From 831c4af9270129d7896d107c8e25c5b14367f5bf Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Wed, 5 Mar 2008 13:46:41 -0800 Subject: [PATCH] client: disable ceph_lookup_open intent stuff for now --- src/kernel/file.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/src/kernel/file.c b/src/kernel/file.c index b659d168f63ad..5ff3c4a58f4c5 100644 --- a/src/kernel/file.c +++ b/src/kernel/file.c @@ -133,6 +133,20 @@ int ceph_open(struct inode *inode, struct file *file) return 0; } + +/* + * FIXME + * + * this whole thing needs some fixing. we need to be sure that + * a successful lookup (that does an open) will be followed by + * a VFS open (that sets up teh file*?). i think there are + * lots of things that can ahppen in between, mainly security + * checks that may not match the server's checks. + * otherwise, it's not safe to just go ahead and do the open + * here or you won't get an eventual release... + */ +#if 0 + int ceph_lookup_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { @@ -200,6 +214,7 @@ out: //ceph_mdsc_put_session(session); return err; } +#endif int ceph_release(struct inode *inode, struct file *file) { -- 2.39.5