Re: [PATCH] remoteproc: imx_rproc: iterate all notifiyids in rx callback

[Mathieu Poirier]

On Sun, Jun 25, 2023 at 08:35:14PM +0800, Peng Fan (OSS) wrote:
> From: Peng Fan <peng.fan@xxxxxxx>
> 
> The current code has an assumption that there is one tx and one rx
> vring, but it is not always true. There maybe more vrings. So iterate
> all notifyids to not miss any vring events.

Can you be more specific on the use case where more than 2 virqueues are
allocated?  The remoteproc core can handle more than 2 but right now the only
configuration I see doesn't support more than that.

> 
> Signed-off-by: Peng Fan <peng.fan@xxxxxxx>
> ---
>  drivers/remoteproc/imx_rproc.c | 14 ++++++++++++--
>  1 file changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
> index f9874fc5a80f..e3f40d0e9f3d 100644
> --- a/drivers/remoteproc/imx_rproc.c
> +++ b/drivers/remoteproc/imx_rproc.c
> @@ -725,13 +725,23 @@ static int imx_rproc_addr_init(struct imx_rproc *priv,
>  	return 0;
>  }
>  
> +static int imx_rproc_notified_idr_cb(int id, void *ptr, void *data)
> +{
> +	struct rproc *rproc = data;
> +
> +	if (rproc_vq_interrupt(rproc, id) == IRQ_NONE)
> +		dev_dbg(&rproc->dev, "no message in vqid: %d\n", id);
> +
> +	return 0;
> +}
> +
>  static void imx_rproc_vq_work(struct work_struct *work)
>  {
>  	struct imx_rproc *priv = container_of(work, struct imx_rproc,
>  					      rproc_work);
> +	struct rproc *rproc = priv->rproc;
>  
> -	rproc_vq_interrupt(priv->rproc, 0);
> -	rproc_vq_interrupt(priv->rproc, 1);
> +	idr_for_each(&rproc->notifyids, imx_rproc_notified_idr_cb, rproc);
>  }
>  
>  static void imx_rproc_rx_callback(struct mbox_client *cl, void *msg)
> -- 
> 2.37.1
>