Hi Michael,
On Fri, 9 May 2025 at 14:29, Michael Dege <michael.dege@xxxxxxxxxxx> wrote:
> There is a slight difference in SerDes hardware behavior between each
> version after resetting. This step is to ensure the stable condition of
> initialization, especially for R-Car S4 v1.1 .
>
> Signed-off-by: Michael Dege <michael.dege@xxxxxxxxxxx>
Thanks for your patch!
> --- a/drivers/phy/renesas/renesas-ether-serdes.c
> +++ b/drivers/phy/renesas/renesas-ether-serdes.c
> @@ -265,7 +265,12 @@ static int renesas_eth_serdes_hw_init(struct renesas_eth_serdes_channel *channel
>
> reset_control_reset(dd->reset);
>
> - usleep_range(1000, 2000);
> + /* There is a slight difference in SerDes hardware behavior between
> + * each version after resetting. This step is to ensure the stable
> + * condition of initialization, especially for R-Car S4 v1.1.
> + */
> + mdelay(1);
> + iowrite32(0, dd->addr + RENESAS_ETH_SERDES_LOCAL_OFFSET);
Which of these two changes fixes the issue?
I suspect/hope the usleep_range() can stay, as sleeping is preferred
over busy-waiting.
>
> ret = renesas_eth_serdes_common_init_ram(dd);
> if (ret)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
