Re: [PATCH v3 4/4] pwm: imx27: return error, if clean PWM setup fails

From: Uwe Kleine-König
Date: Wed Sep 06 2023 - 11:44:18 EST


On Tue, Aug 15, 2023 at 12:43:32PM +0200, Leif Middelschulte wrote:
> Instead of issuing a warning, return an error (as adviced[0]), if the
> FIFO cannot be cleanly set up.
>
> [0] https://lore.kernel.org/lkml/20230310174517.rb7xxrougkse2lrc@xxxxxxxxxxxxxx/T/#ec9560c1f613d9c0d7b77d72ad9051768812f80db
>
> Signed-off-by: Leif Middelschulte <Leif.Middelschulte@xxxxxxxxx>
> ---
> drivers/pwm/pwm-imx27.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/pwm/pwm-imx27.c b/drivers/pwm/pwm-imx27.c
> index 9673e809d212..5fd6d34a7722 100644
> --- a/drivers/pwm/pwm-imx27.c
> +++ b/drivers/pwm/pwm-imx27.c
> @@ -133,14 +133,15 @@ static int pwm_imx27_wait_fifo_empty(struct pwm_chip *chip,
> sr = readl(imx->mmio_base + MX3_PWMSR);
> fifoav = FIELD_GET(MX3_PWMSR_FIFOAV, sr);
> if (fifoav == MX3_PWMSR_FIFOAV_EMPTY)
> - return;
> + return 0;
> /* if the FIFO value does not decrease, there is another problem */
> if (previous_fifoav == fifoav)
> break;
> previous_fifoav = fifoav;
> msleep(period_ms);
> }
> - dev_warn(dev, "FIFO has been refilled concurrently\n");
> +
> + return -EAGAIN;
> }
>
> static int pwm_imx27_get_state(struct pwm_chip *chip,
> @@ -208,7 +209,9 @@ static int pwm_imx27_get_state(struct pwm_chip *chip,
> * been supplied and a period expired between the call of the wait
> * function and the subsequent readl.
> */
> - pwm_imx27_wait_fifo_empty(chip, pwm);
> + ret = pwm_imx27_wait_fifo_empty(chip, pwm);
> + if (ret)
> + return ret;
> val = readl(imx->mmio_base + MX3_PWMSAR);
> } else {
> val = imx->duty_cycle;

I'd squash this into the patch introducing pwm_imx27_wait_fifo_empty,
too.

Best regards
Uwe

--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | https://www.pengutronix.de/ |

Attachment: signature.asc
Description: PGP signature