Commit 11e40f5c authored by Neil Horman's avatar Neil Horman Committed by Jason Gunthorpe

vmw_pvrdma: Release netdev when vmxnet3 module is removed

On repeated module load/unload cycles, its possible for the pvrmda driver
to encounter this crash:

...
[  297.032448] RIP: 0010:[<ffffffff839e4620>]  [<ffffffff839e4620>] netdev_walk_all_upper_dev_rcu+0x50/0xb0
[  297.034078] RSP: 0018:ffff95087780bd08  EFLAGS: 00010286
[  297.034986] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff95087a0c0000
[  297.036196] RDX: ffff95087a0c0000 RSI: ffffffff839e44e0 RDI: ffff950835d0c000
[  297.037421] RBP: ffff95087780bd40 R08: ffff95087a0e0ea0 R09: abddacd03f8e0ea0
[  297.038636] R10: abddacd03f8e0ea0 R11: ffffef5901e9dbc0 R12: ffff95087a0c0000
[  297.039854] R13: ffffffff839e44e0 R14: ffff95087a0c0000 R15: ffff950835d0c828
[  297.041071] FS:  0000000000000000(0000) GS:ffff95087fc00000(0000) knlGS:0000000000000000
[  297.042443] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  297.043429] CR2: ffffffffffffffe8 CR3: 000000007a652000 CR4: 00000000003607f0
[  297.044674] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  297.045893] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  297.047109] Call Trace:
[  297.047545]  [<ffffffff839e4698>] netdev_has_upper_dev_all_rcu+0x18/0x20
[  297.048691]  [<ffffffffc05d31af>] is_eth_port_of_netdev+0x2f/0xa0 [ib_core]
[  297.049886]  [<ffffffffc05d3180>] ? is_eth_active_slave_of_bonding_rcu+0x70/0x70 [ib_core]
...

This occurs because vmw_pvrdma on probe stores a pointer to the netdev
that exists on function 0 of the same bus/device/slot (which represents
the vmxnet3 ethernet driver).  However, it never removes this pointer if
the vmxnet3 module is removed, leading to crashes resulting from use after
free dereferencing incidents like the one above.

The fix is pretty straightforward.  vmw_pvrdma should listen for
NETDEV_REGISTER and NETDEV_UNREGISTER events in its event listener code
block, and update the stored netdev pointer accordingly.  This solution
has been tested by myself and the reporter with successful results.  This
fix also allows the pvrdma driver to find its underlying ethernet device
in the event that vmxnet3 is loaded after pvrdma, which it was not able to
do before.
Signed-off-by: default avatarNeil Horman <nhorman@tuxdriver.com>
Reported-by: ruquin@redhat.com
Tested-by: default avatarAdit Ranadive <aditr@vmware.com>
Acked-by: default avatarAdit Ranadive <aditr@vmware.com>
Signed-off-by: default avatarJason Gunthorpe <jgg@mellanox.com>
parent a93b632c
...@@ -695,8 +695,12 @@ static int pvrdma_del_gid(const struct ib_gid_attr *attr, void **context) ...@@ -695,8 +695,12 @@ static int pvrdma_del_gid(const struct ib_gid_attr *attr, void **context)
} }
static void pvrdma_netdevice_event_handle(struct pvrdma_dev *dev, static void pvrdma_netdevice_event_handle(struct pvrdma_dev *dev,
struct net_device *ndev,
unsigned long event) unsigned long event)
{ {
struct pci_dev *pdev_net;
unsigned int slot;
switch (event) { switch (event) {
case NETDEV_REBOOT: case NETDEV_REBOOT:
case NETDEV_DOWN: case NETDEV_DOWN:
...@@ -714,6 +718,24 @@ static void pvrdma_netdevice_event_handle(struct pvrdma_dev *dev, ...@@ -714,6 +718,24 @@ static void pvrdma_netdevice_event_handle(struct pvrdma_dev *dev,
else else
pvrdma_dispatch_event(dev, 1, IB_EVENT_PORT_ACTIVE); pvrdma_dispatch_event(dev, 1, IB_EVENT_PORT_ACTIVE);
break; break;
case NETDEV_UNREGISTER:
dev_put(dev->netdev);
dev->netdev = NULL;
break;
case NETDEV_REGISTER:
/* vmxnet3 will have same bus, slot. But func will be 0 */
slot = PCI_SLOT(dev->pdev->devfn);
pdev_net = pci_get_slot(dev->pdev->bus,
PCI_DEVFN(slot, 0));
if ((dev->netdev == NULL) &&
(pci_get_drvdata(pdev_net) == ndev)) {
/* this is our netdev */
dev->netdev = ndev;
dev_hold(ndev);
}
pci_dev_put(pdev_net);
break;
default: default:
dev_dbg(&dev->pdev->dev, "ignore netdevice event %ld on %s\n", dev_dbg(&dev->pdev->dev, "ignore netdevice event %ld on %s\n",
event, dev->ib_dev.name); event, dev->ib_dev.name);
...@@ -730,8 +752,11 @@ static void pvrdma_netdevice_event_work(struct work_struct *work) ...@@ -730,8 +752,11 @@ static void pvrdma_netdevice_event_work(struct work_struct *work)
mutex_lock(&pvrdma_device_list_lock); mutex_lock(&pvrdma_device_list_lock);
list_for_each_entry(dev, &pvrdma_device_list, device_link) { list_for_each_entry(dev, &pvrdma_device_list, device_link) {
if (dev->netdev == netdev_work->event_netdev) { if ((netdev_work->event == NETDEV_REGISTER) ||
pvrdma_netdevice_event_handle(dev, netdev_work->event); (dev->netdev == netdev_work->event_netdev)) {
pvrdma_netdevice_event_handle(dev,
netdev_work->event_netdev,
netdev_work->event);
break; break;
} }
} }
...@@ -964,6 +989,7 @@ static int pvrdma_pci_probe(struct pci_dev *pdev, ...@@ -964,6 +989,7 @@ static int pvrdma_pci_probe(struct pci_dev *pdev,
ret = -ENODEV; ret = -ENODEV;
goto err_free_cq_ring; goto err_free_cq_ring;
} }
dev_hold(dev->netdev);
dev_info(&pdev->dev, "paired device to %s\n", dev->netdev->name); dev_info(&pdev->dev, "paired device to %s\n", dev->netdev->name);
...@@ -1036,6 +1062,10 @@ static int pvrdma_pci_probe(struct pci_dev *pdev, ...@@ -1036,6 +1062,10 @@ static int pvrdma_pci_probe(struct pci_dev *pdev,
pvrdma_free_irq(dev); pvrdma_free_irq(dev);
pci_free_irq_vectors(pdev); pci_free_irq_vectors(pdev);
err_free_cq_ring: err_free_cq_ring:
if (dev->netdev) {
dev_put(dev->netdev);
dev->netdev = NULL;
}
pvrdma_page_dir_cleanup(dev, &dev->cq_pdir); pvrdma_page_dir_cleanup(dev, &dev->cq_pdir);
err_free_async_ring: err_free_async_ring:
pvrdma_page_dir_cleanup(dev, &dev->async_pdir); pvrdma_page_dir_cleanup(dev, &dev->async_pdir);
...@@ -1075,6 +1105,11 @@ static void pvrdma_pci_remove(struct pci_dev *pdev) ...@@ -1075,6 +1105,11 @@ static void pvrdma_pci_remove(struct pci_dev *pdev)
flush_workqueue(event_wq); flush_workqueue(event_wq);
if (dev->netdev) {
dev_put(dev->netdev);
dev->netdev = NULL;
}
/* Unregister ib device */ /* Unregister ib device */
ib_unregister_device(&dev->ib_dev); ib_unregister_device(&dev->ib_dev);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment