync ? " datasync" : "");
 	sync_write_wait(inode);
 
-	ret = filemap_write_and_wait(inode->i_mapping);
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
 	if (ret < 0)
 		return ret;
+	mutex_lock(&inode->i_mutex);
 
 	dirty = try_flush_caps(inode, NULL, &flush_tid);
 	dout("fsync dirty caps are %s\n", ceph_cap_string(dirty));
@@ -1841,6 +1842,7 @@ int ceph_fsync(struct file *file, int datasync)
 	}
 
 	dout("fsync %p%s done\n", inode, datasync ? " datasync" : "");
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 0972b457a03f..1065ac779840 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -1118,7 +1118,8 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
  * an fsync() on a dir will wait for any uncommitted directory
  * operations to commit.
  */
-static int ceph_dir_fsync(struct file *file, int datasync)
+static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end,
+			  int datasync)
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1128,6 +1129,11 @@ static int ceph_dir_fsync(struct file *file, int datasync)
 	int ret = 0;
 
 	dout("dir_fsync %p\n", inode);
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	spin_lock(&ci->i_unsafe_lock);
 	if (list_empty(head))
 		goto out;
@@ -1161,6 +1167,8 @@ static int ceph_dir_fsync(struct file *file, int datasync)
 	} while (req->r_tid < last_tid);
 out:
 	spin_unlock(&ci->i_unsafe_lock);
+	mutex_unlock(&inode->i_mutex);
+
 	return ret;
 }
 
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 56c41ef47cad..30446b144e3d 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -728,7 +728,8 @@ extern void ceph_put_cap(struct ceph_mds_client *mdsc,
 
 extern void ceph_queue_caps_release(struct inode *inode);
 extern int ceph_write_inode(struct inode *inode, struct writeback_control *wbc);
-extern int ceph_fsync(struct file *file, int datasync);
+extern int ceph_fsync(struct file *file, loff_t start, loff_t end,
+		      int datasync);
 extern void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
 				    struct ceph_mds_session *session);
 extern struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci,
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 036ca83e5f46..fbd050c8d52a 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -91,8 +91,8 @@ extern ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
 extern ssize_t cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
 				  unsigned long nr_segs, loff_t pos);
 extern int cifs_lock(struct file *, int, struct file_lock *);
-extern int cifs_fsync(struct file *, int);
-extern int cifs_strict_fsync(struct file *, int);
+extern int cifs_fsync(struct file *, loff_t, loff_t, int);
+extern int cifs_strict_fsync(struct file *, loff_t, loff_t, int);
 extern int cifs_flush(struct file *, fl_owner_t id);
 extern int cifs_file_mmap(struct file * , struct vm_area_struct *);
 extern int cifs_file_strict_mmap(struct file * , struct vm_area_struct *);
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index bb71471a4d9d..cef584451113 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -1401,7 +1401,8 @@ static int cifs_write_end(struct file *file, struct address_space *mapping,
 	return rc;
 }
 
-int cifs_strict_fsync(struct file *file, int datasync)
+int cifs_strict_fsync(struct file *file, loff_t start, loff_t end,
+		      int datasync)
 {
 	int xid;
 	int rc = 0;
@@ -1410,6 +1411,11 @@ int cifs_strict_fsync(struct file *file, int datasync)
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 
+	rc = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (rc)
+		return rc;
+	mutex_lock(&inode->i_mutex);
+
 	xid = GetXid();
 
 	cFYI(1, "Sync file - name: %s datasync: 0x%x",
@@ -1428,16 +1434,23 @@ int cifs_strict_fsync(struct file *file, int datasync)
 		rc = CIFSSMBFlush(xid, tcon, smbfile->netfid);
 
 	FreeXid(xid);
+	mutex_unlock(&inode->i_mutex);
 	return rc;
 }
 
-int cifs_fsync(struct file *file, int datasync)
+int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	int xid;
 	int rc = 0;
 	struct cifs_tcon *tcon;
 	struct cifsFileInfo *smbfile = file->private_data;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
+	struct inode *inode = file->f_mapping->host;
+
+	rc = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (rc)
+		return rc;
+	mutex_lock(&inode->i_mutex);
 
 	xid = GetXid();
 
@@ -1449,6 +1462,7 @@ int cifs_fsync(struct file *file, int datasync)
 		rc = CIFSSMBFlush(xid, tcon, smbfile->netfid);
 
 	FreeXid(xid);
+	mutex_unlock(&inode->i_mutex);
 	return rc;
 }
 
diff --git a/fs/coda/coda_int.h b/fs/coda/coda_int.h
index 6b443ff43a19..b7143cf783ac 100644
--- a/fs/coda/coda_int.h
+++ b/fs/coda/coda_int.h
@@ -11,7 +11,7 @@ extern int coda_fake_statfs;
 
 void coda_destroy_inodecache(void);
 int coda_init_inodecache(void);
-int coda_fsync(struct file *coda_file, int datasync);
+int coda_fsync(struct file *coda_file, loff_t start, loff_t end, int datasync);
 void coda_sysctl_init(void);
 void coda_sysctl_clean(void);
 
diff --git a/fs/coda/file.c b/fs/coda/file.c
index 0433057be330..8edd404e6419 100644
--- a/fs/coda/file.c
+++ b/fs/coda/file.c
@@ -199,7 +199,7 @@ int coda_release(struct inode *coda_inode, struct file *coda_file)
 	return 0;
 }
 
-int coda_fsync(struct file *coda_file, int datasync)
+int coda_fsync(struct file *coda_file, loff_t start, loff_t end, int datasync)
 {
 	struct file *host_file;
 	struct inode *coda_inode = coda_file->f_path.dentry->d_inode;
@@ -210,6 +210,11 @@ int coda_fsync(struct file *coda_file, int datasync)
 	      S_ISLNK(coda_inode->i_mode)))
 		return -EINVAL;
 
