aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorKonstantin Belousov <kib@FreeBSD.org>2023-11-16 22:42:25 +0000
committerKonstantin Belousov <kib@FreeBSD.org>2023-11-17 01:51:41 +0000
commit22bac49b09a9f975330dc0d811436599ce740890 (patch)
tree5d10215c6aa534ede2fd71e252ac09388ea3360e
parente256f71389fbf0ef6cf8f547342d5f789e69f918 (diff)
downloadsrc-22bac49b09a9f975330dc0d811436599ce740890.tar.gz
src-22bac49b09a9f975330dc0d811436599ce740890.zip
vn_lock_pair(): reasonably handle vp1 == vp2 case
Lock the vnode in the most exclusive lock mode requested, once. All callers already ensure that vp1 != vp2 or are careful enough to only unlock once otherwise. Reviewed by: markj Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D42642
-rw-r--r--sys/kern/vfs_vnops.c41
1 files changed, 36 insertions, 5 deletions
diff --git a/sys/kern/vfs_vnops.c b/sys/kern/vfs_vnops.c
index 27ce5401f15f..f81516608eeb 100644
--- a/sys/kern/vfs_vnops.c
+++ b/sys/kern/vfs_vnops.c
@@ -4042,10 +4042,10 @@ vn_lock_pair_pause(const char *wmesg)
}
/*
- * Lock pair of vnodes vp1, vp2, avoiding lock order reversal.
- * vp1_locked indicates whether vp1 is locked; if not, vp1 must be
- * unlocked. Same for vp2 and vp2_locked. One of the vnodes can be
- * NULL.
+ * Lock pair of (possibly same) vnodes vp1, vp2, avoiding lock order
+ * reversal. vp1_locked indicates whether vp1 is locked; if not, vp1
+ * must be unlocked. Same for vp2 and vp2_locked. One of the vnodes
+ * can be NULL.
*
* The function returns with both vnodes exclusively or shared locked,
* according to corresponding lkflags, and guarantees that it does not
@@ -4056,12 +4056,14 @@ vn_lock_pair_pause(const char *wmesg)
*
* Only one of LK_SHARED and LK_EXCLUSIVE must be specified.
* LK_NODDLKTREAT can be optionally passed.
+ *
+ * If vp1 == vp2, only one, most exclusive, lock is obtained on it.
*/
void
vn_lock_pair(struct vnode *vp1, bool vp1_locked, int lkflags1,
struct vnode *vp2, bool vp2_locked, int lkflags2)
{
- int error;
+ int error, locked1;
MPASS(((lkflags1 & LK_SHARED) != 0) ^ ((lkflags1 & LK_EXCLUSIVE) != 0));
MPASS((lkflags1 & ~(LK_SHARED | LK_EXCLUSIVE | LK_NODDLKTREAT)) == 0);
@@ -4071,6 +4073,35 @@ vn_lock_pair(struct vnode *vp1, bool vp1_locked, int lkflags1,
if (vp1 == NULL && vp2 == NULL)
return;
+ if (vp1 == vp2) {
+ MPASS(vp1_locked == vp2_locked);
+
+ /* Select the most exclusive mode for lock. */
+ if ((lkflags1 & LK_TYPE_MASK) != (lkflags2 & LK_TYPE_MASK))
+ lkflags1 = (lkflags1 & ~LK_SHARED) | LK_EXCLUSIVE;
+
+ if (vp1_locked) {
+ ASSERT_VOP_LOCKED(vp1, "vp1");
+
+ /* No need to relock if any lock is exclusive. */
+ if ((vp1->v_vnlock->lock_object.lo_flags &
+ LK_NOSHARE) != 0)
+ return;
+
+ locked1 = VOP_ISLOCKED(vp1);
+ if (((lkflags1 & LK_SHARED) != 0 &&
+ locked1 != LK_EXCLUSIVE) ||
+ ((lkflags1 & LK_EXCLUSIVE) != 0 &&
+ locked1 == LK_EXCLUSIVE))
+ return;
+ VOP_UNLOCK(vp1);
+ }
+
+ ASSERT_VOP_UNLOCKED(vp1, "vp1");
+ vn_lock(vp1, lkflags1 | LK_RETRY);
+ return;
+ }
+
if (vp1 != NULL) {
if ((lkflags1 & LK_SHARED) != 0 &&
(vp1->v_vnlock->lock_object.lo_flags & LK_NOSHARE) != 0)