Linux CIFS filesystem development
 help / color / mirror / Atom feed
* [PATCH] smb: client: fix SMB1 TRANS2 multi-response truncation in SendReceive()
@ 2026-08-04 19:21 Frank Sorenson
  0 siblings, 0 replies; only message in thread
From: Frank Sorenson @ 2026-08-04 19:21 UTC (permalink / raw)
  To: linux-cifs, stfrench; +Cc: stable, Paulo Alcantara

When a TRANS2 response is split across multiple secondary packets,
coalesce_t2() assembles the payload into the large response buffer.
Two bugs cause SendReceive() to copy only a small fraction of the
assembled buffer into the caller's output buffer.

This manifests when listing a large directory on an SMB1 share
(observed against Windows XP); the first getdents returns only
partial results, and subsequent getdents returns EINVAL.

Bug 1: coalesce_t2() computes the coalesced size as:

  *pdu_len += total_in_src;

cifs_demultiplex_thread() resets *pdu_len to each secondary's own
pdu_length before calling coalesce_t2(), so this accumulates from
the wrong baseline on every secondary after the first.

Bug 2: after reassembly, cifs_demultiplex_thread() sets
mid->resp_buf_size to the final secondary's raw packet size.  This
value is later used as the memcpy length, so only a portion of the
coalesced response is copied.

Fix both by replacing the stale *pdu_len arithmetic with
smbCalcSize(), which reads the BCC field that coalesce_t2()
maintains correctly throughout reassembly.

Fixes: 83bfbd0bb902 ("cifs: Remove the RFC1002 header from smb_hdr")
Cc: stable@vger.kernel.org
Signed-off-by: Frank Sorenson <sorenson@redhat.com>
Reviewed-by: Paulo Alcantara <pc@manguebit.org>
---
 fs/smb/client/smb1transport.c | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/fs/smb/client/smb1transport.c b/fs/smb/client/smb1transport.c
index 53abb29fe71b..966f2cf83a51 100644
--- a/fs/smb/client/smb1transport.c
+++ b/fs/smb/client/smb1transport.c
@@ -260,9 +260,23 @@ SendReceive(const unsigned int xid, struct cifs_ses *ses,
 		goto out;
 
 	if (out_buf) {
-		*pbytes_returned = resp_iov.iov_len;
-		if (resp_iov.iov_len)
-			memcpy(out_buf, resp_iov.iov_base, resp_iov.iov_len);
+		/* Use smbCalcSize() for both single- and multi-part T2 responses,
+		 * both here and in coalesce_t2().
+		 */
+		unsigned int copy_len;
+		if (WARN_ON_ONCE(!resp_iov.iov_base)) {
+			rc = -EIO;
+			goto out;
+		}
+		copy_len = smbCalcSize(resp_iov.iov_base);
+		if (copy_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) {
+			cifs_dbg(VFS, "response size %u exceeds buffer\n",
+				 copy_len);
+			rc = -ENOBUFS;
+			goto out;
+		}
+		*pbytes_returned = copy_len;
+		memcpy(out_buf, resp_iov.iov_base, copy_len);
 	}
 
 out:
@@ -386,11 +400,13 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr, unsigned int *pdu_len)
 	}
 	put_bcc(byte_count, target_hdr);
 
-	byte_count = *pdu_len;
-	byte_count += total_in_src;
+	/* use smbCalcSize() rather than *pdu_len: the demux loop resets
+	 * *pdu_len to each secondary's pdu_length, making it unreliable.
+	 */
+	byte_count = smbCalcSize(target_hdr);
 	/* don't allow buffer to overflow */
 	if (byte_count > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) {
-		cifs_dbg(FYI, "coalesced BCC exceeds buffer size (%u)\n",
+		cifs_dbg(FYI, "coalesced size exceeds buffer size (%u)\n",
 			 byte_count);
 		return -ENOBUFS;
 	}
-- 
2.55.0


^ permalink raw reply related	[flat|nested] only message in thread

only message in thread, other threads:[~2026-08-04 19:21 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-04 19:21 [PATCH] smb: client: fix SMB1 TRANS2 multi-response truncation in SendReceive() Frank Sorenson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox