On Fri, Mar 31, 2023 at 06:06:34PM +0200, Arnaud Pouliquen wrote:
> The workqueue may execute late even after remoteproc is stopped or
> stopping, some resources (rpmsg device and endpoint) have been
> released in rproc_stop_subdevices(), then rproc_vq_interrupt()
> accessing these resources will cause kennel dump.
>
> Call trace:
> virtqueue_add_inbuf
> virtqueue_add_inbuf
> rpmsg_recv_single
> rpmsg_recv_done
> vring_interrupt
> stm32_rproc_mb_vq_work
> process_one_work
> worker_thread
> kthread
>
> Suggested-by: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>
> Signed-off-by: Arnaud Pouliquen <arnaud.pouliquen@xxxxxxxxxxx>
>
I had forgotten about this issue - applied.
> ---
> This patch is similar to the issue fixed in
> commit 47e6ab07018e ("remoteproc: imx_dsp_rproc: Add mutex protection for workqueue")
> ---
> drivers/remoteproc/stm32_rproc.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/remoteproc/stm32_rproc.c b/drivers/remoteproc/stm32_rproc.c
> index 7d782ed9e589..f618405cf420 100644
> --- a/drivers/remoteproc/stm32_rproc.c
> +++ b/drivers/remoteproc/stm32_rproc.c
> @@ -287,8 +287,16 @@ static void stm32_rproc_mb_vq_work(struct work_struct *work)
> struct stm32_mbox *mb = container_of(work, struct stm32_mbox, vq_work);
> struct rproc *rproc = dev_get_drvdata(mb->client.dev);
>
> + mutex_lock(&rproc->lock);
> +
> + if (rproc->state != RPROC_RUNNING)
> + goto unlock_mutex;
> +
> if (rproc_vq_interrupt(rproc, mb->vq_id) == IRQ_NONE)
> dev_dbg(&rproc->dev, "no message found in vq%d\n", mb->vq_id);
> +
> +unlock_mutex:
> + mutex_unlock(&rproc->lock);
> }
>
> static void stm32_rproc_mb_callback(struct mbox_client *cl, void *data)
> --
> 2.25.1
>
