aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
diff options
context:
space:
mode:
authorLibravatar Linus Torvalds <torvalds@linux-foundation.org>2023-02-21 18:24:12 -0800
committerLibravatar Linus Torvalds <torvalds@linux-foundation.org>2023-02-21 18:24:12 -0800
commit5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 (patch)
treecc5c2d0a898769fd59549594fedb3ee6f84e59a0 /drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
downloadlinux-5b7c4cabbb65f5c469464da6c5f614cbd7f730f2.tar.gz
linux-5b7c4cabbb65f5c469464da6c5f614cbd7f730f2.zip
Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted
Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ...
Diffstat (limited to 'drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c')
-rw-r--r--drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c454
1 files changed, 454 insertions, 0 deletions
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
new file mode 100644
index 000000000..051748b99
--- /dev/null
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c
@@ -0,0 +1,454 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2022 NXP
+ */
+#include <linux/filter.h>
+#include <linux/compiler.h>
+#include <linux/bpf_trace.h>
+#include <net/xdp.h>
+#include <net/xdp_sock_drv.h>
+
+#include "dpaa2-eth.h"
+
+static void dpaa2_eth_setup_consume_func(struct dpaa2_eth_priv *priv,
+ struct dpaa2_eth_channel *ch,
+ enum dpaa2_eth_fq_type type,
+ dpaa2_eth_consume_cb_t *consume)
+{
+ struct dpaa2_eth_fq *fq;
+ int i;
+
+ for (i = 0; i < priv->num_fqs; i++) {
+ fq = &priv->fq[i];
+
+ if (fq->type != type)
+ continue;
+ if (fq->channel != ch)
+ continue;
+
+ fq->consume = consume;
+ }
+}
+
+static u32 dpaa2_xsk_run_xdp(struct dpaa2_eth_priv *priv,
+ struct dpaa2_eth_channel *ch,
+ struct dpaa2_eth_fq *rx_fq,
+ struct dpaa2_fd *fd, void *vaddr)
+{
+ dma_addr_t addr = dpaa2_fd_get_addr(fd);
+ struct bpf_prog *xdp_prog;
+ struct xdp_buff *xdp_buff;
+ struct dpaa2_eth_swa *swa;
+ u32 xdp_act = XDP_PASS;
+ int err;
+
+ xdp_prog = READ_ONCE(ch->xdp.prog);
+ if (!xdp_prog)
+ goto out;
+
+ swa = (struct dpaa2_eth_swa *)(vaddr + DPAA2_ETH_RX_HWA_SIZE +
+ ch->xsk_pool->umem->headroom);
+ xdp_buff = swa->xsk.xdp_buff;
+
+ xdp_buff->data_hard_start = vaddr;
+ xdp_buff->data = vaddr + dpaa2_fd_get_offset(fd);
+ xdp_buff->data_end = xdp_buff->data + dpaa2_fd_get_len(fd);
+ xdp_set_data_meta_invalid(xdp_buff);
+ xdp_buff->rxq = &ch->xdp_rxq;
+
+ xsk_buff_dma_sync_for_cpu(xdp_buff, ch->xsk_pool);
+ xdp_act = bpf_prog_run_xdp(xdp_prog, xdp_buff);
+
+ /* xdp.data pointer may have changed */
+ dpaa2_fd_set_offset(fd, xdp_buff->data - vaddr);
+ dpaa2_fd_set_len(fd, xdp_buff->data_end - xdp_buff->data);
+
+ if (likely(xdp_act == XDP_REDIRECT)) {
+ err = xdp_do_redirect(priv->net_dev, xdp_buff, xdp_prog);
+ if (unlikely(err)) {
+ ch->stats.xdp_drop++;
+ dpaa2_eth_recycle_buf(priv, ch, addr);
+ } else {
+ ch->buf_count--;
+ ch->stats.xdp_redirect++;
+ }
+
+ goto xdp_redir;
+ }
+
+ switch (xdp_act) {
+ case XDP_PASS:
+ break;
+ case XDP_TX:
+ dpaa2_eth_xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid);
+ break;
+ default:
+ bpf_warn_invalid_xdp_action(priv->net_dev, xdp_prog, xdp_act);
+ fallthrough;
+ case XDP_ABORTED:
+ trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act);
+ fallthrough;
+ case XDP_DROP:
+ dpaa2_eth_recycle_buf(priv, ch, addr);
+ ch->stats.xdp_drop++;
+ break;
+ }
+
+xdp_redir:
+ ch->xdp.res |= xdp_act;
+out:
+ return xdp_act;
+}
+
+/* Rx frame processing routine for the AF_XDP fast path */
+static void dpaa2_xsk_rx(struct dpaa2_eth_priv *priv,
+ struct dpaa2_eth_channel *ch,
+ const struct dpaa2_fd *fd,
+ struct dpaa2_eth_fq *fq)
+{
+ dma_addr_t addr = dpaa2_fd_get_addr(fd);
+ u8 fd_format = dpaa2_fd_get_format(fd);
+ struct rtnl_link_stats64 *percpu_stats;
+ u32 fd_length = dpaa2_fd_get_len(fd);
+ struct sk_buff *skb;
+ void *vaddr;
+ u32 xdp_act;
+
+ trace_dpaa2_rx_xsk_fd(priv->net_dev, fd);
+
+ vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
+ percpu_stats = this_cpu_ptr(priv->percpu_stats);
+
+ if (fd_format != dpaa2_fd_single) {
+ WARN_ON(priv->xdp_prog);
+ /* AF_XDP doesn't support any other formats */
+ goto err_frame_format;
+ }
+
+ xdp_act = dpaa2_xsk_run_xdp(priv, ch, fq, (struct dpaa2_fd *)fd, vaddr);
+ if (xdp_act != XDP_PASS) {
+ percpu_stats->rx_packets++;
+ percpu_stats->rx_bytes += dpaa2_fd_get_len(fd);
+ return;
+ }
+
+ /* Build skb */
+ skb = dpaa2_eth_alloc_skb(priv, ch, fd, fd_length, vaddr);
+ if (!skb)
+ /* Nothing else we can do, recycle the buffer and
+ * drop the frame.
+ */
+ goto err_alloc_skb;
+
+ /* Send the skb to the Linux networking stack */
+ dpaa2_eth_receive_skb(priv, ch, fd, vaddr, fq, percpu_stats, skb);
+
+ return;
+
+err_alloc_skb:
+ dpaa2_eth_recycle_buf(priv, ch, addr);
+err_frame_format:
+ percpu_stats->rx_dropped++;
+}
+
+static void dpaa2_xsk_set_bp_per_qdbin(struct dpaa2_eth_priv *priv,
+ struct dpni_pools_cfg *pools_params)
+{
+ int curr_bp = 0, i, j;
+
+ pools_params->pool_options = DPNI_POOL_ASSOC_QDBIN;
+ for (i = 0; i < priv->num_bps; i++) {
+ for (j = 0; j < priv->num_channels; j++)
+ if (priv->bp[i] == priv->channel[j]->bp)
+ pools_params->pools[curr_bp].priority_mask |= (1 << j);
+ if (!pools_params->pools[curr_bp].priority_mask)
+ continue;
+
+ pools_params->pools[curr_bp].dpbp_id = priv->bp[i]->bpid;
+ pools_params->pools[curr_bp].buffer_size = priv->rx_buf_size;
+ pools_params->pools[curr_bp++].backup_pool = 0;
+ }
+ pools_params->num_dpbp = curr_bp;
+}
+
+static int dpaa2_xsk_disable_pool(struct net_device *dev, u16 qid)
+{
+ struct xsk_buff_pool *pool = xsk_get_pool_from_qid(dev, qid);
+ struct dpaa2_eth_priv *priv = netdev_priv(dev);
+ struct dpni_pools_cfg pools_params = { 0 };
+ struct dpaa2_eth_channel *ch;
+ int err;
+ bool up;
+
+ ch = priv->channel[qid];
+ if (!ch->xsk_pool)
+ return -EINVAL;
+
+ up = netif_running(dev);
+ if (up)
+ dev_close(dev);
+
+ xsk_pool_dma_unmap(pool, 0);
+ err = xdp_rxq_info_reg_mem_model(&ch->xdp_rxq,
+ MEM_TYPE_PAGE_ORDER0, NULL);
+ if (err)
+ netdev_err(dev, "xsk_rxq_info_reg_mem_model() failed (err = %d)\n",
+ err);
+
+ dpaa2_eth_free_dpbp(priv, ch->bp);
+
+ ch->xsk_zc = false;
+ ch->xsk_pool = NULL;
+ ch->xsk_tx_pkts_sent = 0;
