On Tue, 15 Dec 2020 at 02:46, Zev Weiss <zev@xxxxxxxxxxxxxxxxx> wrote:
>
> This device seems to have a propensity for asserting interrupts that
> aren't enabled -- in addition to the CAPTURE_COMPLETE and FRAME_COMPLETE
> interrupts squashed in commit 65d270acb2d662c3346793663ac3a759eb4491b8,
> COMP_READY has also been observed. Adding a message diagnosing what
> happened in the event of unhandled interrupt status bits should
> hopefully make the debugging process simpler for any others that pop up
> in the future.
Ryan, is this a known issue with the video engine hardware?
>
> Signed-off-by: Zev Weiss <zev@xxxxxxxxxxxxxxxxx>
> ---
> drivers/media/platform/aspeed-video.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/media/platform/aspeed-video.c b/drivers/media/platform/aspeed-video.c
> index 7d98db1d9b52..eb02043532e3 100644
> --- a/drivers/media/platform/aspeed-video.c
> +++ b/drivers/media/platform/aspeed-video.c
> @@ -562,6 +562,7 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
> {
> struct aspeed_video *video = arg;
> u32 sts = aspeed_video_read(video, VE_INTERRUPT_STATUS);
> + u32 orig_sts = sts;
>
> /*
> * Resolution changed or signal was lost; reset the engine and
> @@ -639,6 +640,10 @@ static irqreturn_t aspeed_video_irq(int irq, void *arg)
> if (sts & VE_INTERRUPT_FRAME_COMPLETE)
> sts &= ~VE_INTERRUPT_FRAME_COMPLETE;
>
> + if (sts)
> + dev_err_ratelimited(video->dev, "unexpected interrupt asserted:"
> + " sts=%08x, orig_sts=%08x", sts, orig_sts);
Do you want to do this before clearing the FRAME and CAPTURE bits?
> +
> return sts ? IRQ_NONE : IRQ_HANDLED;
> }
>
> --
> 2.29.2
>
