Hello.
On 03/13/2014 12:44 PM, Arnd Bergmann wrote:
Since commit d25f06ea466e "vmxnet3: fix netpoll race condition",
the vmxnet3 driver fails to build when CONFIG_PCI_MSI is disabled,
because it unconditionally references the vmxnet3_msix_rx()
function.
To fix this, use the same #ifdef in the caller that exists around
the function definition.
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
Cc: Neil Horman <nhorman@xxxxxxxxxxxxx>
Cc: Shreyas Bhatewara <sbhatewara@xxxxxxxxxx>
Cc: "VMware, Inc." <pv-drivers@xxxxxxxxxx>
Cc: "David S. Miller" <davem@xxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
---
Found this during randconfig testing on ARM. Most of the time when
I report network driver problems, they get fixed in the netdev tree
before I even find them, but since this is for a patch marked "stable",
I made a proper patch anyway.
Please ignore if this is already a known problem, otherwise make sure
the original patch doesn't get backported without addressing this
issue first.
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index cbd898f..28965ad 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1761,13 +1761,16 @@ static void
vmxnet3_netpoll(struct net_device *netdev)
{
struct vmxnet3_adapter *adapter = netdev_priv(netdev);
- int i;
switch (adapter->intr.type) {
- case VMXNET3_IT_MSIX:
+#ifdef CONFIG_PCI_MSI
+ case VMXNET3_IT_MSIX: {
+ int i;
This file's coding style assumes empty line after declaration, so does the
networking coding style in general.
for (i = 0; i < adapter->num_rx_queues; i++)
vmxnet3_msix_rx(0, &adapter->rx_queue[i]);
break;
+ }
+#endif
case VMXNET3_IT_MSI:
default:
vmxnet3_intr(0, adapter->netdev);
WBR, Sergei
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html