[PATCH net] net: emac: release DT PHY resources on teardown

[PATCH net] net: emac: release DT PHY resources on teardown

[Rosen Penev]

Disconnect PHYLIB PHYs and deregister fixed PHYs during both probe
unwind and driver removal. Without this cleanup, unbinding a DT-based
EMAC can leave its PHY attached to the freed netdev, while fixed-link
instances retain their registered fixed PHY.

Fixes: a577ca6badb5 ("net: emac: add support for device-tree based PHY discovery and setup")
Signed-off-by: Rosen Penev <rosenp@gmail.com>
---
 drivers/net/ethernet/ibm/emac/core.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c
index e7043523457c..4124a90b18c4 100644
--- a/drivers/net/ethernet/ibm/emac/core.c
+++ b/drivers/net/ethernet/ibm/emac/core.c
@@ -2810,6 +2810,16 @@ static int emac_init_phy(struct emac_instance *dev)
 	return 0;
 }
 
+static void emac_cleanup_phy(struct emac_instance *dev)
+{
+	struct device_node *np = dev->ofdev->dev.of_node;
+
+	if (dev->ndev->phydev && dev->phy.address >= 0)
+		phy_disconnect(dev->ndev->phydev);
+	if (of_phy_is_fixed_link(np))
+		of_phy_deregister_fixed_link(np);
+}
+
 static int emac_init_config(struct emac_instance *dev)
 {
 	struct device_node *np = dev->ofdev->dev.of_node;
@@ -3197,6 +3207,7 @@ static int emac_probe(struct platform_device *ofdev)
 	if (emac_has_feature(dev, EMAC_FTR_HAS_ZMII))
 		zmii_detach(dev->zmii_dev, dev->zmii_port);
  err_unreg_commac:
+	emac_cleanup_phy(dev);
 	mal_unregister_commac(dev->mal, &dev->commac);
  err_rel_deps:
 	emac_put_deps(dev);
@@ -3220,6 +3231,7 @@ static void emac_remove(struct platform_device *ofdev)
 	unregister_netdev(dev->ndev);
 
 	cancel_work_sync(&dev->reset_work);
+	emac_cleanup_phy(dev);
 
 	if (emac_has_feature(dev, EMAC_FTR_HAS_TAH))
 		tah_detach(dev->tah_dev, dev->tah_port);
-- 
2.56.0

Re: [PATCH net] net: emac: release DT PHY resources on teardown

[netdev-bot+sinfo]

Hi!

This is an automated message. This series looks like a fix, but its
commit messages seem to be missing some information:

 - How the issue was discovered, e.g. hit in production, hit during
   development, syzbot report, manual code inspection, LLM or static
   analysis tool scan.

 - Whether the issue was actually triggered, or is only theoretical
   (e.g. found by code inspection). If it was triggered please include
   the symptoms, like the stack trace or error messages.

 - What hardware the change was tested on. For driver fixes please
   mention the device (and if relevant firmware version) used for
   testing, or say that the change was not tested on real hardware.

Please do not repost the series just to address the above. Instead,
reply to this email with the missing information, so that reviewers
can take it into account. If the series needs another revision for
other reasons, please include the information in the commit messages
then.

The evaluation is done by an LLM so it may be wrong, if you think
that is the case please reply and explain.

Re: [PATCH net] net: emac: release DT PHY resources on teardown

[Rosen Penev]

On Tue, Oct 20, 2026 at 3:33 PM <netdev-bot+sinfo@kernel.org> wrote:
>
> Hi!
>
> This is an automated message. This series looks like a fix, but its
> commit messages seem to be missing some information:
>
>  - How the issue was discovered, e.g. hit in production, hit during
>    development, syzbot report, manual code inspection, LLM or static
>    analysis tool scan.
LLM
>
>  - Whether the issue was actually triggered, or is only theoretical
>    (e.g. found by code inspection). If it was triggered please include
>    the symptoms, like the stack trace or error messages.
>
>  - What hardware the change was tested on. For driver fixes please
>    mention the device (and if relevant firmware version) used for
>    testing, or say that the change was not tested on real hardware.
>
> Please do not repost the series just to address the above. Instead,
> reply to this email with the missing information, so that reviewers
> can take it into account. If the series needs another revision for
> other reasons, please include the information in the commit messages
> then.
>
> The evaluation is done by an LLM so it may be wrong, if you think
> that is the case please reply and explain.

