Hi, David,
On 2013/12/17 0:27, David Sterba wrote:
On Fri, Dec 13, 2013 at 09:51:42AM +0900, Tsutomu Itoh wrote:
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4974,10 +4974,17 @@ static void btrfs_dentry_release(struct dentry *dentry)
static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry,
unsigned int flags)
{
- struct dentry *ret;
+ struct inode *inode;
- ret = d_splice_alias(btrfs_lookup_dentry(dir, dentry), dentry);
- return ret;
+ inode = btrfs_lookup_dentry(dir, dentry);
+ if (IS_ERR(inode)) {
+ if (PTR_ERR(inode) == -ENOENT)
+ inode = NULL;
+ else
+ return ERR_CAST(inode);
+ }
+
+ return d_splice_alias(inode, dentry);
btrfs_lookup used to be a simple d_splice_alias(btrfs_lookup_dentry ...)
and the expanded back and forth with the DCACHE_NEED_LOOKUP flag.
a66e7cc626f42de6c745963fe0d807518fa49d39 added
39e3c9553f34381a1b664c27b0c696a266a5735e removed
d_splice_alias has been made robust in
a9049376ee05bf966bfe2b081b5071326856890a
"make d_splice_alias(ERR_PTR(err), dentry) = ERR_PTR(err)"
you can drop the error handling from btrfs_lookup completely.
d_splice_alias() is called by the following formats if btrfs_lookup_dentry()
returns NULL before my patch is applied.
d_splice_alias(NULL, dentry);
However, d_splice_alias() is called by the following formats when
becoming the same situation after my patch is applied.
d_splice_alias(-ENOENT, dentry);
Therefore, I added the following error handling code.
+ if (IS_ERR(inode)) {
+ if (PTR_ERR(inode) == -ENOENT)
+ inode = NULL;
+ else
+ return ERR_CAST(inode);
+ }
Thanks,
Tsutomu
The rest looks ok.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
