#include <linux/cpu_rmap.h>
#endif /* CONFIG_RFS_ACCEL */
#include <linux/ethtool.h>
-#include <linux/if_vlan.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/numa.h>
#include <net/ip.h>
#include "ena_netdev.h"
+#include <linux/bpf_trace.h>
#include "ena_pci_id_tbl.h"
static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
static void check_for_admin_com_state(struct ena_adapter *adapter);
static void ena_destroy_device(struct ena_adapter *adapter, bool graceful);
static int ena_restore_device(struct ena_adapter *adapter);
+static void ena_down(struct ena_adapter *adapter);
+static int ena_up(struct ena_adapter *adapter);
static void ena_tx_timeout(struct net_device *dev)
{
return ret;
}
+static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+{
+ struct bpf_prog *xdp_prog;
+ u32 verdict = XDP_PASS;
+
+ rcu_read_lock();
+ xdp_prog = READ_ONCE(rx_ring->xdp_bpf_prog);
+
+ if (!xdp_prog)
+ goto out;
+
+ verdict = bpf_prog_run_xdp(xdp_prog, xdp);
+
+ if (unlikely(verdict == XDP_ABORTED))
+ trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict);
+ else if (unlikely(verdict >= XDP_TX))
+ bpf_warn_invalid_xdp_action(verdict);
+out:
+ rcu_read_unlock();
+ return verdict;
+}
+
+void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter,
+ struct bpf_prog *prog,
+ int first,
+ int count)
+{
+ struct ena_ring *rx_ring;
+ int i = 0;
+
+ for (i = first; i < count; i++) {
+ rx_ring = &adapter->rx_ring[i];
+ xchg(&rx_ring->xdp_bpf_prog, prog);
+ if (prog)
+ rx_ring->rx_headroom = XDP_PACKET_HEADROOM;
+ else
+ rx_ring->rx_headroom = 0;
+ }
+}
+
+void ena_xdp_exchange_program(struct ena_adapter *adapter,
+ struct bpf_prog *prog)
+{
+ struct bpf_prog *old_bpf_prog = xchg(&adapter->xdp_bpf_prog, prog);
+
+ ena_xdp_exchange_program_rx_in_range(adapter,
+ prog,
+ 0,
+ adapter->num_io_queues);
+
+ if (old_bpf_prog)
+ bpf_prog_put(old_bpf_prog);
+}
+
+static int ena_xdp_set(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+ struct ena_adapter *adapter = netdev_priv(netdev);
+ struct bpf_prog *prog = bpf->prog;
+ int rc, prev_mtu;
+ bool is_up;
+
+ is_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
+
+ if (ena_xdp_allowed(adapter)) {
+ if (is_up)
+ ena_down(adapter);
+
+ ena_xdp_exchange_program(adapter, prog);
+
+ prev_mtu = netdev->max_mtu;
+ netdev->max_mtu = prog ? ENA_XDP_MAX_MTU : adapter->max_mtu;
+ if (is_up) {
+ rc = ena_up(adapter);
+ if (rc)
+ return rc;
+ }
+ netif_info(adapter, drv, adapter->netdev, "xdp program set, changging the max_mtu from %d to %d",
+ prev_mtu, netdev->max_mtu);
+
+ } else {
+ netif_err(adapter, drv, adapter->netdev, "Failed to set xdp program, the current MTU (%d) is larger than the maximum allowed MTU (%lu) while xdp is on",
+ netdev->mtu, ENA_XDP_MAX_MTU);
+ NL_SET_ERR_MSG_MOD(bpf->extack, "Failed to set xdp program, the current MTU is larger than the maximum allowed MTU. Check the dmesg for more info");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/* This is the main xdp callback, it's used by the kernel to set/unset the xdp
+ * program as well as to query the current xdp program id.
+ */
+static int ena_xdp(struct net_device *netdev, struct netdev_bpf *bpf)
+{
+ struct ena_adapter *adapter = netdev_priv(netdev);
+
+ switch (bpf->command) {
+ case XDP_SETUP_PROG:
+ return ena_xdp_set(netdev, bpf);
+ case XDP_QUERY_PROG:
+ bpf->prog_id = adapter->xdp_bpf_prog ?
