linux-kernel - Re: [PATCH 1/2] net: mdio: add reset deassertion for Aspeed MDIO

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

Date:   Mon, 21 Mar 2022 09:14:38 +0100
From:   Philipp Zabel <p.zabel@...gutronix.de>
To:     Dylan Hung <dylan_hung@...eedtech.com>, robh+dt@...nel.org,
        joel@....id.au, andrew@...id.au, andrew@...n.ch,
        hkallweit1@...il.com, linux@...linux.org.uk, davem@...emloft.net,
        kuba@...nel.org, pabeni@...hat.com, devicetree@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org,
        linux-aspeed@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org
Cc:     BMC-SW@...eedtech.com, stable@...r.kernel.org
Subject: Re: [PATCH 1/2] net: mdio: add reset deassertion for Aspeed MDIO

Hi Dylan,

On Mo, 2022-03-21 at 15:01 +0800, Dylan Hung wrote:
> Add reset deassertion for Aspeed MDIO.  There are 4 MDIO controllers
> embedded in Aspeed AST2600 SOC and share one reset control register
> SCU50[3]. So devm_reset_control_get_shared is used in this change.
> 
> Signed-off-by: Dylan Hung <dylan_hung@...eedtech.com>
> Cc: stable@...r.kernel.org
> ---
>  drivers/net/mdio/mdio-aspeed.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/net/mdio/mdio-aspeed.c b/drivers/net/mdio/mdio-
> aspeed.c
> index e2273588c75b..8ac262a12d13 100644
> --- a/drivers/net/mdio/mdio-aspeed.c
> +++ b/drivers/net/mdio/mdio-aspeed.c
> @@ -3,6 +3,7 @@
>  
>  #include <linux/bitfield.h>
>  #include <linux/delay.h>
> +#include <linux/reset.h>
>  #include <linux/iopoll.h>
>  #include <linux/mdio.h>
>  #include <linux/module.h>
> @@ -37,6 +38,7 @@
>  
>  struct aspeed_mdio {
>         void __iomem *base;
> +       struct reset_control *reset;
>  };
>  
>  static int aspeed_mdio_read(struct mii_bus *bus, int addr, int
> regnum)
> @@ -120,6 +122,12 @@ static int aspeed_mdio_probe(struct
> platform_device *pdev)
>         if (IS_ERR(ctx->base))
>                 return PTR_ERR(ctx->base);
>  
> +       ctx->reset = devm_reset_control_get_shared(&pdev->dev, NULL);

The device tree bindings should have a required reset control property
documented before this is added.

> +       if (IS_ERR(ctx->reset))
> +               return PTR_ERR(ctx->reset);
> +
> +       reset_control_deassert(ctx->reset);
> +

This is missing a corresponding reset_control_assert() in
aspeed_mdio_remove() and in the error path of of_mdiobus_register().
That would allow to assert the reset line again after all MDIO
controllers are unbound.

regards
Philipp