[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMZ6RqJqwyJQ17EJd0eV_8tbsC5b16nbuQj3uD38BeRzSdPc3Q@mail.gmail.com>
Date: Mon, 22 May 2023 16:58:02 +0900
From: Vincent MAILHOL <mailhol.vincent@...adoo.fr>
To: Marcel Hellwig <git@...kiesoft.de>
Cc: Appana Durga Kedareswara rao <appana.durga.rao@...inx.com>,
Naga Sureshkumar Relli <naga.sureshkumar.relli@...inx.com>, linux-can@...r.kernel.org,
netdev@...r.kernel.org, Marcel Hellwig <mhellwig@...-group.com>
Subject: Re: [PATCH] can: dev: add transceiver capabilities to xilinx_can
Hi Marcel,
Style check only (I am not yet familiar enough with the devicetree).
On Mon. 17 Apr. 2023 at 18:01, Marcel Hellwig <git@...kiesoft.de> wrote:
> Currently the xilinx_can driver does not support adding a phy like the
> "ti,tcan1043" to its devicetree.
>
> This code makes it possible to add such phy, so that the kernel makes
> sure that the PHY is in operational state, when the link is set to an
> "up" state.
>
> Signed-off-by: Marcel Hellwig <git@...kiesoft.de>
> ---
> drivers/net/can/xilinx_can.c | 20 ++++++++++++++++++++
> 1 file changed, 20 insertions(+)
>
> diff --git a/drivers/net/can/xilinx_can.c b/drivers/net/can/xilinx_can.c
> index 43c812ea1de0..6a5b805d579a 100644
> --- a/drivers/net/can/xilinx_can.c
> +++ b/drivers/net/can/xilinx_can.c
> @@ -28,6 +28,7 @@
> #include <linux/types.h>
> #include <linux/can/dev.h>
> #include <linux/can/error.h>
> +#include <linux/phy/phy.h>
> #include <linux/pm_runtime.h>
>
> #define DRIVER_NAME "xilinx_can"
> @@ -215,6 +216,7 @@ struct xcan_priv {
> struct clk *bus_clk;
> struct clk *can_clk;
> struct xcan_devtype_data devtype;
> + struct phy *transceiver;
> };
>
> /* CAN Bittiming constants as per Xilinx CAN specs */
> @@ -1419,6 +1421,12 @@ static int xcan_open(struct net_device *ndev)
> struct xcan_priv *priv = netdev_priv(ndev);
> int ret;
>
> + ret = phy_power_on(priv->transceiver);
> + if (ret) {
> + netdev_err(ndev, "%s: phy_power_on failed(%d)\n", __func__, ret);
>From the Linux kernel coding style:
Printing numbers in parentheses (%d) adds no value and should be avoided.
Link: https://www.kernel.org/doc/html/latest/process/coding-style.html#printing-kernel-messages
Also consider %pe to print the mnemotechnic instead of the value:
netdev_err(ndev, "%s: phy_power_on failed: %pe\n", __func__,
ERR_PTR(ret));
> + return ret;
> + }
> +
> ret = pm_runtime_get_sync(priv->dev);
> if (ret < 0) {
> netdev_err(ndev, "%s: pm_runtime_get failed(%d)\n",
It is up to you, but bonus points if you send a clean-up patch to fix
the existing log messages.
> @@ -1461,6 +1469,7 @@ static int xcan_open(struct net_device *ndev)
> err_irq:
> free_irq(ndev->irq, ndev);
> err:
> + phy_power_off(priv->transceiver);
> pm_runtime_put(priv->dev);
>
> return ret;
> @@ -1482,6 +1491,7 @@ static int xcan_close(struct net_device *ndev)
> free_irq(ndev->irq, ndev);
> close_candev(ndev);
>
> + phy_power_off(priv->transceiver);
> pm_runtime_put(priv->dev);
>
> return 0;
> @@ -1713,6 +1723,7 @@ static int xcan_probe(struct platform_device *pdev)
> {
> struct net_device *ndev;
> struct xcan_priv *priv;
> + struct phy *transceiver;
> const struct of_device_id *of_id;
> const struct xcan_devtype_data *devtype = &xcan_axi_data;
> void __iomem *addr;
> @@ -1843,6 +1854,14 @@ static int xcan_probe(struct platform_device *pdev)
> goto err_free;
> }
>
> + transceiver = devm_phy_optional_get(&pdev->dev, NULL);
> + if (IS_ERR(transceiver)) {
> + ret = PTR_ERR(transceiver);
> + dev_err_probe(&pdev->dev, ret, "failed to get phy\n");
> + goto err_free;
> + }
> + priv->transceiver = transceiver;
> +
> priv->write_reg = xcan_write_reg_le;
> priv->read_reg = xcan_read_reg_le;
>
> @@ -1869,6 +1888,7 @@ static int xcan_probe(struct platform_device *pdev)
> goto err_disableclks;
> }
>
> + of_can_transceiver(ndev);
> pm_runtime_put(&pdev->dev);
>
> if (priv->devtype.flags & XCAN_FLAG_CANFD_2) {
> --
> 2.34.1
>
Powered by blists - more mailing lists