diff options
Diffstat (limited to 'drivers/net/ethernet/freescale/dpaa/dpaa_eth.c')
| -rw-r--r-- | drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 937 |
1 files changed, 700 insertions, 237 deletions
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index 2972244e6eb0..3edc8d142dd5 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -1,39 +1,14 @@ -/* Copyright 2008 - 2016 Freescale Semiconductor Inc. +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later +/* + * Copyright 2008 - 2016 Freescale Semiconductor Inc. * Copyright 2020 NXP - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions are met: - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * * Neither the name of Freescale Semiconductor nor the - * names of its contributors may be used to endorse or promote products - * derived from this software without specific prior written permission. - * - * ALTERNATIVELY, this software may be distributed under the terms of the - * GNU General Public License ("GPL") as published by the Free Software - * Foundation, either version 2 of that License or (at your option) any - * later version. - * - * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY - * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED - * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE - * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY - * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES - * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; - * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND - * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS - * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/init.h> +#include <linux/mod_devicetable.h> #include <linux/module.h> -#include <linux/of_platform.h> #include <linux/of_mdio.h> #include <linux/of_net.h> #include <linux/io.h> @@ -42,6 +17,7 @@ #include <linux/icmp.h> #include <linux/ip.h> #include <linux/ipv6.h> +#include <linux/platform_device.h> #include <linux/udp.h> #include <linux/tcp.h> #include <linux/net.h> @@ -52,7 +28,8 @@ #include <linux/percpu.h> #include <linux/dma-mapping.h> #include <linux/sort.h> -#include <linux/phy_fixed.h> +#include <linux/bpf.h> +#include <linux/bpf_trace.h> #include <soc/fsl/bman.h> #include <soc/fsl/qman.h> #include "fman.h" @@ -87,7 +64,7 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define DPAA_MSG_DEFAULT (NETIF_MSG_DRV | NETIF_MSG_PROBE | \ NETIF_MSG_LINK | NETIF_MSG_IFUP | \ - NETIF_MSG_IFDOWN) + NETIF_MSG_IFDOWN | NETIF_MSG_HW) #define DPAA_INGRESS_CS_THRESHOLD 0x10000000 /* Ingress congestion threshold on FMan ports @@ -174,12 +151,17 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define DPAA_PARSE_RESULTS_SIZE sizeof(struct fman_prs_result) #define DPAA_TIME_STAMP_SIZE 8 #define DPAA_HASH_RESULTS_SIZE 8 +#define DPAA_HWA_SIZE (DPAA_PARSE_RESULTS_SIZE + DPAA_TIME_STAMP_SIZE \ + + DPAA_HASH_RESULTS_SIZE) +#define DPAA_RX_PRIV_DATA_DEFAULT_SIZE (DPAA_TX_PRIV_DATA_SIZE + \ + XDP_PACKET_HEADROOM - DPAA_HWA_SIZE) #ifdef CONFIG_DPAA_ERRATUM_A050385 -#define DPAA_RX_PRIV_DATA_SIZE (DPAA_A050385_ALIGN - (DPAA_PARSE_RESULTS_SIZE\ - + DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE)) +#define DPAA_RX_PRIV_DATA_A050385_SIZE (DPAA_A050385_ALIGN - DPAA_HWA_SIZE) +#define DPAA_RX_PRIV_DATA_SIZE (fman_has_errata_a050385() ? \ + DPAA_RX_PRIV_DATA_A050385_SIZE : \ + DPAA_RX_PRIV_DATA_DEFAULT_SIZE) #else -#define DPAA_RX_PRIV_DATA_SIZE (u16)(DPAA_TX_PRIV_DATA_SIZE + \ - dpaa_rx_extra_headroom) +#define DPAA_RX_PRIV_DATA_SIZE DPAA_RX_PRIV_DATA_DEFAULT_SIZE #endif #define DPAA_ETH_PCD_RXQ_NUM 128 @@ -215,12 +197,15 @@ static int dpaa_rx_extra_headroom; #define dpaa_get_max_mtu() \ (dpaa_max_frm - (VLAN_ETH_HLEN + ETH_FCS_LEN)) +static void dpaa_eth_cgr_set_speed(struct mac_device *mac_dev, int speed); + static int dpaa_netdev_init(struct net_device *net_dev, const struct net_device_ops *dpaa_ops, u16 tx_timeout) { struct dpaa_priv *priv = netdev_priv(net_dev); struct device *dev = net_dev->dev.parent; + struct mac_device *mac_dev = priv->mac_dev; struct dpaa_percpu_priv *percpu_priv; const u8 *mac_addr; int i, err; @@ -234,16 +219,16 @@ static int dpaa_netdev_init(struct net_device *net_dev, } net_dev->netdev_ops = dpaa_ops; - mac_addr = priv->mac_dev->addr; + mac_addr = mac_dev->addr; - net_dev->mem_start = priv->mac_dev->res->start; - net_dev->mem_end = priv->mac_dev->res->end; + net_dev->mem_start = (unsigned long)priv->mac_dev->res->start; + net_dev->mem_end = (unsigned long)priv->mac_dev->res->end; net_dev->min_mtu = ETH_MIN_MTU; net_dev->max_mtu = dpaa_get_max_mtu(); net_dev->hw_features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | - NETIF_F_LLTX | NETIF_F_RXHASH); + NETIF_F_RXHASH); net_dev->hw_features |= NETIF_F_SG | NETIF_F_HIGHDMA; /* The kernels enables GSO automatically, if we declare NETIF_F_SG. @@ -253,19 +238,24 @@ static int dpaa_netdev_init(struct net_device *net_dev, net_dev->features |= NETIF_F_RXCSUM; net_dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + net_dev->lltx = true; /* we do not want shared skbs on TX */ net_dev->priv_flags &= ~IFF_TX_SKB_SHARING; net_dev->features |= net_dev->hw_features; net_dev->vlan_features = net_dev->features; + net_dev->xdp_features = NETDEV_XDP_ACT_BASIC | + NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_NDO_XMIT; + if (is_valid_ether_addr(mac_addr)) { memcpy(net_dev->perm_addr, mac_addr, net_dev->addr_len); - memcpy(net_dev->dev_addr, mac_addr, net_dev->addr_len); + eth_hw_addr_set(net_dev, mac_addr); } else { eth_hw_addr_random(net_dev); - err = priv->mac_dev->change_addr(priv->mac_dev->fman_mac, - (enet_addr_t *)net_dev->dev_addr); + err = mac_dev->change_addr(mac_dev->fman_mac, + (const enet_addr_t *)net_dev->dev_addr); if (err) { dev_err(dev, "Failed to set random MAC address\n"); return -EINVAL; @@ -279,12 +269,27 @@ static int dpaa_netdev_init(struct net_device *net_dev, net_dev->needed_headroom = priv->tx_headroom; net_dev->watchdog_timeo = msecs_to_jiffies(tx_timeout); + /* The rest of the config is filled in by the mac device already */ + mac_dev->phylink_config.dev = &net_dev->dev; + mac_dev->phylink_config.type = PHYLINK_NETDEV; + mac_dev->update_speed = dpaa_eth_cgr_set_speed; + mac_dev->phylink = phylink_create(&mac_dev->phylink_config, + dev_fwnode(mac_dev->dev), + mac_dev->phy_if, + mac_dev->phylink_ops); + if (IS_ERR(mac_dev->phylink)) { + err = PTR_ERR(mac_dev->phylink); + dev_err_probe(dev, err, "Could not create phylink\n"); + return err; + } + /* start without the RUNNING flag, phylib controls it later */ netif_carrier_off(net_dev); err = register_netdev(net_dev); if (err < 0) { dev_err(dev, "register_netdev() = %d\n", err); + phylink_destroy(mac_dev->phylink); return err; } @@ -295,7 +300,8 @@ static int dpaa_stop(struct net_device *net_dev) { struct mac_device *mac_dev; struct dpaa_priv *priv; - int i, err, error; + int i, error; + int err = 0; priv = netdev_priv(net_dev); mac_dev = priv->mac_dev; @@ -306,10 +312,8 @@ static int dpaa_stop(struct net_device *net_dev) */ msleep(200); - err = mac_dev->stop(mac_dev); - if (err < 0) - netif_err(priv, ifdown, net_dev, "mac_dev->stop() = %d\n", - err); + phylink_stop(mac_dev->phylink); + mac_dev->disable(mac_dev->fman_mac); for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++) { error = fman_port_disable(mac_dev->port[i]); @@ -317,8 +321,7 @@ static int dpaa_stop(struct net_device *net_dev) err = error; } - if (net_dev->phydev) - phy_disconnect(net_dev->phydev); + phylink_disconnect_phy(mac_dev->phylink); net_dev->phydev = NULL; msleep(200); @@ -368,6 +371,7 @@ static int dpaa_setup_tc(struct net_device *net_dev, enum tc_setup_type type, void *type_data) { struct dpaa_priv *priv = netdev_priv(net_dev); + int num_txqs_per_tc = dpaa_num_txqs_per_tc(); struct tc_mqprio_qopt *mqprio = type_data; u8 num_tc; int i; @@ -395,12 +399,12 @@ static int dpaa_setup_tc(struct net_device *net_dev, enum tc_setup_type type, netdev_set_num_tc(net_dev, num_tc); for (i = 0; i < num_tc; i++) - netdev_set_tc_queue(net_dev, i, DPAA_TC_TXQ_NUM, - i * DPAA_TC_TXQ_NUM); + netdev_set_tc_queue(net_dev, i, num_txqs_per_tc, + i * num_txqs_per_tc); out: priv->num_tc = num_tc ? : 1; - netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM); + netif_set_real_num_tx_queues(net_dev, priv->num_tc * num_txqs_per_tc); return 0; } @@ -445,7 +449,7 @@ static int dpaa_set_mac_address(struct net_device *net_dev, void *addr) mac_dev = priv->mac_dev; err = mac_dev->change_addr(mac_dev->fman_mac, - (enet_addr_t *)net_dev->dev_addr); + (const enet_addr_t *)net_dev->dev_addr); if (err < 0) { netif_err(priv, drv, net_dev, "mac_dev->change_addr() = %d\n", err); @@ -458,6 +462,22 @@ static int dpaa_set_mac_address(struct net_device *net_dev, void *addr) return 0; } +static int dpaa_addr_sync(struct net_device *net_dev, const u8 *addr) +{ + const struct dpaa_priv *priv = netdev_priv(net_dev); + + return priv->mac_dev->add_hash_mac_addr(priv->mac_dev->fman_mac, + (enet_addr_t *)addr); +} + +static int dpaa_addr_unsync(struct net_device *net_dev, const u8 *addr) +{ + const struct dpaa_priv *priv = netdev_priv(net_dev); + + return priv->mac_dev->remove_hash_mac_addr(priv->mac_dev->fman_mac, + (enet_addr_t *)addr); +} + static void dpaa_set_rx_mode(struct net_device *net_dev) { const struct dpaa_priv *priv; @@ -485,9 +505,9 @@ static void dpaa_set_rx_mode(struct net_device *net_dev) err); } - err = priv->mac_dev->set_multi(net_dev, priv->mac_dev); + err = __dev_mc_sync(net_dev, dpaa_addr_sync, dpaa_addr_unsync); if (err < 0) - netif_err(priv, drv, net_dev, "mac_dev->set_multi() = %d\n", + netif_err(priv, drv, net_dev, "dpaa_addr_sync() = %d\n", err); } @@ -646,7 +666,7 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx) fq->wq = 6; break; case FQ_TYPE_TX: - switch (idx / DPAA_TC_TXQ_NUM) { + switch (idx / dpaa_num_txqs_per_tc()) { case 0: /* Low priority (best effort) */ fq->wq = 6; @@ -664,8 +684,8 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx) fq->wq = 0; break; default: - WARN(1, "Too many TX FQs: more than %d!\n", - DPAA_ETH_TXQ_NUM); + WARN(1, "Too many TX FQs: more than %zu!\n", + dpaa_max_num_txqs()); } break; default: @@ -737,7 +757,8 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list, port_fqs->rx_pcdq = &dpaa_fq[0]; - if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX_CONF_MQ)) + if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, + FQ_TYPE_TX_CONF_MQ)) goto fq_alloc_failed; dpaa_fq = dpaa_fq_alloc(dev, 0, 1, list, FQ_TYPE_TX_ERROR); @@ -752,7 +773,7 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list, port_fqs->tx_defq = &dpaa_fq[0]; - if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX)) + if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, FQ_TYPE_TX)) goto fq_alloc_failed; return 0; @@ -844,12 +865,12 @@ static int dpaa_eth_cgr_init(struct dpaa_priv *priv) initcgr.we_mask = cpu_to_be16(QM_CGR_WE_CSCN_EN | QM_CGR_WE_CS_THRES); initcgr.cgr.cscn_en = QM_CGR_EN; - /* Set different thresholds based on the MAC speed. - * This may turn suboptimal if the MAC is reconfigured at a speed - * lower than its max, e.g. if a dTSEC later negotiates a 100Mbps link. - * In such cases, we ought to reconfigure the threshold, too. + /* Set different thresholds based on the configured MAC speed. + * This may turn suboptimal if the MAC is reconfigured at another + * speed, so MACs must call dpaa_eth_cgr_set_speed in their link_up + * callback. */ - if (priv->mac_dev->if_support & SUPPORTED_10000baseT_Full) + if (priv->mac_dev->phylink_config.mac_capabilities & MAC_10000FD) cs_th = DPAA_CS_THRESHOLD_10G; else cs_th = DPAA_CS_THRESHOLD_1G; @@ -876,6 +897,31 @@ out_error: return err; } +static void dpaa_eth_cgr_set_speed(struct mac_device *mac_dev, int speed) +{ + struct net_device *net_dev = to_net_dev(mac_dev->phylink_config.dev); + struct dpaa_priv *priv = netdev_priv(net_dev); + struct qm_mcc_initcgr opts = { }; + u32 cs_th; + int err; + + opts.we_mask = cpu_to_be16(QM_CGR_WE_CS_THRES); + switch (speed) { + case SPEED_10000: + cs_th = DPAA_CS_THRESHOLD_10G; + break; + case SPEED_1000: + default: + cs_th = DPAA_CS_THRESHOLD_1G; + break; + } + qm_cgr_cs_thres_set64(&opts.cgr.cs_thres, cs_th, 1); + + err = qman_update_cgr_safe(&priv->cgr_data.cgr, &opts); + if (err) + netdev_err(net_dev, "could not update speed: %d\n", err); +} + static inline void dpaa_setup_ingress(const struct dpaa_priv *priv, struct dpaa_fq *fq, const struct qman_fq *template) @@ -903,14 +949,18 @@ static inline void dpaa_setup_egress(const struct dpaa_priv *priv, } } -static void dpaa_fq_setup(struct dpaa_priv *priv, - const struct dpaa_fq_cbs *fq_cbs, - struct fman_port *tx_port) +static int dpaa_fq_setup(struct dpaa_priv *priv, + const struct dpaa_fq_cbs *fq_cbs, + struct fman_port *tx_port) { int egress_cnt = 0, conf_cnt = 0, num_portals = 0, portal_cnt = 0, cpu; const cpumask_t *affine_cpus = qman_affine_cpus(); - u16 channels[NR_CPUS]; struct dpaa_fq *fq; + u16 *channels; + + channels = kcalloc(num_possible_cpus(), sizeof(u16), GFP_KERNEL); + if (!channels) + return -ENOMEM; for_each_cpu_and(cpu, affine_cpus, cpu_online_mask) channels[num_portals++] = qman_affine_channel(cpu); @@ -937,15 +987,11 @@ static void dpaa_fq_setup(struct dpaa_priv *priv, case FQ_TYPE_TX: dpaa_setup_egress(priv, fq, tx_port, &fq_cbs->egress_ern); - /* If we have more Tx queues than the number of cores, - * just ignore the extra ones. - */ - if (egress_cnt < DPAA_ETH_TXQ_NUM) - priv->egress_fqs[egress_cnt++] = &fq->fq_base; + priv->egress_fqs[egress_cnt++] = &fq->fq_base; break; case FQ_TYPE_TX_CONF_MQ: priv->conf_fqs[conf_cnt++] = &fq->fq_base; - /* fall through */ + fallthrough; case FQ_TYPE_TX_CONFIRM: dpaa_setup_ingress(priv, fq, &fq_cbs->tx_defq); break; @@ -959,16 +1005,9 @@ static void dpaa_fq_setup(struct dpaa_priv *priv, } } - /* Make sure all CPUs receive a corresponding Tx queue. */ - while (egress_cnt < DPAA_ETH_TXQ_NUM) { - list_for_each_entry(fq, &priv->dpaa_fq_list, list) { - if (fq->fq_type != FQ_TYPE_TX) - continue; - priv->egress_fqs[egress_cnt++] = &fq->fq_base; - if (egress_cnt == DPAA_ETH_TXQ_NUM) - break; - } - } + kfree(channels); + + return 0; } static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv, @@ -976,7 +1015,7 @@ static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv, { int i; - for (i = 0; i < DPAA_ETH_TXQ_NUM; i++) + for (i = 0; i < dpaa_max_num_txqs(); i++) if (priv->egress_fqs[i] == tx_fq) return i; @@ -1123,6 +1162,25 @@ static int dpaa_fq_init(struct dpaa_fq *dpaa_fq, bool td_enable) dpaa_fq->fqid = qman_fq_fqid(fq); + if (dpaa_fq->fq_type == FQ_TYPE_RX_DEFAULT || + dpaa_fq->fq_type == FQ_TYPE_RX_PCD) { + err = xdp_rxq_info_reg(&dpaa_fq->xdp_rxq, dpaa_fq->net_dev, + dpaa_fq->fqid, 0); + if (err) { + dev_err(dev, "xdp_rxq_info_reg() = %d\n", err); + return err; + } + + err = xdp_rxq_info_reg_mem_model(&dpaa_fq->xdp_rxq, + MEM_TYPE_PAGE_ORDER0, NULL); + if (err) { + dev_err(dev, "xdp_rxq_info_reg_mem_model() = %d\n", + err); + xdp_rxq_info_unreg(&dpaa_fq->xdp_rxq); + return err; + } + } + return 0; } @@ -1152,6 +1210,11 @@ static int dpaa_fq_free_entry(struct device *dev, struct qman_fq *fq) } } + if ((dpaa_fq->fq_type == FQ_TYPE_RX_DEFAULT || + dpaa_fq->fq_type == FQ_TYPE_RX_PCD) && + xdp_rxq_info_is_reg(&dpaa_fq->xdp_rxq)) + xdp_rxq_info_unreg(&dpaa_fq->xdp_rxq); + qman_destroy_fq(fq); list_del(&dpaa_fq->list); @@ -1432,13 +1495,8 @@ static int dpaa_enable_tx_csum(struct dpaa_priv *priv, parse_result = (struct fman_prs_result *)parse_results; /* If we're dealing with VLAN, get the real Ethernet type */ - if (ethertype == ETH_P_8021Q) { - /* We can't always assume the MAC header is set correctly - * by the stack, so reset to beginning of skb->data - */ - skb_reset_mac_header(skb); - ethertype = ntohs(vlan_eth_hdr(skb)->h_vlan_encapsulated_proto); - } + if (ethertype == ETH_P_8021Q) + ethertype = ntohs(skb_vlan_eth_hdr(skb)->h_vlan_encapsulated_proto); /* Fill in the relevant L3 parse result fields * and read the L4 protocol type @@ -1594,17 +1652,13 @@ static int dpaa_eth_refill_bpools(struct dpaa_priv *priv) { struct dpaa_bp *dpaa_bp; int *countptr; - int res; dpaa_bp = priv->dpaa_bp; if (!dpaa_bp) return -EINVAL; countptr = this_cpu_ptr(dpaa_bp->percpu_count); - res = dpaa_eth_refill_bpool(dpaa_bp, countptr); - if (res) - return res; - return 0; + return dpaa_eth_refill_bpool(dpaa_bp, countptr); } /* Cleanup function for outgoing frame descriptors that were built on Tx path, @@ -1618,6 +1672,9 @@ static int dpaa_eth_refill_bpools(struct dpaa_priv *priv) * * Return the skb backpointer, since for S/G frames the buffer containing it * gets freed here. + * + * No skb backpointer is set when transmitting XDP frames. Cleanup the buffer + * and return NULL in this case. */ static struct sk_buff *dpaa_cleanup_tx_fd(const struct dpaa_priv *priv, const struct qm_fd *fd, bool ts) @@ -1628,6 +1685,7 @@ static struct sk_buff *dpaa_cleanup_tx_fd(const struct dpaa_priv *priv, dma_addr_t addr = qm_fd_addr(fd); void *vaddr = phys_to_virt(addr); const struct qm_sg_entry *sgt; + struct dpaa_eth_swbp *swbp; struct sk_buff *skb; u64 ns; int i; @@ -1656,11 +1714,20 @@ static struct sk_buff *dpaa_cleanup_tx_fd(const struct dpaa_priv *priv, } } else { dma_unmap_single(priv->tx_dma_dev, addr, - priv->tx_headroom + qm_fd_get_length(fd), + qm_fd_get_offset(fd) + qm_fd_get_length(fd), dma_dir); } - skb = *(struct sk_buff **)vaddr; + swbp = (struct dpaa_eth_swbp *)vaddr; + skb = swbp->skb; + + /* No skb backpointer is set when running XDP. An xdp_frame + * backpointer is saved instead. + */ + if (!skb) { + xdp_return_frame(swbp->xdpf); + return NULL; + } /* DMA unmapping is required before accessing the HW provided info */ if (ts && priv->tx_tstamp && @@ -1726,7 +1793,6 @@ static struct sk_buff *contig_fd_to_skb(const struct dpaa_priv *priv, SKB_DATA_ALIGN(sizeof(struct skb_shared_info))); if (WARN_ONCE(!skb, "Build skb failure on Rx\n")) goto free_buffer; - WARN_ON(fd_off != priv->rx_headroom); skb_reserve(skb, fd_off); skb_put(skb, qm_fd_get_length(fd)); @@ -1753,7 +1819,6 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv, struct page *page, *head_page; struct dpaa_bp *dpaa_bp; void *vaddr, *sg_vaddr; - int frag_off, frag_len; struct sk_buff *skb; dma_addr_t sg_addr; int page_offset; @@ -1796,6 +1861,11 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv, * on Tx, if extra headers are added. */ WARN_ON(fd_off != priv->rx_headroom); + /* The offset to data start within the buffer holding + * the SGT should always be equal to the offset to data + * start within the first buffer holding the frame. + */ + WARN_ON_ONCE(fd_off != qm_sg_entry_get_off(&sgt[i])); skb_reserve(skb, fd_off); skb_put(skb, qm_sg_entry_get_len(&sgt[i])); } else { @@ -1809,21 +1879,23 @@ static struct sk_buff *sg_fd_to_skb(const struct dpaa_priv *priv, page = virt_to_page(sg_vaddr); head_page = virt_to_head_page(sg_vaddr); - /* Compute offset in (possibly tail) page */ + /* Compute offset of sg_vaddr in (possibly tail) page */ page_offset = ((unsigned long)sg_vaddr & (PAGE_SIZE - 1)) + (page_address(page) - page_address(head_page)); - /* page_offset only refers to the beginning of sgt[i]; - * but the buffer itself may have an internal offset. + + /* Non-initial SGT entries should not have a buffer + * offset. */ - frag_off = qm_sg_entry_get_off(&sgt[i]) + page_offset; - frag_len = qm_sg_entry_get_len(&sgt[i]); + WARN_ON_ONCE(qm_sg_entry_get_off(&sgt[i])); + /* skb_add_rx_frag() does no checking on the page; if * we pass it a tail page, we'll end up with - * bad page accounting and eventually with segafults. + * bad page accounting and eventually with segfaults. */ - skb_add_rx_frag(skb, i - 1, head_page, frag_off, - frag_len, dpaa_bp->size); + skb_add_rx_frag(skb, i - 1, head_page, page_offset, + qm_sg_entry_get_len(&sgt[i]), + dpaa_bp->size); } /* Update the pool count for the current {cpu x bpool} */ @@ -1874,8 +1946,8 @@ static int skb_to_contig_fd(struct dpaa_priv *priv, { struct net_device *net_dev = priv->net_dev; enum dma_data_direction dma_dir; + struct dpaa_eth_swbp *swbp; unsigned char *buff_start; - struct sk_buff **skbh; dma_addr_t addr; int err; @@ -1886,8 +1958,8 @@ static int skb_to_contig_fd(struct dpaa_priv *priv, buff_start = skb->data - priv->tx_headroom; dma_dir = DMA_TO_DEVICE; - skbh = (struct sk_buff **)buff_start; - *skbh = skb; + swbp = (struct dpaa_eth_swbp *)buff_start; + swbp->skb = skb; /* Enable L3/L4 hardware checksum computation. * @@ -1926,8 +1998,8 @@ static int skb_to_sg_fd(struct dpaa_priv *priv, const enum dma_data_direction dma_dir = DMA_TO_DEVICE; const int nr_frags = skb_shinfo(skb)->nr_frags; struct net_device *net_dev = priv->net_dev; + struct dpaa_eth_swbp *swbp; struct qm_sg_entry *sgt; - struct sk_buff **skbh; void *buff_start; skb_frag_t *frag; dma_addr_t addr; @@ -2000,8 +2072,8 @@ static int skb_to_sg_fd(struct dpaa_priv *priv, qm_fd_set_sg(fd, priv->tx_headroom, skb->len); /* DMA map the SGT page */ - skbh = (struct sk_buff **)buff_start; - *skbh = skb; + swbp = (struct dpaa_eth_swbp *)buff_start; + swbp->skb = skb; addr = dma_map_page(priv->tx_dma_dev, p, 0, priv->tx_headroom + DPAA_SGT_SIZE, dma_dir); @@ -2062,7 +2134,7 @@ static inline int dpaa_xmit(struct dpaa_priv *priv, } #ifdef CONFIG_DPAA_ERRATUM_A050385 -static int dpaa_a050385_wa(struct net_device *net_dev, struct sk_buff **s) +static int dpaa_a050385_wa_skb(struct net_device *net_dev, struct sk_buff **s) { struct dpaa_priv *priv = netdev_priv(net_dev); struct sk_buff *new_skb, *skb = *s; @@ -2115,6 +2187,15 @@ workaround: skb_copy_header(new_skb, skb); new_skb->dev = skb->dev; + /* Copy relevant timestamp info from the old skb to the new */ + if (priv->tx_tstamp) { + skb_shinfo(new_skb)->tx_flags = skb_shinfo(skb)->tx_flags; + skb_shinfo(new_skb)->hwtstamps = skb_shinfo(skb)->hwtstamps; + skb_shinfo(new_skb)->tskey = skb_shinfo(skb)->tskey; + if (skb->sk) + skb_set_owner_w(new_skb, skb->sk); + } + /* We move the headroom when we align it so we have to reset the * network and transport header offsets relative to the new data * pointer. The checksum offload relies on these offsets. @@ -2122,24 +2203,103 @@ workaround: skb_set_network_header(new_skb, skb_network_offset(skb)); skb_set_transport_header(new_skb, skb_transport_offset(skb)); - /* TODO: does timestamping need the result in the old skb? */ dev_kfree_skb(skb); *s = new_skb; return 0; } + +static int dpaa_a050385_wa_xdpf(struct dpaa_priv *priv, + struct xdp_frame **init_xdpf) +{ + struct xdp_frame *new_xdpf, *xdpf = *init_xdpf; + void *new_buff, *aligned_data; + struct page *p; + u32 data_shift; + int headroom; + + /* Check the data alignment and make sure the headroom is large + * enough to store the xdpf backpointer. Use an aligned headroom + * value. + * + * Due to alignment constraints, we give XDP access to the full 256 + * byte frame headroom. If the XDP program uses all of it, copy the + * data to a new buffer and make room for storing the backpointer. + */ + if (PTR_IS_ALIGNED(xdpf->data, DPAA_FD_DATA_ALIGNMENT) && + xdpf->headroom >= priv->tx_headroom) { + xdpf->headroom = priv->tx_headroom; + return 0; + } + + /* Try to move the data inside the buffer just enough to align it and + * store the xdpf backpointer. If the available headroom isn't large + * enough, resort to allocating a new buffer and copying the data. + */ + aligned_data = PTR_ALIGN_DOWN(xdpf->data, DPAA_FD_DATA_ALIGNMENT); + data_shift = xdpf->data - aligned_data; + + /* The XDP frame's headroom needs to be large enough to accommodate + * shifting the data as well as storing the xdpf backpointer. + */ + if (xdpf->headroom >= data_shift + priv->tx_headroom) { + memmove(aligned_data, xdpf->data, xdpf->len); + xdpf->data = aligned_data; + xdpf->headroom = priv->tx_headroom; + return 0; + } + + /* The new xdp_frame is stored in the new buffer. Reserve enough space + * in the headroom for storing it along with the driver's private + * info. The headroom needs to be aligned to DPAA_FD_DATA_ALIGNMENT to + * guarantee the data's alignment in the buffer. + */ + headroom = ALIGN(sizeof(*new_xdpf) + priv->tx_headroom, + DPAA_FD_DATA_ALIGNMENT); + + /* Assure the extended headroom and data don't overflow the buffer, + * while maintaining the mandatory tailroom. + */ + if (headroom + xdpf->len > DPAA_BP_RAW_SIZE - + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) + return -ENOMEM; + + p = dev_alloc_pages(0); + if (unlikely(!p)) + return -ENOMEM; + + /* Copy the data to the new buffer at a properly aligned offset */ + new_buff = page_address(p); + memcpy(new_buff + headroom, xdpf->data, xdpf->len); + + /* Create an XDP frame around the new buffer in a similar fashion + * to xdp_convert_buff_to_frame. + */ + new_xdpf = new_buff; + new_xdpf->data = new_buff + headroom; + new_xdpf->len = xdpf->len; + new_xdpf->headroom = priv->tx_headroom; + new_xdpf->frame_sz = DPAA_BP_RAW_SIZE; + new_xdpf->mem_type = MEM_TYPE_PAGE_ORDER0; + + /* Release the initial buffer */ + xdp_return_frame_rx_napi(xdpf); + + *init_xdpf = new_xdpf; + return 0; +} #endif static netdev_tx_t dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) { const int queue_mapping = skb_get_queue_mapping(skb); - bool nonlinear = skb_is_nonlinear(skb); struct rtnl_link_stats64 *percpu_stats; struct dpaa_percpu_priv *percpu_priv; struct netdev_queue *txq; struct dpaa_priv *priv; struct qm_fd fd; + bool nonlinear; int offset = 0; int err = 0; @@ -2149,6 +2309,13 @@ dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) qm_fd_clear_fd(&fd); + /* Packet data is always read as 32-bit words, so zero out any part of + * the skb which might be sent if we have to pad the packet + */ + if (__skb_put_padto(skb, ETH_ZLEN, false)) + goto enomem; + + nonlinear = skb_is_nonlinear(skb); if (!nonlinear) { /* We're going to store the skb backpointer at the beginning * of the data buffer, so we need a privately owned skb @@ -2178,7 +2345,7 @@ dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) #ifdef CONFIG_DPAA_ERRATUM_A050385 if (unlikely(fman_has_errata_a050385())) { - if (dpaa_a050385_wa(net_dev, &skb)) + if (dpaa_a050385_wa_skb(net_dev, &skb)) goto enomem; nonlinear = skb_is_nonlinear(skb); } @@ -2198,7 +2365,7 @@ dpaa_start_xmit(struct sk_buff *skb, struct net_device *net_dev) txq = netdev_get_tx_queue(net_dev, queue_mapping); /* LLTX requires to do our own update of trans_start */ - txq->trans_start = jiffies; + txq_trans_cond_update(txq); if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { fd.cmd |= cpu_to_be32(FM_FD_CMD_UPD); @@ -2262,8 +2429,14 @@ static int dpaa_eth_poll(struct napi_struct *napi, int budget) { struct dpaa_napi_portal *np = container_of(napi, struct dpaa_napi_portal, napi); + int cleaned; + + np->xdp_act = 0; - int cleaned = qman_p_poll_dqrr(np->p, budget); + cleaned = qman_p_poll_dqrr(np->p, budget); + + if (np->xdp_act & XDP_REDIRECT) + xdp_do_flush(); if (cleaned < budget) { napi_complete_done(napi, cleaned); @@ -2300,9 +2473,9 @@ static void dpaa_tx_conf(struct net_device *net_dev, } static inline int dpaa_eth_napi_schedule(struct dpaa_percpu_priv *percpu_priv, - struct qman_portal *portal) + struct qman_portal *portal, bool sched_napi) { - if (unlikely(in_irq() || !in_serving_softirq())) { + if (sched_napi) { /* Disable QMan IRQ and invoke NAPI */ qman_p_irqsource_remove(portal, QM_PIRQ_DQRI); @@ -2316,7 +2489,8 @@ static inline int dpaa_eth_napi_schedule(struct dpaa_percpu_priv *percpu_priv, static enum qman_cb_dqrr_result rx_error_dqrr(struct qman_portal *portal, struct qman_fq *fq, - const struct qm_dqrr_entry *dq) + const struct qm_dqrr_entry *dq, + bool sched_napi) { struct dpaa_fq *dpaa_fq = container_of(fq, struct dpaa_fq, fq_base); struct dpaa_percpu_priv *percpu_priv; @@ -2332,7 +2506,7 @@ static enum qman_cb_dqrr_result rx_error_dqrr(struct qman_portal *portal, percpu_priv = this_cpu_ptr(priv->percpu_priv); - if (dpaa_eth_napi_schedule(percpu_priv, portal)) + if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi)) return qman_cb_dqrr_stop; dpaa_eth_refill_bpools(priv); @@ -2341,29 +2515,199 @@ static enum qman_cb_dqrr_result rx_error_dqrr(struct qman_portal *portal, return qman_cb_dqrr_consume; } +static int dpaa_xdp_xmit_frame(struct net_device *net_dev, + struct xdp_frame *xdpf) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct rtnl_link_stats64 *percpu_stats; + struct dpaa_percpu_priv *percpu_priv; + struct dpaa_eth_swbp *swbp; + struct netdev_queue *txq; + void *buff_start; + struct qm_fd fd; + dma_addr_t addr; + int err; + + percpu_priv = this_cpu_ptr(priv->percpu_priv); + percpu_stats = &percpu_priv->stats; + +#ifdef CONFIG_DPAA_ERRATUM_A050385 + if (unlikely(fman_has_errata_a050385())) { + if (dpaa_a050385_wa_xdpf(priv, &xdpf)) { + err = -ENOMEM; + goto out_error; + } + } +#endif + + if (xdpf->headroom < DPAA_TX_PRIV_DATA_SIZE) { + err = -EINVAL; + goto out_error; + } + + buff_start = xdpf->data - xdpf->headroom; + + /* Leave empty the skb backpointer at the start of the buffer. + * Save the XDP frame for easy cleanup on confirmation. + */ + swbp = (struct dpaa_eth_swbp *)buff_start; + swbp->skb = NULL; + swbp->xdpf = xdpf; + + qm_fd_clear_fd(&fd); + fd.bpid = FSL_DPAA_BPID_INV; + fd.cmd |= cpu_to_be32(FM_FD_CMD_FCO); + qm_fd_set_contig(&fd, xdpf->headroom, xdpf->len); + + addr = dma_map_single(priv->tx_dma_dev, buff_start, + xdpf->headroom + xdpf->len, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(priv->tx_dma_dev, addr))) { + err = -EINVAL; + goto out_error; + } + + qm_fd_addr_set64(&fd, addr); + + /* Bump the trans_start */ + txq = netdev_get_tx_queue(net_dev, smp_processor_id()); + txq_trans_cond_update(txq); + + err = dpaa_xmit(priv, percpu_stats, smp_processor_id(), &fd); + if (err) { + dma_unmap_single(priv->tx_dma_dev, addr, + qm_fd_get_offset(&fd) + qm_fd_get_length(&fd), + DMA_TO_DEVICE); + goto out_error; + } + + return 0; + +out_error: + percpu_stats->tx_errors++; + return err; +} + +static u32 dpaa_run_xdp(struct dpaa_priv *priv, struct qm_fd *fd, void *vaddr, + struct dpaa_fq *dpaa_fq, unsigned int *xdp_meta_len) +{ + ssize_t fd_off = qm_fd_get_offset(fd); + struct bpf_prog *xdp_prog; + struct xdp_frame *xdpf; + struct xdp_buff xdp; + u32 xdp_act; + int err; + + xdp_prog = READ_ONCE(priv->xdp_prog); + if (!xdp_prog) + return XDP_PASS; + + xdp_init_buff(&xdp, DPAA_BP_RAW_SIZE - DPAA_TX_PRIV_DATA_SIZE, + &dpaa_fq->xdp_rxq); + xdp_prepare_buff(&xdp, vaddr + fd_off - XDP_PACKET_HEADROOM, + XDP_PACKET_HEADROOM, qm_fd_get_length(fd), true); + + /* We reserve a fixed headroom of 256 bytes under the erratum and we + * offer it all to XDP programs to use. If no room is left for the + * xdpf backpointer on TX, we will need to copy the data. + * Disable metadata support since data realignments might be required + * and the information can be lost. + */ +#ifdef CONFIG_DPAA_ERRATUM_A050385 + if (unlikely(fman_has_errata_a050385())) { + xdp_set_data_meta_invalid(&xdp); + xdp.data_hard_start = vaddr; + xdp.frame_sz = DPAA_BP_RAW_SIZE; + } +#endif + + xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp); + + /* Update the length and the offset of the FD */ + qm_fd_set_contig(fd, xdp.data - vaddr, xdp.data_end - xdp.data); + + switch (xdp_act) { + case XDP_PASS: +#ifdef CONFIG_DPAA_ERRATUM_A050385 + *xdp_meta_len = xdp_data_meta_unsupported(&xdp) ? 