fs: rcu-walk for path lookup

Perform common cases of path lookups without any stores or locking in the
ancestor dentry elements. This is called rcu-walk, as opposed to the current
algorithm which is a refcount based walk, or ref-walk.

This results in far fewer atomic operations on every path element,
significantly improving path lookup performance. It also avoids cacheline
bouncing on common dentries, significantly improving scalability.

The overall design is like this:
* LOOKUP_RCU is set in nd->flags, which distinguishes rcu-walk from ref-walk.
* Take the RCU lock for the entire path walk, starting with the acquiring
  of the starting path (eg. root/cwd/fd-path). So now dentry refcounts are
  not required for dentry persistence.
* synchronize_rcu is called when unregistering a filesystem, so we can
  access d_ops and i_ops during rcu-walk.
* Similarly take the vfsmount lock for the entire path walk. So now mnt
  refcounts are not required for persistence. Also we are free to perform mount
  lookups, and to assume dentry mount points and mount roots are stable up and
  down the path.
* Have a per-dentry seqlock to protect the dentry name, parent, and inode,
  so we can load this tuple atomically, and also check whether any of its
  members have changed.
* Dentry lookups (based on parent, candidate string tuple) recheck the parent
  sequence after the child is found in case anything changed in the parent
  during the path walk.
* inode is also RCU protected so we can load d_inode and use the inode for
  limited things.
* i_mode, i_uid, i_gid can be tested for exec permissions during path walk.
* i_op can be loaded.

When we reach the destination dentry, we lock it, recheck lookup sequence,
and increment its refcount and mountpoint refcount. RCU and vfsmount locks
are dropped. This is termed "dropping rcu-walk". If the dentry refcount does
not match, we can not drop rcu-walk gracefully at the current point in the
lokup, so instead return -ECHILD (for want of a better errno). This signals the
path walking code to re-do the entire lookup with a ref-walk.

Aside from the final dentry, there are other situations that may be encounted
where we cannot continue rcu-walk. In that case, we drop rcu-walk (ie. take
a reference on the last good dentry) and continue with a ref-walk. Again, if
we can drop rcu-walk gracefully, we return -ECHILD and do the whole lookup
using ref-walk. But it is very important that we can continue with ref-walk
for most cases, particularly to avoid the overhead of double lookups, and to
gain the scalability advantages on common path elements (like cwd and root).

The cases where rcu-walk cannot continue are:
* NULL dentry (ie. any uncached path element)
* parent with d_inode->i_op->permission or ACLs
* dentries with d_revalidate
* Following links

In future patches, permission checks and d_revalidate become rcu-walk aware. It
may be possible eventually to make following links rcu-walk aware.

Uncached path elements will always require dropping to ref-walk mode, at the
very least because i_mutex needs to be grabbed, and objects allocated.

Signed-off-by: Nick Piggin <npiggin@kernel.dk>

3 files changed, 46 insertions, 9 deletions

diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index ca648685f0c..c2e7390289c 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -5,6 +5,7 @@
 #include <linux/list.h>
 #include <linux/rculist.h>
 #include <linux/spinlock.h>
+#include <linux/seqlock.h>
 #include <linux/cache.h>
 #include <linux/rcupdate.h>
 
@@ -90,6 +91,7 @@ struct dentry {
 	unsigned int d_count;		/* protected by d_lock */
 	unsigned int d_flags;		/* protected by d_lock */
 	spinlock_t d_lock;		/* per dentry lock */
+	seqcount_t d_seq;		/* per dentry seqlock */
 	int d_mounted;
 	struct inode *d_inode;		/* Where the name belongs to - NULL is
 					 * negative */
@@ -266,9 +268,33 @@ extern void d_move(struct dentry *, struct dentry *);
 extern struct dentry *d_ancestor(struct dentry *, struct dentry *);
 
 /* appendix may either be NULL or be used for transname suffixes */
-extern struct dentry * d_lookup(struct dentry *, struct qstr *);
-extern struct dentry * __d_lookup(struct dentry *, struct qstr *);
-extern struct dentry * d_hash_and_lookup(struct dentry *, struct qstr *);
+extern struct dentry *d_lookup(struct dentry *, struct qstr *);
+extern struct dentry *d_hash_and_lookup(struct dentry *, struct qstr *);
+extern struct dentry *__d_lookup(struct dentry *, struct qstr *);
+extern struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name,
+				unsigned *seq, struct inode **inode);
+
+/**
+ * __d_rcu_to_refcount - take a refcount on dentry if sequence check is ok
+ * @dentry: dentry to take a ref on
+ * @seq: seqcount to verify against
+ * @Returns: 0 on failure, else 1.
+ *
+ * __d_rcu_to_refcount operates on a dentry,seq pair that was returned
+ * by __d_lookup_rcu, to get a reference on an rcu-walk dentry.
+ */
+static inline int __d_rcu_to_refcount(struct dentry *dentry, unsigned seq)
+{
+	int ret = 0;
+
+	assert_spin_locked(&dentry->d_lock);
+	if (!read_seqcount_retry(&dentry->d_seq, seq)) {
+		ret = 1;
+		dentry->d_count++;
+	}
+
+	return ret;
+}
 
 /* validate "insecure" dentry pointer */
 extern int d_validate(struct dentry *, struct dentry *);
diff --git a/include/linux/namei.h b/include/linux/namei.h
index aec730b5393..18d06add0a4 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -19,7 +19,10 @@ struct nameidata {
 	struct path	path;
 	struct qstr	last;
 	struct path	root;
+	struct file	*file;
+	struct inode	*inode; /* path.dentry.d_inode */
 	unsigned int	flags;
+	unsigned	seq;
 	int		last_type;
 	unsigned	depth;
 	char *saved_names[MAX_NESTED_LINKS + 1];
@@ -43,11 +46,13 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
  *  - internal "there are more path components" flag
  *  - dentry cache is untrusted; force a real lookup
  */
-#define LOOKUP_FOLLOW		 1
-#define LOOKUP_DIRECTORY	 2
-#define LOOKUP_CONTINUE		 4
-#define LOOKUP_PARENT		16
-#define LOOKUP_REVAL		64
+#define LOOKUP_FOLLOW		0x0001
+#define LOOKUP_DIRECTORY	0x0002
+#define LOOKUP_CONTINUE		0x0004
+
+#define LOOKUP_PARENT		0x0010
+#define LOOKUP_REVAL		0x0020
+#define LOOKUP_RCU		0x0040
 /*
  * Intent data
  */
diff --git a/include/linux/security.h b/include/linux/security.h
index fd4d55fb884..ed95401970c 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -457,7 +457,6 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	called when the actual read/write operations are performed.
  *	@inode contains the inode structure to check.
  *	@mask contains the permission mask.
- *	@nd contains the nameidata (may be NULL).
  *	Return 0 if permission is granted.
  * @inode_setattr:
  *	Check permission before setting file attributes.  Note that the kernel
@@ -1713,6 +1712,7 @@ 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,
@@ -2102,6 +2102,12 @@ 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)
 {