netdev - Re: [PATCH RFC net-next] net: lan966x: make PHY reset support optional

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 31 Mar 2022 10:30:38 +0200
From:   Horatiu Vultur <horatiu.vultur@...rochip.com>
To:     Michael Walle <michael@...le.cc>
CC:     Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
        "Philipp Zabel" <p.zabel@...gutronix.de>,
        <UNGLinuxDriver@...rochip.com>, <netdev@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH RFC net-next] net: lan966x: make PHY reset support
 optional

The 03/30/2022 13:02, Michael Walle wrote:

Hi Michael,

> 
> The PHY subsystem as well as the MIIM mdio driver (in case of the
> integrated PHYs) will already take care of the resets of any external
> and internal PHY. There is no need for this reset anymore, so mark it
> optionally to be backwards compatible.
> 
> Signed-off-by: Michael Walle <michael@...le.cc>
> ---
> 
> Horatiu, what do you think, should it be removed altogether? 

I think it is OK to remove it altogether. If you get both [1] and [2]
in.

> There is
> no user for that in mainline and I don't know about downstream but the
> reset driver doesn't really work (as it also resets the GPIO/SGPIO)

Yes because I didn't manage to send yet those patches. But if your get
yours in that is fine for me.
My problem was, if after the probe of the MDIO controller it was probed
the SGPIO then the PHYs will be in reset because the SGPIO was resetting
the swich. But you put the reset of the swich on the pinctrl which will
be probed before the MDIO, so that should be fine.

> and conceptionally the property is on the wrong DT node. All of the
> drawbacks should have been addressed by my patches for the miim [1]
> and the pinctrl driver [2].
> 
> [1] https://lore.kernel.org/netdev/20220318201324.1647416-1-michael@walle.cc/
> [2] https://lore.kernel.org/linux-gpio/20220313154640.63813-1-michael@walle.cc/
> 
>  drivers/net/ethernet/microchip/lan966x/lan966x_main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
> index 1f8c67f0261b..0765064d2845 100644
> --- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
> +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c
> @@ -916,7 +916,7 @@ static int lan966x_reset_switch(struct lan966x *lan966x)
>                 return dev_err_probe(lan966x->dev, PTR_ERR(switch_reset),
>                                      "Could not obtain switch reset");
> 
> -       phy_reset = devm_reset_control_get_shared(lan966x->dev, "phy");
> +       phy_reset = devm_reset_control_get_optional_shared(lan966x->dev, "phy");
>         if (IS_ERR(phy_reset))
>                 return dev_err_probe(lan966x->dev, PTR_ERR(phy_reset),
>                                      "Could not obtain phy reset\n");
> --
> 2.30.2
> 

-- 
/Horatiu