Module Name: src
Committed By: hannken
Date: Thu Nov 12 11:35:42 UTC 2015
Modified Files:
src/sys/kern: vfs_vnode.c
Log Message:
Take the vnode lock before the vnode is marked VI_CHANGING and fed
to vclean(). Prevents a deadlock with two null mounts on the same
physical mount where one thread tries to vclean() a layer node and
another thread tries to vget() a layer node pointing to the same
physical node.
Fixes PR kern/50375 layerfs (nullfs) locking problem leading to livelock
To generate a diff of this commit:
cvs rdiff -u -r1.45 -r1.46 src/sys/kern/vfs_vnode.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/sys/kern/vfs_vnode.c
diff -u src/sys/kern/vfs_vnode.c:1.45 src/sys/kern/vfs_vnode.c:1.46
--- src/sys/kern/vfs_vnode.c:1.45 Sun Jul 12 08:11:28 2015
+++ src/sys/kern/vfs_vnode.c Thu Nov 12 11:35:42 2015
@@ -1,4 +1,4 @@
-/* $NetBSD: vfs_vnode.c,v 1.45 2015/07/12 08:11:28 hannken Exp $ */
+/* $NetBSD: vfs_vnode.c,v 1.46 2015/11/12 11:35:42 hannken Exp $ */
/*-
* Copyright (c) 1997-2011 The NetBSD Foundation, Inc.
@@ -116,7 +116,7 @@
*/
#include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: vfs_vnode.c,v 1.45 2015/07/12 08:11:28 hannken Exp $");
+__KERNEL_RCSID(0, "$NetBSD: vfs_vnode.c,v 1.46 2015/11/12 11:35:42 hannken Exp $");
#define _VFS_VNODE_PRIVATE
@@ -325,15 +325,17 @@ try_nextlist:
KASSERT((vp->v_iflag & VI_CLEAN) == 0);
KASSERT(vp->v_freelisthd == listhd);
- if (!mutex_tryenter(vp->v_interlock))
+ if (vn_lock(vp, LK_EXCLUSIVE | LK_NOWAIT) != 0)
continue;
- if ((vp->v_iflag & VI_XLOCK) != 0) {
- mutex_exit(vp->v_interlock);
+ if (!mutex_tryenter(vp->v_interlock)) {
+ VOP_UNLOCK(vp);
continue;
}
+ KASSERT((vp->v_iflag & VI_XLOCK) == 0);
mp = vp->v_mount;
if (fstrans_start_nowait(mp, FSTRANS_SHARED) != 0) {
mutex_exit(vp->v_interlock);
+ VOP_UNLOCK(vp);
continue;
}
break;
@@ -643,6 +645,11 @@ vrelel(vnode_t *vp, int flags)
* Note that VOP_INACTIVE() will drop the vnode lock.
*/
VOP_INACTIVE(vp, &recycle);
+ if (recycle) {
+ /* vclean() below will drop the lock. */
+ if (vn_lock(vp, LK_EXCLUSIVE) != 0)
+ recycle = false;
+ }
mutex_enter(vp->v_interlock);
if (!recycle) {
if (vtryrele(vp)) {
@@ -867,6 +874,7 @@ holdrelel(vnode_t *vp)
/*
* Disassociate the underlying file system from a vnode.
*
+ * Must be called with vnode locked and will return unlocked.
* Must be called with the interlock held, and will return with it held.
*/
static void
@@ -876,26 +884,18 @@ vclean(vnode_t *vp)
bool recycle, active;
int error;
+ KASSERT((vp->v_vflag & VV_LOCKSWORK) == 0 ||
+ VOP_ISLOCKED(vp) == LK_EXCLUSIVE);
KASSERT(mutex_owned(vp->v_interlock));
KASSERT((vp->v_iflag & VI_MARKER) == 0);
+ KASSERT((vp->v_iflag & (VI_XLOCK | VI_CLEAN)) == 0);
KASSERT(vp->v_usecount != 0);
- /* If already clean, nothing to do. */
- if ((vp->v_iflag & VI_CLEAN) != 0) {
- return;
- }
-
active = (vp->v_usecount > 1);
- mutex_exit(vp->v_interlock);
-
- vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
-
/*
* Prevent the vnode from being recycled or brought into use
* while we clean it out.
*/
- mutex_enter(vp->v_interlock);
- KASSERT((vp->v_iflag & (VI_XLOCK | VI_CLEAN)) == 0);
vp->v_iflag |= VI_XLOCK;
if (vp->v_iflag & VI_EXECMAP) {
atomic_add_int(&uvmexp.execpages, -vp->v_uobj.uo_npages);
@@ -972,23 +972,26 @@ bool
vrecycle(vnode_t *vp)
{
+ if (vn_lock(vp, LK_EXCLUSIVE) != 0)
+ return false;
+
mutex_enter(vp->v_interlock);
KASSERT((vp->v_iflag & VI_MARKER) == 0);
if (vp->v_usecount != 1) {
mutex_exit(vp->v_interlock);
+ VOP_UNLOCK(vp);
return false;
}
if ((vp->v_iflag & VI_CHANGING) != 0)
vwait(vp, VI_CHANGING);
if (vp->v_usecount != 1) {
mutex_exit(vp->v_interlock);
+ VOP_UNLOCK(vp);
return false;
- } else if ((vp->v_iflag & VI_CLEAN) != 0) {
- mutex_exit(vp->v_interlock);
- return true;
}
+ KASSERT((vp->v_iflag & VI_CLEAN) == 0);
vp->v_iflag |= VI_CHANGING;
vclean(vp);
vrelel(vp, VRELEL_CHANGING_SET);
@@ -1036,6 +1039,11 @@ void
vgone(vnode_t *vp)
{
+ if (vn_lock(vp, LK_EXCLUSIVE) != 0) {
+ KASSERT((vp->v_iflag & VI_CLEAN) != 0);
+ vrele(vp);
+ }
+
mutex_enter(vp->v_interlock);
if ((vp->v_iflag & VI_CHANGING) != 0)
vwait(vp, VI_CHANGING);
