Add mnt_hashseed to struct mount and initialize it witn PRNG bits, use

it to get better hashing in vfs_hash.

In case of an insert collision in vfs_hash_insert(), put the loosing vnode
on a special list so that vfs_hash_remove() can just assume that it is on
a list.

Drop the VI_HASHED flag.

4 files changed, 19 insertions, 16 deletions

diff --git a/sys/kern/vfs_hash.c b/sys/kern/vfs_hash.c
index 72a0278..8940977 100644
--- a/sys/kern/vfs_hash.c
+++ b/sys/kern/vfs_hash.c
@@ -32,11 +32,13 @@ __FBSDID("$FreeBSD$");
 #include <sys/systm.h>
 #include <sys/kernel.h>
 #include <sys/malloc.h>
+#include <sys/mount.h>
 #include <sys/vnode.h>
 
 static MALLOC_DEFINE(M_VFS_HASH, "VFS hash", "VFS hash table");
 
-static LIST_HEAD(vfs_hashhead, vnode)	*vfs_hash_tbl;
+static LIST_HEAD(vfs_hash_head, vnode)	*vfs_hash_tbl;
+static LIST_HEAD(,vnode)		vfs_hash_side;
 static u_long				vfs_hash_mask;
 static struct mtx			vfs_hash_mtx;
 
@@ -46,11 +48,19 @@ vfs_hashinit(void *dummy __unused)
 
 	vfs_hash_tbl = hashinit(desiredvnodes, M_VFS_HASH, &vfs_hash_mask);
 	mtx_init(&vfs_hash_mtx, "vfs hash", NULL, MTX_DEF);
+	LIST_INIT(&vfs_hash_side);
 }
 
 /* Must be SI_ORDER_SECOND so desiredvnodes is available */
 SYSINIT(vfs_hash, SI_SUB_VFS, SI_ORDER_SECOND, vfs_hashinit, NULL)
 
+static struct vfs_hash_head *
+vfs_hash_index(struct mount *mp, u_int hash)
+{
+
+	return(&vfs_hash_tbl[(hash + mp->mnt_hashseed) & vfs_hash_mask]);
+}
+
 int
 vfs_hash_get(struct mount *mp, u_int hash, int flags, struct thread *td, struct vnode **vpp)
 {
@@ -59,8 +69,7 @@ vfs_hash_get(struct mount *mp, u_int hash, int flags, struct thread *td, struct
 
 	while (1) {
 		mtx_lock(&vfs_hash_mtx);
-		LIST_FOREACH(vp,
-		    &vfs_hash_tbl[hash & vfs_hash_mask], v_hashlist) {
+		LIST_FOREACH(vp, vfs_hash_index(mp, hash), v_hashlist) {
 			if (vp->v_hash != hash)
 				continue;
 			if (vp->v_mount != mp)
@@ -88,11 +97,7 @@ vfs_hash_remove(struct vnode *vp)
 {
 
 	mtx_lock(&vfs_hash_mtx);
-	VI_LOCK(vp);
-	VNASSERT(vp->v_iflag & VI_HASHED, vp, ("vfs_hash_remove: not hashed"));
 	LIST_REMOVE(vp, v_hashlist);
-	vp->v_iflag &= ~VI_HASHED;
-	VI_UNLOCK(vp);
 	mtx_unlock(&vfs_hash_mtx);
 }
 
@@ -107,7 +112,7 @@ vfs_hash_insert(struct vnode *vp, u_int hash, int flags, struct thread *td, stru
 	while (1) {
 		mtx_lock(&vfs_hash_mtx);
 		LIST_FOREACH(vp2,
-		    &vfs_hash_tbl[hash & vfs_hash_mask], v_hashlist) {
+		    vfs_hash_index(vp->v_mount, hash), v_hashlist) {
 			if (vp2->v_hash != hash)
 				continue;
 			if (vp2->v_mount != vp->v_mount)
@@ -117,6 +122,9 @@ vfs_hash_insert(struct vnode *vp, u_int hash, int flags, struct thread *td, stru
 			error = vget(vp2, flags | LK_INTERLOCK, td);
 			if (error == ENOENT)
 				break;
+			mtx_lock(&vfs_hash_mtx);
+			LIST_INSERT_HEAD(&vfs_hash_side, vp, v_hashlist);
+			mtx_unlock(&vfs_hash_mtx);
 			vput(vp);
 			if (!error)
 				*vpp = vp2;
@@ -126,13 +134,8 @@ vfs_hash_insert(struct vnode *vp, u_int hash, int flags, struct thread *td, stru
 			break;
 			
 	}
-	VI_LOCK(vp);
-	VNASSERT(!(vp->v_iflag & VI_HASHED), vp,
-	    ("vfs_hash_insert: already hashed"));
 	vp->v_hash = hash;
-	LIST_INSERT_HEAD(&vfs_hash_tbl[hash & vfs_hash_mask], vp, v_hashlist);
-	vp->v_iflag |= VI_HASHED;
-	VI_UNLOCK(vp);
+	LIST_INSERT_HEAD(vfs_hash_index(vp->v_mount, hash), vp, v_hashlist);
 	mtx_unlock(&vfs_hash_mtx);
 	return (0);
 }
@@ -142,7 +145,6 @@ vfs_hash_rehash(struct vnode *vp, u_int hash)
 {
 
 	mtx_lock(&vfs_hash_mtx);
-	VNASSERT(vp->v_iflag & VI_HASHED, vp, ("vfs_hash_rehash: not hashed"));
 	LIST_REMOVE(vp, v_hashlist);
 	LIST_INSERT_HEAD(&vfs_hash_tbl[hash & vfs_hash_mask], vp, v_hashlist);
 	vp->v_hash = hash;
diff --git a/sys/kern/vfs_mount.c b/sys/kern/vfs_mount.c
index 94e2715..7930386 100644
--- a/sys/kern/vfs_mount.c
+++ b/sys/kern/vfs_mount.c
@@ -428,6 +428,7 @@ vfs_mount_alloc(struct vnode *vp, struct vfsconf *vfsp,
 	mac_init_mount(mp);
 	mac_create_mount(td->td_ucred, mp);
 #endif
+	arc4rand(&mp->mnt_hashseed, sizeof mp->mnt_hashseed, 0);
 	*mpp = mp;
 	return (0);
 }
diff --git a/sys/sys/mount.h b/sys/sys/mount.h
index 527ed44..3c4cf2e 100644
--- a/sys/sys/mount.h
+++ b/sys/sys/mount.h
@@ -159,6 +159,7 @@ struct mount {
 	struct label	*mnt_mntlabel;		/* MAC label for the mount */
 	struct label	*mnt_fslabel;		/* MAC label for the fs */
 	int		mnt_nvnodelistsize;	/* # of vnodes on this mount */
+	u_int		mnt_hashseed;		/* Random seed for vfs_hash */
 };
 
 struct vnode *__mnt_vnode_next(struct vnode **nvp, struct mount *mp);
diff --git a/sys/sys/vnode.h b/sys/sys/vnode.h
index f12427d..920df12 100644
--- a/sys/sys/vnode.h
+++ b/sys/sys/vnode.h
@@ -238,7 +238,6 @@ struct xvnode {
  *	VI flags are protected by interlock and live in v_iflag
  *	VV flags are protected by the vnode lock and live in v_vflag
  */
-#define	VI_HASHED	0x0004	/* vnode is hashed in vfs_hash */
 #define	VI_MOUNT	0x0020	/* Mount in progress */
 #define	VI_AGE		0x0040	/* Insert vnode at head of free list */
 #define	VI_DOOMED	0x0080	/* This vnode is being recycled */