Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Alexander Stein <alexander.stein@ew.tq-group.com>
To:     andrew@lunn.ch, hkallweit1@gmail.com, davem@davemloft.net,
        edumazet@google.com, kuba@kernel.org, pabeni@redhat.com,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux@armlinux.org.uk, Yan Wang <rk.code@outlook.com>
Cc:     Yan Wang <rk.code@outlook.com>
Subject: Re: [PATCH v4] net: mdiobus: Add a function to deassert reset
Date:   Fri, 12 May 2023 09:59:56 +0200
Message-ID: <1858925.CQOukoFCf9@steina-w>
Organization: TQ-Systems GmbH
In-Reply-To: <KL1PR01MB54480925428513803DF3D03AE6749@KL1PR01MB5448.apcprd01.prod.exchangelabs.com>
References: <KL1PR01MB54480925428513803DF3D03AE6749@KL1PR01MB5448.apcprd01.prod.exchangelabs.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="iso-8859-1"
Precedence: bulk

Am Donnerstag, 11. Mai 2023, 08:59:09 CEST schrieb Yan Wang:
> It is possible to mount multiple sub-devices on the mido bus.
> The hardware power-on does not necessarily reset these devices.
> The device may be in an uncertain state, causing the device's ID
> to not be scanned.
>=20
> So,before adding a reset to the scan, make sure the device is in

'So, before'. One space after ,

> normal working mode.
>=20
> Reported-by: kernel test robot <lkp@intel.com>
> Link:
> https://lore.kernel.org/oe-kbuild-all/202305101702.4xW6vT72-lkp@intel.com/
> Signed-off-by: Yan Wang <rk.code@outlook.com>
> ---
> v4:
>   - Get pulse width and settling time from the device tree
>   - Add logic for processing (PTR_ERR(reset) =3D=3D -EPROBE_DEFER)
>   - included <linux/goio/consumer.h>
>   - fixed commit message
> v3:
> https://lore.kernel.org/all/KL1PR01MB5448A33A549CDAD7D68945B9E6779@KL1PR0=
1M
> B5448.apcprd01.prod.exchangelabs.com/ - fixed commit message
> v2:
> https://lore.kernel.org/all/KL1PR01MB54482416A8BE0D80EA27223CE6779@KL1PR0=
1M
> B5448.apcprd01.prod.exchangelabs.com/ - fixed commit message
>   - Using gpiod_ replace gpio_
> v1:
> https://lore.kernel.org/all/KL1PR01MB5448631F2D6F71021602117FE6769@KL1PR0=
1M
> B5448.apcprd01.prod.exchangelabs.com/ - Incorrect description of commit
> message.
>   - The gpio-api too old
> ---
>  drivers/net/mdio/fwnode_mdio.c | 32 ++++++++++++++++++++++++++++++++
>  1 file changed, 32 insertions(+)
>=20
> diff --git a/drivers/net/mdio/fwnode_mdio.c b/drivers/net/mdio/fwnode_mdi=
o.c
> index 1183ef5e203e..9d7df6393059 100644
> --- a/drivers/net/mdio/fwnode_mdio.c
> +++ b/drivers/net/mdio/fwnode_mdio.c
> @@ -11,6 +11,7 @@
>  #include <linux/of.h>
>  #include <linux/phy.h>
>  #include <linux/pse-pd/pse.h>
> +#include <linux/gpio/consumer.h>

Please sort alphabetically.
>=20
>  MODULE_AUTHOR("Calvin Johnson <calvin.johnson@oss.nxp.com>");
>  MODULE_LICENSE("GPL");
> @@ -57,6 +58,35 @@ fwnode_find_mii_timestamper(struct fwnode_handle *fwno=
de)
> return register_mii_timestamper(arg.np, arg.args[0]);
>  }
>=20
> +static void fwnode_mdiobus_pre_enable_phy(struct fwnode_handle *fwnode)
> +{
> +	struct gpio_desc *reset;
> +	unsigned int reset_assert_delay;
> +	unsigned int reset_deassert_delay;
> +
> +	reset =3D fwnode_gpiod_get_index(fwnode, "reset", 0, GPIOD_OUT_LOW,=20
NULL);
> +	if (IS_ERR(reset)) {
> +		if (PTR_ERR(reset) =3D=3D -EPROBE_DEFER)
> +			pr_debug("%pOFn: %s: GPIOs not yet available,=20
retry later\n",
> +				 to_of_node(fwnode), __func__);

You are not dealing with this error at all. pr_debug is (mostly) not even=20
printed in kernel log, despite the code registering the PHY ignoring the=20
deferral completely.

> +		else
> +			pr_err("%pOFn: %s: Can't get reset line=20
property\n",
> +			       to_of_node(fwnode), __func__);
> +

This error also gets ignored.

> +		return;
> +	}
> +	fwnode_property_read_u32(fwnode, "reset-assert-us",
> +				 &reset_assert_delay);
> +	fwnode_property_read_u32(fwnode, "reset-deassert-us",
> +				 &reset_deassert_delay);

These are currently read in fwnode_mdiobus_phy_device_register(). There is =
no=20
need to read them twice. It should be moved to a location where it just nee=
d=20
to be read once.

Regards,
Alexander

> +	gpiod_set_value_cansleep(reset, 1);
> +	fsleep(reset_assert_delay);
> +	gpiod_set_value_cansleep(reset, 0);
> +	fsleep(reset_deassert_delay);
> +	/*Release phy's reset line, mdiobus_register_gpiod() need to request=20
it*/
> +	gpiod_put(reset);
> +}
> +
>  int fwnode_mdiobus_phy_device_register(struct mii_bus *mdio,
>  				       struct phy_device *phy,
>  				       struct fwnode_handle *child,=20
u32 addr)
> @@ -119,6 +149,8 @@ int fwnode_mdiobus_register_phy(struct mii_bus *bus,
>  	u32 phy_id;
>  	int rc;
>=20
> +	fwnode_mdiobus_pre_enable_phy(child);
> +
>  	psec =3D fwnode_find_pse_control(child);
>  	if (IS_ERR(psec))
>  		return PTR_ERR(psec);


=2D-=20
TQ-Systems GmbH | M=FChlstra=DFe 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht M=FCnchen, HRB 105018
Gesch=E4ftsf=FChrer: Detlef Schneider, R=FCdiger Stahl, Stefan Schneider
http://www.tq-group.com/