Messages in this thread Patch in this message | | | Date | Thu, 26 Mar 2015 00:19:32 +0900 (JST) | Subject | Re: [PATCH 5/7] nilfs2: add bmap function to seek a valid key | From | Ryusuke Konishi <> |
| |
Hi Andrew,
On Thu, 12 Mar 2015 23:38:12 +0900, Ryusuke Konishi wrote: > Add a new bmap function, nilfs_bmap_seek_key(), which seeks a valid > entry and returns its key starting from a given key. This function > can be used to skip hole blocks efficiently. > > Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> > --- <snip> > +static int nilfs_btree_seek_key(const struct nilfs_bmap *btree, __u64 start, > + __u64 *keyp) > +{ > + struct nilfs_btree_path *path; > + const int minlevel = NILFS_BTREE_LEVEL_NODE_MIN; > + int ret; > +
> + if (start > NILFS_BTREE_KEY_MAX) > + return -ENOENT;
Could you apply the following amendment ?
I've got a report from Dan Carpenter that this range check causes a warning:
fs/nilfs2/btree.c:1611 nilfs_btree_seek_key() warn: impossible condition '(start > (~0)) => (0-u64max > u64max)'
A range check was needed there in a wip patch, but we no longer need it.
Thanks, Ryusuke Konishi -- diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c index 841d177..059f371 100644 --- a/fs/nilfs2/btree.c +++ b/fs/nilfs2/btree.c @@ -1608,9 +1608,6 @@ static int nilfs_btree_seek_key(const struct nilfs_bmap *btree, __u64 start, const int minlevel = NILFS_BTREE_LEVEL_NODE_MIN; int ret; - if (start > NILFS_BTREE_KEY_MAX) - return -ENOENT; - path = nilfs_btree_alloc_path(); if (!path) return -ENOMEM; -- 1.8.3.1
| |