+ ch->bp = priv->bp[DPAA2_ETH_DEFAULT_BP_IDX];
+
+ dpaa2_eth_setup_consume_func(priv, ch, DPAA2_RX_FQ, dpaa2_eth_rx);
+
+ dpaa2_xsk_set_bp_per_qdbin(priv, &pools_params);
+ err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
+ if (err)
+ netdev_err(dev, "dpni_set_pools() failed\n");
+
+ if (up) {
+ err = dev_open(dev, NULL);
+ if (err)
+ return err;
+ }
+
+ return 0;
+}
+
+static int dpaa2_xsk_enable_pool(struct net_device *dev,
+ struct xsk_buff_pool *pool,
+ u16 qid)
+{
+ struct dpaa2_eth_priv *priv = netdev_priv(dev);
+ struct dpni_pools_cfg pools_params = { 0 };
+ struct dpaa2_eth_channel *ch;
+ int err, err2;
+ bool up;
+
+ if (priv->dpni_attrs.wriop_version < DPAA2_WRIOP_VERSION(3, 0, 0)) {
+ netdev_err(dev, "AF_XDP zero-copy not supported on devices <= WRIOP(3, 0, 0)\n");
+ return -EOPNOTSUPP;
+ }
+
+ if (priv->dpni_attrs.num_queues > 8) {
+ netdev_err(dev, "AF_XDP zero-copy not supported on DPNI with more then 8 queues\n");
+ return -EOPNOTSUPP;
+ }
+
+ up = netif_running(dev);
+ if (up)
+ dev_close(dev);
+
+ err = xsk_pool_dma_map(pool, priv->net_dev->dev.parent, 0);
+ if (err) {
+ netdev_err(dev, "xsk_pool_dma_map() failed (err = %d)\n",
+ err);
+ goto err_dma_unmap;
+ }
+
+ ch = priv->channel[qid];
+ err = xdp_rxq_info_reg_mem_model(&ch->xdp_rxq, MEM_TYPE_XSK_BUFF_POOL, NULL);
+ if (err) {
+ netdev_err(dev, "xdp_rxq_info_reg_mem_model() failed (err = %d)\n", err);
+ goto err_mem_model;
+ }
+ xsk_pool_set_rxq_info(pool, &ch->xdp_rxq);
+
+ priv->bp[priv->num_bps] = dpaa2_eth_allocate_dpbp(priv);
+ if (IS_ERR(priv->bp[priv->num_bps])) {
+ err = PTR_ERR(priv->bp[priv->num_bps]);
+ goto err_bp_alloc;
+ }
+ ch->xsk_zc = true;
+ ch->xsk_pool = pool;
+ ch->bp = priv->bp[priv->num_bps++];
+
+ dpaa2_eth_setup_consume_func(priv, ch, DPAA2_RX_FQ, dpaa2_xsk_rx);
+
+ dpaa2_xsk_set_bp_per_qdbin(priv, &pools_params);
+ err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
+ if (err) {
+ netdev_err(dev, "dpni_set_pools() failed\n");
+ goto err_set_pools;
+ }
+
+ if (up) {
+ err = dev_open(dev, NULL);
+ if (err)
+ return err;
+ }
+
+ return 0;
+
+err_set_pools:
+ err2 = dpaa2_xsk_disable_pool(dev, qid);
+ if (err2)
+ netdev_err(dev, "dpaa2_xsk_disable_pool() failed %d\n", err2);
+err_bp_alloc:
+ err2 = xdp_rxq_info_reg_mem_model(&priv->channel[qid]->xdp_rxq,
+ MEM_TYPE_PAGE_ORDER0, NULL);
+ if (err2)
+ netdev_err(dev, "xsk_rxq_info_reg_mem_model() failed with %d)\n", err2);
+err_mem_model:
+ xsk_pool_dma_unmap(pool, 0);
+err_dma_unmap:
+ if (up)
+ dev_open(dev, NULL);
+
+ return err;
+}
+
+int dpaa2_xsk_setup_pool(struct net_device *dev, struct xsk_buff_pool *pool, u16 qid)
+{
+ return pool ? dpaa2_xsk_enable_pool(dev, pool, qid) :
+ dpaa2_xsk_disable_pool(dev, qid);
+}
+
+int dpaa2_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags)
+{
+ struct dpaa2_eth_priv *priv = netdev_priv(dev);
+ struct dpaa2_eth_channel *ch = priv->channel[qid];
+
+ if (!priv->link_state.up)
+ return -ENETDOWN;
+
+ if (!priv->xdp_prog)
+ return -EINVAL;
+
+ if (!ch->xsk_zc)
+ return -EINVAL;
+
+ /* We do not have access to a per channel SW interrupt, so instead we
+ * schedule a NAPI instance.
