[PATCH v3] vhost-scsi: do not count PI bytes in the data length

[PATCH v3] vhost-scsi: do not count PI bytes in the data length

[Jia Jia]

vhost_scsi_handle_vq() moves pi_bytesout or pi_bytesin into the
protection sgl and subtracts those bytes from exp_data_len.  The length
passed to TCM is still exp_data_len + prot_bytes.

prot_bytes comes from virtio_scsi_cmd_req_pi.  With
VIRTIO_SCSI_F_T10_PI negotiated, vhost reads that header and does not
check that the CDB is a READ or WRITE.  sbc_check_prot() replaces
data_length from the CDB only when the protect field is non-zero.
target_cmd_size_check() leaves a write's data_length unchanged when the
CDB size is smaller, so a command that does not get that replacement
keeps a data_length larger than the data sgl.

The KASAN report is a MODE SELECT whose header had pi_bytesout set.
The data sgl held only the parameter list.  spc_emulate_modeselect()
walked data_length and read the next physical page.

Pass the data sgl length.  Protected READ and WRITE still take their
length from the CDB in sbc_check_prot().

KASAN reports:

  BUG: KASAN: use-after-free in spc_emulate_modeselect+0x145/0x380 [target_core_mod]
  Read of size 1

  spc_emulate_modeselect
  __target_execute_cmd
  target_execute_cmd
  vhost_scsi_write_pending
  transport_generic_new_cmd
  __target_submit
  target_queued_submit_work
  process_one_work
  worker_thread
  kthread
  ret_from_fork
  ret_from_fork_asm

Fixes: 9f977ef7b671 ("vhost-scsi: Include prot_bytes into expected data transfer length")
Signed-off-by: Jia Jia <physicalmtea@gmail.com>
---
v3:
  - Rewrite the inline comment to distinguish the data and protection
    SGLs and explain why TCM receives only the data SGL length.
  - Note the related target-core patches for PI tuple accesses at
    protection-SGL boundaries:
    https://lore.kernel.org/all/20261006093338.27342-5-physicalmtea@gmail.com/
    https://lore.kernel.org/all/20261006093338.27342-6-physicalmtea@gmail.com/
    https://lore.kernel.org/all/20261006093338.27342-7-physicalmtea@gmail.com/

v2:
  - Drop the claim that target used to subtract prot_length.
    9f977ef7b671 cites 14ef9200, which is not in the tree
  - prot_bytes comes from virtio_scsi_cmd_req_pi

 drivers/vhost/scsi.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 4f8c0260bc9e..4bd29c2ceed9 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -1534,9 +1534,17 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 		 * vhost_scsi_queue_data_in() and vhost_scsi_queue_status()
 		 */
 		cmd->tvc_vq_desc = vc.head;
+		/*
+		 * vhost_scsi_mapal() maps the PI bytes into cmd->prot_table.
+		 * It maps the remaining exp_data_len bytes into cmd->table,
+		 * which is the data SGL passed to the target core.  Its
+		 * expected length must therefore exclude prot_bytes.
+		 *
+		 * sbc_check_prot() derives the data length for protected READ
+		 * and WRITE commands from the CDB.
+		 */
 		vhost_scsi_target_queue_cmd(nexus, cmd, cdb, lun, task_attr,
-					    data_direction,
-					    exp_data_len + prot_bytes);
+					    data_direction, exp_data_len);
 		ret = 0;
 err:
 		/*
-- 
2.53.0

Re: [PATCH v3] vhost-scsi: do not count PI bytes in the data length

[sashiko-bot]

> vhost_scsi_handle_vq() moves pi_bytesout or pi_bytesin into the
> protection sgl and subtracts those bytes from exp_data_len.  The length
> passed to TCM is still exp_data_len + prot_bytes.
> 
> prot_bytes comes from virtio_scsi_cmd_req_pi.  With
> VIRTIO_SCSI_F_T10_PI negotiated, vhost reads that header and does not
> check that the CDB is a READ or WRITE.  sbc_check_prot() replaces
> data_length from the CDB only when the protect field is non-zero.
> target_cmd_size_check() leaves a write's data_length unchanged when the
> CDB size is smaller, so a command that does not get that replacement
> keeps a data_length larger than the data sgl.
> 
> The KASAN report is a MODE SELECT whose header had pi_bytesout set.
> The data sgl held only the parameter list.  spc_emulate_modeselect()
> walked data_length and read the next physical page.
> [ ... ]
>   ret_from_fork
>   ret_from_fork_asm
> 
> Fixes: 9f977ef7b671 ("vhost-scsi: Include prot_bytes into expected data transfer length")
> Signed-off-by: Jia Jia <physicalmtea@gmail.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261012103954.136698-1-physicalmtea@gmail.com?part=1