* [WTF?] AT_GETATTR_NOSEC checks
@ 2024-11-01 1:17 Al Viro
2024-11-06 10:27 ` Christian Brauner
0 siblings, 1 reply; 2+ messages in thread
From: Al Viro @ 2024-11-01 1:17 UTC (permalink / raw)
To: Christian Brauner; +Cc: Stefan Berger, linux-fsdevel
AFAICS, since the moment it had been introduced it got passed
to *ALL* ->getattr() calls. Unconditionally. So why are we checking
it in ecryptfs and overlayfs instances?
Look: all direct calls of instances are from other instances, with
query_flags passed unchanged. There is only one call via method -
that in vfs_getattr_nosec(), but that caller explicitly adds
AT_GETATTR_NOSEC to query_flags.
So what the hell are the checks in ecryptfs and overlayfs for?
What am I missing here? What would break if we did the following:
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 5ed1e4cf6c0b..255e60bd7dca 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -1008,14 +1008,6 @@ static int ecryptfs_getattr_link(struct mnt_idmap *idmap,
return rc;
}
-static int ecryptfs_do_getattr(const struct path *path, struct kstat *stat,
- u32 request_mask, unsigned int flags)
-{
- if (flags & AT_GETATTR_NOSEC)
- return vfs_getattr_nosec(path, stat, request_mask, flags);
- return vfs_getattr(path, stat, request_mask, flags);
-}
-
static int ecryptfs_getattr(struct mnt_idmap *idmap,
const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int flags)
@@ -1024,7 +1016,7 @@ static int ecryptfs_getattr(struct mnt_idmap *idmap,
struct kstat lower_stat;
int rc;
- rc = ecryptfs_do_getattr(ecryptfs_dentry_to_lower_path(dentry),
+ rc = vfs_getattr_nosec(ecryptfs_dentry_to_lower_path(dentry),
&lower_stat, request_mask, flags);
if (!rc) {
fsstack_copy_attr_all(d_inode(dentry),
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 0bfe35da4b7b..dde85ec96444 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -415,9 +415,8 @@ static inline bool ovl_open_flags_need_copy_up(int flags)
static inline int ovl_do_getattr(const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int flags)
{
- if (flags & AT_GETATTR_NOSEC)
- return vfs_getattr_nosec(path, stat, request_mask, flags);
- return vfs_getattr(path, stat, request_mask, flags);
+ BUG_ON(!(flags & AT_GETATTR_NOSEC));
+ return vfs_getattr_nosec(path, stat, request_mask, flags);
}
/* util.c */
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [WTF?] AT_GETATTR_NOSEC checks
2024-11-01 1:17 [WTF?] AT_GETATTR_NOSEC checks Al Viro
@ 2024-11-06 10:27 ` Christian Brauner
0 siblings, 0 replies; 2+ messages in thread
From: Christian Brauner @ 2024-11-06 10:27 UTC (permalink / raw)
To: Al Viro; +Cc: Stefan Berger, linux-fsdevel
On Fri, Nov 01, 2024 at 01:17:24AM +0000, Al Viro wrote:
> AFAICS, since the moment it had been introduced it got passed
> to *ALL* ->getattr() calls. Unconditionally. So why are we checking
> it in ecryptfs and overlayfs instances?
>
> Look: all direct calls of instances are from other instances, with
> query_flags passed unchanged. There is only one call via method -
> that in vfs_getattr_nosec(), but that caller explicitly adds
> AT_GETATTR_NOSEC to query_flags.
>
> So what the hell are the checks in ecryptfs and overlayfs for?
> What am I missing here? What would break if we did the following:
Sounds good. I'm confused why that's a WTF moment though.
Anyway, I'm just slowly getting back to the fray. I caught atypical
pneumonia and that has been going on for a few weeks now. So my replies
are delayed.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-11-06 10:27 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-11-01 1:17 [WTF?] AT_GETATTR_NOSEC checks Al Viro
2024-11-06 10:27 ` Christian Brauner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox