Merge branch 'xdp_rxq_info'

Jesper Dangaard Brouer says:

====================
V4:
* Added reviewers/acks to patches
* Fix patch desc in i40e that got out-of-sync with code
* Add SPDX license headers for the two new files added in patch 14

V3:
* Fixed bug in virtio_net driver
* Removed export of xdp_rxq_info_init()

V2:
* Changed API exposed to drivers
  - Removed invocation of "init" in drivers, and only call "reg"
    (Suggested by Saeed)
  - Allow "reg" to fail and handle this in drivers
    (Suggested by David Ahern)
* Removed the SINKQ qtype, instead allow to register as "unused"
* Also fixed some drivers during testing on actual HW (noted in patches)

There is a need for XDP to know more about the RX-queue a given XDP
frames have arrived on.  For both the XDP bpf-prog and kernel side.

Instead of extending struct xdp_buff each time new info is needed,
this patchset takes a different approach.  Struct xdp_buff is only
extended with a pointer to a struct xdp_rxq_info (allowing for easier
extending this later).  This xdp_rxq_info contains information related
to how the driver have setup the individual RX-queue's.  This is
read-mostly information, and all xdp_buff frames (in drivers
napi_poll) point to the same xdp_rxq_info (per RX-queue).

We stress this data/cache-line is for read-mostly info.  This is NOT
for dynamic per packet info, use the data_meta for such use-cases.

This patchset start out small, and only expose ingress_ifindex and the
RX-queue index to the XDP/BPF program. Access to tangible info like
the ingress ifindex and RX queue index, is fairly easy to comprehent.
The other future use-cases could allow XDP frames to be recycled back
to the originating device driver, by providing info on RX device and
queue number.

As XDP doesn't have driver feature flags, and eBPF code due to
bpf-tail-calls cannot determine that XDP driver invoke it, this
patchset have to update every driver that support XDP.

For driver developers (review individual driver patches!):

The xdp_rxq_info is tied to the drivers RX-ring(s). Whenever a RX-ring
modification require (temporary) stopping RX frames, then the
xdp_rxq_info should (likely) also be unregistred and re-registered,
especially if reallocating the pages in the ring. Make sure ethtool
set_channels does the right thing. When replacing XDP prog, if and
only if RX-ring need to be changed, then also re-register the
xdp_rxq_info.

I'm Cc'ing the individual driver patches to the registered maintainers.

Testing:

I've only tested the NIC drivers I have hardware for.  The general
test procedure is to (DUT = Device Under Test):
 (1) run pktgen script pktgen_sample04_many_flows.sh       (against DUT)
 (2) run samples/bpf program xdp_rxq_info --dev $DEV       (on DUT)
 (3) runtime modify number of NIC queues via ethtool -L    (on DUT)
 (4) runtime modify number of NIC ring-size via ethtool -G (on DUT)

Patch based on git tree bpf-next (at commit fb982666e380c1632a):
 https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

6 files changed, 29 insertions, 5 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 21bc17fa3854..8fc51bc29003 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2172,8 +2172,9 @@ static int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
 
 		if (mlx4_en_create_rx_ring(priv, &priv->rx_ring[i],
 					   prof->rx_ring_size, priv->stride,
-					   node))
+					   node, i))
 			goto err;
+
 	}
 
 #ifdef CONFIG_RFS_ACCEL
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 5f9dbc9a7f5b..b4d144e67514 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -262,7 +262,7 @@ void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev)
 
 int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_rx_ring **pring,
-			   u32 size, u16 stride, int node)
+			   u32 size, u16 stride, int node, int queue_index)
 {
 	struct mlx4_en_dev *mdev = priv->mdev;
 	struct mlx4_en_rx_ring *ring;
@@ -286,6 +286,9 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 	ring->log_stride = ffs(ring->stride) - 1;
 	ring->buf_size = ring->size * ring->stride + TXBB_SIZE;
 
+	if (xdp_rxq_info_reg(&ring->xdp_rxq, priv->dev, queue_index) < 0)
+		goto err_ring;
+
 	tmp = size * roundup_pow_of_two(MLX4_EN_MAX_RX_FRAGS *
 					sizeof(struct mlx4_en_rx_alloc));
 	ring->rx_info = vzalloc_node(tmp, node);
@@ -293,7 +296,7 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 		ring->rx_info = vzalloc(tmp);
 		if (!ring->rx_info) {
 			err = -ENOMEM;
-			goto err_ring;
+			goto err_xdp_info;
 		}
 	}
 