0 : + xdp.data - xdp.data_meta; +#else + *xdp_meta_len = xdp.data - xdp.data_meta; +#endif + break; + case XDP_TX: + /* We can access the full headroom when sending the frame + * back out + */ + xdp.data_hard_start = vaddr; + xdp.frame_sz = DPAA_BP_RAW_SIZE; + xdpf = xdp_convert_buff_to_frame(&xdp); + if (unlikely(!xdpf)) { + free_pages((unsigned long)vaddr, 0); + break; + } + + if (dpaa_xdp_xmit_frame(priv->net_dev, xdpf)) + xdp_return_frame_rx_napi(xdpf); + + break; + case XDP_REDIRECT: + /* Allow redirect to use the full headroom */ + xdp.data_hard_start = vaddr; + xdp.frame_sz = DPAA_BP_RAW_SIZE; + + err = xdp_do_redirect(priv->net_dev, &xdp, xdp_prog); + if (err) { + trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act); + free_pages((unsigned long)vaddr, 0); + } + 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: + /* Free the buffer */ + free_pages((unsigned long)vaddr, 0); + break; + } + + return xdp_act; +} + static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal, struct qman_fq *fq, - const struct qm_dqrr_entry *dq) + const struct qm_dqrr_entry *dq, + bool sched_napi) { + bool ts_valid = false, hash_valid = false; struct skb_shared_hwtstamps *shhwtstamps; + unsigned int skb_len, xdp_meta_len = 0; struct rtnl_link_stats64 *percpu_stats; struct dpaa_percpu_priv *percpu_priv; const struct qm_fd *fd = &dq->fd; dma_addr_t addr = qm_fd_addr(fd); + struct dpaa_napi_portal *np; enum qm_fd_format fd_format; struct net_device *net_dev; u32 fd_status, hash_offset; + struct qm_sg_entry *sgt; struct dpaa_bp *dpaa_bp; + struct dpaa_fq *dpaa_fq; struct dpaa_priv *priv; - unsigned int skb_len; struct sk_buff *skb; int *count_ptr; + u32 xdp_act; void *vaddr; + u32 hash; u64 ns; + dpaa_fq = container_of(fq, struct dpaa_fq, fq_base); fd_status = be32_to_cpu(fd->status); fd_format = qm_fd_get_format(fd); - net_dev = ((struct dpaa_fq *)fq)->net_dev; + net_dev = dpaa_fq->net_dev; priv = netdev_priv(net_dev); dpaa_bp = dpaa_bpid2pool(dq->fd.bpid); if (!dpaa_bp) @@ -2374,8 +2718,9 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal, percpu_priv = this_cpu_ptr(priv->percpu_priv); percpu_stats = &percpu_priv->stats; + np = &percpu_priv->np; - if (unlikely(dpaa_eth_napi_schedule(percpu_priv, portal))) + if (unlikely(dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi))) return qman_cb_dqrr_stop; /* Make sure we didn't run out of buffers */ @@ -2414,35 +2759,68 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal, count_ptr = this_cpu_ptr(dpaa_bp->percpu_count); (*count_ptr)--; - if (likely(fd_format == qm_fd_contig)) + /* Extract the timestamp stored in the headroom before running XDP */ + if (priv->rx_tstamp) { + if (!fman_port_get_tstamp(priv->mac_dev->port[RX], vaddr, &ns)) + ts_valid = true; + else + WARN_ONCE(1, "fman_port_get_tstamp failed!\n"); + } + + /* Extract the hash stored in the headroom before running XDP */ + if (net_dev->features & NETIF_F_RXHASH && priv->keygen_in_use && + !fman_port_get_hash_result_offset(priv->mac_dev->port[RX], + &hash_offset)) { + hash = be32_to_cpu(*(__be32 *)(vaddr + hash_offset)); + hash_valid = true; + } + + if (likely(fd_format == qm_fd_contig)) { + xdp_act = dpaa_run_xdp(priv, (struct qm_fd *)fd, vaddr, + dpaa_fq, &xdp_meta_len); + np->xdp_act |= xdp_act; + if (xdp_act != XDP_PASS) { + percpu_stats->rx_packets++; + percpu_stats->rx_bytes += qm_fd_get_length(fd); + return qman_cb_dqrr_consume; + } skb = contig_fd_to_skb(priv, fd); - else + } else { + /* XDP doesn't support S/G frames. Return the fragments to the + * buffer pool and release the SGT. + */ + if (READ_ONCE(priv->xdp_prog)) { + WARN_ONCE(1, "S/G frames not supported under XDP\n"); + sgt = vaddr + qm_fd_get_offset(fd); + dpaa_release_sgt_members(sgt); + free_pages((unsigned long)vaddr, 0); + return qman_cb_dqrr_consume; + } skb = sg_fd_to_skb(priv, fd); + } if (!skb) return qman_cb_dqrr_consume; - if (priv->rx_tstamp) { + if (xdp_meta_len) + skb_metadata_set(skb, xdp_meta_len); + + /* Set the previously extracted timestamp */ + if (ts_valid) { shhwtstamps = skb_hwtstamps(skb); memset(shhwtstamps, 0, sizeof(*shhwtstamps)); - - if (!fman_port_get_tstamp(priv->mac_dev->port[RX], vaddr, &ns)) - shhwtstamps->hwtstamp = ns_to_ktime(ns); - else - dev_warn(net_dev->dev.parent, "fman_port_get_tstamp failed!\n"); + shhwtstamps->hwtstamp = ns_to_ktime(ns); } skb->protocol = eth_type_trans(skb, net_dev); - if (net_dev->features & NETIF_F_RXHASH && priv->keygen_in_use && - !fman_port_get_hash_result_offset(priv->mac_dev->port[RX], - &hash_offset)) { + /* Set the previously extracted hash */ + if (hash_valid) { enum pkt_hash_types type; /* if L4 exists, it was used in the hash generation */ type = be32_to_cpu(fd->status) & FM_FD_STAT_L4CV ? PKT_HASH_TYPE_L4 : PKT_HASH_TYPE_L3; - skb_set_hash(skb, be32_to_cpu(*(u32 *)(vaddr + hash_offset)), - type); + skb_set_hash(skb, hash, type); } skb_len = skb->len; @@ -2460,7 +2838,8 @@ static enum qman_cb_dqrr_result rx_default_dqrr(struct qman_portal *portal, static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal, struct qman_fq *fq, - const struct qm_dqrr_entry *dq) + const struct qm_dqrr_entry *dq, + bool sched_napi) { struct dpaa_percpu_priv *percpu_priv; struct net_device *net_dev; @@ -2471,7 +2850,7 @@ static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal, percpu_priv = this_cpu_ptr(priv->percpu_priv); - if (dpaa_eth_napi_schedule(percpu_priv, portal)) + if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi)) return qman_cb_dqrr_stop; dpaa_tx_error(net_dev, priv, percpu_priv, &dq->fd, fq->fqid); @@ -2481,7 +2860,8 @@ static enum qman_cb_dqrr_result conf_error_dqrr(struct qman_portal *portal, static enum qman_cb_dqrr_result conf_dflt_dqrr(struct qman_portal *portal, struct qman_fq *fq, - const struct qm_dqrr_entry *dq) + const struct qm_dqrr_entry *dq, + bool sched_napi) { struct dpaa_percpu_priv *percpu_priv; struct net_device *net_dev; @@ -2495,7 +2875,7 @@ static enum qman_cb_dqrr_result conf_dflt_dqrr(struct qman_portal *portal, percpu_priv = this_cpu_ptr(priv->percpu_priv); - if (dpaa_eth_napi_schedule(percpu_priv, portal)) + if (dpaa_eth_napi_schedule(percpu_priv, portal, sched_napi)) return qman_cb_dqrr_stop; dpaa_tx_conf(net_dev, priv, percpu_priv, &dq->fd, fq->fqid); @@ -2541,7 +2921,7 @@ static void dpaa_eth_napi_enable(struct dpaa_priv *priv) for_each_online_cpu(i) { percpu_priv = per_cpu_ptr(priv->percpu_priv, i); - percpu_priv->np.down = 0; + percpu_priv->np.down = false; napi_enable(&percpu_priv->np.napi); } } @@ -2554,59 +2934,11 @@ static void dpaa_eth_napi_disable(struct dpaa_priv *priv) for_each_online_cpu(i) { percpu_priv = per_cpu_ptr(priv->percpu_priv, i); - percpu_priv->np.down = 1; + percpu_priv->np.down = true; napi_disable(&percpu_priv->np.napi); } } -static void dpaa_adjust_link(struct net_device *net_dev) -{ - struct mac_device *mac_dev; - struct dpaa_priv *priv; - - priv = netdev_priv(net_dev); - mac_dev = priv->mac_dev; - mac_dev->adjust_link(mac_dev); -} - -/* The Aquantia PHYs are capable of performing rate adaptation */ -#define PHY_VEND_AQUANTIA 0x03a1b400 - -static int dpaa_phy_init(struct net_device *net_dev) -{ - __ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, }; - struct mac_device *mac_dev; - struct phy_device *phy_dev; - struct dpaa_priv *priv; - - priv = netdev_priv(net_dev); - mac_dev = priv->mac_dev; - - phy_dev = of_phy_connect(net_dev, mac_dev->phy_node, - &dpaa_adjust_link, 0, - mac_dev->phy_if); - if (!phy_dev) { - netif_err(priv, ifup, net_dev, "init_phy() failed\n"); - return -ENODEV; - } - - /* Unless the PHY is capable of rate adaptation */ - if (mac_dev->phy_if != PHY_INTERFACE_MODE_XGMII || - ((phy_dev->drv->phy_id & GENMASK(31, 10)) != PHY_VEND_AQUANTIA)) { - /* remove any features not supported by the controller */ - ethtool_convert_legacy_u32_to_link_mode(mask, - mac_dev->if_support); - linkmode_and(phy_dev->supported, phy_dev->supported, mask); - } - - phy_support_asym_pause(phy_dev); - - mac_dev->phy_dev = phy_dev; - net_dev->phydev = phy_dev; - - return 0; -} - static int dpaa_open(struct net_device *net_dev) { struct mac_device *mac_dev; @@ -2617,7 +2949,8 @@ static int dpaa_open(struct net_device *net_dev) mac_dev = priv->mac_dev; dpaa_eth_napi_enable(priv); - err = dpaa_phy_init(net_dev); + err = phylink_of_phy_connect(mac_dev->phylink, + mac_dev->dev->of_node, 0); if (err) goto phy_init_failed; @@ -2627,11 +2960,12 @@ static int dpaa_open(struct net_device *net_dev) goto mac_start_failed; } - err = priv->mac_dev->start(mac_dev); + err = priv->mac_dev->enable(mac_dev->fman_mac); if (err < 0) { - netif_err(priv, ifup, net_dev, "mac_dev->start() = %d\n", err); + netif_err(priv, ifup, net_dev, "mac_dev->enable() = %d\n", err); goto mac_start_failed; } + phylink_start(mac_dev->phylink); netif_tx_start_all_queues(net_dev); @@ -2640,6 +2974,7 @@ static int dpaa_open(struct net_device *net_dev) mac_start_failed: for (i = 0; i < ARRAY_SIZE(mac_dev->port); i++) fman_port_disable(mac_dev->port[i]); + phylink_disconnect_phy(mac_dev->phylink); phy_init_failed: dpaa_eth_napi_disable(priv); @@ -2660,15 +2995,118 @@ static int dpaa_eth_stop(struct net_device *net_dev) return err; } -static int dpaa_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) +static bool xdp_validate_mtu(struct dpaa_priv *priv, int mtu) +{ + int max_contig_data = priv->dpaa_bp->size - priv->rx_headroom; + + /* We do not support S/G fragments when XDP is enabled. + * Limit the MTU in relation to the buffer size. + */ + if (mtu + VLAN_ETH_HLEN + ETH_FCS_LEN > max_contig_data) { + dev_warn(priv->net_dev->dev.parent, + "The maximum MTU for XDP is %d\n", + max_contig_data - VLAN_ETH_HLEN - ETH_FCS_LEN); + return false; + } + + return true; +} + +static int dpaa_change_mtu(struct net_device *net_dev, int new_mtu) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + + if (priv->xdp_prog && !xdp_validate_mtu(priv, new_mtu)) + return -EINVAL; + + WRITE_ONCE(net_dev->mtu, new_mtu); + return 0; +} + +static int dpaa_setup_xdp(struct net_device *net_dev, struct netdev_bpf *bpf) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct bpf_prog *old_prog; + int err; + bool up; + + /* S/G fragments are not supported in XDP-mode */ + if (bpf->prog && !xdp_validate_mtu(priv, net_dev->mtu)) { + NL_SET_ERR_MSG_MOD(bpf->extack, "MTU too large for XDP"); + return -EINVAL; + } + + up = netif_running(net_dev); + + if (up) + dpaa_eth_stop(net_dev); + + old_prog = xchg(&priv->xdp_prog, bpf->prog); + if (old_prog) + bpf_prog_put(old_prog); + + if (up) { + err = dpaa_open(net_dev); + if (err) { + NL_SET_ERR_MSG_MOD(bpf->extack, "dpaa_open() failed"); + return err; + } + } + + return 0; +} + +static int dpaa_xdp(struct net_device *net_dev, struct netdev_bpf *xdp) +{ + switch (xdp->command) { + case XDP_SETUP_PROG: + return dpaa_setup_xdp(net_dev, xdp); + default: + return -EINVAL; + } +} + +static int dpaa_xdp_xmit(struct net_device *net_dev, int n, + struct xdp_frame **frames, u32 flags) +{ + struct xdp_frame *xdpf; + int i, nxmit = 0; + + if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) + return -EINVAL; + + if (!netif_running(net_dev)) + return -ENETDOWN; + + for (i = 0; i < n; i++) { + xdpf = frames[i]; + if (dpaa_xdp_xmit_frame(net_dev, xdpf)) + break; + nxmit++; + } + + return nxmit; +} + +static int dpaa_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) { struct dpaa_priv *priv = netdev_priv(dev); - struct hwtstamp_config config; - if (copy_from_user(&config, rq->ifr_data, sizeof(config))) - return -EFAULT; + config->tx_type = priv->tx_tstamp ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + config->rx_filter = priv->rx_tstamp ? HWTSTAMP_FILTER_ALL : + HWTSTAMP_FILTER_NONE; + + return 0; +} + +static int dpaa_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct dpaa_priv *priv = netdev_priv(dev); - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: /* Couldn't disable rx/tx timestamping separately. * Do nothing here. @@ -2683,7 +3121,7 @@ static int dpaa_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) return -ERANGE; } - if (config.rx_filter == HWTSTAMP_FILTER_NONE) { + if (config->rx_filter == HWTSTAMP_FILTER_NONE) { /* Couldn't disable rx/tx timestamping separately. * Do nothing here. */ @@ -2692,26 +3130,17 @@ static int