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