@@ -317,6 +320,8 @@ int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 err_info:
 	vfree(ring->rx_info);
 	ring->rx_info = NULL;
+err_xdp_info:
+	xdp_rxq_info_unreg(&ring->xdp_rxq);
 err_ring:
 	kfree(ring);
 	*pring = NULL;
@@ -440,6 +445,7 @@ void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv,
 					lockdep_is_held(&mdev->state_lock));
 	if (old_prog)
 		bpf_prog_put(old_prog);
+	xdp_rxq_info_unreg(&ring->xdp_rxq);
 	mlx4_free_hwq_res(mdev->dev, &ring->wqres, size * stride + TXBB_SIZE);
 	vfree(ring->rx_info);
 	ring->rx_info = NULL;
@@ -652,6 +658,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 	int cq_ring = cq->ring;
 	bool doorbell_pending;
 	struct mlx4_cqe *cqe;
+	struct xdp_buff xdp;
 	int polled = 0;
 	int index;
 
@@ -666,6 +673,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 	/* Protect accesses to: ring->xdp_prog, priv->mac_hash list */
 	rcu_read_lock();
 	xdp_prog = rcu_dereference(ring->xdp_prog);
+	xdp.rxq = &ring->xdp_rxq;
 	doorbell_pending = 0;
 
 	/* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx
@@ -750,7 +758,6 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 		 * read bytes but not past the end of the frag.
 		 */
 		if (xdp_prog) {
-			struct xdp_buff xdp;
 			dma_addr_t dma;
 			void *orig_data;
 			u32 act;
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 7db3d0d9bfce..f470ae37d937 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -46,6 +46,7 @@
 #endif
 #include <linux/cpu_rmap.h>
 #include <linux/ptp_clock_kernel.h>
+#include <net/xdp.h>
 
 #include <linux/mlx4/device.h>
 #include <linux/mlx4/qp.h>
@@ -356,6 +357,7 @@ struct mlx4_en_rx_ring {
 	unsigned long dropped;
 	int hwtstamp_rx_filter;
 	cpumask_var_t affinity_mask;
+	struct xdp_rxq_info xdp_rxq;
 };
 
 struct mlx4_en_cq {
@@ -720,7 +722,7 @@ void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev);
 void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv);
 int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_rx_ring **pring,
-			   u32 size, u16 stride, int node);
+			   u32 size, u16 stride, int node, int queue_index);
 void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv,
 			     struct mlx4_en_rx_ring **pring,
 			     u32 size, u16 stride);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 543060c305a0..5299310f2481 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -46,6 +46,7 @@
 #include <linux/mlx5/transobj.h>
 #include <linux/rhashtable.h>
 #include <net/switchdev.h>
+#include <net/xdp.h>
 #include "wq.h"
 #include "mlx5_core.h"
 #include "en_stats.h"
@@ -571,6 +572,9 @@ struct mlx5e_rq {
 	u32                    rqn;
 	struct mlx5_core_dev  *mdev;
 	struct mlx5_core_mkey  umr_mkey;
+
+	/* XDP read-mostly */
+	struct xdp_rxq_info    xdp_rxq;
 } ____cacheline_aligned_in_smp;
 
 struct mlx5e_channel {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 3aa1c90e7c86..539bd1d24396 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -582,6 +582,9 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c,
 		goto err_rq_wq_destroy;
 	}
 
+	if (xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq->ix) < 0)
+		goto err_rq_wq_destroy;
+
 	rq->buff.map_dir = rq->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
 	rq->buff.headroom = params->rq_headroom;
 
@@ -687,6 +690,7 @@ err_destroy_umr_mkey:
 err_rq_wq_destroy:
 	if (rq->xdp_prog)
 		bpf_prog_put(rq->xdp_prog);
+	xdp_rxq_info_unreg(&rq->xdp_rxq);
 	mlx5_wq_destroy(&rq->wq_ctrl);
 
 	return err;
@@ -699,6 +703,8 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq)
 	if (rq->xdp_prog)
 		bpf_prog_put(rq->xdp_prog);
 
+	xdp_rxq_info_unreg(&rq->xdp_rxq);
+
 	switch (rq->wq_type) {
 	case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
 		mlx5e_rq_free_mpwqe_info(rq);
@@ -2766,6 +2772,9 @@ static int mlx5e_alloc_drop_rq(struct mlx5_core_dev *mdev,
 	if (err)
 		return err;
 
+	/* Mark as unused given "Drop-RQ" packets never reach XDP */
+	xdp_rxq_info_unused(&rq->xdp_rxq);
+
 	rq->mdev = mdev;
 
 	return 0;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 5b499c7a698f..7b38480811d4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -812,6 +812,7 @@ static inline int mlx5e_xdp_handle(struct mlx5e_rq *rq,
 	xdp_set_data_meta_invalid(&xdp);
 	xdp.data_end = xdp.data + *len;
 	xdp.data_hard_start = va;
+	xdp.rxq = &rq->xdp_rxq;
 
 	act = bpf_prog_run_xdp(prog, &xdp);
 	switch (act) {