+	err = filemap_write_and_wait_range(coda_inode->i_mapping, start, end);
+	if (err)
+		return err;
+	mutex_lock(&coda_inode->i_mutex);
+
 	cfi = CODA_FTOC(coda_file);
 	BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC);
 	host_file = cfi->cfi_container;
@@ -217,6 +222,7 @@ int coda_fsync(struct file *coda_file, int datasync)
 	err = vfs_fsync(host_file, datasync);
 	if (!err && !datasync)
 		err = venus_fsync(coda_inode->i_sb, coda_i2f(coda_inode));
+	mutex_unlock(&coda_inode->i_mutex);
 
 	return err;
 }
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 4ec9eb00a241..c6ac98cf9baa 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -270,14 +270,15 @@ static int ecryptfs_release(struct inode *inode, struct file *file)
 }
 
 static int
-ecryptfs_fsync(struct file *file, int datasync)
+ecryptfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	int rc = 0;
 
-	rc = generic_file_fsync(file, datasync);
+	rc = generic_file_fsync(file, start, end, datasync);
 	if (rc)
 		goto out;
-	rc = vfs_fsync(ecryptfs_file_to_lower(file), datasync);
+	rc = vfs_fsync_range(ecryptfs_file_to_lower(file), start, end,
+			     datasync);
 out:
 	return rc;
 }
diff --git a/fs/exofs/file.c b/fs/exofs/file.c
index 45ca323d8363..491c6c078e7f 100644
--- a/fs/exofs/file.c
+++ b/fs/exofs/file.c
@@ -42,11 +42,19 @@ static int exofs_release_file(struct inode *inode, struct file *filp)
  *   Note, in exofs all metadata is written as part of inode, regardless.
  *   The writeout is synchronous
  */
-static int exofs_file_fsync(struct file *filp, int datasync)
+static int exofs_file_fsync(struct file *filp, loff_t start, loff_t end,
+			    int datasync)
 {
+	struct inode *inode = filp->f_mapping->host;
 	int ret;
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+
+	mutex_lock(&inode->i_mutex);
 	ret = sync_inode_metadata(filp->f_mapping->host, 1);
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 645be9e7ee47..af9fc89b1b2d 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -150,7 +150,8 @@ extern void ext2_write_super (struct super_block *);
 extern const struct file_operations ext2_dir_operations;
 
 /* file.c */
-extern int ext2_fsync(struct file *file, int datasync);
+extern int ext2_fsync(struct file *file, loff_t start, loff_t end,
+		      int datasync);
 extern const struct inode_operations ext2_file_inode_operations;
 extern const struct file_operations ext2_file_operations;
 extern const struct file_operations ext2_xip_file_operations;
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 49eec9456c5b..82e06321de35 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -40,13 +40,13 @@ static int ext2_release_file (struct inode * inode, struct file * filp)
 	return 0;
 }
 
-int ext2_fsync(struct file *file, int datasync)
+int ext2_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	int ret;
 	struct super_block *sb = file->f_mapping->host->i_sb;
 	struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
 
-	ret = generic_file_fsync(file, datasync);
+	ret = generic_file_fsync(file, start, end, datasync);
 	if (ret == -EIO || test_and_clear_bit(AS_EIO, &mapping->flags)) {
 		/* We don't really know where the IO error happened... */
 		ext2_error(sb, __func__,
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index 09b13bb34c94..0bcf63adb80a 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -43,7 +43,7 @@
  * inode to disk.
  */
 
-int ext3_sync_file(struct file *file, int datasync)
+int ext3_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct ext3_inode_info *ei = EXT3_I(inode);
@@ -54,6 +54,17 @@ int ext3_sync_file(struct file *file, int datasync)
 	if (inode->i_sb->s_flags & MS_RDONLY)
 		return 0;
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+
+	/*
+	 * Taking the mutex here just to keep consistent with how fsync was
+	 * called previously, however it looks like we don't need to take
+	 * i_mutex at all.
+	 */
+	mutex_lock(&inode->i_mutex);
+
 	J_ASSERT(ext3_journal_current_handle() == NULL);
 
 	/*
@@ -70,8 +81,10 @@ int ext3_sync_file(struct file *file, int datasync)
 	 *  (they were dirtied by commit).  But that's OK - the blocks are
 	 *  safe in-journal, which is all fsync() needs to ensure.
 	 */
-	if (ext3_should_journal_data(inode))
+	if (ext3_should_journal_data(inode)) {
+		mutex_unlock(&inode->i_mutex);
 		return ext3_force_commit(inode->i_sb);
+	}
 
 	if (datasync)
 		commit_tid = atomic_read(&ei->i_datasync_tid);
@@ -91,5 +104,6 @@ int ext3_sync_file(struct file *file, int datasync)
 	 */
 	if (needs_barrier)
 		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 1921392cd708..fa44df879711 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1758,7 +1758,7 @@ extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 extern void ext4_htree_free_dir_info(struct dir_private_info *p);
 
 /* fsync.c */
-extern int ext4_sync_file(struct file *, int);
+extern int ext4_sync_file(struct file *, loff_t, loff_t, int);
 extern int ext4_flush_completed_IO(struct inode *);
 
 /* hash.c */
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index ce66d2fe826c..da3bed3e0c29 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -151,6 +151,32 @@ static int ext4_sync_parent(struct inode *inode)
 	return ret;
 }
 
+/**
+ * __sync_file - generic_file_fsync without the locking and filemap_write
+ * @inode:	inode to sync
+ * @datasync:	only sync essential metadata if true
+ *
+ * This is just generic_file_fsync without the locking.  This is needed for
+ * nojournal mode to make sure this inodes data/metadata makes it to disk
+ * properly.  The i_mutex should be held already.
+ */
+static int __sync_inode(struct inode *inode, int datasync)
+{
+	int err;
+	int ret;
+
+	ret = sync_mapping_buffers(inode->i_mapping);
+	if (!(inode->i_state & I_DIRTY))
+		return ret;
+	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+		return ret;
+
+	err = sync_inode_metadata(inode, 1);
+	if (ret == 0)
+		ret = err;
+	return ret;
+}
+
 /*
  * akpm: A new design for ext4_sync_file().
  *
@@ -165,7 +191,7 @@ static int ext4_sync_parent(struct inode *inode)
  * i_mutex lock is held when entering and exiting this function
  */
 
