[PATCH net] net: dsa: mt7530: correct ds->num_ports

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>
---
 drivers/net/dsa/mt7530.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index a3b49abd32f1..dbd15da977b5 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c

@@ -1035,9 +1035,6 @@ mt7530_port_enable(struct dsa_switch *ds, int port,
 {
 	struct mt7530_priv *priv = ds->priv;
 
-	if (!dsa_is_user_port(ds, port))
-		return 0;
-
 	mutex_lock(&priv->reg_mutex);
 
 	/* Allow the user port gets connected to the cpu port and also

@@ -1060,9 +1057,6 @@ mt7530_port_disable(struct dsa_switch *ds, int port)
 {
 	struct mt7530_priv *priv = ds->priv;
 
-	if (!dsa_is_user_port(ds, port))
-		return;
-
 	mutex_lock(&priv->reg_mutex);
 
 	/* Clear up all port matrix which could be restored in the next

@@ -3265,7 +3259,7 @@ mt7530_probe(struct mdio_device *mdiodev)
 		return -ENOMEM;
 
 	priv->ds->dev = &mdiodev->dev;
-	priv->ds->num_ports = DSA_MAX_PORTS;
+	priv->ds->num_ports = MT7530_NUM_PORTS;
 
 	/* Use medatek,mcm property to distinguish hardware type that would
 	 * casues a little bit differences on power-on sequence.

-- 
2.25.1

On 10/15/2021 11:24 PM, DENG Qingfang wrote:

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>

Do you really want to target the net tree for this change?
-- 
Florian

On Sat, Oct 16, 2021 at 07:36:14PM -0700, Florian Fainelli wrote:

On 10/15/2021 11:24 PM, DENG Qingfang wrote:

quoted

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>

Do you really want to target the net tree for this change?

Yes because I consider this a bug fix.

Hello:

This patch was applied to netdev/net.git (master)
by David S. Miller [off-list ref]:

On Sat, 16 Oct 2021 14:24:14 +0800 you wrote:

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>

[...]

Here is the summary with links:
  - [net] net: dsa: mt7530: correct ds->num_ports
    https://git.kernel.org/netdev/net/c/342afce10d6f

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

On 10/18/21 1:42 AM, DENG Qingfang wrote:

On Sat, Oct 16, 2021 at 07:36:14PM -0700, Florian Fainelli wrote:

quoted

On 10/15/2021 11:24 PM, DENG Qingfang wrote:

quoted

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>

Do you really want to target the net tree for this change?

Yes because I consider this a bug fix.

OK, why not provide a Fixes tag to help with targeting the back port
then? This has been applied anyway, so hopefully the auto selection will
do its job and tell you where it stops applying cleanly.
-- 
Florian

On Mon, Oct 18, 2021 at 09:22:49AM -0700, Florian Fainelli wrote:

On 10/18/21 1:42 AM, DENG Qingfang wrote:

quoted

On Sat, Oct 16, 2021 at 07:36:14PM -0700, Florian Fainelli wrote:

quoted

On 10/15/2021 11:24 PM, DENG Qingfang wrote:

quoted

Setting ds->num_ports to DSA_MAX_PORTS made DSA core allocate unnecessary
dsa_port's and call mt7530_port_disable for non-existent ports.

Set it to MT7530_NUM_PORTS to fix that, and dsa_is_user_port check in
port_enable/disable is no longer required.

Cc: stable@vger.kernel.org
Signed-off-by: DENG Qingfang <dqfext@gmail.com>

Do you really want to target the net tree for this change?

Yes because I consider this a bug fix.

OK, why not provide a Fixes tag to help with targeting the back port
then? This has been applied anyway, so hopefully the auto selection will
do its job and tell you where it stops applying cleanly.

Without a "Fixes:" tag, I just backport things as far are they are easy
to go and then stop without an email saying anything fails on older
kernels.

thanks,

greg k-h