+ */
+ if (!napi_if_scheduled_mark_missed(&ch->napi))
+ napi_schedule(&ch->napi);
+
+ return 0;
+}
+
+static int dpaa2_xsk_tx_build_fd(struct dpaa2_eth_priv *priv,
+ struct dpaa2_eth_channel *ch,
+ struct dpaa2_fd *fd,
+ struct xdp_desc *xdp_desc)
+{
+ struct device *dev = priv->net_dev->dev.parent;
+ struct dpaa2_sg_entry *sgt;
+ struct dpaa2_eth_swa *swa;
+ void *sgt_buf = NULL;
+ dma_addr_t sgt_addr;
+ int sgt_buf_size;
+ dma_addr_t addr;
+ int err = 0;
+
+ /* Prepare the HW SGT structure */
+ sgt_buf_size = priv->tx_data_offset + sizeof(struct dpaa2_sg_entry);
+ sgt_buf = dpaa2_eth_sgt_get(priv);
+ if (unlikely(!sgt_buf))
+ return -ENOMEM;
+ sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);
+
+ /* Get the address of the XSK Tx buffer */
+ addr = xsk_buff_raw_get_dma(ch->xsk_pool, xdp_desc->addr);
+ xsk_buff_raw_dma_sync_for_device(ch->xsk_pool, addr, xdp_desc->len);
+
+ /* Fill in the HW SGT structure */
+ dpaa2_sg_set_addr(sgt, addr);
+ dpaa2_sg_set_len(sgt, xdp_desc->len);
+ dpaa2_sg_set_final(sgt, true);
+
+ /* Store the necessary info in the SGT buffer */
+ swa = (struct dpaa2_eth_swa *)sgt_buf;
+ swa->type = DPAA2_ETH_SWA_XSK;
+ swa->xsk.sgt_size = sgt_buf_size;
+
+ /* Separately map the SGT buffer */
+ sgt_addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
+ if (unlikely(dma_mapping_error(dev, sgt_addr))) {
+ err = -ENOMEM;
+ goto sgt_map_failed;
+ }
+
+ /* Initialize FD fields */
+ memset(fd, 0, sizeof(struct dpaa2_fd));
+ dpaa2_fd_set_offset(fd, priv->tx_data_offset);
+ dpaa2_fd_set_format(fd, dpaa2_fd_sg);
+ dpaa2_fd_set_addr(fd, sgt_addr);
+ dpaa2_fd_set_len(fd, xdp_desc->len);
+ dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);
+
+ return 0;
+
+sgt_map_failed:
+ dpaa2_eth_sgt_recycle(priv, sgt_buf);
+
+ return err;
+}
+
+bool dpaa2_xsk_tx(struct dpaa2_eth_priv *priv,
+ struct dpaa2_eth_channel *ch)
+{
+ struct xdp_desc *xdp_descs = ch->xsk_pool->tx_descs;
+ struct dpaa2_eth_drv_stats *percpu_extras;
+ struct rtnl_link_stats64 *percpu_stats;
+ int budget = DPAA2_ETH_TX_ZC_PER_NAPI;
+ int total_enqueued, enqueued;
+ int retries, max_retries;
+ struct dpaa2_eth_fq *fq;
+ struct dpaa2_fd *fds;
+ int batch, i, err;
+
+ percpu_stats = this_cpu_ptr(priv->percpu_stats);
+ percpu_extras = this_cpu_ptr(priv->percpu_extras);
+ fds = (this_cpu_ptr(priv->fd))->array;
+
+ /* Use the FQ with the same idx as the affine CPU */
+ fq = &priv->fq[ch->nctx.desired_cpu];
+
+ batch = xsk_tx_peek_release_desc_batch(ch->xsk_pool, budget);
+ if (!batch)
+ return false;
+
+ /* Create a FD for each XSK frame to be sent */
+ for (i = 0; i < batch; i++) {
+ err = dpaa2_xsk_tx_build_fd(priv, ch, &fds[i], &xdp_descs[i]);
+ if (err) {
+ batch = i;
+ break;
+ }
+
+ trace_dpaa2_tx_xsk_fd(priv->net_dev, &fds[i]);
+ }
+
+ /* Enqueue all the created FDs */
+ max_retries = batch * DPAA2_ETH_ENQUEUE_RETRIES;
+ total_enqueued = 0;
+ enqueued = 0;
+ retries = 0;
+ while (total_enqueued < batch && retries < max_retries) {
+ err = priv->enqueue(priv, fq, &fds[total_enqueued], 0,
+ batch - total_enqueued, &enqueued);
+ if (err == -EBUSY) {
+ retries++;
+ continue;
+ }
+
+ total_enqueued += enqueued;
+ }
+ percpu_extras->tx_portal_busy += retries;
+
+ /* Update statistics */
+ percpu_stats->tx_packets += total_enqueued;
+ for (i = 0; i < total_enqueued; i++)
+ percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]);
+ for (i = total_enqueued; i < batch; i++) {
+ dpaa2_eth_free_tx_fd(priv, ch, fq, &fds[i], false);
+ percpu_stats->tx_errors++;
+ }
+
+ xsk_tx_release(ch->xsk_pool);
+
+ return total_enqueued == budget;
+}