On 7/2/07, Christoph Hellwig <[EMAIL PROTECTED]> wrote: > > But that'll require the single receiver queue version I guess. The > netdevice abuse is the only really major issue I see, although I'd of > course really like to see the driver getting rid of the bitfield abuse > aswell.
The submitted driver code supports single queue version in case of MSIX allocation failures... As I said in the other mail, I feel, restricting to single Rx queue in NAPI mode is better approach till Stephen's and DaveM' work of separating NAPI from netdevice is done.. > Lots of drivers where the interface name is assigned after request_irq > just use an internal name, e.g. ixgbeX in the case of this driver. > This sounds ok to me. With this change, this is the output.. [EMAIL PROTECTED] src]# ip link 1: lo: <LOOPBACK,UP,10000> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: sit0: <NOARP> mtu 1480 qdisc noop link/sit 0.0.0.0 brd 0.0.0.0 3: eth6: <BROADCAST,MULTICAST,UP,10000> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:50:8b:05:5f:95 brd ff:ff:ff:ff:ff:ff 29: eth0: <BROADCAST,MULTICAST,UP,10000> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:1b:21:01:e4:93 brd ff:ff:ff:ff:ff:ff 30: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000 link/ether 00:1b:21:01:e4:92 brd ff:ff:ff:ff:ff:ff [EMAIL PROTECTED] src]# cat /proc/interrupts | grep 29 214: 0 0 PCI-MSI-edge ixgbe29-lsc 215: 11764 80213 PCI-MSI-edge ixgbe29-rx7 216: 80257 0 PCI-MSI-edge ixgbe29-rx6 217: 77331 0 PCI-MSI-edge ixgbe29-rx5 218: 24201 0 PCI-MSI-edge ixgbe29-rx4 219: 52911 0 PCI-MSI-edge ixgbe29-rx3 220: 104591 0 PCI-MSI-edge ixgbe29-rx2 221: 80249 8 PCI-MSI-edge ixgbe29-rx1 222: 14 0 PCI-MSI-edge ixgbe29-rx0 223: 194023 118220 PCI-MSI-edge ixgbe29-tx0 Ayyappan - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html