Author: rmacklem
Date: Sun Jun 17 18:34:04 2012
New Revision: 237200
URL: http://svn.freebsd.org/changeset/base/237200
Log:
Move the nfsrpc_close() call in ncl_reclaim() for the NFSv4 client
to below the vnode_destroy_vobject() call, since that is where
writes are flushed.
Suggested by: kib
MFC after: 1 week
Modified:
head/sys/fs/nfsclient/nfs_clnode.c
Modified: head/sys/fs/nfsclient/nfs_clnode.c
==============================================================================
--- head/sys/fs/nfsclient/nfs_clnode.c Sun Jun 17 16:59:37 2012
(r237199)
+++ head/sys/fs/nfsclient/nfs_clnode.c Sun Jun 17 18:34:04 2012
(r237200)
@@ -257,15 +257,6 @@ ncl_reclaim(struct vop_reclaim_args *ap)
struct nfsnode *np = VTONFS(vp);
struct nfsdmap *dp, *dp2;
- if (NFS_ISV4(vp) && vp->v_type == VREG)
- /*
- * Since mmap()'d files do I/O after VOP_CLOSE(), the NFSv4
- * Close operations are delayed until ncl_inactive().
- * However, since VOP_INACTIVE() is not guaranteed to be
- * called, we need to do it again here.
- */
- (void) nfsrpc_close(vp, 1, ap->a_td);
-
/*
* If the NLM is running, give it a chance to abort pending
* locks.
@@ -278,6 +269,15 @@ ncl_reclaim(struct vop_reclaim_args *ap)
*/
vnode_destroy_vobject(vp);
+ if (NFS_ISV4(vp) && vp->v_type == VREG)
+ /*
+ * We can now safely close any remaining NFSv4 Opens for
+ * this file. Most opens will have already been closed by
+ * ncl_inactive(), but there are cases where it is not
+ * called, so we need to do it again here.
+ */
+ (void) nfsrpc_close(vp, 1, ap->a_td);
+
vfs_hash_remove(vp);
/*
_______________________________________________
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
