diff options
Diffstat (limited to 'drivers/net/ethernet/intel/ice/ice_txrx.c')
-rw-r--r-- | drivers/net/ethernet/intel/ice/ice_txrx.c | 151 |
1 files changed, 102 insertions, 49 deletions
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index bc3ba19dc88f..dbe80e5053a8 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -3,10 +3,12 @@ /* The driver transmit and receive code */ -#include <linux/prefetch.h> #include <linux/mm.h> +#include <linux/netdevice.h> +#include <linux/prefetch.h> #include <linux/bpf_trace.h> #include <net/dsfield.h> +#include <net/mpls.h> #include <net/xdp.h> #include "ice_txrx_lib.h" #include "ice_lib.h" @@ -172,6 +174,8 @@ tx_skip_free: tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; + tx_ring->next_dd = ICE_RING_QUARTER(tx_ring) - 1; + tx_ring->next_rs = ICE_RING_QUARTER(tx_ring) - 1; if (!tx_ring->netdev) return; @@ -219,6 +223,9 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget) struct ice_tx_desc *tx_desc; struct ice_tx_buf *tx_buf; + /* get the bql data ready */ + netdev_txq_bql_complete_prefetchw(txring_txq(tx_ring)); + tx_buf = &tx_ring->tx_buf[i]; tx_desc = ICE_TX_DESC(tx_ring, i); i -= tx_ring->count; @@ -232,6 +239,9 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget) if (!eop_desc) break; + /* follow the guidelines of other drivers */ + prefetchw(&tx_buf->skb->users); + smp_rmb(); /* prevent any other reads prior to eop_desc */ ice_trace(clean_tx_irq, tx_ring, tx_desc, tx_buf); @@ -303,9 +313,7 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget) tx_ring->next_to_clean = i; ice_update_tx_ring_stats(tx_ring, total_pkts, total_bytes); - - netdev_tx_completed_queue(txring_txq(tx_ring), total_pkts, - total_bytes); + netdev_tx_completed_queue(txring_txq(tx_ring), total_pkts, total_bytes); #define TX_WAKE_THRESHOLD ((s16)(DESC_NEEDED * 2)) if (unlikely(total_pkts && netif_carrier_ok(tx_ring->netdev) && @@ -314,11 +322,9 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget) * sees the new next_to_clean. */ smp_mb(); - if (__netif_subqueue_stopped(tx_ring->netdev, - tx_ring->q_index) && + if (netif_tx_queue_stopped(txring_txq(tx_ring)) && !test_bit(ICE_VSI_DOWN, vsi->state)) { - netif_wake_subqueue(tx_ring->netdev, - tx_ring->q_index); + netif_tx_wake_queue(txring_txq(tx_ring)); ++tx_ring->tx_stats.restart_q; } } @@ -419,7 +425,10 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) } rx_skip_free: - memset(rx_ring->rx_buf, 0, sizeof(*rx_ring->rx_buf) * rx_ring->count); + if (rx_ring->xsk_pool) + memset(rx_ring->xdp_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->xdp_buf))); + else + memset(rx_ring->rx_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->rx_buf))); /* Zero out the descriptor ring */ size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), @@ -446,8 +455,13 @@ void ice_free_rx_ring(struct ice_rx_ring *rx_ring) if (xdp_rxq_info_is_reg(&rx_ring->xdp_rxq)) xdp_rxq_info_unreg(&rx_ring->xdp_rxq); rx_ring->xdp_prog = NULL; - devm_kfree(rx_ring->dev, rx_ring->rx_buf); - rx_ring->rx_buf = NULL; + if (rx_ring->xsk_pool) { + kfree(rx_ring->xdp_buf); + rx_ring->xdp_buf = NULL; + } else { + kfree(rx_ring->rx_buf); + rx_ring->rx_buf = NULL; + } if (rx_ring->desc) { size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), @@ -475,8 +489,7 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) /* warn if we are about to overwrite the pointer */ WARN_ON(rx_ring->rx_buf); rx_ring->rx_buf = - devm_kcalloc(dev, sizeof(*rx_ring->rx_buf), rx_ring->count, - GFP_KERNEL); + kcalloc(rx_ring->count, sizeof(*rx_ring->rx_buf), GFP_KERNEL); if (!rx_ring->rx_buf) return -ENOMEM; @@ -505,7 +518,7 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) return 0; err: - devm_kfree(dev, rx_ring->rx_buf); + kfree(rx_ring->rx_buf); rx_ring->rx_buf = NULL; return -ENOMEM; } @@ -561,7 +574,7 @@ ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, goto out_failure; return ICE_XDP_REDIR; default: - bpf_warn_invalid_xdp_action(act); + bpf_warn_invalid_xdp_action(rx_ring->netdev, xdp_prog, act); fallthrough; case XDP_ABORTED: out_failure: @@ -597,7 +610,7 @@ ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, if (test_bit(ICE_VSI_DOWN, vsi->state)) return -ENETDOWN; - if (!ice_is_xdp_ena_vsi(vsi) || queue_index >= vsi->num_xdp_txq) + if (!ice_is_xdp_ena_vsi(vsi)) return -ENXIO; if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) @@ -608,6 +621,9 @@ ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, xdp_ring = vsi->xdp_rings[queue_index]; spin_lock(&xdp_ring->tx_lock); } else { + /* Generally, should not happen */ + if (unlikely(queue_index >= vsi->num_xdp_txq)) + return -ENXIO; xdp_ring = vsi->xdp_rings[queue_index]; } @@ -933,7 +949,7 @@ ice_build_skb(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf, */ net_prefetch(xdp->data_meta); /* build an skb around the page buffer */ - skb = build_skb(xdp->data_hard_start, truesize); + skb = napi_build_skb(xdp->data_hard_start, truesize); if (unlikely(!skb)) return NULL; @@ -968,15 +984,17 @@ static struct sk_buff * ice_construct_skb(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf, struct xdp_buff *xdp) { + unsigned int metasize = xdp->data - xdp->data_meta; unsigned int size = xdp->data_end - xdp->data; unsigned int headlen; struct sk_buff *skb; /* prefetch first cache line of first page */ - net_prefetch(xdp->data); + net_prefetch(xdp->data_meta); /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE, + skb = __napi_alloc_skb(&rx_ring->q_vector->napi, + ICE_RX_HDR_SIZE + metasize, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) return NULL; @@ -988,8 +1006,13 @@ ice_construct_skb(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf, headlen = eth_get_headlen(skb->dev, xdp->data, ICE_RX_HDR_SIZE); /* align pull length to size of long to optimize memcpy performance */ - memcpy(__skb_put(skb, headlen), xdp->data, ALIGN(headlen, - sizeof(long))); + memcpy(__skb_put(skb, headlen + metasize), xdp->data_meta, + ALIGN(headlen + metasize, sizeof(long))); + + if (metasize) { + skb_metadata_set(skb, metasize); + __skb_pull(skb, metasize); + } /* if we exhaust the linear part then add what is left as a frag */ size -= headlen; @@ -1065,7 +1088,7 @@ ice_is_non_eop(struct ice_rx_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc) { /* if we are the last buffer then there is nothing else to do */ #define ICE_RXD_EOF BIT(ICE_RX_FLEX_DESC_STATUS0_EOF_S) - if (likely(ice_test_staterr(rx_desc, ICE_RXD_EOF))) + if (likely(ice_test_staterr(rx_desc->wb.status_error0, ICE_RXD_EOF))) return false; rx_ring->rx_stats.non_eop_descs++; @@ -1127,7 +1150,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) * hardware wrote DD then it will be non-zero */ stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_DD_S); - if (!ice_test_staterr(rx_desc, stat_err_bits)) + if (!ice_test_staterr(rx_desc->wb.status_error0, stat_err_bits)) break; /* This