[PATCH AUTOSEL 6.6 13/21] vhost_task: fix vhost_task_create() documentation

[PATCH AUTOSEL 6.6 13/21] vhost_task: fix vhost_task_create() documentation

[Sasha Levin]

From: Stefano Garzarella <sgarzare@redhat.com>

[ Upstream commit fec0abf52609c20279243699d08b660c142ce0aa ]

Commit cb380909ae3b ("vhost: return task creation error instead of NULL")
changed the return value of vhost_task_create(), but did not update the
documentation.

Reflect the change in the documentation: on an error, vhost_task_create()
returns an ERR_PTR() and no longer NULL.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20250327124435.142831-1-sgarzare@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/vhost_task.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/vhost_task.c b/kernel/vhost_task.c
index 8800f5acc0071..0e4455742190c 100644
--- a/kernel/vhost_task.c
+++ b/kernel/vhost_task.c
@@ -111,7 +111,7 @@ EXPORT_SYMBOL_GPL(vhost_task_stop);
  * @arg: data to be passed to fn and handled_kill
  * @name: the thread's name
  *
- * This returns a specialized task for use by the vhost layer or NULL on
+ * This returns a specialized task for use by the vhost layer or ERR_PTR() on
  * failure. The returned task is inactive, and the caller must fire it up
  * through vhost_task_start().
  */
-- 
2.39.5

Re: [PATCH AUTOSEL 6.6 13/21] vhost_task: fix vhost_task_create() documentation

[Stefano Garzarella]

On Tue, Apr 29, 2025 at 07:52:25PM -0400, Sasha Levin wrote:
>From: Stefano Garzarella <sgarzare@redhat.com>
>
>[ Upstream commit fec0abf52609c20279243699d08b660c142ce0aa ]
>
>Commit cb380909ae3b ("vhost: return task creation error instead of NULL")
>changed the return value of vhost_task_create(), but did not update the
>documentation.
>
>Reflect the change in the documentation: on an error, vhost_task_create()
>returns an ERR_PTR() and no longer NULL.
>
>Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
>Message-Id: <20250327124435.142831-1-sgarzare@redhat.com>
>Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
>Signed-off-by: Sasha Levin <sashal@kernel.org>
>---
> kernel/vhost_task.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)

It looks like 6.6 doesn't contain commit cb380909ae3b ("vhost: return 
task creation error instead of NULL") so I think we should not backport 
this.

BTW, this is just a fix for a comment, so not a big issue if we backport 
or not.

Thanks,
Stefano

>
>diff --git a/kernel/vhost_task.c b/kernel/vhost_task.c
>index 8800f5acc0071..0e4455742190c 100644
>--- a/kernel/vhost_task.c
>+++ b/kernel/vhost_task.c
>@@ -111,7 +111,7 @@ EXPORT_SYMBOL_GPL(vhost_task_stop);
>  * @arg: data to be passed to fn and handled_kill
>  * @name: the thread's name
>  *
>- * This returns a specialized task for use by the vhost layer or NULL on
>+ * This returns a specialized task for use by the vhost layer or ERR_PTR() on
>  * failure. The returned task is inactive, and the caller must fire it up
>  * through vhost_task_start().
>  */
>-- 
>2.39.5
>

[PATCH AUTOSEL 6.6 14/21] vhost-scsi: protect vq->log_used with vq->mutex

[Sasha Levin]

From: Dongli Zhang <dongli.zhang@oracle.com>

[ Upstream commit f591cf9fce724e5075cc67488c43c6e39e8cbe27 ]

The vhost-scsi completion path may access vq->log_base when vq->log_used is
already set to false.

    vhost-thread                       QEMU-thread

vhost_scsi_complete_cmd_work()
-> vhost_add_used()
   -> vhost_add_used_n()
      if (unlikely(vq->log_used))
                                      QEMU disables vq->log_used
                                      via VHOST_SET_VRING_ADDR.
                                      mutex_lock(&vq->mutex);
                                      vq->log_used = false now!
                                      mutex_unlock(&vq->mutex);

				      QEMU gfree(vq->log_base)
        log_used()
        -> log_write(vq->log_base)

Assuming the VMM is QEMU. The vq->log_base is from QEMU userpace and can be
reclaimed via gfree(). As a result, this causes invalid memory writes to
QEMU userspace.

The control queue path has the same issue.

Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Mike Christie <michael.christie@oracle.com>
Message-Id: <20250403063028.16045-2-dongli.zhang@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/vhost/scsi.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 8d8a22504d71f..724dd69c86489 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -560,6 +560,9 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 	int ret;
 
 	llnode = llist_del_all(&svq->completion_list);
+
+	mutex_lock(&svq->vq.mutex);
+
 	llist_for_each_entry_safe(cmd, t, llnode, tvc_completion_list) {
 		se_cmd = &cmd->tvc_se_cmd;
 
@@ -593,6 +596,8 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 		vhost_scsi_release_cmd_res(se_cmd);
 	}
 
+	mutex_unlock(&svq->vq.mutex);
+
 	if (signal)
 		vhost_signal(&svq->vs->dev, &svq->vq);
 }
@@ -1301,8 +1306,11 @@ static void vhost_scsi_tmf_resp_work(struct vhost_work *work)
 		resp_code = VIRTIO_SCSI_S_FUNCTION_REJECTED;
 	}
 
+	mutex_lock(&tmf->svq->vq.mutex);
 	vhost_scsi_send_tmf_resp(tmf->vhost, &tmf->svq->vq, tmf->in_iovs,
 				 tmf->vq_desc, &tmf->resp_iov, resp_code);
+	mutex_unlock(&tmf->svq->vq.mutex);
+
 	vhost_scsi_release_tmf_res(tmf);
 }
 
-- 
2.39.5