From: Chuck Lever <cel@kernel.org>
To: NeilBrown <neil@brown.name>, Jeff Layton <jlayton@kernel.org>,
Olga Kornievskaia <okorniev@redhat.com>,
Dai Ngo <dai.ngo@oracle.com>, Tom Talpey <tom@talpey.com>
Cc: <linux-nfs@vger.kernel.org>, sashiko-bot <sashiko-bot@kernel.org>
Subject: [PATCH] NFSD: Map flex file layout IDs through the request's user namespace
Date: Mon, 20 Jul 2026 12:37:24 -0400 [thread overview]
Message-ID: <20260720163724.810227-1-cel@kernel.org> (raw)
nfsd4_ff_proc_layoutget() and nfsd4_ff_encode_layoutget() translate
the file's owner and group with init_user_ns, but every other identity
nfsd places on the wire goes through nfsd_user_namespace(). When the
transport carries a credential from a non-initial user namespace, the
flex file layout reports host-global IDs. The client copies those IDs
into the AUTH_SYS credential it presents to the data server, and
svcauth_unix_accept() resolves that credential in the transport's
namespace, so data server I/O runs under an identity unrelated to the
file's owner.
Switching to the request's namespace introduces a second hazard.
from_kuid() returns (uid_t)-1 when the target namespace has no
mapping for the owner, and the IOMODE_READ arm adds one to that
result to derive an identity for which the data server denies
writes. The addition would wrap to zero, handing the client uid 0
instead of an identity distinct from the owner.
Translate both IDs in nfsd4_ff_proc_layoutget(), which has the
svc_rqst, and carry the wire values in struct pnfs_ff_layout.
from_kuid_munged() substitutes overflowuid for an unmapped owner and
thus never returns (uid_t)-1, so the increment cannot wrap to zero.
Fixes: 9b9960a0ca47 ("nfsd: Add a super simple flex file server")
Reported-by: sashiko-bot <sashiko-bot@kernel.org>
Closes: https://sashiko.dev/#/patchset/20260720141442.783935-1-cel@kernel.org?part=3
Signed-off-by: Chuck Lever <cel@kernel.org>
---
fs/nfsd/flexfilelayout.c | 20 ++++++++++++--------
fs/nfsd/flexfilelayoutxdr.c | 4 ++--
fs/nfsd/flexfilelayoutxdr.h | 5 +++--
3 files changed, 17 insertions(+), 12 deletions(-)
diff --git a/fs/nfsd/flexfilelayout.c b/fs/nfsd/flexfilelayout.c
index 9f532418cac8..c6e6b9106a83 100644
--- a/fs/nfsd/flexfilelayout.c
+++ b/fs/nfsd/flexfilelayout.c
@@ -15,6 +15,7 @@
#include "nfserr.h"
#include "flexfilelayoutxdr.h"
+#include "auth.h"
#include "pnfs.h"
#include "vfs.h"
@@ -24,10 +25,10 @@ static __be32
nfsd4_ff_proc_layoutget(struct svc_rqst *rqstp, struct inode *inode,
const struct svc_fh *fhp, struct nfsd4_layoutget *args)
{
+ struct user_namespace *userns = nfsd_user_namespace(rqstp);
struct nfsd4_layout_seg *seg = &args->lg_seg;
u32 device_generation = 0;
int error;
- uid_t u;
struct pnfs_ff_layout *fl;
@@ -50,13 +51,16 @@ nfsd4_ff_proc_layoutget(struct svc_rqst *rqstp, struct inode *inode,
fl->flags = FF_FLAGS_NO_LAYOUTCOMMIT | FF_FLAGS_NO_IO_THRU_MDS |
FF_FLAGS_NO_READ_IO;
- /* Do not allow a IOMODE_READ segment to have write pemissions */
- if (seg->iomode == IOMODE_READ) {
- u = from_kuid(&init_user_ns, inode->i_uid) + 1;
- fl->uid = make_kuid(&init_user_ns, u);
- } else
- fl->uid = inode->i_uid;
- fl->gid = inode->i_gid;
+ fl->uid = from_kuid_munged(userns, inode->i_uid);
+ fl->gid = from_kgid_munged(userns, inode->i_gid);
+
+ /*
+ * Do not allow an IOMODE_READ segment to have write permissions.
+ * The group is left intact so group-readable files stay readable;
+ * nfsd_setuser() squashes an unmapped uid to the export's anon ID.
+ */
+ if (seg->iomode == IOMODE_READ)
+ fl->uid++;
error = nfsd4_set_deviceid(&fl->deviceid, fhp, device_generation);
if (error)
diff --git a/fs/nfsd/flexfilelayoutxdr.c b/fs/nfsd/flexfilelayoutxdr.c
index 97d8a28dd3a0..c12bb7c371b0 100644
--- a/fs/nfsd/flexfilelayoutxdr.c
+++ b/fs/nfsd/flexfilelayoutxdr.c
@@ -33,8 +33,8 @@ nfsd4_ff_encode_layoutget(struct xdr_stream *xdr,
fh_len = 4 + xdr_align_size(fl->fh.size);
- uid.len = sprintf(uid.buf, "%u", from_kuid(&init_user_ns, fl->uid));
- gid.len = sprintf(gid.buf, "%u", from_kgid(&init_user_ns, fl->gid));
+ uid.len = sprintf(uid.buf, "%u", fl->uid);
+ gid.len = sprintf(gid.buf, "%u", fl->gid);
/* data server entry: deviceid + efficiency + stateid + fh list +
* user + group + flags + stats_collect_hint
diff --git a/fs/nfsd/flexfilelayoutxdr.h b/fs/nfsd/flexfilelayoutxdr.h
index 6d5a1066a903..3e1876d49db2 100644
--- a/fs/nfsd/flexfilelayoutxdr.h
+++ b/fs/nfsd/flexfilelayoutxdr.h
@@ -35,8 +35,9 @@ struct pnfs_ff_device_addr {
struct pnfs_ff_layout {
u32 flags;
u32 stats_collect_hint;
- kuid_t uid;
- kgid_t gid;
+ /* Values to encode; nfsd4_ff_proc_layoutget() has mapped these */
+ u32 uid;
+ u32 gid;
struct nfsd4_deviceid deviceid;
stateid_t stateid;
struct nfs_fh fh;
--
2.54.0
next reply other threads:[~2026-07-20 16:37 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-20 16:37 Chuck Lever [this message]
2026-07-21 11:25 ` [PATCH] NFSD: Map flex file layout IDs through the request's user namespace Jeff Layton
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260720163724.810227-1-cel@kernel.org \
--to=cel@kernel.org \
--cc=dai.ngo@oracle.com \
--cc=jlayton@kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=neil@brown.name \
--cc=okorniev@redhat.com \
--cc=sashiko-bot@kernel.org \
--cc=tom@talpey.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox