On Mon, 8 May 2023, Tony Lindgren wrote:
> If serial8250_register_8250_port() fails, the SoC can hang as the
> deferred PMQoS work will still run as is not flushed and removed.
>
> Fixes: 61929cf0169d ("tty: serial: Add 8250-core based omap driver")
> Signed-off-by: Tony Lindgren <tony@xxxxxxxxxxx>
> ---
> drivers/tty/serial/8250/8250_omap.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/tty/serial/8250/8250_omap.c b/drivers/tty/serial/8250/8250_omap.c
> --- a/drivers/tty/serial/8250/8250_omap.c
> +++ b/drivers/tty/serial/8250/8250_omap.c
> @@ -1532,7 +1532,9 @@ static int omap8250_probe(struct platform_device *pdev)
> err:
> pm_runtime_dont_use_autosuspend(&pdev->dev);
> pm_runtime_put_sync(&pdev->dev);
> + flush_work(&priv->qos_work);
> pm_runtime_disable(&pdev->dev);
> + cpu_latency_qos_remove_request(&priv->pm_qos_request);
> return ret;
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@xxxxxxxxxxxxxxx>
An unrelated comment to the patch itself, there seems to be somewhat
handwavy and possibly wrong calculation for the pm qos latency. First of
all, I think it would want something based on port->frame_time, and I'm
far from convinced that 64 is right as it matches FIFO size which doesn't
feel correct for a wakeup related time.
--
i.