-int ext4_sync_file(struct file *file, int datasync)
+int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct ext4_inode_info *ei = EXT4_I(inode);
@@ -178,15 +204,20 @@ int ext4_sync_file(struct file *file, int datasync)
 
 	trace_ext4_sync_file_enter(file, datasync);
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	if (inode->i_sb->s_flags & MS_RDONLY)
-		return 0;
+		goto out;
 
 	ret = ext4_flush_completed_IO(inode);
 	if (ret < 0)
 		goto out;
 
 	if (!journal) {
-		ret = generic_file_fsync(file, datasync);
+		ret = __sync_inode(inode, datasync);
 		if (!ret && !list_empty(&inode->i_dentry))
 			ret = ext4_sync_parent(inode);
 		goto out;
@@ -220,6 +251,7 @@ int ext4_sync_file(struct file *file, int datasync)
 	if (needs_barrier)
 		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
  out:
+	mutex_unlock(&inode->i_mutex);
 	trace_ext4_sync_file_exit(inode, ret);
 	return ret;
 }
diff --git a/fs/fat/fat.h b/fs/fat/fat.h
index a975b4147e91..a5d3853822e0 100644
--- a/fs/fat/fat.h
+++ b/fs/fat/fat.h
@@ -310,7 +310,8 @@ extern int fat_setattr(struct dentry * dentry, struct iattr * attr);
 extern void fat_truncate_blocks(struct inode *inode, loff_t offset);
 extern int fat_getattr(struct vfsmount *mnt, struct dentry *dentry,
 		       struct kstat *stat);
-extern int fat_file_fsync(struct file *file, int datasync);
+extern int fat_file_fsync(struct file *file, loff_t start, loff_t end,
+			  int datasync);
 
 /* fat/inode.c */
 extern void fat_attach(struct inode *inode, loff_t i_pos);
diff --git a/fs/fat/file.c b/fs/fat/file.c
index e1587c54d3c1..c118acf16e43 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -149,12 +149,12 @@ static int fat_file_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-int fat_file_fsync(struct file *filp, int datasync)
+int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	int res, err;
 
-	res = generic_file_fsync(filp, datasync);
+	res = generic_file_fsync(filp, start, end, datasync);
 	err = sync_mapping_buffers(MSDOS_SB(inode->i_sb)->fat_inode->i_mapping);
 
 	return res ? res : err;
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 02063dde2728..9f63e493a9b6 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -1176,9 +1176,10 @@ static int fuse_dir_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static int fuse_dir_fsync(struct file *file, int datasync)
+static int fuse_dir_fsync(struct file *file, loff_t start, loff_t end,
+			  int datasync)
 {
-	return fuse_fsync_common(file, datasync, 1);
+	return fuse_fsync_common(file, start, end, datasync, 1);
 }
 
 static bool update_mtime(unsigned ivalid)
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 73b89df20851..7bb685cdd00c 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -400,7 +400,8 @@ static void fuse_sync_writes(struct inode *inode)
 	fuse_release_nowrite(inode);
 }
 
-int fuse_fsync_common(struct file *file, int datasync, int isdir)
+int fuse_fsync_common(struct file *file, loff_t start, loff_t end,
+		      int datasync, int isdir)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct fuse_conn *fc = get_fuse_conn(inode);
@@ -412,9 +413,15 @@ int fuse_fsync_common(struct file *file, int datasync, int isdir)
 	if (is_bad_inode(inode))
 		return -EIO;
 
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
 	if ((!isdir && fc->no_fsync) || (isdir && fc->no_fsyncdir))
 		return 0;
 
+	mutex_lock(&inode->i_mutex);
+
 	/*
 	 * Start writeback against all dirty pages of the inode, then
 	 * wait for all outstanding writes, before sending the FSYNC
@@ -422,13 +429,15 @@ int fuse_fsync_common(struct file *file, int datasync, int isdir)
 	 */
 	err = write_inode_now(inode, 0);
 	if (err)
-		return err;
+		goto out;
 
 	fuse_sync_writes(inode);
 
 	req = fuse_get_req(fc);
-	if (IS_ERR(req))
-		return PTR_ERR(req);
+	if (IS_ERR(req)) {
+		err = PTR_ERR(req);
+		goto out;
+	}
 
 	memset(&inarg, 0, sizeof(inarg));
 	inarg.fh = ff->fh;
@@ -448,12 +457,15 @@ int fuse_fsync_common(struct file *file, int datasync, int isdir)
 			fc->no_fsync = 1;
 		err = 0;
 	}
+out:
+	mutex_unlock(&inode->i_mutex);
 	return err;
 }
 
-static int fuse_fsync(struct file *file, int datasync)
+static int fuse_fsync(struct file *file, loff_t start, loff_t end,
+		      int datasync)
 {
-	return fuse_fsync_common(file, datasync, 0);
+	return fuse_fsync_common(file, start, end, datasync, 0);
 }
 
 void fuse_read_fill(struct fuse_req *req, struct file *file, loff_t pos,
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index b788becada76..c6aa2d4b8517 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -589,7 +589,8 @@ void fuse_release_common(struct file *file, int opcode);
 /**
  * Send FSYNC or FSYNCDIR request
  */
-int fuse_fsync_common(struct file *file, int datasync, int isdir);
+int fuse_fsync_common(struct file *file, loff_t start, loff_t end,
+		      int datasync, int isdir);
 
 /**
  * Notify poll wakeup
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 89c39e53760d..f82cb5e1cb6b 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -544,7 +544,9 @@ static int gfs2_close(struct inode *inode, struct file *file)
 
 /**
  * gfs2_fsync - sync the dirty data for a file (across the cluster)
- * @file: the file that points to the dentry (we ignore this)
+ * @file: the file that points to the dentry
+ * @start: the start position in the file to sync
+ * @end: the end position in the file to sync
  * @datasync: set if we can ignore timestamp changes
  *
  * The VFS will flush data for us. We only need to worry
@@ -553,23 +555,32 @@ static int gfs2_close(struct inode *inode, struct file *file)
  * Returns: errno
  */
 
-static int gfs2_fsync(struct file *file, int datasync)
+static int gfs2_fsync(struct file *file, loff_t start, loff_t end,
+		      int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC);
 	struct gfs2_inode *ip = GFS2_I(inode);
 	int ret;
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	if (datasync)
 		sync_state &= ~I_DIRTY_SYNC;
 
 	if (sync_state) {
 		ret = sync_inode_metadata(inode, 1);
-		if (ret)
+		if (ret) {
+			mutex_unlock(&inode->i_mutex);
 			return ret;
+		}
 		gfs2_ail_flush(ip->i_gl);
 	}
 
