From: Matt Benjamin <matt@linuxbox.com>
Date: Sun, 29 Aug 2010 21:33:03 +0000 (-0400)
Subject: FBSD  try-relax child vnode locking (recurse only)
X-Git-Tag: openafs-devel-1_5_77~30
X-Git-Url: https://git.michaelhowe.org/gitweb/?a=commitdiff_plain;h=88bcd3ed6367280bd97ae56cc2c25308e82d86e0;p=packages%2Fo%2Fopenafs.git

FBSD  try-relax child vnode locking (recurse only)

In cases where afs_vop_lookup would return a child vnode
locked, continue to take an exclusive lock, but allow
recursion (LK_CANRECURSE).  Allow recursion also at
afs_vop_link, where we specifically encountered a conflict
due to recursion.

Change-Id: I5f5b0979ce3c86e517f96ea60b96ebdd26401acb
Reviewed-on: http://gerrit.openafs.org/2625
Reviewed-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
(cherry picked from commit ec85304dd738935d8be3f8934b166f78a10995bf)
Reviewed-on: http://gerrit.openafs.org/2683
---

diff --git a/src/afs/FBSD/osi_vnodeops.c b/src/afs/FBSD/osi_vnodeops.c
index d5c16b285..cbd94d797 100644
--- a/src/afs/FBSD/osi_vnodeops.c
+++ b/src/afs/FBSD/osi_vnodeops.c
@@ -548,7 +548,7 @@ afs_vop_lookup(ap)
 	    MA_VOP_UNLOCK(dvp, 0, p);	/* done with parent. */
 #endif
 	}
-	ma_vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
+	ma_vn_lock(vp, LK_EXCLUSIVE | LK_CANRECURSE | LK_RETRY, p);
 	/* always return the child locked */
     }
     *ap->a_vpp = vp;
@@ -1162,7 +1162,7 @@ afs_vop_link(ap)
 	error = EISDIR;
 	goto out;
     }
-    if ((error = ma_vn_lock(vp, LK_EXCLUSIVE, p)) != 0) {
+    if ((error = ma_vn_lock(vp, LK_CANRECURSE | LK_EXCLUSIVE, p)) != 0) {
 	goto out;
     }
     AFS_GLOCK();