[PATCH] net: dsa: qca8k: reset cpu port on MTU change

[PATCH] net: dsa: qca8k: reset cpu port on MTU change

[Christian Marangi]

commit 386228c694bf1e7a7688e44412cb33500b0ac585 upstream.

It was discovered that the Documentation lacks of a fundamental detail
on how to correctly change the MAX_FRAME_SIZE of the switch.

In fact if the MAX_FRAME_SIZE is changed while the cpu port is on, the
switch panics and cease to send any packet. This cause the mgmt ethernet
system to not receive any packet (the slow fallback still works) and
makes the device not reachable. To recover from this a switch reset is
required.

To correctly handle this, turn off the cpu ports before changing the
MAX_FRAME_SIZE and turn on again after the value is applied.

Fixes: f58d2598cf70 ("net: dsa: qca8k: implement the port MTU callbacks")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://lore.kernel.org/r/20220621151122.10220-1-ansuelsmth@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ backport: fix conflict using the old port_sts struct instead of bitmap ]
---
 drivers/net/dsa/qca8k.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c
index a984f06f6f04..67869c8cbeaa 100644
--- a/drivers/net/dsa/qca8k.c
+++ b/drivers/net/dsa/qca8k.c
@@ -1599,7 +1599,7 @@ static int
 qca8k_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 {
 	struct qca8k_priv *priv = ds->priv;
-	int i, mtu = 0;
+	int ret, i, mtu = 0;
 
 	priv->port_mtu[port] = new_mtu;
 
@@ -1607,8 +1607,27 @@ qca8k_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 		if (priv->port_mtu[i] > mtu)
 			mtu = priv->port_mtu[i];
 
+	/* To change the MAX_FRAME_SIZE the cpu ports must be off or
+	 * the switch panics.
+	 * Turn off both cpu ports before applying the new value to prevent
+	 * this.
+	 */
+	if (priv->port_sts[0].enabled)
+		qca8k_port_set_status(priv, 0, 0);
+
+	if (priv->port_sts[6].enabled)
+		qca8k_port_set_status(priv, 6, 0);
+
 	/* Include L2 header / FCS length */
-	return qca8k_write(priv, QCA8K_MAX_FRAME_SIZE, mtu + ETH_HLEN + ETH_FCS_LEN);
+	ret = qca8k_write(priv, QCA8K_MAX_FRAME_SIZE, mtu + ETH_HLEN + ETH_FCS_LEN);
+
+	if (priv->port_sts[0].enabled)
+		qca8k_port_set_status(priv, 0, 1);
+
+	if (priv->port_sts[6].enabled)
+		qca8k_port_set_status(priv, 6, 1);
+
+	return ret;
 }
 
 static int
-- 
2.36.1

Re: [PATCH] net: dsa: qca8k: reset cpu port on MTU change

[Christian Marangi]

On Tue, Jun 28, 2022 at 04:30:10PM +0200, Christian Marangi wrote:
> commit 386228c694bf1e7a7688e44412cb33500b0ac585 upstream.
> 
> It was discovered that the Documentation lacks of a fundamental detail
> on how to correctly change the MAX_FRAME_SIZE of the switch.
> 
> In fact if the MAX_FRAME_SIZE is changed while the cpu port is on, the
> switch panics and cease to send any packet. This cause the mgmt ethernet
> system to not receive any packet (the slow fallback still works) and
> makes the device not reachable. To recover from this a switch reset is
> required.
> 
> To correctly handle this, turn off the cpu ports before changing the
> MAX_FRAME_SIZE and turn on again after the value is applied.
> 
> Fixes: f58d2598cf70 ("net: dsa: qca8k: implement the port MTU callbacks")
> Cc: stable@vger.kernel.org
> Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
> Link: https://lore.kernel.org/r/20220621151122.10220-1-ansuelsmth@gmail.com
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> [ backport: fix conflict using the old port_sts struct instead of bitmap ]

Cc Greg wonder if this has some problems or I pushed it badly? (first
time I try to push a fixed backport)

> ---
>  drivers/net/dsa/qca8k.c | 23 +++++++++++++++++++++--
>  1 file changed, 21 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c
> index a984f06f6f04..67869c8cbeaa 100644
> --- a/drivers/net/dsa/qca8k.c
> +++ b/drivers/net/dsa/qca8k.c
> @@ -1599,7 +1599,7 @@ static int
>  qca8k_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
>  {
>  	struct qca8k_priv *priv = ds->priv;
> -	int i, mtu = 0;
> +	int ret, i, mtu = 0;
>  
>  	priv->port_mtu[port] = new_mtu;
>  
> @@ -1607,8 +1607,27 @@ qca8k_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
>  		if (priv->port_mtu[i] > mtu)
>  			mtu = priv->port_mtu[i];
>  
> +	/* To change the MAX_FRAME_SIZE the cpu ports must be off or
> +	 * the switch panics.
> +	 * Turn off both cpu ports before applying the new value to prevent
> +	 * this.
> +	 */
> +	if (priv->port_sts[0].enabled)
> +		qca8k_port_set_status(priv, 0, 0);
> +
> +	if (priv->port_sts[6].enabled)
> +		qca8k_port_set_status(priv, 6, 0);
> +
>  	/* Include L2 header / FCS length */
> -	return qca8k_write(priv, QCA8K_MAX_FRAME_SIZE, mtu + ETH_HLEN + ETH_FCS_LEN);
> +	ret = qca8k_write(priv, QCA8K_MAX_FRAME_SIZE, mtu + ETH_HLEN + ETH_FCS_LEN);
> +
> +	if (priv->port_sts[0].enabled)
> +		qca8k_port_set_status(priv, 0, 1);
> +
> +	if (priv->port_sts[6].enabled)
> +		qca8k_port_set_status(priv, 6, 1);
> +
> +	return ret;
>  }
>  
>  static int
> -- 
> 2.36.1
> 

-- 
	Ansuel

Re: [PATCH] net: dsa: qca8k: reset cpu port on MTU change

[Greg KH]

On Tue, Jun 28, 2022 at 04:30:10PM +0200, Christian Marangi wrote:
> commit 386228c694bf1e7a7688e44412cb33500b0ac585 upstream.
> 
> It was discovered that the Documentation lacks of a fundamental detail
> on how to correctly change the MAX_FRAME_SIZE of the switch.
> 
> In fact if the MAX_FRAME_SIZE is changed while the cpu port is on, the
> switch panics and cease to send any packet. This cause the mgmt ethernet
> system to not receive any packet (the slow fallback still works) and
> makes the device not reachable. To recover from this a switch reset is
> required.
> 
> To correctly handle this, turn off the cpu ports before changing the
> MAX_FRAME_SIZE and turn on again after the value is applied.
> 
> Fixes: f58d2598cf70 ("net: dsa: qca8k: implement the port MTU callbacks")
> Cc: stable@vger.kernel.org
> Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
> Link: https://lore.kernel.org/r/20220621151122.10220-1-ansuelsmth@gmail.com
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> [ backport: fix conflict using the old port_sts struct instead of bitmap ]
> ---
>  drivers/net/dsa/qca8k.c | 23 +++++++++++++++++++++--
>  1 file changed, 21 insertions(+), 2 deletions(-)

This only applied to 5.15.y and 5.18.y, what about 5.10.y?  Please give
us a hint as to what kernel(s) a backport is for when sending them.

thanks,

greg k-h