+	mutex_unlock(&inode->i_mutex);
 	return 0;
 }
 
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 5e7c3f309617..96a1b625fc74 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -627,12 +627,18 @@ int hfs_inode_setattr(struct dentry *dentry, struct iattr * attr)
 	return 0;
 }
 
-static int hfs_file_fsync(struct file *filp, int datasync)
+static int hfs_file_fsync(struct file *filp, loff_t start, loff_t end,
+			  int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	struct super_block * sb;
 	int ret, err;
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	/* sync the inode to buffers */
 	ret = write_inode_now(inode, 0);
 
@@ -649,6 +655,7 @@ static int hfs_file_fsync(struct file *filp, int datasync)
 	err = sync_blockdev(sb->s_bdev);
 	if (!ret)
 		ret = err;
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index d6857523336d..38184e360932 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -392,7 +392,8 @@ int hfsplus_cat_read_inode(struct inode *, struct hfs_find_data *);
 int hfsplus_cat_write_inode(struct inode *);
 struct inode *hfsplus_new_inode(struct super_block *, int);
 void hfsplus_delete_inode(struct inode *);
-int hfsplus_file_fsync(struct file *file, int datasync);
+int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end,
+		       int datasync);
 
 /* ioctl.c */
 long hfsplus_ioctl(struct file *filp, unsigned int cmd, unsigned long arg);
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 5b1cb98741cc..30486e01d003 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -308,13 +308,19 @@ static int hfsplus_setattr(struct dentry *dentry, struct iattr *attr)
 	return 0;
 }
 
-int hfsplus_file_fsync(struct file *file, int datasync)
+int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end,
+		       int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	struct hfsplus_sb_info *sbi = HFSPLUS_SB(inode->i_sb);
 	int error = 0, error2;
 
+	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (error)
+		return error;
+	mutex_lock(&inode->i_mutex);
+
 	/*
 	 * Sync inode metadata into the catalog and extent trees.
 	 */
@@ -342,6 +348,8 @@ int hfsplus_file_fsync(struct file *file, int datasync)
 	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
 		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
 
+	mutex_unlock(&inode->i_mutex);
+
 	return error;
 }
 
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 6e449c599b9d..0d22afdd4611 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -362,9 +362,20 @@ retry:
 	return 0;
 }
 
-int hostfs_fsync(struct file *file, int datasync)
+int hostfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	return fsync_file(HOSTFS_I(file->f_mapping->host)->fd, datasync);
+	struct inode *inode = file->f_mapping->host;
+	int ret;
+
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+
+	mutex_lock(&inode->i_mutex);
+	ret = fsync_file(HOSTFS_I(inode)->fd, datasync);
+	mutex_unlock(&inode->i_mutex);
+
+	return ret;
 }
 
 static const struct file_operations hostfs_file_fops = {
diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c
index 89c500ee5213..89d2a5803ae3 100644
--- a/fs/hpfs/file.c
+++ b/fs/hpfs/file.c
@@ -18,9 +18,14 @@ static int hpfs_file_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-int hpfs_file_fsync(struct file *file, int datasync)
+int hpfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
+	int ret;
+
+	ret = filemap_write_and_wait_range(file->f_mapping, start, end);
+	if (ret)
+		return ret;
 	return sync_blockdev(inode->i_sb->s_bdev);
 }
 
diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h
index dd552f862c8f..331b5e234ef3 100644
--- a/fs/hpfs/hpfs_fn.h
+++ b/fs/hpfs/hpfs_fn.h
@@ -258,7 +258,7 @@ void hpfs_set_ea(struct inode *, struct fnode *, const char *,
 
 /* file.c */
 
-int hpfs_file_fsync(struct file *, int);
+int hpfs_file_fsync(struct file *, loff_t, loff_t, int);
 extern const struct file_operations hpfs_file_ops;
 extern const struct inode_operations hpfs_file_iops;
 extern const struct address_space_operations hpfs_aops;
diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index 85c098a499f3..8635be5ffd97 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -573,9 +573,10 @@ static int hppfs_readdir(struct file *file, void *ent, filldir_t filldir)
 	return err;
 }
 
-static int hppfs_fsync(struct file *file, int datasync)
+static int hppfs_fsync(struct file *file, loff_t start, loff_t end,
+		       int datasync)
 {
-	return 0;
+	return filemap_write_and_wait_range(file->f_mapping, start, end);
 }
 
 static const struct file_operations hppfs_dir_fops = {
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 1c0a08d711aa..3989f7e09f7f 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -27,13 +27,20 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 			struct page **pagep, void **fsdata);
 static int jffs2_readpage (struct file *filp, struct page *pg);
 
-int jffs2_fsync(struct file *filp, int datasync)
+int jffs2_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	struct jffs2_sb_info *c = JFFS2_SB_INFO(inode->i_sb);
+	int ret;
+
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
 
+	mutex_lock(&inode->i_mutex);
 	/* Trigger GC to flush any pending writes for this inode */
 	jffs2_flush_wbuf_gc(c, inode->i_ino);
+	mutex_unlock(&inode->i_mutex);
 
 	return 0;
 }
diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h
index 65c6c43ca482..9c252835e8e5 100644
--- a/fs/jffs2/os-linux.h
+++ b/fs/jffs2/os-linux.h
@@ -158,7 +158,7 @@ extern const struct inode_operations jffs2_dir_inode_operations;
 extern const struct file_operations jffs2_file_operations;
 extern const struct inode_operations jffs2_file_inode_operations;
 extern const struct address_space_operations jffs2_file_address_operations;
-int jffs2_fsync(struct file *, int);
+int jffs2_fsync(struct file *, loff_t, loff_t, int);
 int jffs2_do_readpage_unlock (struct inode *inode, struct page *pg);
 
 /* ioctl.c */
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index 9f32315acef1..7527855b5cc6 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -28,19 +28,26 @@
 #include "jfs_acl.h"
 #include "jfs_debug.h"
 
-int jfs_fsync(struct file *file, int datasync)
+int jfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	int rc = 0;
 
