netdev - Re: [PATCH net-next 3/4] device property: add a helper for loading netdev->dev

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]

Message-ID: <YV1TkRtiu/q2vm/S@kuha.fi.intel.com>
Date:   Wed, 6 Oct 2021 10:43:13 +0300
From:   Heikki Krogerus <heikki.krogerus@...ux.intel.com>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     davem@...emloft.net, netdev@...r.kernel.org,
        gregkh@...uxfoundation.org, rafael@...nel.org, andrew@...n.ch,
        hkallweit1@...il.com, linux@...linux.org.uk, robh+dt@...nel.org,
        frowand.list@...il.com, devicetree@...r.kernel.org
Subject: Re: [PATCH net-next 3/4] device property: add a helper for loading
 netdev->dev_addr

On Tue, Oct 05, 2021 at 08:53:20AM -0700, Jakub Kicinski wrote:
> Commit 406f42fa0d3c ("net-next: When a bond have a massive amount
> of VLANs...") introduced a rbtree for faster Ethernet address look
> up. To maintain netdev->dev_addr in this tree we need to make all
> the writes to it got through appropriate helpers.
> 
> There is a handful of drivers which pass netdev->dev_addr as
> the destination buffer to device_get_mac_address(). Add a helper
> which takes a dev pointer instead, so it can call an appropriate
> helper.
> 
> Signed-off-by: Jakub Kicinski <kuba@...nel.org>
> ---
>  drivers/base/property.c  | 20 ++++++++++++++++++++
>  include/linux/property.h |  2 ++
>  2 files changed, 22 insertions(+)
> 
> diff --git a/drivers/base/property.c b/drivers/base/property.c
> index 453918eb7390..1c8d4676addc 100644
> --- a/drivers/base/property.c
> +++ b/drivers/base/property.c
> @@ -997,6 +997,26 @@ void *device_get_mac_address(struct device *dev, char *addr, int alen)
>  }
>  EXPORT_SYMBOL(device_get_mac_address);
>  
> +/**
> + * device_get_ethdev_addr - Set netdev's MAC address from a given device
> + * @dev:	Pointer to the device
> + * @netdev:	Pointer to netdev to write the address to
> + *
> + * Wrapper around device_get_mac_address() which writes the address
> + * directly to netdev->dev_addr.
> + */
> +void *device_get_ethdev_addr(struct device *dev, struct net_device *netdev)
> +{
> +	u8 addr[ETH_ALEN];
> +	void *ret;
> +
> +	ret = device_get_mac_address(dev, addr, ETH_ALEN);
> +	if (ret)
> +		eth_hw_addr_set(netdev, addr);
> +	return ret;
> +}
> +EXPORT_SYMBOL(device_get_ethdev_addr);

Is there some reason why can't this be in net/ethernet/eth.c?

I would really prefer that we don't add any more subsystem specific
functions into this file (drivers/base/property.c).

Shouldn't actually fwnode_get_mac_addr() and fwnode_get_mac_address()
be moved to net/ethernet/eth.c as well?

>  /**
>   * fwnode_irq_get - Get IRQ directly from a fwnode
>   * @fwnode:	Pointer to the firmware node
> diff --git a/include/linux/property.h b/include/linux/property.h
> index 357513a977e5..24dc4d2b9dbd 100644
> --- a/include/linux/property.h
> +++ b/include/linux/property.h
> @@ -15,6 +15,7 @@
>  #include <linux/types.h>
>  
>  struct device;
> +struct net_device;
>  
>  enum dev_prop_type {
>  	DEV_PROP_U8,
> @@ -390,6 +391,7 @@ const void *device_get_match_data(struct device *dev);
>  int device_get_phy_mode(struct device *dev);
>  
>  void *device_get_mac_address(struct device *dev, char *addr, int alen);
> +void *device_get_ethdev_addr(struct device *dev, struct net_device *netdev);

So this you would then introduce in linux/etherdevce.h (?)
But not here, in include/linux/property.h

thanks,

-- 
heikki