Re: [PATCH net] net: emac: release DT PHY resources on teardown

[Andrew Lunn]

On Tue, Oct 20, 2026 at 03:37:57PM -0700, Rosen Penev wrote:
> On Tue, Oct 20, 2026 at 3:33 PM <netdev-bot+sinfo@kernel.org> wrote:
> >
> > Hi!
> >
> > This is an automated message. This series looks like a fix, but its
> > commit messages seem to be missing some information:
> >
> >  - How the issue was discovered, e.g. hit in production, hit during
> >    development, syzbot report, manual code inspection, LLM or static
> >    analysis tool scan.
> LLM

Ideally, you should learn from these emails, and put it into the
commit message.

       Andrew

Re: [PATCH net] net: emac: release DT PHY resources on teardown

[Andrew Lunn]

On Tue, Oct 20, 2026 at 03:30:02PM -0700, Rosen Penev wrote:
> Disconnect PHYLIB PHYs and deregister fixed PHYs during both probe
> unwind and driver removal. Without this cleanup, unbinding a DT-based
> EMAC can leave its PHY attached to the freed netdev, while fixed-link
> instances retain their registered fixed PHY.

This driver is quite convoluted with its PHY handling.

Please split this into two, phy_disconnect() and
of_phy_deregister_fixed_link() and add better commit messages.

	if ((dev->phy_address == 0xffffffff && dev->phy_map == 0xffffffff) ||
	    of_phy_is_fixed_link(np)) {

		/* PHY-less configuration. */
		dev->phy.address = -1;
...
		if (of_phy_is_fixed_link(np)) {
			int res = emac_dt_mdio_probe(dev);

			if (res)
				return res;

			res = of_phy_register_fixed_link(np);

The logic here is more complex than 

> +	if (of_phy_is_fixed_link(np))
> +		of_phy_deregister_fixed_link(np);

So you need to explain in the commit message why it is correct.

> +static void emac_cleanup_phy(struct emac_instance *dev)
> +{
> +	struct device_node *np = dev->ofdev->dev.of_node;
> +
> +	if (dev->ndev->phydev && dev->phy.address >= 0)
> +		phy_disconnect(dev->ndev->phydev);

It is also not clear this is correct, since a fixed-link should be
connected to, yet a fixed link has dev->phy.address == -1?

	Andrew

Re: [PATCH net] net: emac: release DT PHY resources on teardown

[Rosen Penev]

On Tue Oct 20, 2026 at 6:05 PM PDT, Andrew Lunn wrote:
> On Tue, Oct 20, 2026 at 03:30:02PM -0700, Rosen Penev wrote:
>> Disconnect PHYLIB PHYs and deregister fixed PHYs during both probe
>> unwind and driver removal. Without this cleanup, unbinding a DT-based
>> EMAC can leave its PHY attached to the freed netdev, while fixed-link
>> instances retain their registered fixed PHY.
>
> This driver is quite convoluted with its PHY handling.
I agree. I got this LLM to convert the whole thing to phylink. It tested
a bunch of stuff over serial and ethernet, which works. That's net-next
material though. It'll probably get rejected.
>
> Please split this into two, phy_disconnect() and
> of_phy_deregister_fixed_link() and add better commit messages.
>
> 	if ((dev->phy_address == 0xffffffff && dev->phy_map == 0xffffffff) ||
> 	    of_phy_is_fixed_link(np)) {
>
> 		/* PHY-less configuration. */
> 		dev->phy.address = -1;
> ...
> 		if (of_phy_is_fixed_link(np)) {
> 			int res = emac_dt_mdio_probe(dev);
>
> 			if (res)
> 				return res;
>
> 			res = of_phy_register_fixed_link(np);
>
> The logic here is more complex than
>
>> +	if (of_phy_is_fixed_link(np))
>> +		of_phy_deregister_fixed_link(np);
>
> So you need to explain in the commit message why it is correct.
>
>> +static void emac_cleanup_phy(struct emac_instance *dev)
>> +{
>> +	struct device_node *np = dev->ofdev->dev.of_node;
>> +
>> +	if (dev->ndev->phydev && dev->phy.address >= 0)
>> +		phy_disconnect(dev->ndev->phydev);
>
> It is also not clear this is correct, since a fixed-link should be
> connected to, yet a fixed link has dev->phy.address == -1?
>
> 	Andrew