- Remove the backtrace() call from the *_vinvalbuf() functions. Thanks to a

   stack trace supplied by phk, I now understand what's going on here.  The
   check for VI_XLOCK stops us from calling vinvalbuf once the vnode has been
   partially torn down in vclean().  It is not clear that this would cause
   a problem.  Document this in nfs_bio.c, which is where the other two
   filesystems copied this code from.

1 files changed, 6 insertions, 5 deletions

diff --git a/sys/nfsclient/nfs_bio.c b/sys/nfsclient/nfs_bio.c
index 9e613ad..a0d507e 100644
--- a/sys/nfsclient/nfs_bio.c
+++ b/sys/nfsclient/nfs_bio.c
@@ -1067,12 +1067,13 @@ nfs_vinvalbuf(struct vnode *vp, int flags, struct ucred *cred,
 
 	ASSERT_VOP_LOCKED(vp, "nfs_vinvalbuf");
 
-	if (vp->v_iflag & VI_XLOCK) {
-#ifdef INVARIANTS
-		backtrace();
-#endif
+	/*
+	 * XXX This check stops us from needlessly doing a vinvalbuf when
+	 * being called through vclean().  It is not clear that this is
+	 * unsafe.
+	 */
+	if (vp->v_iflag & VI_XLOCK)
 		return (0);
-	}
 
 	if ((nmp->nm_flag & NFSMNT_INT) == 0)
 		intrflg = 0;