+	rc = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (rc)
+		return rc;
+
+	mutex_lock(&inode->i_mutex);
 	if (!(inode->i_state & I_DIRTY) ||
 	    (datasync && !(inode->i_state & I_DIRTY_DATASYNC))) {
 		/* Make sure committed changes hit the disk */
 		jfs_flush_journal(JFS_SBI(inode->i_sb)->log, 1);
+		mutex_unlock(&inode->i_mutex);
 		return rc;
 	}
 
 	rc |= jfs_commit_inode(inode, 1);
+	mutex_unlock(&inode->i_mutex);
 
 	return rc ? -EIO : 0;
 }
diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h
index ec2fb8b945fc..9271cfe4a149 100644
--- a/fs/jfs/jfs_inode.h
+++ b/fs/jfs/jfs_inode.h
@@ -21,7 +21,7 @@
 struct fid;
 
 extern struct inode *ialloc(struct inode *, umode_t);
-extern int jfs_fsync(struct file *, int);
+extern int jfs_fsync(struct file *, loff_t, loff_t, int);
 extern long jfs_ioctl(struct file *, unsigned int, unsigned long);
 extern long jfs_compat_ioctl(struct file *, unsigned int, unsigned long);
 extern struct inode *jfs_iget(struct super_block *, unsigned long);
diff --git a/fs/libfs.c b/fs/libfs.c
index bd50b11f92da..8f2271a5df53 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -905,21 +905,29 @@ EXPORT_SYMBOL_GPL(generic_fh_to_parent);
  * filesystems which track all non-inode metadata in the buffers list
  * hanging off the address_space structure.
  */
-int generic_file_fsync(struct file *file, int datasync)
+int generic_file_fsync(struct file *file, loff_t start, loff_t end,
+		       int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	int err;
 	int ret;
 
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
+	mutex_lock(&inode->i_mutex);
 	ret = sync_mapping_buffers(inode->i_mapping);
 	if (!(inode->i_state & I_DIRTY))
-		return ret;
+		goto out;
 	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
-		return ret;
+		goto out;
 
 	err = sync_inode_metadata(inode, 1);
 	if (ret == 0)
 		ret = err;
+out:
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 EXPORT_SYMBOL(generic_file_fsync);
@@ -956,7 +964,7 @@ EXPORT_SYMBOL(generic_check_addressable);
 /*
  * No-op implementation of ->fsync for in-memory filesystems.
  */
-int noop_fsync(struct file *file, int datasync)
+int noop_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	return 0;
 }
diff --git a/fs/logfs/file.c b/fs/logfs/file.c
index c2ad7028def4..b548c87a86f1 100644
--- a/fs/logfs/file.c
+++ b/fs/logfs/file.c
@@ -219,11 +219,20 @@ long logfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	}
 }
 
-int logfs_fsync(struct file *file, int datasync)
+int logfs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct super_block *sb = file->f_mapping->host->i_sb;
+	struct inode *inode = file->f_mapping->host;
+	int ret;
+
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
 
+	mutex_lock(&inode->i_mutex);
 	logfs_write_anchor(sb);
+	mutex_unlock(&inode->i_mutex);
+
 	return 0;
 }
 
diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h
index 57afd4a6fabb..f22d108bfa5d 100644
--- a/fs/logfs/logfs.h
+++ b/fs/logfs/logfs.h
@@ -506,7 +506,7 @@ extern const struct file_operations logfs_reg_fops;
 extern const struct address_space_operations logfs_reg_aops;
 int logfs_readpage(struct file *file, struct page *page);
 long logfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
-int logfs_fsync(struct file *file, int datasync);
+int logfs_fsync(struct file *file, loff_t start, loff_t end, int datasync);
 
 /* gc.c */
 u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec);
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index 0ed65e0c3dfe..64a326418aa2 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -20,9 +20,9 @@
 
 #include "ncp_fs.h"
 
-static int ncp_fsync(struct file *file, int datasync)
+static int ncp_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	return 0;
+	return filemap_write_and_wait_range(file->f_mapping, start, end);
 }
 
 /*
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 8a45e6d1f6a4..57f578e2560a 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -56,7 +56,7 @@ static int nfs_link(struct dentry *, struct inode *, struct dentry *);
 static int nfs_mknod(struct inode *, struct dentry *, int, dev_t);
 static int nfs_rename(struct inode *, struct dentry *,
 		      struct inode *, struct dentry *);
-static int nfs_fsync_dir(struct file *, int);
+static int nfs_fsync_dir(struct file *, loff_t, loff_t, int);
 static loff_t nfs_llseek_dir(struct file *, loff_t, int);
 static void nfs_readdir_clear_array(struct page*);
 
@@ -945,15 +945,19 @@ out:
  * All directory operations under NFS are synchronous, so fsync()
  * is a dummy operation.
  */
-static int nfs_fsync_dir(struct file *filp, int datasync)
+static int nfs_fsync_dir(struct file *filp, loff_t start, loff_t end,
+			 int datasync)
 {
 	struct dentry *dentry = filp->f_path.dentry;
+	struct inode *inode = dentry->d_inode;
 
 	dfprintk(FILE, "NFS: fsync dir(%s/%s) datasync %d\n",
 			dentry->d_parent->d_name.name, dentry->d_name.name,
 			datasync);
 
+	mutex_lock(&inode->i_mutex);
 	nfs_inc_stats(dentry->d_inode, NFSIOS_VFSFSYNC);
+	mutex_unlock(&inode->i_mutex);
 	return 0;
 }
 
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 2c1705b6acd7..28b8c3f3cda3 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -55,7 +55,7 @@ static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
 static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos);
 static int  nfs_file_flush(struct file *, fl_owner_t id);
-static int  nfs_file_fsync(struct file *, int datasync);
+static int  nfs_file_fsync(struct file *, loff_t, loff_t, int datasync);
 static int nfs_check_flags(int flags);
 static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl);
 static int nfs_flock(struct file *filp, int cmd, struct file_lock *fl);
@@ -308,7 +308,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
  * fall back to doing a synchronous write.
  */
 static int
-nfs_file_fsync(struct file *file, int datasync)
+nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct dentry *dentry = file->f_path.dentry;
 	struct nfs_open_context *ctx = nfs_file_open_context(file);
@@ -316,11 +316,15 @@ nfs_file_fsync(struct file *file, int datasync)
 	int have_error, status;
 	int ret = 0;
 