memory barrier is needed to keep us from reading @@ -1141,7 +1164,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) struct ice_vsi *ctrl_vsi = rx_ring->vsi; if (rx_desc->wb.rxdid == FDIR_DESC_RXDID && - ctrl_vsi->vf_id != ICE_INVAL_VFID) + ctrl_vsi->vf) ice_vc_fdir_irq_handler(ctrl_vsi, rx_desc); ice_put_rx_buf(rx_ring, NULL, 0); cleaned_count++; @@ -1213,14 +1236,13 @@ construct_skb: continue; stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); - if (unlikely(ice_test_staterr(rx_desc, stat_err_bits))) { + if (unlikely(ice_test_staterr(rx_desc->wb.status_error0, + stat_err_bits))) { dev_kfree_skb_any(skb); continue; } - stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_L2TAG1P_S); - if (ice_test_staterr(rx_desc, stat_err_bits)) - vlan_tag = le16_to_cpu(rx_desc->wb.l2tag1); + vlan_tag = ice_get_vlan_tag_from_rx_desc(rx_desc); /* pad the skb if needed, to make a valid ethernet frame */ if (eth_skb_pad(skb)) { @@ -1445,7 +1467,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) bool wd; if (tx_ring->xsk_pool) - wd = ice_clean_tx_irq_zc(tx_ring, budget); + wd = ice_xmit_zc(tx_ring); else if (ice_ring_is_xdp(tx_ring)) wd = true; else @@ -1498,7 +1520,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) /* Exit the polling mode, but don't re-enable interrupts if stack might * poll us due to busy-polling */ - if (likely(napi_complete_done(napi, work_done))) { + if (napi_complete_done(napi, work_done)) { ice_net_dim(q_vector); ice_enable_interrupt(q_vector); } else { @@ -1517,7 +1539,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) */ static int __ice_maybe_stop_tx(struct ice_tx_ring *tx_ring, unsigned int size) { - netif_stop_subqueue(tx_ring->netdev, tx_ring->q_index); + netif_tx_stop_queue(txring_txq(tx_ring)); /* Memory barrier before checking head and tail */ smp_mb(); @@ -1525,8 +1547,8 @@ static int __ice_maybe_stop_tx(struct ice_tx_ring *tx_ring, unsigned int size) if (likely(ICE_DESC_UNUSED(tx_ring) < size)) return -EBUSY; - /* A reprieve! - use start_subqueue because it doesn't call schedule */ - netif_start_subqueue(tx_ring->netdev, tx_ring->q_index); + /* A reprieve! - use start_queue because it doesn't call schedule */ + netif_tx_start_queue(txring_txq(tx_ring)); ++tx_ring->tx_stats.restart_q; return 0; } @@ -1568,6 +1590,7 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first, struct sk_buff *skb; skb_frag_t *frag; dma_addr_t dma; + bool kick; td_tag = off->td_l2tag1; td_cmd = off->td_cmd; @@ -1649,9 +1672,6 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first, tx_buf = &tx_ring->tx_buf[i]; } - /* record bytecount for BQL */ - netdev_tx_sent_queue(txring_txq(tx_ring), first->bytecount); - /* record SW timestamp if HW timestamp is not available */ skb_tx_timestamp(first->skb); @@ -1680,7 +1700,10 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first, ice_maybe_stop_tx(tx_ring, DESC_NEEDED); /* notify HW of packet */ - if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more()) + kick = __netdev_tx_sent_queue(txring_txq(tx_ring), first->bytecount, + netdev_xmit_more()); + if (kick) + /* notify HW of packet */ writel(i, tx_ring->tail); return; @@ -1729,18 +1752,26 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off) if (skb->ip_summed != CHECKSUM_PARTIAL) return 0; - ip.hdr = skb_network_header(skb); - l4.hdr = skb_transport_header(skb); + protocol = vlan_get_protocol(skb); + + if (eth_p_mpls(protocol)) { + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_checksum_start(skb); + } else { + ip.hdr = skb_network_header(skb); + l4.hdr = skb_transport_header(skb); + } /* compute outer L2 header size */ l2_len = ip.hdr - skb->data; offset = (l2_len / 2) << ICE_TX_DESC_LEN_MACLEN_S; - protocol = vlan_get_protocol(skb); - - if (protocol == htons(ETH_P_IP)) + /* set the tx_flags to indicate the IP protocol type. this is + * required so that checksum header computation below is accurate. + */ + if (ip.v4->version == 4) first->tx_flags |= ICE_TX_FLAGS_IPV4; - else if (protocol == htons(ETH_P_IPV6)) + else if (ip.v6->version == 6) first->tx_flags |= ICE_TX_FLAGS_IPV6; if (skb->encapsulation) { @@ -1901,12 +1932,16 @@ ice_tx_prepare_vlan_flags(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first) if (!skb_vlan_tag_present(skb) && eth_type_vlan(skb->protocol)) return; - /* currently, we always assume 802.1Q for VLAN insertion as VLAN - * insertion for 802.1AD is not supported + /* the VLAN ethertype/tpid is determined by VSI configuration and netdev + * feature flags, which the driver only allows either 802.1Q or 802.1ad + * VLAN offloads exclusively so we only care about the VLAN ID here */ if (skb_vlan_tag_present(skb)) { first->tx_flags |= skb_vlan_tag_get(skb) << ICE_TX_FLAGS_VLAN_S; - first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; + if (tx_ring->flags & ICE_TX_FLAGS_RING_VLAN_L2TAG2) + first->tx_flags |= ICE_TX_FLAGS_HW_OUTER_SINGLE_VLAN; + else + first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; } ice_tx_prepare_vlan_flags_dcb(tx_ring, first); @@ -1934,6 +1969,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) unsigned char *hdr; } l4; u64 cd_mss, cd_tso_len; + __be16 protocol; u32 paylen; u8 l4_start; int err; @@ -1949,8 +1985,13 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) return err; /* cppcheck-suppress unreadVariable */ - ip.hdr = skb_network_header(skb); - l4.hdr = skb_transport_header(skb); + protocol = vlan_get_protocol(skb); + + if (eth_p_mpls(protocol)) + ip.hdr = skb_inner_network_header(skb); + else + ip.hdr = skb_network_header(skb); + l4.hdr = skb_checksum_start(skb); /* initialize outer IP header fields */ if (ip.v4->version == 4) { @@ -2217,8 +2258,10 @@ ice_tstamp(struct ice_tx_ring *tx_ring, struct sk_buff *skb, /* Grab an open timestamp slot */ idx = ice_ptp_request_ts(tx_ring->tx_tstamps, skb); - if (idx < 0) + if (idx < 0) { + tx_ring->vsi->back->ptp.tx_hwtstamp_skipped++; return; + } off->cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX | (ICE_TX_CTX_DESC_TSYN << ICE_TXD_CTX_QW1_CMD_S) | @@ -2265,6 +2308,9 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_tx_ring *tx_ring) return NETDEV_TX_BUSY; } + /* prefetch for bql data which is infrequently used */ + netdev_txq_bql_enqueue_prefetchw(txring_txq(tx_ring)); + offload.tx_ring = tx_ring; /* record the location of the first descriptor for this packet */ @@ -2276,6 +2322,13 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_tx_ring *tx_ring) /* prepare the VLAN tagging flags for Tx */ ice_tx_prepare_vlan_flags(tx_ring, first); + if (first->tx_flags & ICE_TX_FLAGS_HW_OUTER_SINGLE_VLAN) { + offload.cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX | + (ICE_TX_CTX_DESC_IL2TAG2 << + ICE_TXD_CTX_QW1_CMD_S)); + offload.cd_l2tag2 = (first->tx_flags & ICE_TX_FLAGS_VLAN_M) >> + ICE_TX_FLAGS_VLAN_S; + } /* set up TSO offload */ tso = ice_tso(first, &offload); |