[bpf-next,V1-RFC,03/14] i40e: setup xdp_rxq_info

The i40e driver have a special "FDIR" RX-ring (I40E_VSI_FDIR) which is
a sideband channel for configuring/updating the flow director tables.
This (i40e_vsi_)type does not invoke XDP-ebpf code. Thus, mark this
type as a XDP RX-queue type RXQ_TYPE_SINK.

Driver hook points for xdp_rxq_info:
 * init+reg: i40e_setup_rx_descriptors (via i40e_vsi_setup_rx_resources)
 * unreg   : i40e_free_rx_resources    (via i40e_vsi_free_rx_resources)

Tested on actual hardware with samples/bpf program.

Cc: intel-wired-lan@lists.osuosl.org
Cc: Björn Töpel <bjorn.topel@intel.com>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c |   18 +++++++++++++++++-
 drivers/net/ethernet/intel/i40e/i40e_txrx.h |    3 +++
 2 files changed, 20 insertions(+), 1 deletion(-)

2017-12-13 12:19 GMT+01:00 Jesper Dangaard Brouer <brouer@redhat.com>:
> The i40e driver have a special "FDIR" RX-ring (I40E_VSI_FDIR) which is
> a sideband channel for configuring/updating the flow director tables.
> This (i40e_vsi_)type does not invoke XDP-ebpf code. Thus, mark this
> type as a XDP RX-queue type RXQ_TYPE_SINK.
>
> Driver hook points for xdp_rxq_info:
>  * init+reg: i40e_setup_rx_descriptors (via i40e_vsi_setup_rx_resources)
>  * unreg   : i40e_free_rx_resources    (via i40e_vsi_free_rx_resources)
>

Firstly, thanks for working on this, Jesper. This is very valuable to
the AF_PACKET V4^W^WAF_XDP work.

> Tested on actual hardware with samples/bpf program.
>
> Cc: intel-wired-lan@lists.osuosl.org
> Cc: Björn Töpel <bjorn.topel@intel.com>
> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
> ---
>  drivers/net/ethernet/intel/i40e/i40e_txrx.c |   18 +++++++++++++++++-
>  drivers/net/ethernet/intel/i40e/i40e_txrx.h |    3 +++
>  2 files changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> index 4566d66ffc7c..d1bfc4232ca9 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> @@ -27,6 +27,7 @@
>  #include <linux/prefetch.h>
>  #include <net/busy_poll.h>
>  #include <linux/bpf_trace.h>
> +#include <net/xdp.h>
>  #include "i40e.h"
>  #include "i40e_trace.h"
>  #include "i40e_prototype.h"
> @@ -1236,6 +1237,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring)
>  void i40e_free_rx_resources(struct i40e_ring *rx_ring)
>  {
>         i40e_clean_rx_ring(rx_ring);
> +       xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
>         rx_ring->xdp_prog = NULL;
>         kfree(rx_ring->rx_bi);
>         rx_ring->rx_bi = NULL;
> @@ -1283,6 +1285,18 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring)
>         rx_ring->next_to_clean = 0;
>         rx_ring->next_to_use = 0;
>
> +       /* XDP RX-queue info */
> +       xdp_rxq_info_init(&rx_ring->xdp_rxq);
> +
> +       /* Flow director side channel does not invoke XDP/bpf */
> +       if (rx_ring->vsi->type == I40E_VSI_FDIR)
> +               xdp_rxq_info_type(&rx_ring->xdp_rxq, RXQ_TYPE_SINK);

For me, it doesn't make sense to expose the FD Rx ring outside the
scope of i40e driver. Check against VSI_MAIN in setup/cleanup so the
FD ring is not exposed outside the driver. Maybe you can get rid of
the RXQ_TYPE_SINK as well then.

> +       else
> +               rx_ring->xdp_rxq.dev = rx_ring->netdev;
> +
> +       rx_ring->xdp_rxq.queue_index = rx_ring->queue_index;
> +       xdp_rxq_info_reg(&rx_ring->xdp_rxq);
> +
>         rx_ring->xdp_prog = rx_ring->vsi->xdp_prog;
>
>         return 0;
> @@ -2068,11 +2082,13 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
>         struct sk_buff *skb = rx_ring->skb;
>         u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
>         bool failure = false, xdp_xmit = false;
> +       struct xdp_buff xdp;
> +
> +       xdp.rxq = &rx_ring->xdp_rxq;
>
>         while (likely(total_rx_packets < (unsigned int)budget)) {
>                 struct i40e_rx_buffer *rx_buffer;
>                 union i40e_rx_desc *rx_desc;
> -               struct xdp_buff xdp;
>                 unsigned int size;
>                 u16 vlan_tag;
>                 u8 rx_ptype;
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
> index fbae1182e2ea..2d08760fc4ce 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
> @@ -27,6 +27,8 @@
>  #ifndef _I40E_TXRX_H_
>  #define _I40E_TXRX_H_
>
> +#include <net/xdp.h>
> +
>  /* Interrupt Throttling and Rate Limiting Goodies */
>
>  #define I40E_MAX_ITR               0x0FF0  /* reg uses 2 usec resolution */
> @@ -428,6 +430,7 @@ struct i40e_ring {
>                                          */
>
>         struct i40e_channel *ch;
> +       struct xdp_rxq_info xdp_rxq;
>  } ____cacheline_internodealigned_in_smp;
>
>  static inline bool ring_uses_build_skb(struct i40e_ring *ring)
>
> _______________________________________________
> Intel-wired-lan mailing list
> Intel-wired-lan@osuosl.org
> https://lists.osuosl.org/mailman/listinfo/intel-wired-lan

On Mon, 18 Dec 2017 11:52:08 +0100
Björn Töpel <bjorn.topel@gmail.com> wrote:

> > +       /* Flow director side channel does not invoke XDP/bpf */
> > +       if (rx_ring->vsi->type == I40E_VSI_FDIR)
> > +               xdp_rxq_info_type(&rx_ring->xdp_rxq, RXQ_TYPE_SINK);  
> 
> For me, it doesn't make sense to expose the FD Rx ring outside the
> scope of i40e driver. Check against VSI_MAIN in setup/cleanup so the
> FD ring is not exposed outside the driver. Maybe you can get rid of
> the RXQ_TYPE_SINK as well then.

Okay, took this advice and changed patch to instead check against
I40E_VSI_MAIN.