-
 	dprintk("NFS: fsync file(%s/%s) datasync %d\n",
 			dentry->d_parent->d_name.name, dentry->d_name.name,
 			datasync);
 
+	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (ret)
+		return ret;
+	mutex_lock(&inode->i_mutex);
+
 	nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
 	have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
 	status = nfs_commit_inode(inode, FLUSH_SYNC);
@@ -332,6 +336,7 @@ nfs_file_fsync(struct file *file, int datasync)
 	if (!ret && !datasync)
 		/* application has asked for meta-data sync */
 		ret = pnfs_layoutcommit_inode(inode, true);
+	mutex_unlock(&inode->i_mutex);
 	return ret;
 }
 
diff --git a/fs/nilfs2/file.c b/fs/nilfs2/file.c
index d7eeca62febd..26601529dc17 100644
--- a/fs/nilfs2/file.c
+++ b/fs/nilfs2/file.c
@@ -27,7 +27,7 @@
 #include "nilfs.h"
 #include "segment.h"
 
-int nilfs_sync_file(struct file *file, int datasync)
+int nilfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	/*
 	 * Called from fsync() system call
@@ -40,8 +40,15 @@ int nilfs_sync_file(struct file *file, int datasync)
 	struct inode *inode = file->f_mapping->host;
 	int err;
 
-	if (!nilfs_inode_dirty(inode))
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+	mutex_lock(&inode->i_mutex);
+
+	if (!nilfs_inode_dirty(inode)) {
+		mutex_unlock(&inode->i_mutex);
 		return 0;
+	}
 
 	if (datasync)
 		err = nilfs_construct_dsync_segment(inode->i_sb, inode, 0,
@@ -49,6 +56,7 @@ int nilfs_sync_file(struct file *file, int datasync)
 	else
 		err = nilfs_construct_segment(inode->i_sb);
 
+	mutex_unlock(&inode->i_mutex);
 	return err;
 }
 
diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h
index 6fb7511c0328..255d5e1c03b7 100644
--- a/fs/nilfs2/nilfs.h
+++ b/fs/nilfs2/nilfs.h
@@ -235,7 +235,7 @@ extern void nilfs_set_link(struct inode *, struct nilfs_dir_entry *,
 			   struct page *, struct inode *);
 
 /* file.c */
-extern int nilfs_sync_file(struct file *, int);
+extern int nilfs_sync_file(struct file *, loff_t, loff_t, int);
 
 /* ioctl.c */
 long nilfs_ioctl(struct file *, unsigned int, unsigned long);
diff --git a/fs/ntfs/dir.c b/fs/ntfs/dir.c
index 0f48e7c5d9e1..99e36107ff60 100644
--- a/fs/ntfs/dir.c
+++ b/fs/ntfs/dir.c
@@ -1527,13 +1527,20 @@ static int ntfs_dir_open(struct inode *vi, struct file *filp)
  * this problem for now.  We do write the $BITMAP attribute if it is present
  * which is the important one for a directory so things are not too bad.
  */
-static int ntfs_dir_fsync(struct file *filp, int datasync)
+static int ntfs_dir_fsync(struct file *filp, loff_t start, loff_t end,
+			  int datasync)
 {
 	struct inode *bmp_vi, *vi = filp->f_mapping->host;
 	int err, ret;
 	ntfs_attr na;
 
 	ntfs_debug("Entering for inode 0x%lx.", vi->i_ino);
+
+	err = filemap_write_and_wait_range(vi->i_mapping, start, end);
+	if (err)
+		return err;
+	mutex_lock(&vi->i_mutex);
+
 	BUG_ON(!S_ISDIR(vi->i_mode));
 	/* If the bitmap attribute inode is in memory sync it, too. */
 	na.mft_no = vi->i_ino;
@@ -1555,6 +1562,7 @@ static int ntfs_dir_fsync(struct file *filp, int datasync)
 	else
 		ntfs_warning(vi->i_sb, "Failed to f%ssync inode 0x%lx.  Error "
 				"%u.", datasync ? "data" : "", vi->i_ino, -ret);
+	mutex_unlock(&vi->i_mutex);
 	return ret;
 }
 
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index b59f5ac26bef..c587e2d27183 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -2152,12 +2152,19 @@ static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
  * with this inode but since we have no simple way of getting to them we ignore
  * this problem for now.
  */
-static int ntfs_file_fsync(struct file *filp, int datasync)
+static int ntfs_file_fsync(struct file *filp, loff_t start, loff_t end,
+			   int datasync)
 {
 	struct inode *vi = filp->f_mapping->host;
 	int err, ret = 0;
 
 	ntfs_debug("Entering for inode 0x%lx.", vi->i_ino);
+
+	err = filemap_write_and_wait_range(vi->i_mapping, start, end);
+	if (err)
+		return err;
+	mutex_lock(&vi->i_mutex);
+
 	BUG_ON(S_ISDIR(vi->i_mode));
 	if (!datasync || !NInoNonResident(NTFS_I(vi)))
 		ret = __ntfs_write_inode(vi, 1);
@@ -2175,6 +2182,7 @@ static int ntfs_file_fsync(struct file *filp, int datasync)
 	else
 		ntfs_warning(vi->i_sb, "Failed to f%ssync inode 0x%lx.  Error "
 				"%u.", datasync ? "data" : "", vi->i_ino, -ret);
+	mutex_unlock(&vi->i_mutex);
 	return ret;
 }
 
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 22d604601957..0fc2bd34039d 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -171,7 +171,8 @@ static int ocfs2_dir_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static int ocfs2_sync_file(struct file *file, int datasync)
+static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end,
+			   int datasync)
 {
 	int err = 0;
 	journal_t *journal;
@@ -184,6 +185,16 @@ static int ocfs2_sync_file(struct file *file, int datasync)
 			      file->f_path.dentry->d_name.name,
 			      (unsigned long long)datasync);
 
