Re: [PATCH 1/7] cxl/mbox: Allow for IRQ_NONE case in the isr

[Dave Jiang <dave.jiang@intel.com>]

On 6/12/23 11:10, Davidlohr Bueso wrote:
> For cases when the mailbox background operation is not complete,
> do not "handle" the interrupt, as it was not from this device.
> And furthermore there are no racy scenarios such as the hw being
> out of sync with the driver and starting a new background op
> behind its back.
>
> Reported-by: Jonathan Cameron <Jonathan.Cameron@Huawei.com>
> Fixes: ccadf1310fb (cxl/mbox: Add background cmd handling machinery)
> Signed-off-by: Davidlohr Bueso <dave@stgolabs.net>


Reviewed-by: Dave Jiang <dave.jiang@intel.com>


> ---
>   drivers/cxl/pci.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
> index a78e40e6d0e0..4b2575502f49 100644
> --- a/drivers/cxl/pci.c
> +++ b/drivers/cxl/pci.c
> @@ -118,9 +118,11 @@ static irqreturn_t cxl_pci_mbox_irq(int irq, void *id)
>   	struct cxl_dev_id *dev_id = id;
>   	struct cxl_dev_state *cxlds = dev_id->cxlds;
>   
> +	if (!cxl_mbox_background_complete(cxlds))
> +		return IRQ_NONE;
> +
>   	/* short-circuit the wait in __cxl_pci_mbox_send_cmd() */
> -	if (cxl_mbox_background_complete(cxlds))
> -		rcuwait_wake_up(&cxlds->mbox_wait);
> +	rcuwait_wake_up(&cxlds->mbox_wait);
>   
>   	return IRQ_HANDLED;
>   }