Linux kernel -stable discussions
 help / color / mirror / Atom feed
From: Salvatore Bonaccorso <carnil@debian.org>
To: Greg KH <gregkh@linuxfoundation.org>
Cc: manizada <manizada@pm.me>,
	"stable@vger.kernel.org" <stable@vger.kernel.org>
Subject: Re: Please apply 3da1fdf4efbc to stable
Date: Wed, 27 May 2026 14:06:29 +0200	[thread overview]
Message-ID: <ahbeRcvu5qN55gaE@eldamar.lan> (raw)
In-Reply-To: <2026052742-discharge-smudge-6453@gregkh>

Hi Greg,

Not Asmin here, but saw the following.

On Wed, May 27, 2026 at 10:08:03AM +0200, Greg KH wrote:
> On Tue, May 26, 2026 at 05:08:43PM +0000, manizada wrote:
> > Hi stable team,
> > 
> > Please apply the following upstream commit to the supported stable trees:
> > 
> >   3da1fdf4efbc490041eb4f836bf596201203f8f2
> >   smb: client: reject userspace cifs.spnego descriptions
> > 
> > Reason:
> >   cifs.spnego descriptions contain authority-bearing fields consumed by
> >   cifs.upcall. This commit prevents userspace from creating trusted
> >   cifs.spnego descriptions via request_key(2)/add_key(2).
> > 
> > Requested branches:
> >   Please apply to all currently supported stable/LTS branches where it is
> >   applicable, including 7.0.y, 6.18.y, 6.12.y, 6.6.y, 6.1.y, 5.15.y, and
> >   5.10.y.
> 
> This does not apply to the 5.15.y or 5.10.y tree, please provide a
> backported version that can apply there.

There was 38c8a9a52082 ("smb: move client and server files to common
directory fs/smb") moving the files around (and which got as well
backported to the 6.1.y series). So the following should do the trick.

Regards,
Salvatore

From f89a8b4dfcdb7967b2f306b5629f7e5b92f74a26 Mon Sep 17 00:00:00 2001
From: Asim Viladi Oglu Manizada <manizada@pm.me>
Date: Sat, 16 May 2026 21:15:39 +0000
Subject: [PATCH] smb: client: reject userspace cifs.spnego descriptions

commit 3da1fdf4efbc490041eb4f836bf596201203f8f2 upstream.

cifs.spnego key descriptions contain authority-bearing fields such as
pid, uid, creduid, and upcall_target that cifs.upcall treats as
kernel-originating inputs. However, userspace can also create keys of
this type through request_key(2) or add_key(2), allowing those fields to
be supplied without CIFS origin.

Only accept cifs.spnego descriptions while CIFS is using its private
spnego_cred to request the key.

Fixes: f1d662a7d5e5 ("[CIFS] Add upcall files for cifs to use spnego/kerberos")
Assisted-by: avom-custom-harness:gpt-5.5-qwen3.6-mod-mix
Reviewed-by: David Howells <dhowells@redhat.com>
Signed-off-by: Asim Viladi Oglu Manizada <manizada@pm.me>
Signed-off-by: Steve French <stfrench@microsoft.com>
[Salvatore Bonaccorso: Apply changes to fs/cifs/cifs_spnego.c instead of
fs/smb/client/cifs_spnego.c before 38c8a9a52082 ("smb: move client and server
files to common directory fs/smb") in v6.4-rc1 and backported to v6.1.36]
Signed-off-by: Salvatore Bonaccorso <carnil@debian.org>
---
 fs/cifs/cifs_spnego.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c
index 4f9d08ac9dde..5b7614451033 100644
--- a/fs/cifs/cifs_spnego.c
+++ b/fs/cifs/cifs_spnego.c
@@ -20,6 +20,7 @@
  */
 
 #include <linux/list.h>
+#include <linux/cred.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <keys/user-type.h>
@@ -58,12 +59,27 @@ cifs_spnego_key_destroy(struct key *key)
 	kfree(key->payload.data[0]);
 }
 
+static int
+cifs_spnego_key_vet_description(const char *description)
+{
+	/*
+	 * cifs.spnego descriptions are authority-bearing inputs to cifs.upcall.
+	 * They are only valid when produced by CIFS while using the private
+	 * spnego_cred installed below.  Do not let userspace create this type
+	 * of key through request_key(2)/add_key(2), since the helper treats
+	 * pid/uid/creduid/upcall_target as kernel-originating fields.
+	 */
+	if (current_cred() != spnego_cred)
+		return -EPERM;
+	return 0;
+}
 
 /*
  * keytype for CIFS spnego keys
  */
 struct key_type cifs_spnego_key_type = {
 	.name		= "cifs.spnego",
+	.vet_description = cifs_spnego_key_vet_description,
 	.instantiate	= cifs_spnego_key_instantiate,
 	.destroy	= cifs_spnego_key_destroy,
 	.describe	= user_describe,
-- 
2.53.0

  reply	other threads:[~2026-05-27 12:06 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-26 17:08 Please apply 3da1fdf4efbc to stable manizada
2026-05-27  8:08 ` Greg KH
2026-05-27 12:06   ` Salvatore Bonaccorso [this message]
2026-05-27 12:16     ` Greg KH

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=ahbeRcvu5qN55gaE@eldamar.lan \
    --to=carnil@debian.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=manizada@pm.me \
    --cc=stable@vger.kernel.org \
    /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