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
+	/*
+	 * Probably don't need the i_mutex at all in here, just putting it here
+	 * to be consistent with how fsync used to be called, someone more
+	 * familiar with the fs could possibly remove it.
+	 */
+	mutex_lock(&inode->i_mutex);
 	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) {
 		/*
 		 * We still have to flush drive's caches to get data to the
@@ -200,6 +211,7 @@ static int ocfs2_sync_file(struct file *file, int datasync)
 bail:
 	if (err)
 		mlog_errno(err);
+	mutex_unlock(&inode->i_mutex);
 
 	return (err < 0) ? -EIO : 0;
 }
diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c
index 198dabf1b2bb..133e9355dc6f 100644
--- a/fs/reiserfs/dir.c
+++ b/fs/reiserfs/dir.c
@@ -14,7 +14,8 @@
 extern const struct reiserfs_key MIN_KEY;
 
 static int reiserfs_readdir(struct file *, void *, filldir_t);
-static int reiserfs_dir_fsync(struct file *filp, int datasync);
+static int reiserfs_dir_fsync(struct file *filp, loff_t start, loff_t end,
+			      int datasync);
 
 const struct file_operations reiserfs_dir_operations = {
 	.llseek = generic_file_llseek,
@@ -27,13 +28,21 @@ const struct file_operations reiserfs_dir_operations = {
 #endif
 };
 
-static int reiserfs_dir_fsync(struct file *filp, int datasync)
+static int reiserfs_dir_fsync(struct file *filp, loff_t start, loff_t end,
+			      int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	int err;
+
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
+	mutex_lock(&inode->i_mutex);
 	reiserfs_write_lock(inode->i_sb);
 	err = reiserfs_commit_for_inode(inode);
 	reiserfs_write_unlock(inode->i_sb);
+	mutex_unlock(&inode->i_mutex);
 	if (err < 0)
 		return err;
 	return 0;
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index bbf31003d308..c7156dc39ce7 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -140,12 +140,18 @@ static void reiserfs_vfs_truncate_file(struct inode *inode)
  * be removed...
  */
 
-static int reiserfs_sync_file(struct file *filp, int datasync)
+static int reiserfs_sync_file(struct file *filp, loff_t start, loff_t end,
+			      int datasync)
 {
 	struct inode *inode = filp->f_mapping->host;
 	int err;
 	int barrier_done;
 
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+
+	mutex_lock(&inode->i_mutex);
 	BUG_ON(!S_ISREG(inode->i_mode));
 	err = sync_mapping_buffers(inode->i_mapping);
 	reiserfs_write_lock(inode->i_sb);
@@ -153,6 +159,7 @@ static int reiserfs_sync_file(struct file *filp, int datasync)
 	reiserfs_write_unlock(inode->i_sb);
 	if (barrier_done != 1 && reiserfs_barrier_flush(inode->i_sb))
 		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+	mutex_unlock(&inode->i_mutex);
 	if (barrier_done < 0)
 		return barrier_done;
 	return (err < 0) ? -EIO : 0;
diff --git a/fs/sync.c b/fs/sync.c
index c38ec163da6c..c98a7477edfd 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -165,28 +165,9 @@ SYSCALL_DEFINE1(syncfs, int, fd)
  */
 int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync)
 {
-	struct address_space *mapping = file->f_mapping;
-	int err, ret;
-
-	if (!file->f_op || !file->f_op->fsync) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	ret = filemap_write_and_wait_range(mapping, start, end);
-
-	/*
-	 * We need to protect against concurrent writers, which could cause
-	 * livelocks in fsync_buffers_list().
-	 */
-	mutex_lock(&mapping->host->i_mutex);
-	err = file->f_op->fsync(file, datasync);
-	if (!ret)
-		ret = err;
-	mutex_unlock(&mapping->host->i_mutex);
-
-out:
-	return ret;
+	if (!file->f_op || !file->f_op->fsync)
+		return -EINVAL;
+	return file->f_op->fsync(file, start, end, datasync);
 }
 EXPORT_SYMBOL(vfs_fsync_range);
 
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 5e7fccfc4b29..89ef9a2f7837 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -1304,7 +1304,7 @@ static void *ubifs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	return NULL;
 }
 
-int ubifs_fsync(struct file *file, int datasync)
+int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	struct ubifs_info *c = inode->i_sb->s_fs_info;
@@ -1319,14 +1319,16 @@ int ubifs_fsync(struct file *file, int datasync)
 		 */
 		return 0;
 
-	/*
-	 * VFS has already synchronized dirty pages for this inode. Synchronize
-	 * the inode unless this is a 'datasync()' call.
-	 */
+	err = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (err)
+		return err;
+	mutex_lock(&inode->i_mutex);
+
+	/* Synchronize the inode unless this is a 'datasync()' call. */
 	if (!datasync || (inode->i_state & I_DIRTY_DATASYNC)) {
 		err = inode->i_sb->s_op->write_inode(inode, NULL);
 		if (err)
-			return err;
+			goto out;
 	}
 
 	/*
@@ -1334,10 +1336,9 @@ int ubifs_fsync(struct file *file, int datasync)
 	 * them.
 	 */
 	err = ubifs_sync_wbufs_by_inode(c, inode);
-	if (err)
-		return err;
-
-	return 0;
+out:
+	mutex_unlock(&inode->i_mutex);
+	return err;
 }
 
 /**
diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h
index f79983d6f860..4cd648501fa4 100644
--- a/fs/ubifs/ubifs.h
+++ b/fs/ubifs/ubifs.h
@@ -1720,7 +1720,7 @@ const struct ubifs_lprops *ubifs_fast_find_frdi_idx(struct ubifs_info *c);
 int ubifs_calc_dark(const struct ubifs_info *c, int spc);
 
 /* file.c */
-int ubifs_fsync(struct file *file, int datasync);
+int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync);
 int ubifs_setattr(struct dentry *dentry, struct iattr *attr);
 
 /* dir.c */
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 7f782af286bf..fbbf657df0cd 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -127,6 +127,8 @@ xfs_iozero(
 STATIC int
 xfs_file_fsync(
 	struct file		*file,
+	loff_t			start,
+	loff_t			end,
 	int			datasync)
 {
 	struct inode		*inode = file->f_mapping->host;
@@ -138,6 +140,10 @@ xfs_file_fsync(
 
 	trace_xfs_file_fsync(ip);
 
+	error = filemap_write_and_wait_range(inode->i_mapping, start, end);
+	if (error)
+		return error;
+
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -XFS_ERROR(EIO);
 
