From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 3 Apr 2021 20:49:44 +0000 (-0400)
Subject: namei: make sure nd->depth is always valid
X-Git-Tag: v5.15~813^2
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=7962c7d196e36aa597fadb78c1cb4fe7e209f803;p=platform%2Fkernel%2Flinux-starfive.git

namei: make sure nd->depth is always valid

Zero it in set_nameidata() rather than in path_init().  That way
it always matches the number of valid nd->stack[] entries.
Since terminate_walk() does zero it (after having emptied the
stack), we don't need to reinitialize it in subsequent path_init().

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---

diff --git a/fs/namei.c b/fs/namei.c
index 40ffb24..d38b17a 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -581,6 +581,7 @@ static void __set_nameidata(struct nameidata *p, int dfd, struct filename *name)
 {
 	struct nameidata *old = current->nameidata;
 	p->stack = p->internal;
+	p->depth = 0;
 	p->dfd = dfd;
 	p->name = name;
 	p->path.mnt = NULL;
@@ -2320,7 +2321,6 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 
 	nd->flags = flags;
 	nd->state |= ND_JUMPED;
-	nd->depth = 0;
 
 	nd->m_seq = __read_seqcount_begin(&mount_lock.seqcount);
 	nd->r_seq = __read_seqcount_begin(&rename_lock.seqcount);