From: Ian Bridges <icb@fastmail.org>
To: Namjae Jeon <linkinjeon@kernel.org>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>,
Tom Talpey <tom@talpey.com>,
ChenXiaoSong <chenxiaosong@chenxiaosong.com>,
linux-cifs@vger.kernel.org, linux-kernel@vger.kernel.org,
Aamir Ahmed <elb12345@hotmail.co.uk>,
Ian Bridges <icb@fastmail.org>,
stable@vger.kernel.org
Subject: [PATCH 1/2] ksmbd: let the owner read and change the DACL when no ACE names it
Date: Mon, 2 Nov 2026 00:15:36 -0600 [thread overview]
Message-ID: <20261102061537.830714-2-icb@fastmail.org> (raw)
In-Reply-To: <20261102061537.830714-1-icb@fastmail.org>
smb_check_perm_dacl() refuses the owner when neither its DACL lookup nor
the POSIX ACL fallback matches. The owner then cannot read or replace
the DACL over SMB. MS-DTYP 2.5.3.2 grants the owner READ_CONTROL and
WRITE_DAC unless an OWNER RIGHTS ACE applies.
An empty DACL allows the owner's explicit requests for these rights but
refuses MAXIMUM_ALLOWED. smb2_open() uses that check for the maximal
access reply (MS-SMB2 3.3.5.9.5), so even a READ_CONTROL open fails when
it asks for that reply.
Grant the owner's implicit rights on both paths, for explicit requests
and MAXIMUM_ALLOWED, unless an OWNER RIGHTS ACE applies.
Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3")
Cc: stable@vger.kernel.org
Signed-off-by: Ian Bridges <icb@fastmail.org>
---
Notes:
Reachable on a share with "vfs objects = acl_xattr", for example after
the owner replaces a file's DACL with one that grants only another
user.
The empty-DACL check this patch changes came with de0cffef897c ("ksmbd:
deny access for an empty DACL instead of granting the request"), merged
for v7.4, so an older stable tree needs that commit first. The check
before it refused MAXIMUM_ALLOWED on an empty DACL as well.
Tested on ksmbd-for-next d13da43938d1 with KASAN. Only the owner's
results change. The owner of a file whose only ACE grants another user
can now open it for READ_CONTROL or WRITE_DAC, and is still refused
READ_DATA. On an empty DACL, the owner's MAXIMUM_ALLOWED handle gets
READ_CONTROL and WRITE_DAC when POSIX permits the open. A READ_CONTROL
open with a maximal access context also succeeds. Its handle holds
READ_CONTROL, and its reply reports both rights. Without POSIX read
permission, MAXIMUM_ALLOWED still fails in smb2_open(). An ACE stored
after a DACL whose ACE count is zero still grants nothing.
A reproducer is available on request.
This fix was developed with LLM assistance.
fs/smb/server/smbacl.c | 34 ++++++++++++++++++++++++++++++----
1 file changed, 30 insertions(+), 4 deletions(-)
diff --git a/fs/smb/server/smbacl.c b/fs/smb/server/smbacl.c
index 664f01263611..83974643aa88 100644
--- a/fs/smb/server/smbacl.c
+++ b/fs/smb/server/smbacl.c
@@ -1477,6 +1477,7 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path,
int i, found = 0;
unsigned int access_bits = 0, denied = 0;
struct smb_ace *others_ace = NULL;
+ bool or_ace = false;
struct posix_acl_entry *pa_entry;
unsigned int sid_type = SIDOWNER;
unsigned short ace_size;
@@ -1509,6 +1510,8 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path,
is_owner = uid == from_kuid(&init_user_ns, vfsuid_into_kuid(vfsuid));
if (!pdacl->num_aces) {
+ unsigned int allowed = 0;
+
/*
* An empty (present, zero-ACE) DACL grants no access to
* anyone except the object owner's implicit READ_CONTROL
@@ -1517,9 +1520,14 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path,
* not let trailing bytes after a zero-ACE DACL become an
* implicit grant.
*/
- if (!is_owner ||
- (*pdaccess & ~(FILE_READ_CONTROL_LE | FILE_WRITE_DAC_LE)))
+ if (is_owner)
+ allowed = READ_CONTROL | WRITE_DAC;
+ if (*pdaccess & FILE_MAXIMAL_ACCESS_LE)
+ requested |= allowed;
+ if (!allowed || requested & ~allowed)
rc = -EACCES;
+ else
+ *pdaccess = cpu_to_le32(requested);
goto err_out;
}
@@ -1604,6 +1612,9 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path,
found = 1;
break;
}
+ if (!compare_sids(&sid_owner_rights, &ace->sid) &&
+ !(ace->flags & INHERIT_ONLY_ACE))
+ or_ace = true;
if (!compare_sids(&sid_everyone, &ace->sid) ||
!compare_sids(&sid_authusers, &ace->sid))
others_ace = ace;
@@ -1645,8 +1656,23 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path,
if (others_ace) {
ace = others_ace;
} else {
- ksmbd_debug(SMB, "Can't find corresponding sid\n");
- rc = -EACCES;
+ unsigned int allowed = 0;
+
+ /*
+ * No ACE names the owner, who may still read and
+ * change the DACL unless OWNER RIGHTS applies
+ * (MS-DTYP 2.5.3.2).
+ */
+ if (is_owner && !or_ace)
+ allowed = READ_CONTROL | WRITE_DAC;
+ if (*pdaccess & FILE_MAXIMAL_ACCESS_LE)
+ requested |= allowed;
+ if (!allowed || requested & ~allowed) {
+ ksmbd_debug(SMB, "Can't find corresponding sid\n");
+ rc = -EACCES;
+ goto err_out;
+ }
+ *pdaccess = cpu_to_le32(requested);
goto err_out;
}
}
--
2.47.3
next prev parent reply other threads:[~2026-11-02 6:16 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-02 6:15 [PATCH 0/2] ksmbd: fix DACL checks for owners and groups Ian Bridges
2026-11-02 6:15 ` Ian Bridges [this message]
2026-11-02 9:25 ` [PATCH 1/2] ksmbd: let the owner read and change the DACL when no ACE names it Namjae Jeon
2026-11-02 6:15 ` [PATCH 2/2] ksmbd: grant a caller what its groups and OWNER RIGHTS ACEs allow Ian Bridges
2026-11-02 9:15 ` Namjae Jeon
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=20261102061537.830714-2-icb@fastmail.org \
--to=icb@fastmail.org \
--cc=chenxiaosong@chenxiaosong.com \
--cc=elb12345@hotmail.co.uk \
--cc=linkinjeon@kernel.org \
--cc=linux-cifs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=senozhatsky@chromium.org \
--cc=stable@vger.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
all inboxes | Powered by JetHome®