@@ -875,18 +881,11 @@ xfs_file_aio_write(
 	/* Handle various SYNC-type writes */
 	if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
 		loff_t end = pos + ret - 1;
-		int error, error2;
 
 		xfs_rw_iunlock(ip, iolock);
-		error = filemap_write_and_wait_range(mapping, pos, end);
+		ret = -xfs_file_fsync(file, pos, end,
+				      (file->f_flags & __O_SYNC) ? 0 : 1);
 		xfs_rw_ilock(ip, iolock);
-
-		error2 = -xfs_file_fsync(file,
-					 (file->f_flags & __O_SYNC) ? 0 : 1);
-		if (error)
-			ret = error;
-		else if (error2)
-			ret = error2;
 	}
 
 out_unlock:
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 5e06acf95d0f..0c473fd79acb 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -877,7 +877,7 @@ extern int ext3_htree_store_dirent(struct file *dir_file, __u32 hash,
 extern void ext3_htree_free_dir_info(struct dir_private_info *p);
 
 /* fsync.c */
-extern int ext3_sync_file(struct file *, int);
+extern int ext3_sync_file(struct file *, loff_t, loff_t, int);
 
 /* hash.c */
 extern int ext3fs_dirhash(const char *name, int len, struct
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 6a8274877171..1d6836c498dd 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -1043,7 +1043,8 @@ extern void fb_deferred_io_open(struct fb_info *info,
 				struct inode *inode,
 				struct file *file);
 extern void fb_deferred_io_cleanup(struct fb_info *info);
-extern int fb_deferred_io_fsync(struct file *file, int datasync);
+extern int fb_deferred_io_fsync(struct file *file, loff_t start,
+				loff_t end, int datasync);
 
 static inline bool fb_be_math(struct fb_info *info)
 {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4a61f98823a6..9cd2075c4a39 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1572,7 +1572,7 @@ struct file_operations {
 	int (*open) (struct inode *, struct file *);
 	int (*flush) (struct file *, fl_owner_t id);
 	int (*release) (struct inode *, struct file *);
-	int (*fsync) (struct file *, int datasync);
+	int (*fsync) (struct file *, loff_t, loff_t, int datasync);
 	int (*aio_fsync) (struct kiocb *, int datasync);
 	int (*fasync) (int, struct file *, int);
 	int (*lock) (struct file *, int, struct file_lock *);
@@ -2360,7 +2360,8 @@ extern int generic_segment_checks(const struct iovec *iov,
 /* fs/block_dev.c */
 extern ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos);
-extern int blkdev_fsync(struct file *filp, int datasync);
+extern int blkdev_fsync(struct file *filp, loff_t start, loff_t end,
+			int datasync);
 
 /* fs/splice.c */
 extern ssize_t generic_file_splice_read(struct file *, loff_t *,
@@ -2490,7 +2491,7 @@ extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
 extern int simple_rename(struct inode *, struct dentry *, struct inode *, struct dentry *);
-extern int noop_fsync(struct file *, int);
+extern int noop_fsync(struct file *, loff_t, loff_t, int);
 extern int simple_empty(struct dentry *);
 extern int simple_readpage(struct file *file, struct page *page);
 extern int simple_write_begin(struct file *file, struct address_space *mapping,
@@ -2515,7 +2516,7 @@ extern ssize_t simple_read_from_buffer(void __user *to, size_t count,
 extern ssize_t simple_write_to_buffer(void *to, size_t available, loff_t *ppos,
 		const void __user *from, size_t count);
 
-extern int generic_file_fsync(struct file *, int);
+extern int generic_file_fsync(struct file *, loff_t, loff_t, int);
 
 extern int generic_check_addressable(unsigned, u64);
 
diff --git a/ipc/shm.c b/ipc/shm.c
index ab3385a21b27..27884adb1a90 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -277,13 +277,13 @@ static int shm_release(struct inode *ino, struct file *file)
 	return 0;
 }
 
-static int shm_fsync(struct file *file, int datasync)
+static int shm_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct shm_file_data *sfd = shm_file_data(file);
 
 	if (!sfd->file->f_op->fsync)
 		return -EINVAL;
-	return sfd->file->f_op->fsync(sfd->file, datasync);
+	return sfd->file->f_op->fsync(sfd->file, start, end, datasync);
 }
 
 static unsigned long shm_get_unmapped_area(struct file *file,
-- 
cgit v1.2.3


From 295cc522c2b2834b302424e41ebb5846df6d61bd Mon Sep 17 00:00:00 2001
From: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Date: Tue, 19 Jul 2011 09:48:39 +0800
Subject: fs:update the NOTE of the file_operations structure

Big kernel lock had been removed and setlease now use the lock_flocks()
to hold a special spin lock file_lock_lock by Matthew.
So just remove the out-of-date NOTE.

Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/fs.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9cd2075c4a39..e0569e4ab2d5 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1552,11 +1552,6 @@ struct block_device_operations;
 #define HAVE_COMPAT_IOCTL 1
 #define HAVE_UNLOCKED_IOCTL 1
 
-/*
- * NOTE:
- * all file operations except setlease can be called without
- * the big kernel lock held in all filesystems.
- */
 struct file_operations {
 	struct module *owner;
 	loff_t (*llseek) (struct file *, loff_t, int);
-- 
cgit v1.2.3


From ed70afcd6e795e3de98df56f1cd0f898fbf641a7 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Thu, 21 Jul 2011 13:55:37 -0700
Subject: mm/truncate.c: fix build for CONFIG_BLOCK not enabled

Fix build error when CONFIG_BLOCK is not enabled by providing a stub
inode_dio_wait() function.

mm/truncate.c:612: error: implicit declaration of function 'inode_dio_wait'

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/fs.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index e0569e4ab2d5..b224dc468a23 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2423,6 +2423,10 @@ static inline ssize_t blockdev_direct_IO(int rw, struct kiocb *iocb,
 				    offset, nr_segs, get_block, NULL, NULL,
 				    DIO_LOCKING | DIO_SKIP_HOLES);
 }
+#else
+static inline void inode_dio_wait(struct inode *inode)
+{
+}
 #endif
 
 extern const struct file_operations generic_ro_fops;
-- 
cgit v1.2.3