FAILED: patch "[PATCH] smb: client: reject short SESSION_SETUP responses" failed to apply to 6.6-stable tree

The patch below does not apply to the 6.6-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y
git checkout FETCH_HEAD
git cherry-pick -x 4d8930afa3446041358a7db33e7b23a0bb9459b5
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026110430-unearned-antennae-d4ee@gregkh' --subject-prefix 'PATCH 6.6.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 4d8930afa3446041358a7db33e7b23a0bb9459b5 Mon Sep 17 00:00:00 2001
From: Abdifatah Suruur <[email protected]>
Date: Thu, 22 Oct 2026 06:53:14 +0300
Subject: [PATCH] smb: client: reject short SESSION_SETUP responses

smb2_check_message() accepts frames shorter than struct smb2_pdu when the SMB2 status is non-zero, before the minimum-PDU check added by b4694f269e66d. A 64-byte STATUS_MORE_PROCESSING_REQUIRED SESSION_SETUP response is therefore accepted even though it cannot contain the SecurityBufferOffset and SecurityBufferLength fields.

The parser then reads those fields from stale bytes left in the reused receive buffer by an earlier frame. With a long stale SecurityBufferLength, decode_ntlmssp_challenge() copies up to about 64 KB from beyond the receive object and sends the data back to the server in the NTLMv2 AUTHENTICATE target info.

Reject short MPR SESSION_SETUP responses before the short-error path, and validate SecurityBufferOffset + SecurityBufferLength against the received response length before decoding the challenge.

The ordering bypass is not new: the short-error early return predates b4694f269e66d, and the unvalidated SecurityBufferLength read predates the 2016 raw NTLMSSP split in 166cea4dc3a4, so all supported stable trees are affected. Reproduced on 6.6.87.2; the path is unchanged on mainline.

index 5e5cf92d1eb3..5f13bd7ce73c 100644
@@ -225,6 +225,22 @@ smb2_check_message(char *buf, unsigned int pdu_len, unsigned int len,
 		return 1;
 	}
 
+	/*
+	 * A SESSION_SETUP response that asks for more processing carries a
+	 * security blob and is parsed as a normal reply. Do not let the
+	 * short-error path below accept a frame that cannot hold the fixed
+	 * response fields; the minimum-PDU check later is unreachable once
+	 * the early return has accepted the frame.
+	 */
+	if (command == SMB2_SESSION_SETUP_HE &&
+	    shdr->Status == STATUS_MORE_PROCESSING_REQUIRED &&
+	    len < smb2_min_pdu_len[command]) {
+		cifs_server_dbg(VFS,
+				"%s: session setup response too short: %u < %zu\n",
+				__func__, len, smb2_min_pdu_len[command]);
+		return 1;
+	}
+
 	if (len < pdu_size) {
 		if ((len >= hdr_size)
 		    && (shdr->Status != 0)) {

index 5d892300dabf..06a6ca0dc098 100644
@@ -1857,14 +1857,20 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data)
 		goto out;
 
 	u16 boff = le16_to_cpu(rsp->SecurityBufferOffset);
+	u16 blen = le16_to_cpu(rsp->SecurityBufferLength);
 
 	if (offsetof(struct smb2_sess_setup_rsp, Buffer) != boff) {
 		cifs_dbg(VFS, "Invalid security buffer offset %d\n", boff);
 		rc = smb_EIO1(smb_eio_trace_sess_buf_off, boff);
 		goto out;
 	}
-	rc = decode_ntlmssp_challenge(rsp->Buffer,
-			le16_to_cpu(rsp->SecurityBufferLength), ses);
+	if ((u32)boff + blen > sess_data->iov[0].iov_len) {
+		cifs_dbg(VFS, "%s: invalid security buffer length: %u\n",
+			 __func__, blen);
+		rc = -EINVAL;
+		goto out;
+	}
+	rc = decode_ntlmssp_challenge(rsp->Buffer, blen, ses);
 	if (rc)
 		goto out;