On 17-04-24 12:29 PM, Jesper Dangaard Brouer wrote:
> On Sun, 23 Apr 2017 18:31:36 -0700
> John Fastabend <john.fastab...@gmail.com> wrote:
>
>> +static int ixgbe_xmit_xdp_ring(struct ixgbe_adapter *adapter,
>> + struct xdp_buff *xdp)
>> +{
>> + struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
>
> I was about to question whether is it always true that the array size
> can match the number of CPUs in the system, but I can see later in
> ixgbe_xdp_setup() that you reject XDP program if the system have more
> CPUs that MAX_XDP_QUEUES.
Yep.
[...]
>> +
>> + tx_buffer->next_to_watch = tx_desc;
>> + ring->next_to_use = i;
>> +
>> + writel(i, ring->tail);
>
> A tailptr write for every XDP_TX packet is not going be fast, but you
> already mentioned that this is not optimal yet, so I guess you are aware.
>
There is another patch on Jeff's tree to only kick the tail ptr once per
receive path invocation.
https://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue.git/commit/?h=dev-queue&id=24043a662d11e048de903e12bf86059844c207e2
That patch brings packet rates up to near line rate @ 64 bytes.
Thanks,
John