+ adapter->xdp_bpf_prog->aux->id : 0;
+ break;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+}
+
static int ena_init_rx_cpu_rmap(struct ena_adapter *adapter)
{
#ifdef CONFIG_RFS_ACCEL
vfree(rx_ring->free_ids);
rx_ring->free_ids = NULL;
+
+ xdp_rxq_info_unreg_mem_model(&rx_ring->xdp_rxq);
+ xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
}
/* ena_setup_all_rx_resources - allocate I/O Rx queues resources for all queues
rx_info->page = page;
rx_info->page_offset = 0;
ena_buf = &rx_info->ena_buf;
- ena_buf->paddr = dma;
+ ena_buf->paddr = dma + rx_ring->rx_headroom;
ena_buf->len = ENA_PAGE_SIZE;
return 0;
}
}
+int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp)
+{
+ struct ena_rx_buffer *rx_info;
+ int ret;
+
+ rx_info = &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id];
+ xdp->data = page_address(rx_info->page) +
+ rx_info->page_offset + rx_ring->rx_headroom;
+ xdp_set_data_meta_invalid(xdp);
+ xdp->data_hard_start = page_address(rx_info->page);
+ xdp->data_end = xdp->data + rx_ring->ena_bufs[0].len;
+ /* If for some reason we received a bigger packet than
+ * we expect, then we simply drop it
+ */
+ if (unlikely(rx_ring->ena_bufs[0].len > ENA_XDP_MAX_MTU))
+ return XDP_DROP;
+
+ ret = ena_xdp_execute(rx_ring, xdp);
+
+ /* The xdp program might expand the headers */
+ if (ret == XDP_PASS) {
+ rx_info->page_offset = xdp->data - xdp->data_hard_start;
+ rx_ring->ena_bufs[0].len = xdp->data_end - xdp->data;
+ }
+
+ return ret;
+}
/* ena_clean_rx_irq - Cleanup RX irq
* @rx_ring: RX ring to clean
* @napi: napi handler
struct ena_com_rx_ctx ena_rx_ctx;
struct ena_adapter *adapter;
+ int rx_copybreak_pkt = 0;
+ int refill_threshold;
struct sk_buff *skb;
int refill_required;
- int refill_threshold;
- int rc = 0;
+ struct xdp_buff xdp;
int total_len = 0;
- int rx_copybreak_pkt = 0;
+ int xdp_verdict;
+ int rc = 0;
int i;
netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev,
"%s qid %d\n", __func__, rx_ring->qid);
res_budget = budget;
-
+ xdp.rxq = &rx_ring->xdp_rxq;
do {
+ xdp_verdict = XDP_PASS;
+ skb = NULL;
ena_rx_ctx.ena_bufs = rx_ring->ena_bufs;
ena_rx_ctx.max_bufs = rx_ring->sgl_size;
ena_rx_ctx.descs = 0;
rx_ring->qid, ena_rx_ctx.descs, ena_rx_ctx.l3_proto,
ena_rx_ctx.l4_proto, ena_rx_ctx.hash);
+ if (ena_xdp_present_ring(rx_ring))
+ xdp_verdict = ena_xdp_handle_buff(rx_ring, &xdp);
+
/* allocate skb and fill it */
- skb = ena_rx_skb(rx_ring, rx_ring->ena_bufs, ena_rx_ctx.descs,
- &next_to_clean);
+ if (xdp_verdict == XDP_PASS)
+ skb = ena_rx_skb(rx_ring,
+ rx_ring->ena_bufs,
+ ena_rx_ctx.descs,
+ &next_to_clean);
- /* exit if we failed to retrieve a buffer */
if (unlikely(!skb)) {
for (i = 0; i < ena_rx_ctx.descs; i++) {
rx_ring->free_ids[next_to_clean] =
ENA_RX_RING_IDX_NEXT(next_to_clean,
rx_ring->ring_size);
}
+ if (xdp_verdict == XDP_DROP)
+ continue;
break;
}
netif_err(adapter, ifup, adapter->netdev,
"Failed to get RX queue handlers. RX queue num %d rc: %d\n",
qid, rc);
- ena_com_destroy_io_queue(ena_dev, ena_qid);
- return rc;
+ goto err;
}
ena_com_update_numa_node(rx_ring->ena_com_io_cq, ctx.numa_node);
+ rc = xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev, qid);
+
+ if (rc) {
+ netif_err(adapter, ifup, adapter->netdev,
+ "Failed to register xdp rx queue info. RX queue num %d rc: %d\n",
+ qid, rc);
+ goto err;
+ }
+
+ rc = xdp_rxq_info_reg_mem_model(&rx_ring->xdp_rxq, MEM_TYPE_PAGE_SHARED,
+ NULL);
+
+ if (rc) {
+ netif_err(adapter, ifup, adapter->netdev,
+ "Failed to register xdp rx queue info memory model. RX queue num %d rc: %d\n",
+ qid, rc);
+ xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
+ goto err;
+ }
+
+ return rc;
+err:
+ ena_com_destroy_io_queue(ena_dev, ena_qid);
return rc;
}
int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count)
{
struct ena_com_dev *ena_dev = adapter->ena_dev;
+ int prev_channel_count;
bool dev_was_up;
dev_was_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags);
ena_close(adapter->netdev);
+ prev_channel_count = adapter->num_io_queues;
adapter->num_io_queues = new_channel_count;
+
+ if (ena_xdp_present(adapter) && ena_xdp_allowed(adapter)) {
+ if (prev_channel_count > new_channel_count)
+ ena_xdp_exchange_program_rx_in_range(adapter,
+ NULL,
+ new_channel_count,
+ prev_channel_count);
+ else
+ ena_xdp_exchange_program_rx_in_range(adapter,
+ adapter->xdp_bpf_prog,
+ prev_channel_count,
+ new_channel_count);
+ }
+
/* We need to destroy the rss table so that the indirection
* table will be reinitialized by ena_up()
*/
.ndo_change_mtu = ena_change_mtu,
.ndo_set_mac_address = NULL,
.ndo_validate_addr = eth_validate_addr,
+ .ndo_bpf = ena_xdp,
};
static int ena_device_validate_params(struct ena_adapter *adapter,
projects / linux.git / commitdiff