dpaa_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) priv->mac_dev->set_tstamp(priv->mac_dev->fman_mac, true); priv->rx_tstamp = true; /* TS is set for all frame types, not only those requested */ - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; } - return copy_to_user(rq->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; + return 0; } static int dpaa_ioctl(struct net_device *net_dev, struct ifreq *rq, int cmd) { - int ret = -EINVAL; - - if (cmd == SIOCGMIIREG) { - if (net_dev->phydev) - return phy_mii_ioctl(net_dev->phydev, rq, cmd); - } - - if (cmd == SIOCSHWTSTAMP) - return dpaa_ts_ioctl(net_dev, rq, cmd); + struct dpaa_priv *priv = netdev_priv(net_dev); - return ret; + return phylink_mii_ioctl(priv->mac_dev->phylink, rq, cmd); } static const struct net_device_ops dpaa_ops = { @@ -2720,12 +3149,16 @@ static const struct net_device_ops dpaa_ops = { .ndo_stop = dpaa_eth_stop, .ndo_tx_timeout = dpaa_tx_timeout, .ndo_get_stats64 = dpaa_get_stats64, - .ndo_change_carrier = fixed_phy_change_carrier, .ndo_set_mac_address = dpaa_set_mac_address, .ndo_validate_addr = eth_validate_addr, .ndo_set_rx_mode = dpaa_set_rx_mode, - .ndo_do_ioctl = dpaa_ioctl, + .ndo_eth_ioctl = dpaa_ioctl, .ndo_setup_tc = dpaa_setup_tc, + .ndo_change_mtu = dpaa_change_mtu, + .ndo_bpf = dpaa_xdp, + .ndo_xdp_xmit = dpaa_xdp_xmit, + .ndo_hwtstamp_get = dpaa_hwtstamp_get, + .ndo_hwtstamp_set = dpaa_hwtstamp_set, }; static int dpaa_napi_add(struct net_device *net_dev) @@ -2737,8 +3170,7 @@ static int dpaa_napi_add(struct net_device *net_dev) for_each_possible_cpu(cpu) { percpu_priv = per_cpu_ptr(priv->percpu_priv, cpu); - netif_napi_add(net_dev, &percpu_priv->np.napi, - dpaa_eth_poll, NAPI_POLL_WEIGHT); + netif_napi_add(net_dev, &percpu_priv->np.napi, dpaa_eth_poll); } return 0; @@ -2753,8 +3185,9 @@ static void dpaa_napi_del(struct net_device *net_dev) for_each_possible_cpu(cpu) { percpu_priv = per_cpu_ptr(priv->percpu_priv, cpu); - netif_napi_del(&percpu_priv->np.napi); + __netif_napi_del(&percpu_priv->np.napi); } + synchronize_net(); } static inline void dpaa_bp_free_pf(const struct dpaa_bp *bp, @@ -2840,7 +3273,8 @@ out_error: return err; } -static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl) +static u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl, + enum port_type port) { u16 headroom; @@ -2854,10 +3288,18 @@ static inline u16 dpaa_get_headroom(struct dpaa_buffer_layout *bl) * * Also make sure the headroom is a multiple of data_align bytes */ - headroom = (u16)(bl->priv_data_size + DPAA_PARSE_RESULTS_SIZE + - DPAA_TIME_STAMP_SIZE + DPAA_HASH_RESULTS_SIZE); + headroom = (u16)(bl[port].priv_data_size + DPAA_HWA_SIZE); - return ALIGN(headroom, DPAA_FD_DATA_ALIGNMENT); + if (port == RX) { +#ifdef CONFIG_DPAA_ERRATUM_A050385 + if (unlikely(fman_has_errata_a050385())) + headroom = XDP_PACKET_HEADROOM; +#endif + + return ALIGN(headroom, DPAA_FD_RX_DATA_ALIGNMENT); + } else { + return ALIGN(headroom, DPAA_FD_DATA_ALIGNMENT); + } } static int dpaa_eth_probe(struct platform_device *pdev) @@ -2907,7 +3349,7 @@ static int dpaa_eth_probe(struct platform_device *pdev) /* Allocate this early, so we can store relevant information in * the private area */ - net_dev = alloc_etherdev_mq(sizeof(*priv), DPAA_ETH_TXQ_NUM); + net_dev = alloc_etherdev_mq(sizeof(*priv), dpaa_max_num_txqs()); if (!net_dev) { dev_err(dev, "alloc_etherdev_mq() failed\n"); return -ENOMEM; @@ -2922,6 +3364,22 @@ static int dpaa_eth_probe(struct platform_device *pdev) priv->msg_enable = netif_msg_init(debug, DPAA_MSG_DEFAULT); + priv->egress_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(), + sizeof(*priv->egress_fqs), + GFP_KERNEL); + if (!priv->egress_fqs) { + err = -ENOMEM; + goto free_netdev; + } + + priv->conf_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(), + sizeof(*priv->conf_fqs), + GFP_KERNEL); + if (!priv->conf_fqs) { + err = -ENOMEM; + goto free_netdev; + } + mac_dev = dpaa_mac_dev_get(pdev); if (IS_ERR(mac_dev)) { netdev_err(net_dev, "dpaa_mac_dev_get() failed\n"); @@ -2938,7 +3396,7 @@ static int dpaa_eth_probe(struct platform_device *pdev) DMA_BIT_MASK(40)); if (err) { netdev_err(net_dev, "dma_coerce_mask_and_coherent() failed\n"); - return err; + goto free_netdev; } /* If fsl_fm_max_frm is set to a higher value than the all-common 1500, @@ -2999,7 +3457,9 @@ static int dpaa_eth_probe(struct platform_device *pdev) */ dpaa_eth_add_channel(priv->channel, &pdev->dev); - dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]); + err = dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]); + if (err) + goto free_dpaa_bps; /* Create a congestion group for this netdev, with * dynamically-allocated CGR ID. @@ -3025,8 +3485,8 @@ static int dpaa_eth_probe(struct platform_device *pdev) goto free_dpaa_fqs; } - priv->tx_headroom = dpaa_get_headroom(&priv->buf_layout[TX]); - priv->rx_headroom = dpaa_get_headroom(&priv->buf_layout[RX]); + priv->tx_headroom = dpaa_get_headroom(priv->buf_layout, TX); + priv->rx_headroom = dpaa_get_headroom(priv->buf_layout, RX); /* All real interfaces need their ports initialized */ err = dpaa_eth_init_ports(mac_dev, dpaa_bp, &port_fqs, @@ -3045,7 +3505,8 @@ static int dpaa_eth_probe(struct platform_device *pdev) } priv->num_tc = 1; - netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM); + netif_set_real_num_tx_queues(net_dev, + priv->num_tc * dpaa_num_txqs_per_tc()); /* Initialize NAPI */ err = dpaa_napi_add(net_dev); @@ -3081,7 +3542,7 @@ free_netdev: return err; } -static int dpaa_remove(struct platform_device *pdev) +static void dpaa_remove(struct platform_device *pdev) { struct net_device *net_dev; struct dpaa_priv *priv; @@ -3097,8 +3558,12 @@ static int dpaa_remove(struct platform_device *pdev) dev_set_drvdata(dev, NULL); unregister_netdev(net_dev); + phylink_destroy(priv->mac_dev->phylink); err = dpaa_fq_free(dev, &priv->dpaa_fq_list); + if (err) + dev_err(dev, "Failed to free FQs on remove (%pE)\n", + ERR_PTR(err)); qman_delete_cgr_safe(&priv->ingress_cgr); qman_release_cgrid(priv->ingress_cgr.cgrid); @@ -3110,8 +3575,6 @@ static int dpaa_remove(struct platform_device *pdev) dpaa_bps_free(priv); free_netdev(net_dev); - - return err; } static const struct platform_device_id dpaa_devtype[] = { |