Skip to contentSkip to conversation

net: dsa: Do not leave DSA master with NULL netdev_ops

PATCH In review Florian Fainelli 4 participants 7 messages May 4, 2020 – May 7, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net: dsa: Do not leave DSA master with NULL netdev_ops

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")
---
 net/dsa/master.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/dsa/master.c +2 -1
index b5c535af63a3..a621367c6e8c 100644
@@ -289,7 +289,8 @@ static void dsa_master_ndo_teardown(struct net_device *dev)
 {
 	struct dsa_port *cpu_dp = dev->dsa_ptr;
 
-	dev->netdev_ops = cpu_dp->orig_ndo_ops;
+	if (cpu_dp->orig_ndo_ops)
+		dev->netdev_ops = cpu_dp->orig_ndo_ops;
 	cpu_dp->orig_ndo_ops = NULL;
 }
 
Signature · 1 line
2.20.1
maintainer
#1 lore raw

Hi Florian,

Quoted 8 lines · Florian Fainelli

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 8 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

---

The fix makes complete sense.
But on another note, if we don't overlay an ndo_get_phys_port_name if the master already has one, doesn't that render the entire mechanism of having a reliable way for user space to determine the CPU port number pointless?

Quoted 18 lines
 net/dsa/master.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/dsa/master.c +2 -2
index b5c535af63a3..a621367c6e8c 100644
@@ -289,7 +289,8 @@ static void dsa_master_ndo_teardown(struct net_device *dev)
 {
        struct dsa_port *cpu_dp = dev->dsa_ptr;
 
-       dev->netdev_ops = cpu_dp->orig_ndo_ops;
+       if (cpu_dp->orig_ndo_ops)
+               dev->netdev_ops = cpu_dp->orig_ndo_ops;
        cpu_dp->orig_ndo_ops = NULL;
 }
 
--
Quoted 2 lines

2.20.1

Regards,
-Vladimir

#2 lore raw
Quoted 20 lines · Vladimir Oltean

Hi Florian,

Quoted 8 lines · Florian Fainelli

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 8 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

---

Quoted 5 lines

The fix makes complete sense.
But on another note, if we don't overlay an ndo_get_phys_port_name if the master already has one, doesn't that render the entire mechanism of having a reliable way for user space to determine the CPU port number pointless?

For the CPU port I would consider ndo_get_phys_port_name() to be more best effort than an absolute need unlike the user facing ports, where this is necessary for a variety of actions (e.g.: determining queues/port numbers etc.) which is why there was no overlay being done in that case. There is not a good way to cascade the information other than do something like pX.Y and defining what the X and Y are, what do you think?

Signature · 1 line
Florian
maintainer
#3 lore raw
Quoted 30 lines · Florian Fainelli
Quoted 20 lines · Vladimir Oltean

Hi Florian,

Quoted 8 lines · Florian Fainelli

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 8 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

---

Quoted 5 lines

The fix makes complete sense.
But on another note, if we don't overlay an ndo_get_phys_port_name if the master already has one, doesn't that render the entire mechanism of having a reliable way for user space to determine the CPU port number pointless?

Quoted 9 lines

For the CPU port I would consider ndo_get_phys_port_name() to be more best effort than an absolute need unlike the user facing ports, where this is necessary for a variety of actions (e.g.: determining queues/port numbers etc.) which is why there was no overlay being done in that case. There is not a good way to cascade the information other than do something like pX.Y and defining what the X and Y are, what do you think?
--
Florian

For the CPU/master port I am not actually sure who is the final consumer of the ndo_get_phys_port_name, I thought it is simply informational, with the observation that it may be unreliable in transmitting that information over.
Speaking of which, if "informational" is the only purpose, could this not be used?

devlink port | grep "flavour cpu" pci/0000:00:00.5/4: type notset flavour cpu port 4 spi/spi2.0/4: type notset flavour cpu port 4 spi/spi2.1/4: type notset flavour cpu port 4

Thanks,
-Vladimir

#4 lore raw
Quoted 41 lines · Vladimir Oltean
Quoted 30 lines · Florian Fainelli
Quoted 20 lines · Vladimir Oltean

Hi Florian,

Quoted 8 lines · Florian Fainelli

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 8 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

---

Quoted 5 lines

The fix makes complete sense.
But on another note, if we don't overlay an ndo_get_phys_port_name if the master already has one, doesn't that render the entire mechanism of having a reliable way for user space to determine the CPU port number pointless?

Quoted 9 lines

For the CPU port I would consider ndo_get_phys_port_name() to be more best effort than an absolute need unlike the user facing ports, where this is necessary for a variety of actions (e.g.: determining queues/port numbers etc.) which is why there was no overlay being done in that case. There is not a good way to cascade the information other than do something like pX.Y and defining what the X and Y are, what do you think?
--
Florian

Quoted 6 lines

For the CPU/master port I am not actually sure who is the final consumer of the ndo_get_phys_port_name, I thought it is simply informational, with the observation that it may be unreliable in transmitting that information over.
Speaking of which, if "informational" is the only purpose, could this not be used?

Yes, I had not considered devlink would expose that information, ndo_phys_port_name() is there now though and since it is exposed through sysfs so reverting would be an ABI breakage.

Quoted 9 lines

devlink port | grep "flavour cpu" pci/0000:00:00.5/4: type notset flavour cpu port 4 spi/spi2.0/4: type notset flavour cpu port 4 spi/spi2.1/4: type notset flavour cpu port 4

Thanks,
-Vladimir

Signature · 1 line
Florian
Tested-by on v1 1/1
#5 lore raw
In reply to Florian Fainelli
Quoted 7 lines

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 7 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

Thank you Florain.

Quoted 19 lines
---
  net/dsa/master.c | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)
net/dsa/master.c +2 -1
index b5c535af63a3..a621367c6e8c 100644
@@ -289,7 +289,8 @@ static void dsa_master_ndo_teardown(struct net_device *dev)
  {
  	struct dsa_port *cpu_dp = dev->dsa_ptr;
 
-	dev->netdev_ops = cpu_dp->orig_ndo_ops;
+	if (cpu_dp->orig_ndo_ops)
+		dev->netdev_ops = cpu_dp->orig_ndo_ops;
  	cpu_dp->orig_ndo_ops = NULL;
  }
 
maintainer
#6 lore raw
In reply to Florian Fainelli

From: Florian Fainelli <[email protected]>
Date: Mon, 4 May 2020 13:18:06 -0700

Quoted 7 lines

When ndo_get_phys_port_name() for the CPU port was added we introduced an early check for when the DSA master network device in dsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When we perform the teardown operation in dsa_master_ndo_teardown() we would not be checking that cpu_dp->orig_ndo_ops was successfully allocated and non-NULL initialized.

Quoted 7 lines

With network device drivers such as virtio_net, this leads to a NPD as soon as the DSA switch hanging off of it gets torn down because we are now assigning the virtio_net device's netdev_ops a NULL pointer.

Fixes da7b9e9b00d4 ("net: dsa: Add ndo_get_phys_port_name() for CPU port")

Applied and queued up for -stable, thanks Florian.