Introduce a global vnode_mtx and use it to make vn_lock() safe to be called

without the KERNEL_LOCK.
This moves VXLOCK and VXWANT to a mutex protected v_lflag field and also
v_lockcount is protected by this mutex.

The vn_lock() dance is overly complex and all of this should probably replaced
by a proper lock on the vnode but such a diff is a lot more complex. This
is an intermediate step so that at least some calls can be modified to grab
the KERNEL_LOCK later or not at all.

OK mpi@

1 files changed, 6 insertions, 4 deletions

diff --git a/sys/miscfs/deadfs/dead_vnops.c b/sys/miscfs/deadfs/dead_vnops.c
index 54083f509b2..8aada87b09e 100644
--- a/sys/miscfs/deadfs/dead_vnops.c
+++ b/sys/miscfs/deadfs/dead_vnops.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: dead_vnops.c,v 1.34 2021/03/24 16:11:32 semarie Exp $	*/
+/*	$OpenBSD: dead_vnops.c,v 1.35 2021/04/28 09:53:53 claudio Exp $	*/
 /*	$NetBSD: dead_vnops.c,v 1.16 1996/02/13 13:12:48 mycroft Exp $	*/
 
 /*
@@ -285,10 +285,12 @@ chkvnlock(struct vnode *vp)
 {
 	int locked = 0;
 
-	while (vp->v_flag & VXLOCK) {
-		vp->v_flag |= VXWANT;
-		tsleep_nsec(vp, PINOD, "chkvnlock", INFSLP);
+	mtx_enter(&vnode_mtx);
+	while (vp->v_lflag & VXLOCK) {
+		vp->v_lflag |= VXWANT;
+		msleep_nsec(vp, &vnode_mtx, PINOD, "chkvnlock", INFSLP);
 		locked = 1;
 	}
+	mtx_leave(&vnode_mtx);
 	return (locked);
 }