From: Heiner Kallweit <hkallweit1@gmail.com>
To: hhtracer@gmail.com, andrew@lunn.ch, linux@armlinux.org.uk
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
huhai <huhai@kylinos.cn>
Subject: Re: [PATCH v2] net: phy: Fix return value when !CONFIG_PHYLIB
Date: Sun, 13 Apr 2025 15:50:03 +0200 [thread overview]
Message-ID: <0fd79f9e-fe6d-4c85-8dda-7f64e917129c@gmail.com> (raw)
In-Reply-To: <20250413133709.5784-1-huhai@kylinos.cn>
On 13.04.2025 15:37, hhtracer@gmail.com wrote:
> From: huhai <huhai@kylinos.cn>
>
> Many call sites of get_phy_device() and fwnode_get_phy_node(), such as
> sfp_sm_probe_phy(), phylink_fwnode_phy_connect(), etc., rely on IS_ERR()
> to check for errors in the returned pointer.
>
> Furthermore, the implementations of get_phy_device() and
> fwnode_get_phy_node() themselves use ERR_PTR() to return error codes.
>
> Therefore, when CONFIG_PHYLIB is disabled, returning NULL is incorrect,
> as this would bypass IS_ERR() checks and may lead to NULL pointer
> dereference.
>
Is there actually any call site which doesn't select PHYLIB directly or
indirectly? When briefly checking I didn't find one.
So my question would be rather: Do we need/want stubs for the following
functions at all?
fwnode_get_phy_id
fwnode_mdio_find_device
fwnode_phy_find_device
device_phy_find_device
fwnode_get_phy_node
get_phy_device
phy_device_register
And a formal remark:
Your v2 has no change log, and please allow 24h before sending a new version.
> Returning ERR_PTR(-ENXIO) is the correct and consistent way to indicate
> that PHY support is not available, and it avoids such issues.
>
> Signed-off-by: huhai <huhai@kylinos.cn>
> ---
> include/linux/phy.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/phy.h b/include/linux/phy.h
> index a2bfae80c449..be299c572d73 100644
> --- a/include/linux/phy.h
> +++ b/include/linux/phy.h
> @@ -1787,13 +1787,13 @@ static inline struct phy_device *device_phy_find_device(struct device *dev)
> static inline
> struct fwnode_handle *fwnode_get_phy_node(struct fwnode_handle *fwnode)
> {
> - return NULL;
> + return ERR_PTR(-ENXIO);
> }
>
> static inline
> struct phy_device *get_phy_device(struct mii_bus *bus, int addr, bool is_c45)
> {
> - return NULL;
> + return ERR_PTR(-ENXIO);
> }
>
> static inline int phy_device_register(struct phy_device *phy)
next prev parent reply other threads:[~2025-04-13 13:49 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-13 13:37 hhtracer
2025-04-13 13:50 ` Heiner Kallweit [this message]
2025-04-13 14:13 ` Russell King (Oracle)
2025-04-14 2:42 ` 胡海
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0fd79f9e-fe6d-4c85-8dda-7f64e917129c@gmail.com \
--to=hkallweit1@gmail.com \
--cc=andrew@lunn.ch \
--cc=hhtracer@gmail.com \
--cc=huhai@kylinos.cn \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@armlinux.org.uk \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®