Javier Martinez Canillas <[email protected]> writes:
> Report these failures only once, instead of keep logging the warnings for
> the same condition every time that a SDIO read or write is attempted. This
> behaviour is spammy and unnecessarily pollutes the kernel log buffer.
Removing error messages is not usually a good idea, it would be much
better to fix the root cause.
> For example, on an AM625 BeaglePlay board where accessing a SDIO WiFi chip
> fails with an -110 error:
>
> $ dmesg | grep "sdio write\|read failed (-110)" | wc -l
> 39
-110 is -ETIMEDOUT. Why is it timing out?
> Signed-off-by: Javier Martinez Canillas <[email protected]>
> ---
>
> drivers/net/wireless/ti/wlcore/sdio.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/wireless/ti/wlcore/sdio.c b/drivers/net/wireless/ti/wlcore/sdio.c
> index eb5482ed76ae..47ecf33a0fbe 100644
> --- a/drivers/net/wireless/ti/wlcore/sdio.c
> +++ b/drivers/net/wireless/ti/wlcore/sdio.c
> @@ -75,8 +75,8 @@ static int __must_check wl12xx_sdio_raw_read(struct device *child, int addr,
>
> sdio_release_host(func);
>
> - if (WARN_ON(ret))
> - dev_err(child->parent, "sdio read failed (%d)\n", ret);
> + if (WARN_ON_ONCE(ret))
> + dev_err_once(child->parent, "sdio read failed (%d)\n", ret);
WARN_ON() feels excessive here, maybe remove that entirely? But
dev_err_ratelimited() feels more approriate than printing the error just
once.
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches