diff options
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5/core/en_main.c')
-rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4617 |
1 files changed, 2582 insertions, 2035 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 21de4764d4c0..e3a4f01bcceb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -31,25 +31,23 @@ */ #include <net/tc_act/tc_gact.h> -#include <net/pkt_cls.h> #include <linux/mlx5/fs.h> #include <net/vxlan.h> #include <net/geneve.h> #include <linux/bpf.h> #include <linux/if_bridge.h> +#include <linux/filter.h> #include <net/page_pool.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include "eswitch.h" #include "en.h" #include "en/txrx.h" #include "en_tc.h" #include "en_rep.h" #include "en_accel/ipsec.h" -#include "en_accel/ipsec_rxtx.h" +#include "en_accel/macsec.h" #include "en_accel/en_accel.h" -#include "en_accel/tls.h" -#include "accel/ipsec.h" -#include "accel/tls.h" +#include "en_accel/ktls.h" #include "lib/vxlan.h" #include "lib/clock.h" #include "en/port.h" @@ -58,87 +56,54 @@ #include "en/monitor_stats.h" #include "en/health.h" #include "en/params.h" -#include "en/xsk/umem.h" +#include "en/xsk/pool.h" #include "en/xsk/setup.h" #include "en/xsk/rx.h" #include "en/xsk/tx.h" #include "en/hv_vhca_stats.h" +#include "en/devlink.h" #include "lib/mlx5.h" +#include "en/ptp.h" +#include "en/htb.h" +#include "qos.h" +#include "en/trap.h" - -bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev) +bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev, u8 page_shift, + enum mlx5e_mpwrq_umr_mode umr_mode) { - bool striding_rq_umr = MLX5_CAP_GEN(mdev, striding_rq) && - MLX5_CAP_GEN(mdev, umr_ptr_rlky) && - MLX5_CAP_ETH(mdev, reg_umr_sq); - u16 max_wqe_sz_cap = MLX5_CAP_GEN(mdev, max_wqe_sz_sq); - bool inline_umr = MLX5E_UMR_WQE_INLINE_SZ <= max_wqe_sz_cap; + u16 umr_wqebbs, max_wqebbs; + bool striding_rq_umr; + striding_rq_umr = MLX5_CAP_GEN(mdev, striding_rq) && MLX5_CAP_GEN(mdev, umr_ptr_rlky) && + MLX5_CAP_ETH(mdev, reg_umr_sq); if (!striding_rq_umr) return false; - if (!inline_umr) { - mlx5_core_warn(mdev, "Cannot support Striding RQ: UMR WQE size (%d) exceeds maximum supported (%d).\n", - (int)MLX5E_UMR_WQE_INLINE_SZ, max_wqe_sz_cap); - return false; - } - return true; -} - -void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, - struct mlx5e_params *params) -{ - params->log_rq_mtu_frames = is_kdump_kernel() ? - MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE : - MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE; - mlx5_core_info(mdev, "MLX5E: StrdRq(%d) RqSz(%ld) StrdSz(%ld) RxCqeCmprss(%d)\n", - params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ, - params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ ? - BIT(mlx5e_mpwqe_get_log_rq_size(params, NULL)) : - BIT(params->log_rq_mtu_frames), - BIT(mlx5e_mpwqe_get_log_stride_size(mdev, params, NULL)), - MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)); -} - -bool mlx5e_striding_rq_possible(struct mlx5_core_dev *mdev, - struct mlx5e_params *params) -{ - if (!mlx5e_check_fragmented_striding_rq_cap(mdev)) - return false; - - if (MLX5_IPSEC_DEV(mdev)) + umr_wqebbs = mlx5e_mpwrq_umr_wqebbs(mdev, page_shift, umr_mode); + max_wqebbs = mlx5e_get_max_sq_aligned_wqebbs(mdev); + /* Sanity check; should never happen, because mlx5e_mpwrq_umr_wqebbs is + * calculated from mlx5e_get_max_sq_aligned_wqebbs. + */ + if (WARN_ON(umr_wqebbs > max_wqebbs)) return false; - if (params->xdp_prog) { - /* XSK params are not considered here. If striding RQ is in use, - * and an XSK is being opened, mlx5e_rx_mpwqe_is_linear_skb will - * be called with the known XSK params. - */ - if (!mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL)) - return false; - } - return true; } -void mlx5e_set_rq_type(struct mlx5_core_dev *mdev, struct mlx5e_params *params) -{ - params->rq_wq_type = mlx5e_striding_rq_possible(mdev, params) && - MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ) ? - MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ : - MLX5_WQ_TYPE_CYCLIC; -} - void mlx5e_update_carrier(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; u8 port_state; + bool up; port_state = mlx5_query_vport_state(mdev, MLX5_VPORT_STATE_OP_MOD_VNIC_VPORT, 0); - if (port_state == VPORT_STATE_UP) { + up = port_state == VPORT_STATE_UP; + if (up == netif_carrier_ok(priv->netdev)) + netif_carrier_event(priv->netdev); + if (up) { netdev_info(priv->netdev, "Link up\n"); netif_carrier_on(priv->netdev); } else { @@ -159,16 +124,6 @@ static void mlx5e_update_carrier_work(struct work_struct *work) mutex_unlock(&priv->state_lock); } -void mlx5e_update_ndo_stats(struct mlx5e_priv *priv) -{ - int i; - - for (i = mlx5e_nic_stats_grps_num(priv) - 1; i >= 0; i--) - if (mlx5e_nic_stats_grps[i]->update_stats_mask & - MLX5E_NDO_UPDATE_STATS) - mlx5e_nic_stats_grps[i]->update_stats(priv); -} - static void mlx5e_update_stats_work(struct work_struct *work) { struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, @@ -221,49 +176,167 @@ static void mlx5e_disable_async_events(struct mlx5e_priv *priv) mlx5_notifier_unregister(priv->mdev, &priv->events_nb); } +static int blocking_event(struct notifier_block *nb, unsigned long event, void *data) +{ + struct mlx5e_priv *priv = container_of(nb, struct mlx5e_priv, blocking_events_nb); + int err; + + switch (event) { + case MLX5_DRIVER_EVENT_TYPE_TRAP: + err = mlx5e_handle_trap_event(priv, data); + break; + default: + netdev_warn(priv->netdev, "Sync event: Unknown event %ld\n", event); + err = -EINVAL; + } + return err; +} + +static void mlx5e_enable_blocking_events(struct mlx5e_priv *priv) +{ + priv->blocking_events_nb.notifier_call = blocking_event; + mlx5_blocking_notifier_register(priv->mdev, &priv->blocking_events_nb); +} + +static void mlx5e_disable_blocking_events(struct mlx5e_priv *priv) +{ + mlx5_blocking_notifier_unregister(priv->mdev, &priv->blocking_events_nb); +} + +static u16 mlx5e_mpwrq_umr_octowords(u32 entries, enum mlx5e_mpwrq_umr_mode umr_mode) +{ + u8 umr_entry_size = mlx5e_mpwrq_umr_entry_size(umr_mode); + + WARN_ON_ONCE(entries * umr_entry_size % MLX5_OCTWORD); + + return entries * umr_entry_size / MLX5_OCTWORD; +} + static inline void mlx5e_build_umr_wqe(struct mlx5e_rq *rq, struct mlx5e_icosq *sq, struct mlx5e_umr_wqe *wqe) { struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; struct mlx5_wqe_umr_ctrl_seg *ucseg = &wqe->uctrl; - u8 ds_cnt = DIV_ROUND_UP(MLX5E_UMR_WQE_INLINE_SZ, MLX5_SEND_WQE_DS); + u16 octowords; + u8 ds_cnt; + + ds_cnt = DIV_ROUND_UP(mlx5e_mpwrq_umr_wqe_sz(rq->mdev, rq->mpwqe.page_shift, + rq->mpwqe.umr_mode), + MLX5_SEND_WQE_DS); cseg->qpn_ds = cpu_to_be32((sq->sqn << MLX5_WQE_CTRL_QPN_SHIFT) | ds_cnt); - cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; - cseg->imm = rq->mkey_be; + cseg->umr_mkey = rq->mpwqe.umr_mkey_be; ucseg->flags = MLX5_UMR_TRANSLATION_OFFSET_EN | MLX5_UMR_INLINE; - ucseg->xlt_octowords = - cpu_to_be16(MLX5_MTT_OCTW(MLX5_MPWRQ_PAGES_PER_WQE)); + octowords = mlx5e_mpwrq_umr_octowords(rq->mpwqe.pages_per_wqe, rq->mpwqe.umr_mode); + ucseg->xlt_octowords = cpu_to_be16(octowords); ucseg->mkey_mask = cpu_to_be64(MLX5_MKEY_MASK_FREE); } -static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, - struct mlx5e_channel *c) +static int mlx5e_rq_shampo_hd_alloc(struct mlx5e_rq *rq, int node) +{ + rq->mpwqe.shampo = kvzalloc_node(sizeof(*rq->mpwqe.shampo), + GFP_KERNEL, node); + if (!rq->mpwqe.shampo) + return -ENOMEM; + return 0; +} + +static void mlx5e_rq_shampo_hd_free(struct mlx5e_rq *rq) +{ + kvfree(rq->mpwqe.shampo); +} + +static int mlx5e_rq_shampo_hd_info_alloc(struct mlx5e_rq *rq, int node) +{ + struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo; + + shampo->bitmap = bitmap_zalloc_node(shampo->hd_per_wq, GFP_KERNEL, + node); + if (!shampo->bitmap) + return -ENOMEM; + + shampo->info = kvzalloc_node(array_size(shampo->hd_per_wq, + sizeof(*shampo->info)), + GFP_KERNEL, node); + if (!shampo->info) { + kvfree(shampo->bitmap); + return -ENOMEM; + } + return 0; +} + +static void mlx5e_rq_shampo_hd_info_free(struct mlx5e_rq *rq) +{ + kvfree(rq->mpwqe.shampo->bitmap); + kvfree(rq->mpwqe.shampo->info); +} + +static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node) { int wq_sz = mlx5_wq_ll_get_size(&rq->mpwqe.wq); + size_t alloc_size; + + alloc_size = array_size(wq_sz, struct_size(rq->mpwqe.info, alloc_units, + rq->mpwqe.pages_per_wqe)); - rq->mpwqe.info = kvzalloc_node(array_size(wq_sz, - sizeof(*rq->mpwqe.info)), - GFP_KERNEL, cpu_to_node(c->cpu)); + rq->mpwqe.info = kvzalloc_node(alloc_size, GFP_KERNEL, node); if (!rq->mpwqe.info) return -ENOMEM; - mlx5e_build_umr_wqe(rq, &c->icosq, &rq->mpwqe.umr_wqe); + mlx5e_build_umr_wqe(rq, rq->icosq, &rq->mpwqe.umr_wqe); return 0; } -static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, - u64 npages, u8 page_shift, - struct mlx5_core_mkey *umr_mkey) + +static u8 mlx5e_mpwrq_access_mode(enum mlx5e_mpwrq_umr_mode umr_mode) { - int inlen = MLX5_ST_SZ_BYTES(create_mkey_in); + switch (umr_mode) { + case MLX5E_MPWRQ_UMR_MODE_ALIGNED: + return MLX5_MKC_ACCESS_MODE_MTT; + case MLX5E_MPWRQ_UMR_MODE_UNALIGNED: + return MLX5_MKC_ACCESS_MODE_KSM; + case MLX5E_MPWRQ_UMR_MODE_OVERSIZED: + return MLX5_MKC_ACCESS_MODE_KLMS; + case MLX5E_MPWRQ_UMR_MODE_TRIPLE: + return MLX5_MKC_ACCESS_MODE_KSM; + } + WARN_ONCE(1, "MPWRQ UMR mode %d is not known\n", umr_mode); + return 0; +} + +static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, + u32 npages, u8 page_shift, u32 *umr_mkey, + dma_addr_t filler_addr, + enum mlx5e_mpwrq_umr_mode umr_mode, + u32 xsk_chunk_size) +{ + struct mlx5_mtt *mtt; + struct mlx5_ksm *ksm; + struct mlx5_klm *klm; + u32 octwords; + int inlen; void *mkc; u32 *in; int err; + int i; + + if ((umr_mode == MLX5E_MPWRQ_UMR_MODE_UNALIGNED || + umr_mode == MLX5E_MPWRQ_UMR_MODE_TRIPLE) && + !MLX5_CAP_GEN(mdev, fixed_buffer_size)) { + mlx5_core_warn(mdev, "Unaligned AF_XDP requires fixed_buffer_size capability\n"); + return -EINVAL; + } + + octwords = mlx5e_mpwrq_umr_octowords(npages, umr_mode); + + inlen = MLX5_FLEXIBLE_INLEN(mdev, MLX5_ST_SZ_BYTES(create_mkey_in), + MLX5_OCTWORD, octwords); + if (inlen < 0) + return inlen; in = kvzalloc(inlen, GFP_KERNEL); if (!in) @@ -275,15 +348,99 @@ static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, MLX5_SET(mkc, mkc, umr_en, 1); MLX5_SET(mkc, mkc, lw, 1); MLX5_SET(mkc, mkc, lr, 1); - MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_MTT); - + MLX5_SET(mkc, mkc, access_mode_1_0, mlx5e_mpwrq_access_mode(umr_mode)); + mlx5e_mkey_set_relaxed_ordering(mdev, mkc); MLX5_SET(mkc, mkc, qpn, 0xffffff); - MLX5_SET(mkc, mkc, pd, mdev->mlx5e_res.pdn); + MLX5_SET(mkc, mkc, pd, mdev->mlx5e_res.hw_objs.pdn); MLX5_SET64(mkc, mkc, len, npages << page_shift); - MLX5_SET(mkc, mkc, translations_octword_size, - MLX5_MTT_OCTW(npages)); - MLX5_SET(mkc, mkc, log_page_size, page_shift); + MLX5_SET(mkc, mkc, translations_octword_size, octwords); + if (umr_mode == MLX5E_MPWRQ_UMR_MODE_TRIPLE) + MLX5_SET(mkc, mkc, log_page_size, page_shift - 2); + else if (umr_mode != MLX5E_MPWRQ_UMR_MODE_OVERSIZED) + MLX5_SET(mkc, mkc, log_page_size, page_shift); + MLX5_SET(create_mkey_in, in, translations_octword_actual_size, octwords); + + /* Initialize the mkey with all MTTs pointing to a default + * page (filler_addr). When the channels are activated, UMR + * WQEs will redirect the RX WQEs to the actual memory from + * the RQ's pool, while the gaps (wqe_overflow) remain mapped + * to the default page. + */ + switch (umr_mode) { + case MLX5E_MPWRQ_UMR_MODE_OVERSIZED: + klm = MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); + for (i = 0; i < npages; i++) { + klm[i << 1] = (struct mlx5_klm) { + .va = cpu_to_be64(filler_addr), + .bcount = cpu_to_be32(xsk_chunk_size), + .key = cpu_to_be32(mdev->mlx5e_res.hw_objs.mkey), + }; + klm[(i << 1) + 1] = (struct mlx5_klm) { + .va = cpu_to_be64(filler_addr), + .bcount = cpu_to_be32((1 << page_shift) - xsk_chunk_size), + .key = cpu_to_be32(mdev->mlx5e_res.hw_objs.mkey), + }; + } + break; + case MLX5E_MPWRQ_UMR_MODE_UNALIGNED: + ksm = MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); + for (i = 0; i < npages; i++) + ksm[i] = (struct mlx5_ksm) { + .key = cpu_to_be32(mdev->mlx5e_res.hw_objs.mkey), + .va = cpu_to_be64(filler_addr), + }; + break; + case MLX5E_MPWRQ_UMR_MODE_ALIGNED: + mtt = MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); + for (i = 0; i < npages; i++) + mtt[i] = (struct mlx5_mtt) { + .ptag = cpu_to_be64(filler_addr), + }; + break; + case MLX5E_MPWRQ_UMR_MODE_TRIPLE: + ksm = MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); + for (i = 0; i < npages * 4; i++) { + ksm[i] = (struct mlx5_ksm) { + .key = cpu_to_be32(mdev->mlx5e_res.hw_objs.mkey), + .va = cpu_to_be64(filler_addr), + }; + } + break; + } + + err = mlx5_core_create_mkey(mdev, umr_mkey, in, inlen); + + kvfree(in); + return err; +} + +static int mlx5e_create_umr_klm_mkey(struct mlx5_core_dev *mdev, + u64 nentries, + u32 *umr_mkey) +{ + int inlen; + void *mkc; + u32 *in; + int err; + + inlen = MLX5_ST_SZ_BYTES(create_mkey_in); + in = kvzalloc(inlen, GFP_KERNEL); + if (!in) + return -ENOMEM; + + mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry); + + MLX5_SET(mkc, mkc, free, 1); + MLX5_SET(mkc, mkc, umr_en, 1); + MLX5_SET(mkc, mkc, lw, 1); + MLX5_SET(mkc, mkc, lr, 1); + MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_KLMS); + mlx5e_mkey_set_relaxed_ordering(mdev, mkc); + MLX5_SET(mkc, mkc, qpn, 0xffffff); + MLX5_SET(mkc, mkc, pd, mdev->mlx5e_res.hw_objs.pdn); + MLX5_SET(mkc, mkc, translations_octword_size, nentries); + MLX5_SET(mkc, mkc, length64, 1); err = mlx5_core_create_mkey(mdev, umr_mkey, in, inlen); kvfree(in); @@ -292,14 +449,41 @@ static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, static int mlx5e_create_rq_umr_mkey(struct mlx5_core_dev *mdev, struct mlx5e_rq *rq) { - u64 num_mtts = MLX5E_REQUIRED_MTTS(mlx5_wq_ll_get_size(&rq->mpwqe.wq)); + u32 xsk_chunk_size = rq->xsk_pool ? rq->xsk_pool->chunk_size : 0; + u32 wq_size = mlx5_wq_ll_get_size(&rq->mpwqe.wq); + u32 num_entries, max_num_entries; + u32 umr_mkey; + int err; + + max_num_entries = mlx5e_mpwrq_max_num_entries(mdev, rq->mpwqe.umr_mode); - return mlx5e_create_umr_mkey(mdev, num_mtts, PAGE_SHIFT, &rq->umr_mkey); + /* Shouldn't overflow, the result is at most MLX5E_MAX_RQ_NUM_MTTS. */ + if (WARN_ON_ONCE(check_mul_overflow(wq_size, (u32)rq->mpwqe.mtts_per_wqe, + &num_entries) || + num_entries > max_num_entries)) + mlx5_core_err(mdev, "%s: multiplication overflow: %u * %u > %u\n", + __func__, wq_size, rq->mpwqe.mtts_per_wqe, + max_num_entries); + + err = mlx5e_create_umr_mkey(mdev, num_entries, rq->mpwqe.page_shift, + &umr_mkey, rq->wqe_overflow.addr, + rq->mpwqe.umr_mode, xsk_chunk_size); + rq->mpwqe.umr_mkey_be = cpu_to_be32(umr_mkey); + return err; } -static inline u64 mlx5e_get_mpwqe_offset(struct mlx5e_rq *rq, u16 wqe_ix) +static int mlx5e_create_rq_hd_umr_mkey(struct mlx5_core_dev *mdev, + struct mlx5e_rq *rq) { - return (wqe_ix << MLX5E_LOG_ALIGNED_MPWQE_PPW) << PAGE_SHIFT; + u32 max_klm_size = BIT(MLX5_CAP_GEN(mdev, log_max_klm_list_size)); + + if (max_klm_size < rq->mpwqe.shampo->hd_per_wq) { + mlx5_core_err(mdev, "max klm list size 0x%x is smaller than shampo header buffer list size 0x%x\n", + max_klm_size, rq->mpwqe.shampo->hd_per_wq); + return -EINVAL; + } + return mlx5e_create_umr_klm_mkey(mdev, rq->mpwqe.shampo->hd_per_wq, + &rq->mpwqe.shampo->mkey); } static void mlx5e_init_frags_partition(struct mlx5e_rq *rq) @@ -308,7 +492,14 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq) struct mlx5e_wqe_frag_info *prev = NULL; int i; - next_frag.di = &rq->wqe.di[0]; + if (rq->xsk_pool) { + /* Assumptions used by XSK batched allocator. */ + WARN_ON(rq->wqe.info.num_frags != 1); + WARN_ON(rq->wqe.info.log_num_frags != 0); + WARN_ON(rq->wqe.info.arr[0].frag_stride != PAGE_SIZE); + } + + next_frag.au = &rq->wqe.alloc_units[0]; for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++) { struct mlx5e_rq_frag_info *frag_info = &rq->wqe.info.arr[0]; @@ -318,7 +509,7 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq) for (f = 0; f < rq->wqe.info.num_frags; f++, frag++) { if (next_frag.offset + frag_info[f].frag_stride > PAGE_SIZE) { - next_frag.di++; + next_frag.au++; next_frag.offset = 0; if (prev) prev->last_in_page = true; @@ -335,14 +526,13 @@ static void mlx5e_init_frags_partition(struct mlx5e_rq *rq) prev->last_in_page = true; } -static int mlx5e_init_di_list(struct mlx5e_rq *rq, - int wq_sz, int cpu) +static int mlx5e_init_au_list(struct mlx5e_rq *rq, int wq_sz, int node) { int len = wq_sz << rq->wqe.info.log_num_frags; - rq->wqe.di = kvzalloc_node(array_size(len, sizeof(*rq->wqe.di)), - GFP_KERNEL, cpu_to_node(cpu)); - if (!rq->wqe.di) + rq->wqe.alloc_units = kvzalloc_node(array_size(len, sizeof(*rq->wqe.alloc_units)), + GFP_KERNEL, node); + if (!rq->wqe.alloc_units) return -ENOMEM; mlx5e_init_frags_partition(rq); @@ -350,9 +540,9 @@ static int mlx5e_init_di_list(struct mlx5e_rq *rq, return 0; } -static void mlx5e_free_di_list(struct mlx5e_rq *rq) +static void mlx5e_free_au_list(struct mlx5e_rq *rq) { - kvfree(rq->wqe.di); + kvfree(rq->wqe.alloc_units); } static void mlx5e_rq_err_cqe_work(struct work_struct *recover_work) @@ -362,183 +552,229 @@ static void mlx5e_rq_err_cqe_work(struct work_struct *recover_work) mlx5e_reporter_rq_cqe_err(rq); } -static int mlx5e_alloc_rq(struct mlx5e_channel *c, - struct mlx5e_params *params, +static int mlx5e_alloc_mpwqe_rq_drop_page(struct mlx5e_rq *rq) +{ + rq->wqe_overflow.page = alloc_page(GFP_KERNEL); + if (!rq->wqe_overflow.page) + return -ENOMEM; + + rq->wqe_overflow.addr = dma_map_page(rq->pdev, rq->wqe_overflow.page, 0, + PAGE_SIZE, rq->buff.map_dir); + if (dma_mapping_error(rq->pdev, rq->wqe_overflow.addr)) { + __free_page(rq->wqe_overflow.page); + return -ENOMEM; + } + return 0; +} + +static void mlx5e_free_mpwqe_rq_drop_page(struct mlx5e_rq *rq) +{ + dma_unmap_page(rq->pdev, rq->wqe_overflow.addr, PAGE_SIZE, + rq->buff.map_dir); + __free_page(rq->wqe_overflow.page); +} + +static int mlx5e_init_rxq_rq(struct mlx5e_channel *c, struct mlx5e_params *params, + struct mlx5e_rq *rq) +{ + struct mlx5_core_dev *mdev = c->mdev; + int err; + + rq->wq_type = params->rq_wq_type; + rq->pdev = c->pdev; + rq->netdev = c->netdev; + rq->priv = c->priv; + rq->tstamp = c->tstamp; + rq->clock = &mdev->clock; + rq->icosq = &c->icosq; + rq->ix = c->ix; + rq->channel = c; + rq->mdev = mdev; + rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); + rq->xdpsq = &c->rq_xdpsq; + rq->stats = &c->priv->channel_stats[c->ix]->rq; + rq->ptp_cyc2time = mlx5_rq_ts_translator(mdev); + err = mlx5e_rq_set_handlers(rq, params, NULL); + if (err) + return err; + + return xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq->ix, c->napi.napi_id); +} + +static int mlx5_rq_shampo_alloc(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, + struct mlx5e_rq_param *rqp, + struct mlx5e_rq *rq, + u32 *pool_size, + int node) +{ + void *wqc = MLX5_ADDR_OF(rqc, rqp->rqc, wq); + int wq_size; + int err; + + if (!test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) + return 0; + err = mlx5e_rq_shampo_hd_alloc(rq, node); + if (err) + goto out; + rq->mpwqe.shampo->hd_per_wq = + mlx5e_shampo_hd_per_wq(mdev, params, rqp); + err = mlx5e_create_rq_hd_umr_mkey(mdev, rq); + if (err) + goto err_shampo_hd; + err = mlx5e_rq_shampo_hd_info_alloc(rq, node); + if (err) + goto err_shampo_info; + rq->hw_gro_data = kvzalloc_node(sizeof(*rq->hw_gro_data), GFP_KERNEL, node); + if (!rq->hw_gro_data) { + err = -ENOMEM; + goto err_hw_gro_data; + } + rq->mpwqe.shampo->key = + cpu_to_be32(rq->mpwqe.shampo->mkey); + rq->mpwqe.shampo->hd_per_wqe = + mlx5e_shampo_hd_per_wqe(mdev, params, rqp); + wq_size = BIT(MLX5_GET(wq, wqc, log_wq_sz)); + *pool_size += (rq->mpwqe.shampo->hd_per_wqe * wq_size) / + MLX5E_SHAMPO_WQ_HEADER_PER_PAGE; + return 0; + +err_hw_gro_data: + mlx5e_rq_shampo_hd_info_free(rq); +err_shampo_info: + mlx5_core_destroy_mkey(mdev, rq->mpwqe.shampo->mkey); +err_shampo_hd: + mlx5e_rq_shampo_hd_free(rq); +out: + return err; +} + +static void mlx5e_rq_free_shampo(struct mlx5e_rq *rq) +{ + if (!test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) + return; + + kvfree(rq->hw_gro_data); + mlx5e_rq_shampo_hd_info_free(rq); + mlx5_core_destroy_mkey(rq->mdev, rq->mpwqe.shampo->mkey); + mlx5e_rq_shampo_hd_free(rq); +} + +static int mlx5e_alloc_rq(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk, - struct xdp_umem *umem, struct mlx5e_rq_param *rqp, - struct mlx5e_rq *rq) + int node, struct mlx5e_rq *rq) { struct page_pool_params pp_params = { 0 }; - struct mlx5_core_dev *mdev = c->mdev; + struct mlx5_core_dev *mdev = rq->mdev; void *rqc = rqp->rqc; void *rqc_wq = MLX5_ADDR_OF(rqc, rqc, wq); - u32 num_xsk_frames = 0; - u32 rq_xdp_ix; u32 pool_size; int wq_sz; int err; int i; - rqp->wq.db_numa_node = cpu_to_node(c->cpu); - - rq->wq_type = params->rq_wq_type; - rq->pdev = c->pdev; - rq->netdev = c->netdev; - rq->tstamp = c->tstamp; - rq->clock = &mdev->clock; - rq->channel = c; - rq->ix = c->ix; - rq->mdev = mdev; - rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); - rq->xdpsq = &c->rq_xdpsq; - rq->umem = umem; - - if (rq->umem) - rq->stats = &c->priv->channel_stats[c->ix].xskrq; - else - rq->stats = &c->priv->channel_stats[c->ix].rq; + rqp->wq.db_numa_node = node; INIT_WORK(&rq->recover_work, mlx5e_rq_err_cqe_work); if (params->xdp_prog) bpf_prog_inc(params->xdp_prog); - rq->xdp_prog = params->xdp_prog; - - rq_xdp_ix = rq->ix; - if (xsk) - rq_xdp_ix += params->num_channels * MLX5E_RQ_GROUP_XSK; - err = xdp_rxq_info_reg(&rq->xdp_rxq, rq->netdev, rq_xdp_ix); - if (err < 0) - goto err_rq_wq_destroy; + RCU_INIT_POINTER(rq->xdp_prog, params->xdp_prog); - rq->buff.map_dir = rq->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; + rq->buff.map_dir = params->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; rq->buff.headroom = mlx5e_get_rq_headroom(mdev, params, xsk); - rq->buff.umem_headroom = xsk ? xsk->headroom : 0; pool_size = 1 << params->log_rq_mtu_frames; + rq->mkey_be = cpu_to_be32(mdev->mlx5e_res.hw_objs.mkey); + switch (rq->wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: err = mlx5_wq_ll_create(mdev, &rqp->wq, rqc_wq, &rq->mpwqe.wq, &rq->wq_ctrl); if (err) - return err; + goto err_rq_xdp_prog; + + err = mlx5e_alloc_mpwqe_rq_drop_page(rq); + if (err) + goto err_rq_wq_destroy; rq->mpwqe.wq.db = &rq->mpwqe.wq.db[MLX5_RCV_DBR]; wq_sz = mlx5_wq_ll_get_size(&rq->mpwqe.wq); - if (xsk) - num_xsk_frames = wq_sz << - mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk); - - pool_size = MLX5_MPWRQ_PAGES_PER_WQE << - mlx5e_mpwqe_get_log_rq_size(params, xsk); - - rq->post_wqes = mlx5e_post_rx_mpwqes; - rq->dealloc_wqe = mlx5e_dealloc_rx_mpwqe; - - rq->handle_rx_cqe = c->priv->profile->rx_handlers.handle_rx_cqe_mpwqe; -#ifdef CONFIG_MLX5_EN_IPSEC - if (MLX5_IPSEC_DEV(mdev)) { - err = -EINVAL; - netdev_err(c->netdev, "MPWQE RQ with IPSec offload not supported\n"); - goto err_rq_wq_destroy; - } -#endif - if (!rq->handle_rx_cqe) { - err = -EINVAL; - netdev_err(c->netdev, "RX handler of MPWQE RQ is not set, err %d\n", err); - goto err_rq_wq_destroy; - } - - rq->mpwqe.skb_from_cqe_mpwrq = xsk ? - mlx5e_xsk_skb_from_cqe_mpwrq_linear : - mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL) ? - mlx5e_skb_from_cqe_mpwrq_linear : - mlx5e_skb_from_cqe_mpwrq_nonlinear; + rq->mpwqe.page_shift = mlx5e_mpwrq_page_shift(mdev, xsk); + rq->mpwqe.umr_mode = mlx5e_mpwrq_umr_mode(mdev, xsk); + rq->mpwqe.pages_per_wqe = + mlx5e_mpwrq_pages_per_wqe(mdev, rq->mpwqe.page_shift, + rq->mpwqe.umr_mode); + rq->mpwqe.umr_wqebbs = + mlx5e_mpwrq_umr_wqebbs(mdev, rq->mpwqe.page_shift, + rq->mpwqe.umr_mode); + rq->mpwqe.mtts_per_wqe = + mlx5e_mpwrq_mtts_per_wqe(mdev, rq->mpwqe.page_shift, + rq->mpwqe.umr_mode); + + pool_size = rq->mpwqe.pages_per_wqe << + mlx5e_mpwqe_get_log_rq_size(mdev, params, xsk); rq->mpwqe.log_stride_sz = mlx5e_mpwqe_get_log_stride_size(mdev, params, xsk); rq->mpwqe.num_strides = BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk)); + rq->mpwqe.min_wqe_bulk = mlx5e_mpwqe_get_min_wqe_bulk(wq_sz); + + rq->buff.frame0_sz = (1 << rq->mpwqe.log_stride_sz); err = mlx5e_create_rq_umr_mkey(mdev, rq); if (err) - goto err_rq_wq_destroy; - rq->mkey_be = cpu_to_be32(rq->umr_mkey.key); + goto err_rq_drop_page; + + err = mlx5e_rq_alloc_mpwqe_info(rq, node); + if (err) + goto err_rq_mkey; - err = mlx5e_rq_alloc_mpwqe_info(rq, c); + err = mlx5_rq_shampo_alloc(mdev, params, rqp, rq, &pool_size, node); if (err) - goto err_free; + goto err_free_mpwqe_info; + break; default: /* MLX5_WQ_TYPE_CYCLIC */ err = mlx5_wq_cyc_create(mdev, &rqp->wq, rqc_wq, &rq->wqe.wq, &rq->wq_ctrl); if (err) - return err; + goto err_rq_xdp_prog; rq->wqe.wq.db = &rq->wqe.wq.db[MLX5_RCV_DBR]; wq_sz = mlx5_wq_cyc_get_size(&rq->wqe.wq); - if (xsk) - num_xsk_frames = wq_sz << rq->wqe.info.log_num_frags; - rq->wqe.info = rqp->frags_info; + rq->buff.frame0_sz = rq->wqe.info.arr[0].frag_stride; + rq->wqe.frags = kvzalloc_node(array_size(sizeof(*rq->wqe.frags), (wq_sz << rq->wqe.info.log_num_frags)), - GFP_KERNEL, cpu_to_node(c->cpu)); + GFP_KERNEL, node); if (!rq->wqe.frags) { err = -ENOMEM; - goto err_free; + goto err_rq_wq_destroy; } - err = mlx5e_init_di_list(rq, wq_sz, c->cpu); + err = mlx5e_init_au_list(rq, wq_sz, node); if (err) - goto err_free; - - rq->post_wqes = mlx5e_post_rx_wqes; - rq->dealloc_wqe = mlx5e_dealloc_rx_wqe; - -#ifdef CONFIG_MLX5_EN_IPSEC - if (c->priv->ipsec) - rq->handle_rx_cqe = mlx5e_ipsec_handle_rx_cqe; - else -#endif - rq->handle_rx_cqe = c->priv->profile->rx_handlers.handle_rx_cqe; - if (!rq->handle_rx_cqe) { - err = -EINVAL; - netdev_err(c->netdev, "RX handler of RQ is not set, err %d\n", err); - goto err_free; - } - - rq->wqe.skb_from_cqe = xsk ? - mlx5e_xsk_skb_from_cqe_linear : - mlx5e_rx_is_linear_skb(params, NULL) ? - mlx5e_skb_from_cqe_linear : - mlx5e_skb_from_cqe_nonlinear; - rq->mkey_be = c->mkey_be; + goto err_rq_frags; } if (xsk) { - err = mlx5e_xsk_resize_reuseq(umem, num_xsk_frames); - if (unlikely(err)) { - mlx5_core_err(mdev, "Unable to allocate the Reuse Ring for %u frames\n", - num_xsk_frames); - goto err_free; - } - - rq->zca.free = mlx5e_xsk_zca_free; err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq, - MEM_TYPE_ZERO_COPY, - &rq->zca); + MEM_TYPE_XSK_BUFF_POOL, NULL); + xsk_pool_set_rxq_info(rq->xsk_pool, &rq->xdp_rxq); } else { /* Create a page_pool and register it with rxq */ pp_params.order = 0; pp_params.flags = 0; /* No-internal DMA mapping in page_pool */ pp_params.pool_size = pool_size; - pp_params.nid = cpu_to_node(c->cpu); - pp_params.dev = c->pdev; + pp_params.nid = node; + pp_params.dev = rq->pdev; pp_params.dma_dir = rq->buff.map_dir; /* page_pool can be used even when there is no rq->xdp_prog, @@ -550,13 +786,14 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, if (IS_ERR(rq->page_pool)) { err = PTR_ERR(rq->page_pool); rq->page_pool = NULL; - goto err_free; + goto err_free_by_rq_type; } - err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq, - MEM_TYPE_PAGE_POOL, rq->page_pool); + if (xdp_rxq_info_is_reg(&rq->xdp_rxq)) + err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq, + MEM_TYPE_PAGE_POOL, rq->page_pool); } if (err) - goto err_free; + goto err_destroy_page_pool; for (i = 0; i < wq_sz; i++) { if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { @@ -564,11 +801,14 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, mlx5_wq_ll_get_wqe(&rq->mpwqe.wq, i); u32 byte_count = rq->mpwqe.num_strides << rq->mpwqe.log_stride_sz; - u64 dma_offset = mlx5e_get_mpwqe_offset(rq, i); + u64 dma_offset = mul_u32_u32(i, rq->mpwqe.mtts_per_wqe) << + rq->mpwqe.page_shift; + u16 headroom = test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state) ? + 0 : rq->buff.headroom; - wqe->data[0].addr = cpu_to_be64(dma_offset + rq->buff.headroom); + wqe->data[0].addr = cpu_to_be64(dma_offset + headroom); wqe->data[0].byte_count = cpu_to_be32(byte_count); - wqe->data[0].lkey = rq->mkey_be; + wqe->data[0].lkey = rq->mpwqe.umr_mkey_be; } else { struct mlx5e_rx_wqe_cyc *wqe = mlx5_wq_cyc_get_wqe(&rq->wqe.wq, i); @@ -606,53 +846,64 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, return 0; -err_free: +err_destroy_page_pool: + page_pool_destroy(rq->page_pool); +err_free_by_rq_type: switch (rq->wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: + mlx5e_rq_free_shampo(rq); +err_free_mpwqe_info: kvfree(rq->mpwqe.info); - mlx5_core_destroy_mkey(mdev, &rq->umr_mkey); +err_rq_mkey: + mlx5_core_destroy_mkey(mdev, be32_to_cpu(rq->mpwqe.umr_mkey_be)); +err_rq_drop_page: + mlx5e_free_mpwqe_rq_drop_page(rq); break; default: /* MLX5_WQ_TYPE_CYCLIC */ + mlx5e_free_au_list(rq); +err_rq_frags: kvfree(rq->wqe.frags); - mlx5e_free_di_list(rq); } - err_rq_wq_destroy: - if (rq->xdp_prog) - bpf_prog_put(rq->xdp_prog); - xdp_rxq_info_unreg(&rq->xdp_rxq); - page_pool_destroy(rq->page_pool); mlx5_wq_destroy(&rq->wq_ctrl); +err_rq_xdp_prog: + if (params->xdp_prog) + bpf_prog_put(params->xdp_prog); return err; } static void mlx5e_free_rq(struct mlx5e_rq *rq) { + struct bpf_prog *old_prog; int i; - if (rq->xdp_prog) - bpf_prog_put(rq->xdp_prog); + if (xdp_rxq_info_is_reg(&rq->xdp_rxq)) { + old_prog = rcu_dereference_protected(rq->xdp_prog, + lockdep_is_held(&rq->priv->state_lock)); + if (old_prog) + bpf_prog_put(old_prog); + } switch (rq->wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: kvfree(rq->mpwqe.info); - mlx5_core_destroy_mkey(rq->mdev, &rq->umr_mkey); + mlx5_core_destroy_mkey(rq->mdev, be32_to_cpu(rq->mpwqe.umr_mkey_be)); + mlx5e_free_mpwqe_rq_drop_page(rq); + mlx5e_rq_free_shampo(rq); break; default: /* MLX5_WQ_TYPE_CYCLIC */ kvfree(rq->wqe.frags); - mlx5e_free_di_list(rq); + mlx5e_free_au_list(rq); } for (i = rq->page_cache.head; i != rq->page_cache.tail; i = (i + 1) & (MLX5E_CACHE_SIZE - 1)) { - struct mlx5e_dma_info *dma_info = &rq->page_cache.page_cache[i]; - /* With AF_XDP, page_cache is not used, so this loop is not * entered, and it's safe to call mlx5e_page_release_dynamic * directly. */ - mlx5e_page_release_dynamic(rq, dma_info, false); + mlx5e_page_release_dynamic(rq, rq->page_cache.page_cache[i], false); } xdp_rxq_info_unreg(&rq->xdp_rxq); @@ -660,11 +911,10 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq) mlx5_wq_destroy(&rq->wq_ctrl); } -static int mlx5e_create_rq(struct mlx5e_rq *rq, - struct mlx5e_rq_param *param) +int mlx5e_create_rq(struct mlx5e_rq *rq, struct mlx5e_rq_param *param) { struct mlx5_core_dev *mdev = rq->mdev; - + u8 ts_format; void *in; void *rqc; void *wq; @@ -677,6 +927,9 @@ static int mlx5e_create_rq(struct mlx5e_rq *rq, if (!in) return -ENOMEM; + ts_format = mlx5_is_real_time_rq(mdev) ? + MLX5_TIMESTAMP_FORMAT_REAL_TIME : + MLX5_TIMESTAMP_FORMAT_FREE_RUNNING; rqc = MLX5_ADDR_OF(create_rq_in, in, ctx); wq = MLX5_ADDR_OF(rqc, rqc, wq); @@ -684,10 +937,17 @@ static int mlx5e_create_rq(struct mlx5e_rq *rq, MLX5_SET(rqc, rqc, cqn, rq->cq.mcq.cqn); MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RST); + MLX5_SET(rqc, rqc, ts_format, ts_format); MLX5_SET(wq, wq, log_wq_pg_sz, rq->wq_ctrl.buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT); MLX5_SET64(wq, wq, dbr_addr, rq->wq_ctrl.db.dma); + if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) { + MLX5_SET(wq, wq, log_headers_buffer_entry_num, + order_base_2(rq->mpwqe.shampo->hd_per_wq)); + MLX5_SET(wq, wq, headers_mkey, rq->mpwqe.shampo->mkey); + } + mlx5_fill_page_frag_array(&rq->wq_ctrl.buf, (__be64 *)MLX5_ADDR_OF(wq, wq, pas)); @@ -698,7 +958,7 @@ static int mlx5e_create_rq(struct mlx5e_rq *rq, return err; } -int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state) +static int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state) { struct mlx5_core_dev *mdev = rq->mdev; @@ -720,18 +980,42 @@ int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state) MLX5_SET(modify_rq_in, in, rq_state, curr_state); MLX5_SET(rqc, rqc, state, next_state); - err = mlx5_core_modify_rq(mdev, rq->rqn, in, inlen); + err = mlx5_core_modify_rq(mdev, rq->rqn, in); kvfree(in); return err; } +static int mlx5e_rq_to_ready(struct mlx5e_rq *rq, int curr_state) +{ + struct net_device *dev = rq->netdev; + int err; + + err = mlx5e_modify_rq_state(rq, curr_state, MLX5_RQC_STATE_RST); + if (err) { + netdev_err(dev, "Failed to move rq 0x%x to reset\n", rq->rqn); + return err; + } + err = mlx5e_modify_rq_state(rq, MLX5_RQC_STATE_RST, MLX5_RQC_STATE_RDY); + if (err) { + netdev_err(dev, "Failed to move rq 0x%x to ready\n", rq->rqn); + return err; + } + + return 0; +} + +int mlx5e_flush_rq(struct mlx5e_rq *rq, int curr_state) +{ + mlx5e_free_rx_descs(rq); + + return mlx5e_rq_to_ready(rq, curr_state); +} + static int mlx5e_modify_rq_scatter_fcs(struct mlx5e_rq *rq, bool enable) { - struct mlx5e_channel *c = rq->channel; - struct mlx5e_priv *priv = c->priv; - struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_core_dev *mdev = rq->mdev; void *in; void *rqc; @@ -751,7 +1035,7 @@ static int mlx5e_modify_rq_scatter_fcs(struct mlx5e_rq *rq, bool enable) MLX5_SET(rqc, rqc, scatter_fcs, enable); MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RDY); - err = mlx5_core_modify_rq(mdev, rq->rqn, in, inlen); + err = mlx5_core_modify_rq(mdev, rq->rqn, in); kvfree(in); @@ -760,8 +1044,7 @@ static int mlx5e_modify_rq_scatter_fcs(struct mlx5e_rq *rq, bool enable) static int mlx5e_modify_rq_vsd(struct mlx5e_rq *rq, bool vsd) { - struct mlx5e_channel *c = rq->channel; - struct mlx5_core_dev *mdev = c->mdev; + struct mlx5_core_dev *mdev = rq->mdev; void *in; void *rqc; int inlen; @@ -780,14 +1063,14 @@ static int mlx5e_modify_rq_vsd(struct mlx5e_rq *rq, bool vsd) MLX5_SET(rqc, rqc, vsd, vsd); MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RDY); - err = mlx5_core_modify_rq(mdev, rq->rqn, in, inlen); + err = mlx5_core_modify_rq(mdev, rq->rqn, in); kvfree(in); return err; } -static void mlx5e_destroy_rq(struct mlx5e_rq *rq) +void mlx5e_destroy_rq(struct mlx5e_rq *rq) { mlx5_core_destroy_rq(rq->mdev, rq->rqn); } @@ -795,7 +1078,6 @@ static void mlx5e_destroy_rq(struct mlx5e_rq *rq) int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq, int wait_time) { unsigned long exp_time = jiffies + msecs_to_jiffies(wait_time); - struct mlx5e_channel *c = rq->channel; u16 min_wqes = mlx5_min_rx_wqes(rq->wq_type, mlx5e_rqwq_get_size(rq)); @@ -806,13 +1088,45 @@ int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq, int wait_time) msleep(20); } while (time_before(jiffies, exp_time)); - netdev_warn(c->netdev, "Failed to get min RX wqes on Channel[%d] RQN[0x%x] wq cur_sz(%d) min_rx_wqes(%d)\n", - c->ix, rq->rqn, mlx5e_rqwq_get_cur_sz(rq), min_wqes); + netdev_warn(rq->netdev, "Failed to get min RX wqes on Channel[%d] RQN[0x%x] wq cur_sz(%d) min_rx_wqes(%d)\n", + rq->ix, rq->rqn, mlx5e_rqwq_get_cur_sz(rq), min_wqes); mlx5e_reporter_rx_timeout(rq); return -ETIMEDOUT; } +void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq) +{ + struct mlx5_wq_ll *wq; + u16 head; + int i; + + if (rq->wq_type != MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) + return; + + wq = &rq->mpwqe.wq; + head = wq->head; + + /* Outstanding UMR WQEs (in progress) start at wq->head */ + for (i = 0; i < rq->mpwqe.umr_in_progress; i++) { + rq->dealloc_wqe(rq, head); + head = mlx5_wq_ll_get_wqe_next_ix(wq, head); + } + + if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) { + u16 len; + + len = (rq->mpwqe.shampo->pi - rq->mpwqe.shampo->ci) & + (rq->mpwqe.shampo->hd_per_wq - 1); + mlx5e_shampo_dealloc_hd(rq, len, rq->mpwqe.shampo->ci, false); + rq->mpwqe.shampo->pi = rq->mpwqe.shampo->ci; + } + + rq->mpwqe.actual_wq_head = wq->head; + rq->mpwqe.umr_in_progress = 0; + rq->mpwqe.umr_completed = 0; +} + void mlx5e_free_rx_descs(struct mlx5e_rq *rq) { __be16 wqe_ix_be; @@ -820,14 +1134,8 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq) if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { struct mlx5_wq_ll *wq = &rq->mpwqe.wq; - u16 head = wq->head; - int i; - /* Outstanding UMR WQEs (in progress) start at wq->head */ - for (i = 0; i < rq->mpwqe.umr_in_progress; i++) { - rq->dealloc_wqe(rq, head); - head = mlx5_wq_ll_get_wqe_next_ix(wq, head); - } + mlx5e_free_rx_in_progress_descs(rq); while (!mlx5_wq_ll_is_empty(wq)) { struct mlx5e_rx_wqe_ll *wqe; @@ -839,6 +1147,10 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq) mlx5_wq_ll_pop(wq, wqe_ix_be, &wqe->next.next_wqe_index); } + + if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) + mlx5e_shampo_dealloc_hd(rq, rq->mpwqe.shampo->hd_per_wq, + 0, true); } else { struct mlx5_wq_cyc *wq = &rq->wqe.wq; @@ -851,13 +1163,17 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq) } -int mlx5e_open_rq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct mlx5e_rq_param *param, struct mlx5e_xsk_param *xsk, - struct xdp_umem *umem, struct mlx5e_rq *rq) +int mlx5e_open_rq(struct mlx5e_params *params, struct mlx5e_rq_param *param, + struct mlx5e_xsk_param *xsk, int node, + struct mlx5e_rq *rq) { + struct mlx5_core_dev *mdev = rq->mdev; int err; - err = mlx5e_alloc_rq(c, params, xsk, umem, param, rq); + if (params->packet_merge.type == MLX5E_PACKET_MERGE_SHAMPO) + __set_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state); + + err = mlx5e_alloc_rq(params, xsk, param, node, rq); if (err) return err; @@ -869,18 +1185,25 @@ int mlx5e_open_rq(struct mlx5e_channel *c, struct mlx5e_params *params, if (err) goto err_destroy_rq; - if (MLX5_CAP_ETH(c->mdev, cqe_checksum_full)) - __set_bit(MLX5E_RQ_STATE_CSUM_FULL, &c->rq.state); + if (MLX5_CAP_ETH(mdev, cqe_checksum_full)) + __set_bit(MLX5E_RQ_STATE_CSUM_FULL, &rq->state); if (params->rx_dim_enabled) - __set_bit(MLX5E_RQ_STATE_AM, &c->rq.state); + __set_bit(MLX5E_RQ_STATE_AM, &rq->state); /* We disable csum_complete when XDP is enabled since * XDP programs might manipulate packets which will render * skb->checksum incorrect. */ - if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_NO_CSUM_COMPLETE) || c->xdp) - __set_bit(MLX5E_RQ_STATE_NO_CSUM_COMPLETE, &c->rq.state); + if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_NO_CSUM_COMPLETE) || params->xdp_prog) + __set_bit(MLX5E_RQ_STATE_NO_CSUM_COMPLETE, &rq->state); + + /* For CQE compression on striding RQ, use stride index provided by + * HW if capability is supported. + */ + if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ) && + MLX5_CAP_GEN(mdev, mini_cqe_resp_stride_index)) + __set_bit(MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, &rq->state); return 0; @@ -895,19 +1218,17 @@ err_free_rq: void mlx5e_activate_rq(struct mlx5e_rq *rq) { set_bit(MLX5E_RQ_STATE_ENABLED, &rq->state); - mlx5e_trigger_irq(&rq->channel->icosq); } void mlx5e_deactivate_rq(struct mlx5e_rq *rq) { clear_bit(MLX5E_RQ_STATE_ENABLED, &rq->state); - napi_synchronize(&rq->channel->napi); /* prevent mlx5e_post_rx_wqes */ + synchronize_net(); /* Sync with NAPI to prevent mlx5e_post_rx_wqes. */ } void mlx5e_close_rq(struct mlx5e_rq *rq) { cancel_work_sync(&rq->dim.work); - cancel_work_sync(&rq->channel->icosq.recover_work); cancel_work_sync(&rq->recover_work); mlx5e_destroy_rq(rq); mlx5e_free_rx_descs(rq); @@ -925,9 +1246,10 @@ static int mlx5e_alloc_xdpsq_fifo(struct mlx5e_xdpsq *sq, int numa) struct mlx5e_xdp_info_fifo *xdpi_fifo = &sq->db.xdpi_fifo; int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); int dsegs_per_wq = wq_sz * MLX5_SEND_WQEBB_NUM_DS; + size_t size; - xdpi_fifo->xi = kvzalloc_node(sizeof(*xdpi_fifo->xi) * dsegs_per_wq, - GFP_KERNEL, numa); + size = array_size(sizeof(*xdpi_fifo->xi), dsegs_per_wq); + xdpi_fifo->xi = kvzalloc_node(size, GFP_KERNEL, numa); if (!xdpi_fifo->xi) return -ENOMEM; @@ -941,10 +1263,11 @@ static int mlx5e_alloc_xdpsq_fifo(struct mlx5e_xdpsq *sq, int numa) static int mlx5e_alloc_xdpsq_db(struct mlx5e_xdpsq *sq, int numa) { int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); + size_t size; int err; - sq->db.wqe_info = kvzalloc_node(sizeof(*sq->db.wqe_info) * wq_sz, - GFP_KERNEL, numa); + size = array_size(sizeof(*sq->db.wqe_info), wq_sz); + sq->db.wqe_info = kvzalloc_node(size, GFP_KERNEL, numa); if (!sq->db.wqe_info) return -ENOMEM; @@ -959,7 +1282,7 @@ static int mlx5e_alloc_xdpsq_db(struct mlx5e_xdpsq *sq, int numa) static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct xdp_umem *umem, + struct xsk_buff_pool *xsk_pool, struct mlx5e_sq_param *param, struct mlx5e_xdpsq *sq, bool is_redirect) @@ -972,16 +1295,19 @@ static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, sq->pdev = c->pdev; sq->mkey_be = c->mkey_be; sq->channel = c; - sq->uar_map = mdev->mlx5e_res.bfreg.map; + sq->uar_map = mdev->mlx5e_res.hw_objs.bfreg.map; sq->min_inline_mode = params->tx_min_inline_mode; sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); - sq->umem = umem; + sq->xsk_pool = xsk_pool; - sq->stats = sq->umem ? - &c->priv->channel_stats[c->ix].xsksq : + sq->stats = sq->xsk_pool ? + &c->priv->channel_stats[c->ix]->xsksq : is_redirect ? - &c->priv->channel_stats[c->ix].xdpsq : - &c->priv->channel_stats[c->ix].rq_xdpsq; + &c->priv->channel_stats[c->ix]->xdpsq : + &c->priv->channel_stats[c->ix]->rq_xdpsq; + sq->stop_room = param->is_mpw ? mlx5e_stop_room_for_mpwqe(mdev) : + mlx5e_stop_room_for_max_wqe(mdev); + sq->max_sq_mpw_wqebbs = mlx5e_get_max_sq_aligned_wqebbs(mdev); param->wq.db_numa_node = cpu_to_node(c->cpu); err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, wq, &sq->wq_ctrl); @@ -1009,17 +1335,17 @@ static void mlx5e_free_xdpsq(struct mlx5e_xdpsq *sq) static void mlx5e_free_icosq_db(struct mlx5e_icosq *sq) { - kvfree(sq->db.ico_wqe); + kvfree(sq->db.wqe_info); } static int mlx5e_alloc_icosq_db(struct mlx5e_icosq *sq, int numa) { int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); + size_t size; - sq->db.ico_wqe = kvzalloc_node(array_size(wq_sz, - sizeof(*sq->db.ico_wqe)), - GFP_KERNEL, numa); - if (!sq->db.ico_wqe) + size = array_size(wq_sz, sizeof(*sq->db.wqe_info)); + sq->db.wqe_info = kvzalloc_node(size, GFP_KERNEL, numa); + if (!sq->db.wqe_info) return -ENOMEM; return 0; @@ -1033,9 +1359,20 @@ static void mlx5e_icosq_err_cqe_work(struct work_struct *recover_work) mlx5e_reporter_icosq_cqe_err(sq); } +static void mlx5e_async_icosq_err_cqe_work(struct work_struct *recover_work) +{ + struct mlx5e_icosq *sq = container_of(recover_work, struct mlx5e_icosq, + recover_work); + + /* Not implemented yet. */ + + netdev_warn(sq->channel->netdev, "async_icosq recovery is not implemented\n"); +} + static int mlx5e_alloc_icosq(struct mlx5e_channel *c, struct mlx5e_sq_param *param, - struct mlx5e_icosq *sq) + struct mlx5e_icosq *sq, + work_func_t recover_work_func) { void *sqc_wq = MLX5_ADDR_OF(sqc, param->sqc, wq); struct mlx5_core_dev *mdev = c->mdev; @@ -1043,7 +1380,8 @@ static int mlx5e_alloc_icosq(struct mlx5e_channel *c, int err; sq->channel = c; - sq->uar_map = mdev->mlx5e_res.bfreg.map; + sq->uar_map = mdev->mlx5e_res.hw_objs.bfreg.map; + sq->reserved_room = param->stop_room; param->wq.db_numa_node = cpu_to_node(c->cpu); err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, wq, &sq->wq_ctrl); @@ -1055,7 +1393,7 @@ static int mlx5e_alloc_icosq(struct mlx5e_channel *c, if (err) goto err_sq_wq_destroy; - INIT_WORK(&sq->recover_work, mlx5e_icosq_err_cqe_work); + INIT_WORK(&sq->recover_work, recover_work_func); return 0; @@ -1071,13 +1409,14 @@ static void mlx5e_free_icosq(struct mlx5e_icosq *sq) mlx5_wq_destroy(&sq->wq_ctrl); } -static void mlx5e_free_txqsq_db(struct mlx5e_txqsq *sq) +void mlx5e_free_txqsq_db(struct mlx5e_txqsq *sq) { kvfree(sq->db.wqe_info); + kvfree(sq->db.skb_fifo.fifo); kvfree(sq->db.dma_fifo); } -static int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) +int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) { int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); int df_sz = wq_sz * MLX5_SEND_WQEBB_NUM_DS; @@ -1085,20 +1424,26 @@ static int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) sq->db.dma_fifo = kvzalloc_node(array_size(df_sz, sizeof(*sq->db.dma_fifo)), GFP_KERNEL, numa); + sq->db.skb_fifo.fifo = kvzalloc_node(array_size(df_sz, + sizeof(*sq->db.skb_fifo.fifo)), + GFP_KERNEL, numa); sq->db.wqe_info = kvzalloc_node(array_size(wq_sz, sizeof(*sq->db.wqe_info)), GFP_KERNEL, numa); - if (!sq->db.dma_fifo || !sq->db.wqe_info) { + if (!sq->db.dma_fifo || !sq->db.skb_fifo.fifo || !sq->db.wqe_info) { mlx5e_free_txqsq_db(sq); return -ENOMEM; } sq->dma_fifo_mask = df_sz - 1; + sq->db.skb_fifo.pc = &sq->skb_fifo_pc; + sq->db.skb_fifo.cc = &sq->skb_fifo_cc; + sq->db.skb_fifo.mask = df_sz - 1; + return 0; } -static void mlx5e_tx_err_cqe_work(struct work_struct *recover_work); static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, int txq_ix, struct mlx5e_params *params, @@ -1112,30 +1457,26 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, int err; sq->pdev = c->pdev; - sq->tstamp = c->tstamp; sq->clock = &mdev->clock; sq->mkey_be = c->mkey_be; - sq->channel = c; + sq->netdev = c->netdev; + sq->mdev = c->mdev; + sq->priv = c->priv; sq->ch_ix = c->ix; sq->txq_ix = txq_ix; - sq->uar_map = mdev->mlx5e_res.bfreg.map; + sq->uar_map = mdev->mlx5e_res.hw_objs.bfreg.map; sq->min_inline_mode = params->tx_min_inline_mode; sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); - sq->stats = &c->priv->channel_stats[c->ix].sq[tc]; - sq->stop_room = MLX5E_SQ_STOP_ROOM; + sq->max_sq_mpw_wqebbs = mlx5e_get_max_sq_aligned_wqebbs(mdev); INIT_WORK(&sq->recover_work, mlx5e_tx_err_cqe_work); if (!MLX5_CAP_ETH(mdev, wqe_vlan_insert)) set_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state); - if (MLX5_IPSEC_DEV(c->priv->mdev)) + if (mlx5_ipsec_device_caps(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state); -#ifdef CONFIG_MLX5_EN_TLS - if (mlx5_accel_is_tls_device(c->priv->mdev)) { - set_bit(MLX5E_SQ_STATE_TLS, &sq->state); - sq->stop_room += MLX5E_SQ_TLS_ROOM + - mlx5e_ktls_dumps_num_wqebbs(sq, MAX_SKB_FRAGS, - TLS_MAX_PAYLOAD_SIZE); - } -#endif + if (param->is_mpw) + set_bit(MLX5E_SQ_STATE_MPWQE, &sq->state); + sq->stop_room = param->stop_room; + sq->ptp_cyc2time = mlx5_sq_ts_translator(mdev); param->wq.db_numa_node = cpu_to_node(c->cpu); err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, wq, &sq->wq_ctrl); @@ -1158,25 +1499,18 @@ err_sq_wq_destroy: return err; } -static void mlx5e_free_txqsq(struct mlx5e_txqsq *sq) +void mlx5e_free_txqsq(struct mlx5e_txqsq *sq) { mlx5e_free_txqsq_db(sq); mlx5_wq_destroy(&sq->wq_ctrl); } -struct mlx5e_create_sq_param { - struct mlx5_wq_ctrl *wq_ctrl; - u32 cqn; - u32 tisn; - u8 tis_lst_sz; - u8 min_inline_mode; -}; - static int mlx5e_create_sq(struct mlx5_core_dev *mdev, struct mlx5e_sq_param *param, struct mlx5e_create_sq_param *csp, u32 *sqn) { + u8 ts_format; void *in; void *sqc; void *wq; @@ -1189,6 +1523,9 @@ static int mlx5e_create_sq(struct mlx5_core_dev *mdev, if (!in) return -ENOMEM; + ts_format = mlx5_is_real_time_sq(mdev) ? + MLX5_TIMESTAMP_FORMAT_REAL_TIME : + MLX5_TIMESTAMP_FORMAT_FREE_RUNNING; sqc = MLX5_ADDR_OF(create_sq_in, in, ctx); wq = MLX5_ADDR_OF(sqc, sqc, wq); @@ -1196,6 +1533,9 @@ static int mlx5e_create_sq(struct mlx5_core_dev *mdev, MLX5_SET(sqc, sqc, tis_lst_sz, csp->tis_lst_sz); MLX5_SET(sqc, sqc, tis_num_0, csp->tisn); MLX5_SET(sqc, sqc, cqn, csp->cqn); + MLX5_SET(sqc, sqc, ts_cqe_to_dest_cqn, csp->ts_cqe_to_dest_cqn); + MLX5_SET(sqc, sqc, ts_format, ts_format); + if (MLX5_CAP_ETH(mdev, wqe_inline_mode) == MLX5_CAP_INLINE_MODE_VPORT_CONTEXT) MLX5_SET(sqc, sqc, min_wqe_inline_mode, csp->min_inline_mode); @@ -1204,7 +1544,7 @@ static int mlx5e_create_sq(struct mlx5_core_dev *mdev, MLX5_SET(sqc, sqc, flush_in_error_en, 1); MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_CYCLIC); - MLX5_SET(wq, wq, uar_page, mdev->mlx5e_res.bfreg.index); + MLX5_SET(wq, wq, uar_page, mdev->mlx5e_res.hw_objs.bfreg.index); MLX5_SET(wq, wq, log_wq_pg_sz, csp->wq_ctrl->buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT); MLX5_SET64(wq, wq, dbr_addr, csp->wq_ctrl->db.dma); @@ -1222,6 +1562,7 @@ static int mlx5e_create_sq(struct mlx5_core_dev *mdev, int mlx5e_modify_sq(struct mlx5_core_dev *mdev, u32 sqn, struct mlx5e_modify_sq_param *p) { + u64 bitmask = 0; void *in; void *sqc; int inlen; @@ -1237,11 +1578,16 @@ int mlx5e_modify_sq(struct mlx5_core_dev *mdev, u32 sqn, MLX5_SET(modify_sq_in, in, sq_state, p->curr_state); MLX5_SET(sqc, sqc, state, p->next_state); if (p->rl_update && p->next_state == MLX5_SQC_STATE_RDY) { - MLX5_SET64(modify_sq_in, in, modify_bitmask, 1); - MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, p->rl_index); + bitmask |= 1; + MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, p->rl_index); } + if (p->qos_update && p->next_state == MLX5_SQC_STATE_RDY) { + bitmask |= 1 << 2; + MLX5_SET(sqc, sqc, qos_queue_group_id, p->qos_queue_group_id); + } + MLX5_SET64(modify_sq_in, in, modify_bitmask, bitmask); - err = mlx5_core_modify_sq(mdev, sqn, in, inlen); + err = mlx5_core_modify_sq(mdev, sqn, in); kvfree(in); @@ -1253,10 +1599,11 @@ static void mlx5e_destroy_sq(struct mlx5_core_dev *mdev, u32 sqn) mlx5_core_destroy_sq(mdev, sqn); } -static int mlx5e_create_sq_rdy(struct mlx5_core_dev *mdev, - struct mlx5e_sq_param *param, - struct mlx5e_create_sq_param *csp, - u32 *sqn) +int mlx5e_create_sq_rdy(struct mlx5_core_dev *mdev, + struct mlx5e_sq_param *param, + struct mlx5e_create_sq_param *csp, + u16 qos_queue_group_id, + u32 *sqn) { struct mlx5e_modify_sq_param msp = {0}; int err; @@ -1267,6 +1614,10 @@ static int mlx5e_create_sq_rdy(struct mlx5_core_dev *mdev, msp.curr_state = MLX5_SQC_STATE_RST; msp.next_state = MLX5_SQC_STATE_RDY; + if (qos_queue_group_id) { + msp.qos_update = true; + msp.qos_queue_group_id = qos_queue_group_id; + } err = mlx5e_modify_sq(mdev, *sqn, &msp); if (err) mlx5e_destroy_sq(mdev, *sqn); @@ -1277,13 +1628,10 @@ static int mlx5e_create_sq_rdy(struct mlx5_core_dev *mdev, static int mlx5e_set_sq_maxrate(struct net_device *dev, struct mlx5e_txqsq *sq, u32 rate); -static int mlx5e_open_txqsq(struct mlx5e_channel *c, - u32 tisn, - int txq_ix, - struct mlx5e_params *params, - struct mlx5e_sq_param *param, - struct mlx5e_txqsq *sq, - int tc) +int mlx5e_open_txqsq(struct mlx5e_channel *c, u32 tisn, int txq_ix, + struct mlx5e_params *params, struct mlx5e_sq_param *param, + struct mlx5e_txqsq *sq, int tc, u16 qos_queue_group_id, + struct mlx5e_sq_stats *sq_stats) { struct mlx5e_create_sq_param csp = {}; u32 tx_rate; @@ -1293,12 +1641,14 @@ static int mlx5e_open_txqsq(struct mlx5e_channel *c, if (err) return err; + sq->stats = sq_stats; + csp.tisn = tisn; csp.tis_lst_sz = 1; csp.cqn = sq->cq.mcq.cqn; csp.wq_ctrl = &sq->wq_ctrl; csp.min_inline_mode = sq->min_inline_mode; - err = mlx5e_create_sq_rdy(c->mdev, param, &csp, &sq->sqn); + err = mlx5e_create_sq_rdy(c->mdev, param, &csp, qos_queue_group_id, &sq->sqn); if (err) goto err_free_txqsq; @@ -1319,7 +1669,7 @@ err_free_txqsq: void mlx5e_activate_txqsq(struct mlx5e_txqsq *sq) { - sq->txq = netdev_get_tx_queue(sq->channel->netdev, sq->txq_ix); + sq->txq = netdev_get_tx_queue(sq->netdev, sq->txq_ix); set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); netdev_tx_reset_queue(sq->txq); netif_tx_start_queue(sq->txq); @@ -1332,36 +1682,32 @@ void mlx5e_tx_disable_queue(struct netdev_queue *txq) __netif_tx_unlock_bh(txq); } -static void mlx5e_deactivate_txqsq(struct mlx5e_txqsq *sq) +void mlx5e_deactivate_txqsq(struct mlx5e_txqsq *sq) { - struct mlx5e_channel *c = sq->channel; struct mlx5_wq_cyc *wq = &sq->wq; clear_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); - /* prevent netif_tx_wake_queue */ - napi_synchronize(&c->napi); + synchronize_net(); /* Sync with NAPI to prevent netif_tx_wake_queue. */ mlx5e_tx_disable_queue(sq->txq); /* last doorbell out, godspeed .. */ if (mlx5e_wqc_has_room_for(wq, sq->cc, sq->pc, 1)) { u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - struct mlx5e_tx_wqe_info *wi; struct mlx5e_tx_wqe *nop; - wi = &sq->db.wqe_info[pi]; + sq->db.wqe_info[pi] = (struct mlx5e_tx_wqe_info) { + .num_wqebbs = 1, + }; - memset(wi, 0, sizeof(*wi)); - wi->num_wqebbs = 1; nop = mlx5e_post_nop(wq, sq->sqn, &sq->pc); mlx5e_notify_hw(wq, sq->pc, sq->uar_map, &nop->ctrl); } } -static void mlx5e_close_txqsq(struct mlx5e_txqsq *sq) +void mlx5e_close_txqsq(struct mlx5e_txqsq *sq) { - struct mlx5e_channel *c = sq->channel; - struct mlx5_core_dev *mdev = c->mdev; + struct mlx5_core_dev *mdev = sq->mdev; struct mlx5_rate_limit rl = {0}; cancel_work_sync(&sq->dim.work); @@ -1375,7 +1721,7 @@ static void mlx5e_close_txqsq(struct mlx5e_txqsq *sq) mlx5e_free_txqsq(sq); } -static void mlx5e_tx_err_cqe_work(struct work_struct *recover_work) +void mlx5e_tx_err_cqe_work(struct work_struct *recover_work) { struct mlx5e_txqsq *sq = container_of(recover_work, struct mlx5e_txqsq, recover_work); @@ -1383,25 +1729,35 @@ static void mlx5e_tx_err_cqe_work(struct work_struct *recover_work) mlx5e_reporter_tx_err_cqe(sq); } -int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct mlx5e_sq_param *param, struct mlx5e_icosq *sq) +static int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params, + struct mlx5e_sq_param *param, struct mlx5e_icosq *sq, + work_func_t recover_work_func) { struct mlx5e_create_sq_param csp = {}; int err; - err = mlx5e_alloc_icosq(c, param, sq); + err = mlx5e_alloc_icosq(c, param, sq, recover_work_func); if (err) return err; csp.cqn = sq->cq.mcq.cqn; csp.wq_ctrl = &sq->wq_ctrl; csp.min_inline_mode = params->tx_min_inline_mode; - err = mlx5e_create_sq_rdy(c->mdev, param, &csp, &sq->sqn); + err = mlx5e_create_sq_rdy(c->mdev, param, &csp, 0, &sq->sqn); if (err) goto err_free_icosq; + if (param->is_tls) { + sq->ktls_resync = mlx5e_ktls_rx_resync_create_resp_list(); + if (IS_ERR(sq->ktls_resync)) { + err = PTR_ERR(sq->ktls_resync); + goto err_destroy_icosq; + } + } return 0; +err_destroy_icosq: + mlx5e_destroy_sq(c->mdev, sq->sqn); err_free_icosq: mlx5e_free_icosq(sq); @@ -1415,28 +1771,29 @@ void mlx5e_activate_icosq(struct mlx5e_icosq *icosq) void mlx5e_deactivate_icosq(struct mlx5e_icosq *icosq) { - struct mlx5e_channel *c = icosq->channel; - clear_bit(MLX5E_SQ_STATE_ENABLED, &icosq->state); - napi_synchronize(&c->napi); + synchronize_net(); /* Sync with NAPI. */ } -void mlx5e_close_icosq(struct mlx5e_icosq *sq) +static void mlx5e_close_icosq(struct mlx5e_icosq *sq) { struct mlx5e_channel *c = sq->channel; + if (sq->ktls_resync) + mlx5e_ktls_rx_resync_destroy_resp_list(sq->ktls_resync); mlx5e_destroy_sq(c->mdev, sq->sqn); + mlx5e_free_icosq_descs(sq); mlx5e_free_icosq(sq); } int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct mlx5e_sq_param *param, struct xdp_umem *umem, + struct mlx5e_sq_param *param, struct xsk_buff_pool *xsk_pool, struct mlx5e_xdpsq *sq, bool is_redirect) { struct mlx5e_create_sq_param csp = {}; int err; - err = mlx5e_alloc_xdpsq(c, params, umem, param, sq, is_redirect); + err = mlx5e_alloc_xdpsq(c, params, xsk_pool, param, sq, is_redirect); if (err) return err; @@ -1446,14 +1803,22 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, csp.wq_ctrl = &sq->wq_ctrl; csp.min_inline_mode = sq->min_inline_mode; set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); - err = mlx5e_create_sq_rdy(c->mdev, param, &csp, &sq->sqn); + + /* Don't enable multi buffer on XDP_REDIRECT SQ, as it's not yet + * supported by upstream, and there is no defined trigger to allow + * transmitting redirected multi-buffer frames. + */ + if (param->is_xdp_mb && !is_redirect) + set_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state); + + err = mlx5e_create_sq_rdy(c->mdev, param, &csp, 0, &sq->sqn); if (err) goto err_free_xdpsq; mlx5e_set_xmit_fp(sq, param->is_mpw); - if (!param->is_mpw) { - unsigned int ds_cnt = MLX5E_XDP_TX_DS_COUNT; + if (!param->is_mpw && !test_bit(MLX5E_SQ_STATE_XDP_MULTIBUF, &sq->state)) { + unsigned int ds_cnt = MLX5E_TX_WQE_EMPTY_DS_COUNT + 1; unsigned int inline_hdr_sz = 0; int i; @@ -1464,20 +1829,21 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, /* Pre initialize fixed WQE fields */ for (i = 0; i < mlx5_wq_cyc_get_size(&sq->wq); i++) { - struct mlx5e_xdp_wqe_info *wi = &sq->db.wqe_info[i]; struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(&sq->wq, i); struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; struct mlx5_wqe_eth_seg *eseg = &wqe->eth; struct mlx5_wqe_data_seg *dseg; + sq->db.wqe_info[i] = (struct mlx5e_xdp_wqe_info) { + .num_wqebbs = 1, + .num_pkts = 1, + }; + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt); eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz); dseg = (struct mlx5_wqe_data_seg *)cseg + (ds_cnt - 1); dseg->lkey = sq->mkey_be; - - wi->num_wqebbs = 1; - wi->num_pkts = 1; } } @@ -1495,27 +1861,22 @@ void mlx5e_close_xdpsq(struct mlx5e_xdpsq *sq) struct mlx5e_channel *c = sq->channel; clear_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); - napi_synchronize(&c->napi); + synchronize_net(); /* Sync with NAPI. */ mlx5e_destroy_sq(c->mdev, sq->sqn); mlx5e_free_xdpsq_descs(sq); mlx5e_free_xdpsq(sq); } -static int mlx5e_alloc_cq_common(struct mlx5_core_dev *mdev, +static int mlx5e_alloc_cq_common(struct mlx5e_priv *priv, struct mlx5e_cq_param *param, struct mlx5e_cq *cq) { + struct mlx5_core_dev *mdev = priv->mdev; struct mlx5_core_cq *mcq = &cq->mcq; - int eqn_not_used; - unsigned int irqn; int err; u32 i; - err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn_not_used, &irqn); - if (err) - return err; - err = mlx5_cqwq_create(mdev, ¶m->wq, param->cqc, &cq->wq, &cq->wq_ctrl); if (err) @@ -1529,7 +1890,6 @@ static int mlx5e_alloc_cq_common(struct mlx5_core_dev *mdev, mcq->vector = param->eq_ix; mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; - mcq->irqn = irqn; for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) { struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); @@ -1538,25 +1898,27 @@ static int mlx5e_alloc_cq_common(struct mlx5_core_dev *mdev, } cq->mdev = mdev; + cq->netdev = priv->netdev; + cq->priv = priv; return 0; } -static int mlx5e_alloc_cq(struct mlx5e_channel *c, +static int mlx5e_alloc_cq(struct mlx5e_priv *priv, struct mlx5e_cq_param *param, + struct mlx5e_create_cq_param *ccp, struct mlx5e_cq *cq) { - struct mlx5_core_dev *mdev = c->priv->mdev; int err; - param->wq.buf_numa_node = cpu_to_node(c->cpu); - param->wq.db_numa_node = cpu_to_node(c->cpu); - param->eq_ix = c->ix; + param->wq.buf_numa_node = ccp->node; + param->wq.db_numa_node = ccp->node; + param->eq_ix = ccp->ix; - err = mlx5e_alloc_cq_common(mdev, param, cq); + err = mlx5e_alloc_cq_common(priv, param, cq); - cq->napi = &c->napi; - cq->channel = c; + cq->napi = ccp->napi; + cq->ch_stats = ccp->ch_stats; return err; } @@ -1575,11 +1937,10 @@ static int mlx5e_create_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) void *in; void *cqc; int inlen; - unsigned int irqn_not_used; int eqn; int err; - err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); + err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn); if (err) return err; @@ -1597,7 +1958,7 @@ static int mlx5e_create_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) (__be64 *)MLX5_ADDR_OF(create_cq_in, in, pas)); MLX5_SET(cqc, cqc, cq_period_mode, param->cq_period_mode); - MLX5_SET(cqc, cqc, c_eqn, eqn); + MLX5_SET(cqc, cqc, c_eqn_or_apu_element, eqn); MLX5_SET(cqc, cqc, uar_page, mdev->priv.uar->index); MLX5_SET(cqc, cqc, log_page_size, cq->wq_ctrl.buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT); @@ -1620,13 +1981,14 @@ static void mlx5e_destroy_cq(struct mlx5e_cq *cq) mlx5_core_destroy_cq(cq->mdev, &cq->mcq); } -int mlx5e_open_cq(struct mlx5e_channel *c, struct dim_cq_moder moder, - struct mlx5e_cq_param *param, struct mlx5e_cq *cq) +int mlx5e_open_cq(struct mlx5e_priv *priv, struct dim_cq_moder moder, + struct mlx5e_cq_param *param, struct mlx5e_create_cq_param *ccp, + struct mlx5e_cq *cq) { - struct mlx5_core_dev *mdev = c->mdev; + struct mlx5_core_dev *mdev = priv->mdev; int err; - err = mlx5e_alloc_cq(c, param, cq); + err = mlx5e_alloc_cq(priv, param, ccp, cq); if (err) return err; @@ -1652,14 +2014,15 @@ void mlx5e_close_cq(struct mlx5e_cq *cq) static int mlx5e_open_tx_cqs(struct mlx5e_channel *c, struct mlx5e_params *params, + struct mlx5e_create_cq_param *ccp, struct mlx5e_channel_param *cparam) { int err; int tc; for (tc = 0; tc < c->num_tc; tc++) { - err = mlx5e_open_cq(c, params->tx_cq_moderation, - &cparam->tx_cq, &c->sq[tc].cq); + err = mlx5e_open_cq(c->priv, params->tx_cq_moderation, &cparam->txq_sq.cqp, + ccp, &c->sq[tc].cq); if (err) goto err_close_tx_cqs; } @@ -1681,17 +2044,60 @@ static void mlx5e_close_tx_cqs(struct mlx5e_channel *c) mlx5e_close_cq(&c->sq[tc].cq); } +static int mlx5e_mqprio_txq_to_tc(struct netdev_tc_txq *tc_to_txq, unsigned int txq) +{ + int tc; + + for (tc = 0; tc < TC_MAX_QUEUE; tc++) + if (txq - tc_to_txq[tc].offset < tc_to_txq[tc].count) + return tc; + + WARN(1, "Unexpected TCs configuration. No match found for txq %u", txq); + return -ENOENT; +} + +static int mlx5e_txq_get_qos_node_hw_id(struct mlx5e_params *params, int txq_ix, + u32 *hw_id) +{ + int tc; + + if (params->mqprio.mode != TC_MQPRIO_MODE_CHANNEL) { + *hw_id = 0; + return 0; + } + + tc = mlx5e_mqprio_txq_to_tc(params->mqprio.tc_to_txq, txq_ix); + if (tc < 0) + return tc; + + if (tc >= params->mqprio.num_tc) { + WARN(1, "Unexpected TCs configuration. tc %d is out of range of %u", + tc, params->mqprio.num_tc); + return -EINVAL; + } + + *hw_id = params->mqprio.channel.hw_id[tc]; + return 0; +} + static int mlx5e_open_sqs(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_channel_param *cparam) { int err, tc; - for (tc = 0; tc < params->num_tc; tc++) { + for (tc = 0; tc < mlx5e_get_dcb_num_tc(params); tc++) { int txq_ix = c->ix + tc * params->num_channels; + u32 qos_queue_group_id; + + err = mlx5e_txq_get_qos_node_hw_id(params, txq_ix, &qos_queue_group_id); + if (err) + goto err_close_sqs; err = mlx5e_open_txqsq(c, c->priv->tisn[c->lag_port][tc], txq_ix, - params, &cparam->sq, &c->sq[tc], tc); + params, &cparam->txq_sq, &c->sq[tc], tc, + qos_queue_group_id, + &c->priv->channel_stats[c->ix]->sq[tc]); if (err) goto err_close_sqs; } @@ -1794,27 +2200,16 @@ static int mlx5e_set_tx_maxrate(struct net_device *dev, int index, u32 rate) return err; } -static int mlx5e_alloc_xps_cpumask(struct mlx5e_channel *c, - struct mlx5e_params *params) +static int mlx5e_open_rxq_rq(struct mlx5e_channel *c, struct mlx5e_params *params, + struct mlx5e_rq_param *rq_params) { - int num_comp_vectors = mlx5_comp_vectors_count(c->mdev); - int irq; - - if (!zalloc_cpumask_var(&c->xps_cpumask, GFP_KERNEL)) - return -ENOMEM; - - for (irq = c->ix; irq < num_comp_vectors; irq += params->num_channels) { - int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(c->mdev, irq)); - - cpumask_set_cpu(cpu, c->xps_cpumask); - } + int err; - return 0; -} + err = mlx5e_init_rxq_rq(c, params, &c->rq); + if (err) + return err; -static void mlx5e_free_xps_cpumask(struct mlx5e_channel *c) -{ - free_cpumask_var(c->xps_cpumask); + return mlx5e_open_rq(params, rq_params, NULL, cpu_to_node(c->cpu), &c->rq); } static int mlx5e_open_queues(struct mlx5e_channel *c, @@ -1822,73 +2217,92 @@ static int mlx5e_open_queues(struct mlx5e_channel *c, struct mlx5e_channel_param *cparam) { struct dim_cq_moder icocq_moder = {0, 0}; + struct mlx5e_create_cq_param ccp; int err; - err = mlx5e_open_cq(c, icocq_moder, &cparam->icosq_cq, &c->icosq.cq); + mlx5e_build_create_cq_param(&ccp, c); + + err = mlx5e_open_cq(c->priv, icocq_moder, &cparam->async_icosq.cqp, &ccp, + &c->async_icosq.cq); if (err) return err; - err = mlx5e_open_tx_cqs(c, params, cparam); + err = mlx5e_open_cq(c->priv, icocq_moder, &cparam->icosq.cqp, &ccp, + &c->icosq.cq); + if (err) + goto err_close_async_icosq_cq; + + err = mlx5e_open_tx_cqs(c, params, &ccp, cparam); if (err) goto err_close_icosq_cq; - err = mlx5e_open_cq(c, params->tx_cq_moderation, &cparam->tx_cq, &c->xdpsq.cq); + err = mlx5e_open_cq(c->priv, params->tx_cq_moderation, &cparam->xdp_sq.cqp, &ccp, + &c->xdpsq.cq); if (err) goto err_close_tx_cqs; - err = mlx5e_open_cq(c, params->rx_cq_moderation, &cparam->rx_cq, &c->rq.cq); + err = mlx5e_open_cq(c->priv, params->rx_cq_moderation, &cparam->rq.cqp, &ccp, + &c->rq.cq); if (err) goto err_close_xdp_tx_cqs; - /* XDP SQ CQ params are same as normal TXQ sq CQ params */ - err = c->xdp ? mlx5e_open_cq(c, params->tx_cq_moderation, - &cparam->tx_cq, &c->rq_xdpsq.cq) : 0; + err = c->xdp ? mlx5e_open_cq(c->priv, params->tx_cq_moderation, &cparam->xdp_sq.cqp, + &ccp, &c->rq_xdpsq.cq) : 0; if (err) goto err_close_rx_cq; - napi_enable(&c->napi); + spin_lock_init(&c->async_icosq_lock); + + err = mlx5e_open_icosq(c, params, &cparam->async_icosq, &c->async_icosq, + mlx5e_async_icosq_err_cqe_work); + if (err) + goto err_close_xdpsq_cq; + + mutex_init(&c->icosq_recovery_lock); - err = mlx5e_open_icosq(c, params, &cparam->icosq, &c->icosq); + err = mlx5e_open_icosq(c, params, &cparam->icosq, &c->icosq, + mlx5e_icosq_err_cqe_work); if (err) - goto err_disable_napi; + goto err_close_async_icosq; err = mlx5e_open_sqs(c, params, cparam); if (err) goto err_close_icosq; + err = mlx5e_open_rxq_rq(c, params, &cparam->rq); + if (err) + goto err_close_sqs; + if (c->xdp) { err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, NULL, &c->rq_xdpsq, false); if (err) - goto err_close_sqs; + goto err_close_rq; } - err = mlx5e_open_rq(c, params, &cparam->rq, NULL, NULL, &c->rq); - if (err) - goto err_close_xdp_sq; - err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, NULL, &c->xdpsq, true); if (err) - goto err_close_rq; + goto err_close_xdp_sq; return 0; -err_close_rq: - mlx5e_close_rq(&c->rq); - err_close_xdp_sq: if (c->xdp) mlx5e_close_xdpsq(&c->rq_xdpsq); +err_close_rq: + mlx5e_close_rq(&c->rq); + err_close_sqs: mlx5e_close_sqs(c); err_close_icosq: mlx5e_close_icosq(&c->icosq); -err_disable_napi: - napi_disable(&c->napi); +err_close_async_icosq: + mlx5e_close_icosq(&c->async_icosq); +err_close_xdpsq_cq: if (c->xdp) mlx5e_close_cq(&c->rq_xdpsq.cq); @@ -1904,24 +2318,31 @@ err_close_tx_cqs: err_close_icosq_cq: mlx5e_close_cq(&c->icosq.cq); +err_close_async_icosq_cq: + mlx5e_close_cq(&c->async_icosq.cq); + return err; } static void mlx5e_close_queues(struct mlx5e_channel *c) { mlx5e_close_xdpsq(&c->xdpsq); - mlx5e_close_rq(&c->rq); if (c->xdp) mlx5e_close_xdpsq(&c->rq_xdpsq); + /* The same ICOSQ is used for UMRs for both RQ and XSKRQ. */ + cancel_work_sync(&c->icosq.recover_work); + mlx5e_close_rq(&c->rq); mlx5e_close_sqs(c); mlx5e_close_icosq(&c->icosq); - napi_disable(&c->napi); + mutex_destroy(&c->icosq_recovery_lock); + mlx5e_close_icosq(&c->async_icosq); if (c->xdp) mlx5e_close_cq(&c->rq_xdpsq.cq); mlx5e_close_cq(&c->rq.cq); mlx5e_close_cq(&c->xdpsq.cq); mlx5e_close_tx_cqs(c); mlx5e_close_cq(&c->icosq.cq); + mlx5e_close_cq(&c->async_icosq.cq); } static u8 mlx5e_enumerate_lag_port(struct mlx5_core_dev *mdev, int ix) @@ -1931,10 +2352,48 @@ static u8 mlx5e_enumerate_lag_port(struct mlx5_core_dev *mdev, int ix) return (ix + port_aff_bias) % mlx5e_get_num_lag_ports(mdev); } +static int mlx5e_channel_stats_alloc(struct mlx5e_priv *priv, int ix, int cpu) +{ + if (ix > priv->stats_nch) { + netdev_warn(priv->netdev, "Unexpected channel stats index %d > %d\n", ix, + priv->stats_nch); + return -EINVAL; + } + + if (priv->channel_stats[ix]) + return 0; + + /* Asymmetric dynamic memory allocation. + * Freed in mlx5e_priv_arrays_free, not on channel closure. + */ + mlx5e_dbg(DRV, priv, "Creating channel stats %d\n", ix); + priv->channel_stats[ix] = kvzalloc_node(sizeof(**priv->channel_stats), + GFP_KERNEL, cpu_to_node(cpu)); + if (!priv->channel_stats[ix]) + return -ENOMEM; + priv->stats_nch++; + + return 0; +} + +void mlx5e_trigger_napi_icosq(struct mlx5e_channel *c) +{ + spin_lock_bh(&c->async_icosq_lock); + mlx5e_trigger_irq(&c->async_icosq); + spin_unlock_bh(&c->async_icosq_lock); +} + +void mlx5e_trigger_napi_sched(struct napi_struct *napi) +{ + local_bh_disable(); + napi_schedule(napi); + local_bh_enable(); +} + static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_params *params, struct mlx5e_channel_param *cparam, - struct xdp_umem *umem, + struct xsk_buff_pool *xsk_pool, struct mlx5e_channel **cp) { int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(priv->mdev, ix)); @@ -1943,9 +2402,12 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_channel *c; unsigned int irq; int err; - int eqn; - err = mlx5_vector2eqn(priv->mdev, ix, &eqn, &irq); + err = mlx5_vector2irqn(priv->mdev, ix, &irq); + if (err) + return err; + + err = mlx5e_channel_stats_alloc(priv, ix, cpu); if (err) return err; @@ -1958,28 +2420,24 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->tstamp = &priv->tstamp; c->ix = ix; c->cpu = cpu; - c->pdev = priv->mdev->device; + c->pdev = mlx5_core_dma_dev(priv->mdev); c->netdev = priv->netdev; - c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.mkey.key); - c->num_tc = params->num_tc; + c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.hw_objs.mkey); + c->num_tc = mlx5e_get_dcb_num_tc(params); c->xdp = !!params->xdp_prog; - c->stats = &priv->channel_stats[ix].ch; - c->irq_desc = irq_to_desc(irq); + c->stats = &priv->channel_stats[ix]->ch; + c->aff_mask = irq_get_effective_affinity_mask(irq); c->lag_port = mlx5e_enumerate_lag_port(priv->mdev, ix); - err = mlx5e_alloc_xps_cpumask(c, params); - if (err) - goto err_free_channel; - - netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); + netif_napi_add(netdev, &c->napi, mlx5e_napi_poll); err = mlx5e_open_queues(c, params, cparam); if (unlikely(err)) goto err_napi_del; - if (umem) { - mlx5e_build_xsk_param(umem, &xsk); - err = mlx5e_open_xsk(priv, params, &xsk, umem, c); + if (xsk_pool) { + mlx5e_build_xsk_param(xsk_pool, &xsk); + err = mlx5e_open_xsk(priv, params, &xsk, xsk_pool, c); if (unlikely(err)) goto err_close_queues; } @@ -1993,9 +2451,7 @@ err_close_queues: err_napi_del: netif_napi_del(&c->napi); - mlx5e_free_xps_cpumask(c); -err_free_channel: kvfree(c); return err; @@ -2005,14 +2461,19 @@ static void mlx5e_activate_channel(struct mlx5e_channel *c) { int tc; + napi_enable(&c->napi); + for (tc = 0; tc < c->num_tc; tc++) mlx5e_activate_txqsq(&c->sq[tc]); mlx5e_activate_icosq(&c->icosq); - mlx5e_activate_rq(&c->rq); - netif_set_xps_queue(c->netdev, c->xps_cpumask, c->ix); + mlx5e_activate_icosq(&c->async_icosq); if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) mlx5e_activate_xsk(c); + else + mlx5e_activate_rq(&c->rq); + + mlx5e_trigger_napi_icosq(c); } static void mlx5e_deactivate_channel(struct mlx5e_channel *c) @@ -2021,11 +2482,16 @@ static void mlx5e_deactivate_channel(struct mlx5e_channel *c) if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) mlx5e_deactivate_xsk(c); + else + mlx5e_deactivate_rq(&c->rq); - mlx5e_deactivate_rq(&c->rq); + mlx5e_deactivate_icosq(&c->async_icosq); mlx5e_deactivate_icosq(&c->icosq); for (tc = 0; tc < c->num_tc; tc++) mlx5e_deactivate_txqsq(&c->sq[tc]); + mlx5e_qos_deactivate_queues(c); + + napi_disable(&c->napi); } static void mlx5e_close_channel(struct mlx5e_channel *c) @@ -2033,288 +2499,12 @@ static void mlx5e_close_channel(struct mlx5e_channel *c) if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) mlx5e_close_xsk(c); mlx5e_close_queues(c); + mlx5e_qos_close_queues(c); netif_napi_del(&c->napi); - mlx5e_free_xps_cpumask(c); kvfree(c); } -#define DEFAULT_FRAG_SIZE (2048) - -static void mlx5e_build_rq_frags_info(struct mlx5_core_dev *mdev, - struct mlx5e_params *params, - struct mlx5e_xsk_param *xsk, - struct mlx5e_rq_frags_info *info) -{ - u32 byte_count = MLX5E_SW2HW_MTU(params, params->sw_mtu); - int frag_size_max = DEFAULT_FRAG_SIZE; - u32 buf_size = 0; - int i; - -#ifdef CONFIG_MLX5_EN_IPSEC - if (MLX5_IPSEC_DEV(mdev)) - byte_count += MLX5E_METADATA_ETHER_LEN; -#endif - - if (mlx5e_rx_is_linear_skb(params, xsk)) { - int frag_stride; - - frag_stride = mlx5e_rx_get_linear_frag_sz(params, xsk); - frag_stride = roundup_pow_of_two(frag_stride); - - info->arr[0].frag_size = byte_count; - info->arr[0].frag_stride = frag_stride; - info->num_frags = 1; - info->wqe_bulk = PAGE_SIZE / frag_stride; - goto out; - } - - if (byte_count > PAGE_SIZE + - (MLX5E_MAX_RX_FRAGS - 1) * frag_size_max) - frag_size_max = PAGE_SIZE; - - i = 0; - while (buf_size < byte_count) { - int frag_size = byte_count - buf_size; - - if (i < MLX5E_MAX_RX_FRAGS - 1) - frag_size = min(frag_size, frag_size_max); - - info->arr[i].frag_size = frag_size; - info->arr[i].frag_stride = roundup_pow_of_two(frag_size); - - buf_size += frag_size; - i++; - } - info->num_frags = i; - /* number of different wqes sharing a page */ - info->wqe_bulk = 1 + (info->num_frags % 2); - -out: - info->wqe_bulk = max_t(u8, info->wqe_bulk, 8); - info->log_num_frags = order_base_2(info->num_frags); -} - -static inline u8 mlx5e_get_rqwq_log_stride(u8 wq_type, int ndsegs) -{ - int sz = sizeof(struct mlx5_wqe_data_seg) * ndsegs; - - switch (wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - sz += sizeof(struct mlx5e_rx_wqe_ll); - break; - default: /* MLX5_WQ_TYPE_CYCLIC */ - sz += sizeof(struct mlx5e_rx_wqe_cyc); - } - - return order_base_2(sz); -} - -static u8 mlx5e_get_rq_log_wq_sz(void *rqc) -{ - void *wq = MLX5_ADDR_OF(rqc, rqc, wq); - - return MLX5_GET(wq, wq, log_wq_sz); -} - -void mlx5e_build_rq_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_xsk_param *xsk, - struct mlx5e_rq_param *param) -{ - struct mlx5_core_dev *mdev = priv->mdev; - void *rqc = param->rqc; - void *wq = MLX5_ADDR_OF(rqc, rqc, wq); - int ndsegs = 1; - - switch (params->rq_wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - MLX5_SET(wq, wq, log_wqe_num_of_strides, - mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk) - - MLX5_MPWQE_LOG_NUM_STRIDES_BASE); - MLX5_SET(wq, wq, log_wqe_stride_size, - mlx5e_mpwqe_get_log_stride_size(mdev, params, xsk) - - MLX5_MPWQE_LOG_STRIDE_SZ_BASE); - MLX5_SET(wq, wq, log_wq_sz, mlx5e_mpwqe_get_log_rq_size(params, xsk)); - break; - default: /* MLX5_WQ_TYPE_CYCLIC */ - MLX5_SET(wq, wq, log_wq_sz, params->log_rq_mtu_frames); - mlx5e_build_rq_frags_info(mdev, params, xsk, ¶m->frags_info); - ndsegs = param->frags_info.num_frags; - } - - MLX5_SET(wq, wq, wq_type, params->rq_wq_type); - MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); - MLX5_SET(wq, wq, log_wq_stride, - mlx5e_get_rqwq_log_stride(params->rq_wq_type, ndsegs)); - MLX5_SET(wq, wq, pd, mdev->mlx5e_res.pdn); - MLX5_SET(rqc, rqc, counter_set_id, priv->q_counter); - MLX5_SET(rqc, rqc, vsd, params->vlan_strip_disable); - MLX5_SET(rqc, rqc, scatter_fcs, params->scatter_fcs_en); - - param->wq.buf_numa_node = dev_to_node(mdev->device); -} - -static void mlx5e_build_drop_rq_param(struct mlx5e_priv *priv, - struct mlx5e_rq_param *param) -{ - struct mlx5_core_dev *mdev = priv->mdev; - void *rqc = param->rqc; - void *wq = MLX5_ADDR_OF(rqc, rqc, wq); - - MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_CYCLIC); - MLX5_SET(wq, wq, log_wq_stride, - mlx5e_get_rqwq_log_stride(MLX5_WQ_TYPE_CYCLIC, 1)); - MLX5_SET(rqc, rqc, counter_set_id, priv->drop_rq_q_counter); - - param->wq.buf_numa_node = dev_to_node(mdev->device); -} - -void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, - struct mlx5e_sq_param *param) -{ - void *sqc = param->sqc; - void *wq = MLX5_ADDR_OF(sqc, sqc, wq); - - MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); - MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); - - param->wq.buf_numa_node = dev_to_node(priv->mdev->device); -} - -static void mlx5e_build_sq_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_sq_param *param) -{ - void *sqc = param->sqc; - void *wq = MLX5_ADDR_OF(sqc, sqc, wq); - bool allow_swp; - - allow_swp = mlx5_geneve_tx_allowed(priv->mdev) || - !!MLX5_IPSEC_DEV(priv->mdev); - mlx5e_build_sq_param_common(priv, param); - MLX5_SET(wq, wq, log_wq_sz, params->log_sq_size); - MLX5_SET(sqc, sqc, allow_swp, allow_swp); -} - -static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, - struct mlx5e_cq_param *param) -{ - void *cqc = param->cqc; - - MLX5_SET(cqc, cqc, uar_page, priv->mdev->priv.uar->index); - if (MLX5_CAP_GEN(priv->mdev, cqe_128_always) && cache_line_size() >= 128) - MLX5_SET(cqc, cqc, cqe_sz, CQE_STRIDE_128_PAD); -} - -void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_xsk_param *xsk, - struct mlx5e_cq_param *param) -{ - struct mlx5_core_dev *mdev = priv->mdev; - void *cqc = param->cqc; - u8 log_cq_size; - - switch (params->rq_wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - log_cq_size = mlx5e_mpwqe_get_log_rq_size(params, xsk) + - mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk); - break; - default: /* MLX5_WQ_TYPE_CYCLIC */ - log_cq_size = params->log_rq_mtu_frames; - } - - MLX5_SET(cqc, cqc, log_cq_size, log_cq_size); - if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) { - MLX5_SET(cqc, cqc, mini_cqe_res_format, MLX5_CQE_FORMAT_CSUM); - MLX5_SET(cqc, cqc, cqe_comp_en, 1); - } - - mlx5e_build_common_cq_param(priv, param); - param->cq_period_mode = params->rx_cq_moderation.cq_period_mode; -} - -void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_cq_param *param) -{ - void *cqc = param->cqc; - - MLX5_SET(cqc, cqc, log_cq_size, params->log_sq_size); - - mlx5e_build_common_cq_param(priv, param); - param->cq_period_mode = params->tx_cq_moderation.cq_period_mode; -} - -void mlx5e_build_ico_cq_param(struct mlx5e_priv *priv, - u8 log_wq_size, - struct mlx5e_cq_param *param) -{ - void *cqc = param->cqc; - - MLX5_SET(cqc, cqc, log_cq_size, log_wq_size); - - mlx5e_build_common_cq_param(priv, param); - - param->cq_period_mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; -} - -void mlx5e_build_icosq_param(struct mlx5e_priv *priv, - u8 log_wq_size, - struct mlx5e_sq_param *param) -{ - void *sqc = param->sqc; - void *wq = MLX5_ADDR_OF(sqc, sqc, wq); - - mlx5e_build_sq_param_common(priv, param); - - MLX5_SET(wq, wq, log_wq_sz, log_wq_size); - MLX5_SET(sqc, sqc, reg_umr, MLX5_CAP_ETH(priv->mdev, reg_umr_sq)); -} - -void mlx5e_build_xdpsq_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_sq_param *param) -{ - void *sqc = param->sqc; - void *wq = MLX5_ADDR_OF(sqc, sqc, wq); - - mlx5e_build_sq_param_common(priv, param); - MLX5_SET(wq, wq, log_wq_sz, params->log_sq_size); - param->is_mpw = MLX5E_GET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE); -} - -static u8 mlx5e_build_icosq_log_wq_sz(struct mlx5e_params *params, - struct mlx5e_rq_param *rqp) -{ - switch (params->rq_wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - return order_base_2(MLX5E_UMR_WQEBBS) + - mlx5e_get_rq_log_wq_sz(rqp->rqc); - default: /* MLX5_WQ_TYPE_CYCLIC */ - return MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE; - } -} - -static void mlx5e_build_channel_param(struct mlx5e_priv *priv, - struct mlx5e_params *params, - struct mlx5e_channel_param *cparam) -{ - u8 icosq_log_wq_sz; - - mlx5e_build_rq_param(priv, params, NULL, &cparam->rq); - - icosq_log_wq_sz = mlx5e_build_icosq_log_wq_sz(params, &cparam->rq); - - mlx5e_build_sq_param(priv, params, &cparam->sq); - mlx5e_build_xdpsq_param(priv, params, &cparam->xdp_sq); - mlx5e_build_icosq_param(priv, icosq_log_wq_sz, &cparam->icosq); - mlx5e_build_rx_cq_param(priv, params, NULL, &cparam->rx_cq); - mlx5e_build_tx_cq_param(priv, params, &cparam->tx_cq); - mlx5e_build_ico_cq_param(priv, icosq_log_wq_sz, &cparam->icosq_cq); -} - int mlx5e_open_channels(struct mlx5e_priv *priv, struct mlx5e_channels *chs) { @@ -2329,22 +2519,41 @@ int mlx5e_open_channels(struct mlx5e_priv *priv, if (!chs->c || !cparam) goto err_free; - mlx5e_build_channel_param(priv, &chs->params, cparam); + err = mlx5e_build_channel_param(priv->mdev, &chs->params, priv->q_counter, cparam); + if (err) + goto err_free; + for (i = 0; i < chs->num; i++) { - struct xdp_umem *umem = NULL; + struct xsk_buff_pool *xsk_pool = NULL; if (chs->params.xdp_prog) - umem = mlx5e_xsk_get_umem(&chs->params, chs->params.xsk, i); + xsk_pool = mlx5e_xsk_get_pool(&chs->params, chs->params.xsk, i); + + err = mlx5e_open_channel(priv, i, &chs->params, cparam, xsk_pool, &chs->c[i]); + if (err) + goto err_close_channels; + } - err = mlx5e_open_channel(priv, i, &chs->params, cparam, umem, &chs->c[i]); + if (MLX5E_GET_PFLAG(&chs->params, MLX5E_PFLAG_TX_PORT_TS) || chs->params.ptp_rx) { + err = mlx5e_ptp_open(priv, &chs->params, chs->c[0]->lag_port, &chs->ptp); if (err) goto err_close_channels; } + if (priv->htb) { + err = mlx5e_qos_open_queues(priv, chs); + if (err) + goto err_close_ptp; + } + mlx5e_health_channels_update(priv); kvfree(cparam); return 0; +err_close_ptp: + if (chs->ptp) + mlx5e_ptp_close(chs->ptp); + err_close_channels: for (i--; i >= 0; i--) mlx5e_close_channel(chs->c[i]); @@ -2362,9 +2571,10 @@ static void mlx5e_activate_channels(struct mlx5e_channels *chs) for (i = 0; i < chs->num; i++) mlx5e_activate_channel(chs->c[i]); -} -#define MLX5E_RQ_WQES_TIMEOUT 20000 /* msecs */ + if (chs->ptp) + mlx5e_ptp_activate_channel(chs->ptp); +} static int mlx5e_wait_channels_min_rx_wqes(struct mlx5e_channels *chs) { @@ -2373,8 +2583,12 @@ static int mlx5e_wait_channels_min_rx_wqes(struct mlx5e_channels *chs) for (i = 0; i < chs->num; i++) { int timeout = err ? 0 : MLX5E_RQ_WQES_TIMEOUT; + struct mlx5e_channel *c = chs->c[i]; + + if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) + continue; - err |= mlx5e_wait_for_min_rx_wqes(&chs->c[i]->rq, timeout); + err |= mlx5e_wait_for_min_rx_wqes(&c->rq, timeout); /* Don't wait on the XSK RQ, because the newer xdpsock sample * doesn't provide any Fill Ring entries at the setup stage. @@ -2388,6 +2602,9 @@ static void mlx5e_deactivate_channels(struct mlx5e_channels *chs) { int i; + if (chs->ptp) + mlx5e_ptp_deactivate_channel(chs->ptp); + for (i = 0; i < chs->num; i++) mlx5e_deactivate_channel(chs->c[i]); } @@ -2396,6 +2613,10 @@ void mlx5e_close_channels(struct mlx5e_channels *chs) { int i; + if (chs->ptp) { + mlx5e_ptp_close(chs->ptp); + chs->ptp = NULL; + } for (i = 0; i < chs->num; i++) mlx5e_close_channel(chs->c[i]); @@ -2403,586 +2624,402 @@ void mlx5e_close_channels(struct mlx5e_channels *chs) chs->num = 0; } -static int -mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, struct mlx5e_rqt *rqt) +static int mlx5e_modify_tirs_packet_merge(struct mlx5e_priv *priv) { - struct mlx5_core_dev *mdev = priv->mdev; - void *rqtc; - int inlen; - int err; - u32 *in; - int i; - - inlen = MLX5_ST_SZ_BYTES(create_rqt_in) + sizeof(u32) * sz; - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) - return -ENOMEM; + struct mlx5e_rx_res *res = priv->rx_res; - rqtc = MLX5_ADDR_OF(create_rqt_in, in, rqt_context); - - MLX5_SET(rqtc, rqtc, rqt_actual_size, sz); - MLX5_SET(rqtc, rqtc, rqt_max_size, sz); - - for (i = 0; i < sz; i++) - MLX5_SET(rqtc, rqtc, rq_num[i], priv->drop_rq.rqn); - - err = mlx5_core_create_rqt(mdev, in, inlen, &rqt->rqtn); - if (!err) - rqt->enabled = true; - - kvfree(in); - return err; + return mlx5e_rx_res_packet_merge_set_param(res, &priv->channels.params.packet_merge); } -void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) -{ - rqt->enabled = false; - mlx5_core_destroy_rqt(priv->mdev, rqt->rqtn); -} +static MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_modify_tirs_packet_merge); -int mlx5e_create_indirect_rqt(struct mlx5e_priv *priv) +static int mlx5e_set_mtu(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, u16 mtu) { - struct mlx5e_rqt *rqt = &priv->indir_rqt; + u16 hw_mtu = MLX5E_SW2HW_MTU(params, mtu); int err; - err = mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, rqt); + err = mlx5_set_port_mtu(mdev, hw_mtu, 1); if (err) - mlx5_core_warn(priv->mdev, "create indirect rqts failed, %d\n", err); - return err; -} - -int mlx5e_create_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs) -{ - int err; - int ix; - - for (ix = 0; ix < priv->max_nch; ix++) { - err = mlx5e_create_rqt(priv, 1 /*size */, &tirs[ix].rqt); - if (unlikely(err)) - goto err_destroy_rqts; - } + return err; + /* Update vport context MTU */ + mlx5_modify_nic_vport_mtu(mdev, hw_mtu); return 0; - -err_destroy_rqts: - mlx5_core_warn(priv->mdev, "create rqts failed, %d\n", err); - for (ix--; ix >= 0; ix--) - mlx5e_destroy_rqt(priv, &tirs[ix].rqt); - - return err; -} - -void mlx5e_destroy_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs) -{ - int i; - - for (i = 0; i < priv->max_nch; i++) - mlx5e_destroy_rqt(priv, &tirs[i].rqt); -} - -static int mlx5e_rx_hash_fn(int hfunc) -{ - return (hfunc == ETH_RSS_HASH_TOP) ? - MLX5_RX_HASH_FN_TOEPLITZ : - MLX5_RX_HASH_FN_INVERTED_XOR8; } -int mlx5e_bits_invert(unsigned long a, int size) -{ - int inv = 0; - int i; - - for (i = 0; i < size; i++) - inv |= (test_bit(size - i - 1, &a) ? 1 : 0) << i; - - return inv; -} - -static void mlx5e_fill_rqt_rqns(struct mlx5e_priv *priv, int sz, - struct mlx5e_redirect_rqt_param rrp, void *rqtc) +static void mlx5e_query_mtu(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, u16 *mtu) { - int i; - - for (i = 0; i < sz; i++) { - u32 rqn; - - if (rrp.is_rss) { - int ix = i; + u16 hw_mtu = 0; + int err; - if (rrp.rss.hfunc == ETH_RSS_HASH_XOR) - ix = mlx5e_bits_invert(i, ilog2(sz)); + err = mlx5_query_nic_vport_mtu(mdev, &hw_mtu); + if (err || !hw_mtu) /* fallback to port oper mtu */ + mlx5_query_port_oper_mtu(mdev, &hw_mtu, 1); - ix = priv->rss_params.indirection_rqt[ix]; - rqn = rrp.rss.channels->c[ix]->rq.rqn; - } else { - rqn = rrp.rqn; - } - MLX5_SET(rqtc, rqtc, rq_num[i], rqn); - } + *mtu = MLX5E_HW2SW_MTU(params, hw_mtu); } -int mlx5e_redirect_rqt(struct mlx5e_priv *priv, u32 rqtn, int sz, - struct mlx5e_redirect_rqt_param rrp) +int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) { + struct mlx5e_params *params = &priv->channels.params; + struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; - void *rqtc; - int inlen; - u32 *in; + u16 mtu; int err; - inlen = MLX5_ST_SZ_BYTES(modify_rqt_in) + sizeof(u32) * sz; - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) - return -ENOMEM; - - rqtc = MLX5_ADDR_OF(modify_rqt_in, in, ctx); + err = mlx5e_set_mtu(mdev, params, params->sw_mtu); + if (err) + return err; - MLX5_SET(rqtc, rqtc, rqt_actual_size, sz); - MLX5_SET(modify_rqt_in, in, bitmask.rqn_list, 1); - mlx5e_fill_rqt_rqns(priv, sz, rrp, rqtc); - err = mlx5_core_modify_rqt(mdev, rqtn, in, inlen); + mlx5e_query_mtu(mdev, params, &mtu); + if (mtu != params->sw_mtu) + netdev_warn(netdev, "%s: VPort MTU %d is different than netdev mtu %d\n", + __func__, mtu, params->sw_mtu); - kvfree(in); - return err; + params->sw_mtu = mtu; + return 0; } -static u32 mlx5e_get_direct_rqn(struct mlx5e_priv *priv, int ix, - struct mlx5e_redirect_rqt_param rrp) -{ - if (!rrp.is_rss) - return rrp.rqn; - - if (ix >= rrp.rss.channels->num) - return priv->drop_rq.rqn; - - return rrp.rss.channels->c[ix]->rq.rqn; -} +MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_set_dev_port_mtu); -static void mlx5e_redirect_rqts(struct mlx5e_priv *priv, - struct mlx5e_redirect_rqt_param rrp) +void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv) { - u32 rqtn; - int ix; - - if (priv->indir_rqt.enabled) { - /* RSS RQ table */ - rqtn = priv->indir_rqt.rqtn; - mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, rrp); - } - - for (ix = 0; ix < priv->max_nch; ix++) { - struct mlx5e_redirect_rqt_param direct_rrp = { - .is_rss = false, - { - .rqn = mlx5e_get_direct_rqn(priv, ix, rrp) - }, - }; - - /* Direct RQ Tables */ - if (!priv->direct_tir[ix].rqt.enabled) - continue; - - rqtn = priv->direct_tir[ix].rqt.rqtn; - mlx5e_redirect_rqt(priv, rqtn, 1, direct_rrp); - } -} + struct mlx5e_params *params = &priv->channels.params; + struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = priv->mdev; + u16 max_mtu; -static void mlx5e_redirect_rqts_to_channels(struct mlx5e_priv *priv, - struct mlx5e_channels *chs) -{ - struct mlx5e_redirect_rqt_param rrp = { - .is_rss = true, - { - .rss = { - .channels = chs, - .hfunc = priv->rss_params.hfunc, - } - }, - }; - - mlx5e_redirect_rqts(priv, rrp); -} - -static void mlx5e_redirect_rqts_to_drop(struct mlx5e_priv *priv) -{ - struct mlx5e_redirect_rqt_param drop_rrp = { - .is_rss = false, - { - .rqn = priv->drop_rq.rqn, - }, - }; - - mlx5e_redirect_rqts(priv, drop_rrp); -} - -static const struct mlx5e_tirc_config tirc_default_config[MLX5E_NUM_INDIR_TIRS] = { - [MLX5E_TT_IPV4_TCP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV4, - .l4_prot_type = MLX5_L4_PROT_TYPE_TCP, - .rx_hash_fields = MLX5_HASH_IP_L4PORTS, - }, - [MLX5E_TT_IPV6_TCP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV6, - .l4_prot_type = MLX5_L4_PROT_TYPE_TCP, - .rx_hash_fields = MLX5_HASH_IP_L4PORTS, - }, - [MLX5E_TT_IPV4_UDP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV4, - .l4_prot_type = MLX5_L4_PROT_TYPE_UDP, - .rx_hash_fields = MLX5_HASH_IP_L4PORTS, - }, - [MLX5E_TT_IPV6_UDP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV6, - .l4_prot_type = MLX5_L4_PROT_TYPE_UDP, - .rx_hash_fields = MLX5_HASH_IP_L4PORTS, - }, - [MLX5E_TT_IPV4_IPSEC_AH] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV4, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP_IPSEC_SPI, - }, - [MLX5E_TT_IPV6_IPSEC_AH] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV6, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP_IPSEC_SPI, - }, - [MLX5E_TT_IPV4_IPSEC_ESP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV4, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP_IPSEC_SPI, - }, - [MLX5E_TT_IPV6_IPSEC_ESP] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV6, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP_IPSEC_SPI, - }, - [MLX5E_TT_IPV4] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV4, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP, - }, - [MLX5E_TT_IPV6] = { .l3_prot_type = MLX5_L3_PROT_TYPE_IPV6, - .l4_prot_type = 0, - .rx_hash_fields = MLX5_HASH_IP, - }, -}; + /* MTU range: 68 - hw-specific max */ + netdev->min_mtu = ETH_MIN_MTU; -struct mlx5e_tirc_config mlx5e_tirc_get_default_config(enum mlx5e_traffic_types tt) -{ - return tirc_default_config[tt]; + mlx5_query_port_max_mtu(mdev, &max_mtu, 1); + netdev->max_mtu = min_t(unsigned int, MLX5E_HW2SW_MTU(params, max_mtu), + ETH_MAX_MTU); } -static void mlx5e_build_tir_ctx_lro(struct mlx5e_params *params, void *tirc) +static int mlx5e_netdev_set_tcs(struct net_device *netdev, u16 nch, u8 ntc, + struct netdev_tc_txq *tc_to_txq) { - if (!params->lro_en) - return; + int tc, err; -#define ROUGH_MAX_L2_L3_HDR_SZ 256 + netdev_reset_tc(netdev); - MLX5_SET(tirc, tirc, lro_enable_mask, - MLX5_TIRC_LRO_ENABLE_MASK_IPV4_LRO | - MLX5_TIRC_LRO_ENABLE_MASK_IPV6_LRO); - MLX5_SET(tirc, tirc, lro_max_ip_payload_size, - (MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ - ROUGH_MAX_L2_L3_HDR_SZ) >> 8); - MLX5_SET(tirc, tirc, lro_timeout_period_usecs, params->lro_timeout); -} + if (ntc == 1) + return 0; -void mlx5e_build_indir_tir_ctx_hash(struct mlx5e_rss_params *rss_params, - const struct mlx5e_tirc_config *ttconfig, - void *tirc, bool inner) -{ - void *hfso = inner ? MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_inner) : - MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); + err = netdev_set_num_tc(netdev, ntc); + if (err) { + netdev_WARN(netdev, "netdev_set_num_tc failed (%d), ntc = %d\n", err, ntc); + return err; + } - MLX5_SET(tirc, tirc, rx_hash_fn, mlx5e_rx_hash_fn(rss_params->hfunc)); - if (rss_params->hfunc == ETH_RSS_HASH_TOP) { - void *rss_key = MLX5_ADDR_OF(tirc, tirc, - rx_hash_toeplitz_key); - size_t len = MLX5_FLD_SZ_BYTES(tirc, - rx_hash_toeplitz_key); + for (tc = 0; tc < ntc; tc++) { + u16 count, offset; - MLX5_SET(tirc, tirc, rx_hash_symmetric, 1); - memcpy(rss_key, rss_params->toeplitz_hash_key, len); + count = tc_to_txq[tc].count; + offset = tc_to_txq[tc].offset; + netdev_set_tc_queue(netdev, tc, count, offset); } - MLX5_SET(rx_hash_field_select, hfso, l3_prot_type, - ttconfig->l3_prot_type); - MLX5_SET(rx_hash_field_select, hfso, l4_prot_type, - ttconfig->l4_prot_type); - MLX5_SET(rx_hash_field_select, hfso, selected_fields, - ttconfig->rx_hash_fields); -} -static void mlx5e_update_rx_hash_fields(struct mlx5e_tirc_config *ttconfig, - enum mlx5e_traffic_types tt, - u32 rx_hash_fields) -{ - *ttconfig = tirc_default_config[tt]; - ttconfig->rx_hash_fields = rx_hash_fields; + return 0; } -void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen) +int mlx5e_update_tx_netdev_queues(struct mlx5e_priv *priv) { - void *tirc = MLX5_ADDR_OF(modify_tir_in, in, ctx); - struct mlx5e_rss_params *rss = &priv->rss_params; - struct mlx5_core_dev *mdev = priv->mdev; - int ctxlen = MLX5_ST_SZ_BYTES(tirc); - struct mlx5e_tirc_config ttconfig; - int tt; + int nch, ntc, num_txqs, err; + int qos_queues = 0; - MLX5_SET(modify_tir_in, in, bitmask.hash, 1); + if (priv->htb) + qos_queues = mlx5e_htb_cur_leaf_nodes(priv->htb); - for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - memset(tirc, 0, ctxlen); - mlx5e_update_rx_hash_fields(&ttconfig, tt, - rss->rx_hash_fields[tt]); - mlx5e_build_indir_tir_ctx_hash(rss, &ttconfig, tirc, false); - mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, inlen); - } + nch = priv->channels.params.num_channels; + ntc = mlx5e_get_dcb_num_tc(&priv->channels.params); + num_txqs = nch * ntc + qos_queues; + if (MLX5E_GET_PFLAG(&priv->channels.params, MLX5E_PFLAG_TX_PORT_TS)) + num_txqs += ntc; - if (!mlx5e_tunnel_inner_ft_supported(priv->mdev)) - return; + mlx5e_dbg(DRV, priv, "Setting num_txqs %d\n", num_txqs); + err = netif_set_real_num_tx_queues(priv->netdev, num_txqs); + if (err) + netdev_warn(priv->netdev, "netif_set_real_num_tx_queues failed, %d\n", err); - for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - memset(tirc, 0, ctxlen); - mlx5e_update_rx_hash_fields(&ttconfig, tt, - rss->rx_hash_fields[tt]); - mlx5e_build_indir_tir_ctx_hash(rss, &ttconfig, tirc, true); - mlx5_core_modify_tir(mdev, priv->inner_indir_tir[tt].tirn, in, - inlen); - } + return err; } -static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv) +static int mlx5e_update_netdev_queues(struct mlx5e_priv *priv) { - struct mlx5_core_dev *mdev = priv->mdev; - - void *in; - void *tirc; - int inlen; + struct netdev_tc_txq old_tc_to_txq[TC_MAX_QUEUE], *tc_to_txq; + struct net_device *netdev = priv->netdev; + int old_num_txqs, old_ntc; + int nch, ntc; int err; - int tt; - int ix; - - inlen = MLX5_ST_SZ_BYTES(modify_tir_in); - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) - return -ENOMEM; + int i; - MLX5_SET(modify_tir_in, in, bitmask.lro, 1); - tirc = MLX5_ADDR_OF(modify_tir_in, in, ctx); + old_num_txqs = netdev->real_num_tx_queues; + old_ntc = netdev->num_tc ? : 1; + for (i = 0; i < ARRAY_SIZE(old_tc_to_txq); i++) + old_tc_to_txq[i] = netdev->tc_to_txq[i]; - mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc); + nch = priv->channels.params.num_channels; + ntc = priv->channels.params.mqprio.num_tc; + tc_to_txq = priv->channels.params.mqprio.tc_to_txq; - for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - err = mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, - inlen); - if (err) - goto free_in; + err = mlx5e_netdev_set_tcs(netdev, nch, ntc, tc_to_txq); + if (err) + goto err_out; + err = mlx5e_update_tx_netdev_queues(priv); + if (err) + goto err_tcs; + err = netif_set_real_num_rx_queues(netdev, nch); + if (err) { + netdev_warn(netdev, "netif_set_real_num_rx_queues failed, %d\n", err); + goto err_txqs; } - for (ix = 0; ix < priv->max_nch; ix++) { - err = mlx5_core_modify_tir(mdev, priv->direct_tir[ix].tirn, - in, inlen); - if (err) - goto free_in; - } + return 0; -free_in: - kvfree(in); +err_txqs: + /* netif_set_real_num_rx_queues could fail only when nch increased. Only + * one of nch and ntc is changed in this function. That means, the call + * to netif_set_real_num_tx_queues below should not fail, because it + * decreases the number of TX queues. + */ + WARN_ON_ONCE(netif_set_real_num_tx_queues(netdev, old_num_txqs)); +err_tcs: + WARN_ON_ONCE(mlx5e_netdev_set_tcs(netdev, old_num_txqs / old_ntc, old_ntc, + old_tc_to_txq)); +err_out: return err; } -static int mlx5e_set_mtu(struct mlx5_core_dev *mdev, - struct mlx5e_params *params, u16 mtu) +static MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_update_netdev_queues); + +static void mlx5e_set_default_xps_cpumasks(struct mlx5e_priv *priv, + struct mlx5e_params *params) { - u16 hw_mtu = MLX5E_SW2HW_MTU(params, mtu); - int err; + struct mlx5_core_dev *mdev = priv->mdev; + int num_comp_vectors, ix, irq; - err = mlx5_set_port_mtu(mdev, hw_mtu, 1); - if (err) - return err; + num_comp_vectors = mlx5_comp_vectors_count(mdev); - /* Update vport context MTU */ - mlx5_modify_nic_vport_mtu(mdev, hw_mtu); - return 0; -} + for (ix = 0; ix < params->num_channels; ix++) { + cpumask_clear(priv->scratchpad.cpumask); -static void mlx5e_query_mtu(struct mlx5_core_dev *mdev, - struct mlx5e_params *params, u16 *mtu) -{ - u16 hw_mtu = 0; - int err; + for (irq = ix; irq < num_comp_vectors; irq += params->num_channels) { + int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(mdev, irq)); - err = mlx5_query_nic_vport_mtu(mdev, &hw_mtu); - if (err || !hw_mtu) /* fallback to port oper mtu */ - mlx5_query_port_oper_mtu(mdev, &hw_mtu, 1); + cpumask_set_cpu(cpu, priv->scratchpad.cpumask); + } - *mtu = MLX5E_HW2SW_MTU(params, hw_mtu); + netif_set_xps_queue(priv->netdev, priv->scratchpad.cpumask, ix); + } } -int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) +static int mlx5e_num_channels_changed(struct mlx5e_priv *priv) { - struct mlx5e_params *params = &priv->channels.params; - struct net_device *netdev = priv->netdev; - struct mlx5_core_dev *mdev = priv->mdev; - u16 mtu; + u16 count = priv->channels.params.num_channels; int err; - err = mlx5e_set_mtu(mdev, params, params->sw_mtu); + err = mlx5e_update_netdev_queues(priv); if (err) return err; - mlx5e_query_mtu(mdev, params, &mtu); - if (mtu != params->sw_mtu) - netdev_warn(netdev, "%s: VPort MTU %d is different than netdev mtu %d\n", - __func__, mtu, params->sw_mtu); - - params->sw_mtu = mtu; - return 0; -} - -void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv) -{ - struct mlx5e_params *params = &priv->channels.params; - struct net_device *netdev = priv->netdev; - struct mlx5_core_dev *mdev = priv->mdev; - u16 max_mtu; + mlx5e_set_default_xps_cpumasks(priv, &priv->channels.params); - /* MTU range: 68 - hw-specific max */ - netdev->min_mtu = ETH_MIN_MTU; + /* This function may be called on attach, before priv->rx_res is created. */ + if (!netif_is_rxfh_configured(priv->netdev) && priv->rx_res) + mlx5e_rx_res_rss_set_indir_uniform(priv->rx_res, count); - mlx5_query_port_max_mtu(mdev, &max_mtu, 1); - netdev->max_mtu = min_t(unsigned int, MLX5E_HW2SW_MTU(params, max_mtu), - ETH_MAX_MTU); + return 0; } -static void mlx5e_netdev_set_tcs(struct net_device *netdev) -{ - struct mlx5e_priv *priv = netdev_priv(netdev); - int nch = priv->channels.params.num_channels; - int ntc = priv->channels.params.num_tc; - int tc; - - netdev_reset_tc(netdev); - - if (ntc == 1) - return; - - netdev_set_num_tc(netdev, ntc); - - /* Map netdev TCs to offset 0 - * We have our own UP to TXQ mapping for QoS - */ - for (tc = 0; tc < ntc; tc++) - netdev_set_tc_queue(netdev, tc, nch, 0); -} +MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX(mlx5e_num_channels_changed); static void mlx5e_build_txq_maps(struct mlx5e_priv *priv) { - int i, ch; + int i, ch, tc, num_tc; ch = priv->channels.num; + num_tc = mlx5e_get_dcb_num_tc(&priv->channels.params); for (i = 0; i < ch; i++) { - int tc; - - for (tc = 0; tc < priv->channels.params.num_tc; tc++) { + for (tc = 0; tc < num_tc; tc++) { struct mlx5e_channel *c = priv->channels.c[i]; struct mlx5e_txqsq *sq = &c->sq[tc]; priv->txq2sq[sq->txq_ix] = sq; - priv->channel_tc2realtxq[i][tc] = i + tc * ch; } } + + if (!priv->channels.ptp) + goto out; + + if (!test_bit(MLX5E_PTP_STATE_TX, priv->channels.ptp->state)) + goto out; + + for (tc = 0; tc < num_tc; tc++) { + struct mlx5e_ptp *c = priv->channels.ptp; + struct mlx5e_txqsq *sq = &c->ptpsq[tc].txqsq; + + priv->txq2sq[sq->txq_ix] = sq; + } + +out: + /* Make the change to txq2sq visible before the queue is started. + * As mlx5e_xmit runs under a spinlock, there is an implicit ACQUIRE, + * which pairs with this barrier. + */ + smp_wmb(); } void mlx5e_activate_priv_channels(struct mlx5e_priv *priv) { - int num_txqs = priv->channels.num * priv->channels.params.num_tc; - int num_rxqs = priv->channels.num * priv->profile->rq_groups; - struct net_device *netdev = priv->netdev; - - mlx5e_netdev_set_tcs(netdev); - netif_set_real_num_tx_queues(netdev, num_txqs); - netif_set_real_num_rx_queues(netdev, num_rxqs); - mlx5e_build_txq_maps(priv); mlx5e_activate_channels(&priv->channels); + if (priv->htb) + mlx5e_qos_activate_queues(priv); mlx5e_xdp_tx_enable(priv); + + /* dev_watchdog() wants all TX queues to be started when the carrier is + * OK, including the ones in range real_num_tx_queues..num_tx_queues-1. + * Make it happy to avoid TX timeout false alarms. + */ netif_tx_start_all_queues(priv->netdev); if (mlx5e_is_vport_rep(priv)) - mlx5e_add_sqs_fwd_rules(priv); + mlx5e_rep_activate_channels(priv); mlx5e_wait_channels_min_rx_wqes(&priv->channels); - mlx5e_redirect_rqts_to_channels(priv, &priv->channels); - mlx5e_xsk_redirect_rqts_to_channels(priv, &priv->channels); + if (priv->rx_res) + mlx5e_rx_res_channels_activate(priv->rx_res, &priv->channels); } void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv) { - mlx5e_xsk_redirect_rqts_to_drop(priv, &priv->channels); - - mlx5e_redirect_rqts_to_drop(priv); + if (priv->rx_res) + mlx5e_rx_res_channels_deactivate(priv->rx_res); if (mlx5e_is_vport_rep(priv)) - mlx5e_remove_sqs_fwd_rules(priv); + mlx5e_rep_deactivate_channels(priv); - /* FIXME: This is a W/A only for tx timeout watch dog false alarm when - * polling for inactive tx queues. + /* The results of ndo_select_queue are unreliable, while netdev config + * is being changed (real_num_tx_queues, num_tc). Stop all queues to + * prevent ndo_start_xmit from being called, so that it can assume that + * the selected queue is always valid. */ - netif_tx_stop_all_queues(priv->netdev); netif_tx_disable(priv->netdev); + mlx5e_xdp_tx_disable(priv); mlx5e_deactivate_channels(&priv->channels); } -static void mlx5e_switch_priv_channels(struct mlx5e_priv *priv, - struct mlx5e_channels *new_chs, - mlx5e_fp_hw_modify hw_modify) +static int mlx5e_switch_priv_params(struct mlx5e_priv *priv, + struct mlx5e_params *new_params, + mlx5e_fp_preactivate preactivate, + void *context) +{ + struct mlx5e_params old_params; + + old_params = priv->channels.params; + priv->channels.params = *new_params; + + if (preactivate) { + int err; + + err = preactivate(priv, context); + if (err) { + priv->channels.params = old_params; + return err; + } + } + + return 0; +} + +static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, + struct mlx5e_channels *new_chs, + mlx5e_fp_preactivate preactivate, + void *context) { struct net_device *netdev = priv->netdev; - int new_num_txqs; + struct mlx5e_channels old_chs; int carrier_ok; - - new_num_txqs = new_chs->num * new_chs->params.num_tc; + int err = 0; carrier_ok = netif_carrier_ok(netdev); netif_carrier_off(netdev); - if (new_num_txqs < netdev->real_num_tx_queues) - netif_set_real_num_tx_queues(netdev, new_num_txqs); - mlx5e_deactivate_priv_channels(priv); - mlx5e_close_channels(&priv->channels); + old_chs = priv->channels; priv->channels = *new_chs; - /* New channels are ready to roll, modify HW settings if needed */ - if (hw_modify) - hw_modify(priv); + /* New channels are ready to roll, call the preactivate hook if needed + * to modify HW settings or update kernel parameters. + */ + if (preactivate) { + err = preactivate(priv, context); + if (err) { + priv->channels = old_chs; + goto out; + } + } + mlx5e_close_channels(&old_chs); priv->profile->update_rx(priv); + + mlx5e_selq_apply(&priv->selq); +out: mlx5e_activate_priv_channels(priv); /* return carrier back if needed */ if (carrier_ok) netif_carrier_on(netdev); + + return err; } -int mlx5e_safe_switch_channels(struct mlx5e_priv *priv, - struct mlx5e_channels *new_chs, - mlx5e_fp_hw_modify hw_modify) +int mlx5e_safe_switch_params(struct mlx5e_priv *priv, + struct mlx5e_params *params, + mlx5e_fp_preactivate preactivate, + void *context, bool reset) { + struct mlx5e_channels new_chs = {}; int err; - err = mlx5e_open_channels(priv, new_chs); + reset &= test_bit(MLX5E_STATE_OPENED, &priv->state); + if (!reset) + return mlx5e_switch_priv_params(priv, params, preactivate, context); + + new_chs.params = *params; + + mlx5e_selq_prepare_params(&priv->selq, &new_chs.params); + + err = mlx5e_open_channels(priv, &new_chs); if (err) - return err; + goto err_cancel_selq; + + err = mlx5e_switch_priv_channels(priv, &new_chs, preactivate, context); + if (err) + goto err_close; - mlx5e_switch_priv_channels(priv, new_chs, hw_modify); return 0; + +err_close: + mlx5e_close_channels(&new_chs); + +err_cancel_selq: + mlx5e_selq_cancel(&priv->selq); + return err; } int mlx5e_safe_reopen_channels(struct mlx5e_priv *priv) { - struct mlx5e_channels new_channels = {}; - - new_channels.params = priv->channels.params; - return mlx5e_safe_switch_channels(priv, &new_channels, NULL); + return mlx5e_safe_switch_params(priv, &priv->channels.params, NULL, NULL, true); } void mlx5e_timestamp_init(struct mlx5e_priv *priv) @@ -2991,11 +3028,33 @@ void mlx5e_timestamp_init(struct mlx5e_priv *priv) priv->tstamp.rx_filter = HWTSTAMP_FILTER_NONE; } +static void mlx5e_modify_admin_state(struct mlx5_core_dev *mdev, + enum mlx5_port_status state) +{ + struct mlx5_eswitch *esw = mdev->priv.eswitch; + int vport_admin_state; + + mlx5_set_port_admin_status(mdev, state); + + if (mlx5_eswitch_mode(mdev) == MLX5_ESWITCH_OFFLOADS || + !MLX5_CAP_GEN(mdev, uplink_follow)) + return; + + if (state == MLX5_PORT_UP) + vport_admin_state = MLX5_VPORT_ADMIN_STATE_AUTO; + else + vport_admin_state = MLX5_VPORT_ADMIN_STATE_DOWN; + + mlx5_eswitch_set_vport_state(esw, MLX5_VPORT_UPLINK, vport_admin_state); +} + int mlx5e_open_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; + mlx5e_selq_prepare_params(&priv->selq, &priv->channels.params); + set_bit(MLX5E_STATE_OPENED, &priv->state); err = mlx5e_open_channels(priv, &priv->channels); @@ -3003,7 +3062,9 @@ int mlx5e_open_locked(struct net_device *netdev) goto err_clear_state_opened_flag; priv->profile->update_rx(priv); + mlx5e_selq_apply(&priv->selq); mlx5e_activate_priv_channels(priv); + mlx5e_apply_traps(priv, true); if (priv->profile->update_carrier) priv->profile->update_carrier(priv); @@ -3012,6 +3073,7 @@ int mlx5e_open_locked(struct net_device *netdev) err_clear_state_opened_flag: clear_bit(MLX5E_STATE_OPENED, &priv->state); + mlx5e_selq_cancel(&priv->selq); return err; } @@ -3023,12 +3085,9 @@ int mlx5e_open(struct net_device *netdev) mutex_lock(&priv->state_lock); err = mlx5e_open_locked(netdev); if (!err) - mlx5_set_port_admin_status(priv->mdev, MLX5_PORT_UP); + mlx5e_modify_admin_state(priv->mdev, MLX5_PORT_UP); mutex_unlock(&priv->state_lock); - if (mlx5_vxlan_allowed(priv->mdev->vxlan)) - udp_tunnel_get_rx_info(netdev); - return err; } @@ -3042,6 +3101,7 @@ int mlx5e_close_locked(struct net_device *netdev) if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) return 0; + mlx5e_apply_traps(priv, false); clear_bit(MLX5E_STATE_OPENED, &priv->state); netif_carrier_off(priv->netdev); @@ -3060,13 +3120,18 @@ int mlx5e_close(struct net_device *netdev) return -ENODEV; mutex_lock(&priv->state_lock); - mlx5_set_port_admin_status(priv->mdev, MLX5_PORT_DOWN); + mlx5e_modify_admin_state(priv->mdev, MLX5_PORT_DOWN); err = mlx5e_close_locked(netdev); mutex_unlock(&priv->state_lock); return err; } +static void mlx5e_free_drop_rq(struct mlx5e_rq *rq) +{ + mlx5_wq_destroy(&rq->wq_ctrl); +} + static int mlx5e_alloc_drop_rq(struct mlx5_core_dev *mdev, struct mlx5e_rq *rq, struct mlx5e_rq_param *param) @@ -3090,14 +3155,16 @@ static int mlx5e_alloc_drop_rq(struct mlx5_core_dev *mdev, return 0; } -static int mlx5e_alloc_drop_cq(struct mlx5_core_dev *mdev, +static int mlx5e_alloc_drop_cq(struct mlx5e_priv *priv, struct mlx5e_cq *cq, struct mlx5e_cq_param *param) { - param->wq.buf_numa_node = dev_to_node(mdev->device); - param->wq.db_numa_node = dev_to_node(mdev->device); + struct mlx5_core_dev *mdev = priv->mdev; + + param->wq.buf_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); + param->wq.db_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); - return mlx5e_alloc_cq_common(mdev, param, cq); + return mlx5e_alloc_cq_common(priv, param, cq); } int mlx5e_open_drop_rq(struct mlx5e_priv *priv, @@ -3109,9 +3176,9 @@ int mlx5e_open_drop_rq(struct mlx5e_priv *priv, struct mlx5e_cq *cq = &drop_rq->cq; int err; - mlx5e_build_drop_rq_param(priv, &rq_param); + mlx5e_build_drop_rq_param(mdev, priv->drop_rq_q_counter, &rq_param); - err = mlx5e_alloc_drop_cq(mdev, cq, &cq_param); + err = mlx5e_alloc_drop_cq(priv, cq, &cq_param); if (err) return err; @@ -3134,7 +3201,7 @@ int mlx5e_open_drop_rq(struct mlx5e_priv *priv, return 0; err_free_rq: - mlx5e_free_rq(drop_rq); + mlx5e_free_drop_rq(drop_rq); err_destroy_cq: mlx5e_destroy_cq(cq); @@ -3148,7 +3215,7 @@ err_free_cq: void mlx5e_close_drop_rq(struct mlx5e_rq *drop_rq) { mlx5e_destroy_rq(drop_rq); - mlx5e_free_rq(drop_rq); + mlx5e_free_drop_rq(drop_rq); mlx5e_destroy_cq(&drop_rq->cq); mlx5e_free_cq(&drop_rq->cq); } @@ -3157,15 +3224,15 @@ int mlx5e_create_tis(struct mlx5_core_dev *mdev, void *in, u32 *tisn) { void *tisc = MLX5_ADDR_OF(create_tis_in, in, ctx); - MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn); + MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.hw_objs.td.tdn); if (MLX5_GET(tisc, tisc, tls_en)) - MLX5_SET(tisc, tisc, pd, mdev->mlx5e_res.pdn); + MLX5_SET(tisc, tisc, pd, mdev->mlx5e_res.hw_objs.pdn); if (mlx5_lag_is_lacp_owner(mdev)) MLX5_SET(tisc, tisc, strict_lag_tx_port_affinity, 1); - return mlx5_core_create_tis(mdev, in, MLX5_ST_SZ_BYTES(create_tis_in), tisn); + return mlx5_core_create_tis(mdev, in, tisn); } void mlx5e_destroy_tis(struct mlx5_core_dev *mdev, u32 tisn) @@ -3224,258 +3291,292 @@ err_close_tises: static void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) { + if (priv->mqprio_rl) { + mlx5e_mqprio_rl_cleanup(priv->mqprio_rl); + mlx5e_mqprio_rl_free(priv->mqprio_rl); + priv->mqprio_rl = NULL; + } + mlx5e_accel_cleanup_tx(priv); mlx5e_destroy_tises(priv); } -static void mlx5e_build_indir_tir_ctx_common(struct mlx5e_priv *priv, - u32 rqtn, u32 *tirc) +static int mlx5e_modify_channels_scatter_fcs(struct mlx5e_channels *chs, bool enable) { - MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); - MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT); - MLX5_SET(tirc, tirc, indirect_table, rqtn); - MLX5_SET(tirc, tirc, tunneled_offload_en, - priv->channels.params.tunneled_offload_en); + int err = 0; + int i; + + for (i = 0; i < chs->num; i++) { + err = mlx5e_modify_rq_scatter_fcs(&chs->c[i]->rq, enable); + if (err) + return err; + } - mlx5e_build_tir_ctx_lro(&priv->channels.params, tirc); + return 0; } -static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, - enum mlx5e_traffic_types tt, - u32 *tirc) +static int mlx5e_modify_channels_vsd(struct mlx5e_channels *chs, bool vsd) { - mlx5e_build_indir_tir_ctx_common(priv, priv->indir_rqt.rqtn, tirc); - mlx5e_build_indir_tir_ctx_hash(&priv->rss_params, - &tirc_default_config[tt], tirc, false); + int err; + int i; + + for (i = 0; i < chs->num; i++) { + err = mlx5e_modify_rq_vsd(&chs->c[i]->rq, vsd); + if (err) + return err; + } + if (chs->ptp && test_bit(MLX5E_PTP_STATE_RX, chs->ptp->state)) + return mlx5e_modify_rq_vsd(&chs->ptp->rq, vsd); + + return 0; } -static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 rqtn, u32 *tirc) +static void mlx5e_mqprio_build_default_tc_to_txq(struct netdev_tc_txq *tc_to_txq, + int ntc, int nch) { - mlx5e_build_indir_tir_ctx_common(priv, rqtn, tirc); - MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_INVERTED_XOR8); + int tc; + + memset(tc_to_txq, 0, sizeof(*tc_to_txq) * TC_MAX_QUEUE); + + /* Map netdev TCs to offset 0. + * We have our own UP to TXQ mapping for DCB mode of QoS + */ + for (tc = 0; tc < ntc; tc++) { + tc_to_txq[tc] = (struct netdev_tc_txq) { + .count = nch, + .offset = 0, + }; + } } -static void mlx5e_build_inner_indir_tir_ctx(struct mlx5e_priv *priv, - enum mlx5e_traffic_types tt, - u32 *tirc) +static void mlx5e_mqprio_build_tc_to_txq(struct netdev_tc_txq *tc_to_txq, + struct tc_mqprio_qopt *qopt) { - mlx5e_build_indir_tir_ctx_common(priv, priv->indir_rqt.rqtn, tirc); - mlx5e_build_indir_tir_ctx_hash(&priv->rss_params, - &tirc_default_config[tt], tirc, true); + int tc; + + for (tc = 0; tc < TC_MAX_QUEUE; tc++) { + tc_to_txq[tc] = (struct netdev_tc_txq) { + .count = qopt->count[tc], + .offset = qopt->offset[tc], + }; + } } -int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc) +static void mlx5e_params_mqprio_dcb_set(struct mlx5e_params *params, u8 num_tc) { - struct mlx5e_tir *tir; - void *tirc; - int inlen; - int i = 0; - int err; - u32 *in; - int tt; - - inlen = MLX5_ST_SZ_BYTES(create_tir_in); - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) - return -ENOMEM; + params->mqprio.mode = TC_MQPRIO_MODE_DCB; + params->mqprio.num_tc = num_tc; + mlx5e_mqprio_build_default_tc_to_txq(params->mqprio.tc_to_txq, num_tc, + params->num_channels); +} - for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - memset(in, 0, inlen); - tir = &priv->indir_tir[tt]; - tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); - mlx5e_build_indir_tir_ctx(priv, tt, tirc); - err = mlx5e_create_tir(priv->mdev, tir, in, inlen); - if (err) { - mlx5_core_warn(priv->mdev, "create indirect tirs failed, %d\n", err); - goto err_destroy_inner_tirs; - } - } +static void mlx5e_mqprio_rl_update_params(struct mlx5e_params *params, + struct mlx5e_mqprio_rl *rl) +{ + int tc; - if (!inner_ttc || !mlx5e_tunnel_inner_ft_supported(priv->mdev)) - goto out; + for (tc = 0; tc < TC_MAX_QUEUE; tc++) { + u32 hw_id = 0; - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) { - memset(in, 0, inlen); - tir = &priv->inner_indir_tir[i]; - tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); - mlx5e_build_inner_indir_tir_ctx(priv, i, tirc); - err = mlx5e_create_tir(priv->mdev, tir, in, inlen); - if (err) { - mlx5_core_warn(priv->mdev, "create inner indirect tirs failed, %d\n", err); - goto err_destroy_inner_tirs; - } + if (rl) + mlx5e_mqprio_rl_get_node_hw_id(rl, tc, &hw_id); + params->mqprio.channel.hw_id[tc] = hw_id; } +} -out: - kvfree(in); - - return 0; +static void mlx5e_params_mqprio_channel_set(struct mlx5e_params *params, + struct tc_mqprio_qopt_offload *mqprio, + struct mlx5e_mqprio_rl *rl) +{ + int tc; -err_destroy_inner_tirs: - for (i--; i >= 0; i--) - mlx5e_destroy_tir(priv->mdev, &priv->inner_indir_tir[i]); + params->mqprio.mode = TC_MQPRIO_MODE_CHANNEL; + params->mqprio.num_tc = mqprio->qopt.num_tc; - for (tt--; tt >= 0; tt--) - mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[tt]); + for (tc = 0; tc < TC_MAX_QUEUE; tc++) + params->mqprio.channel.max_rate[tc] = mqprio->max_rate[tc]; - kvfree(in); + mlx5e_mqprio_rl_update_params(params, rl); + mlx5e_mqprio_build_tc_to_txq(params->mqprio.tc_to_txq, &mqprio->qopt); +} - return err; +static void mlx5e_params_mqprio_reset(struct mlx5e_params *params) +{ + mlx5e_params_mqprio_dcb_set(params, 1); } -int mlx5e_create_direct_tirs(struct mlx5e_priv *priv, struct mlx5e_tir *tirs) +static int mlx5e_setup_tc_mqprio_dcb(struct mlx5e_priv *priv, + struct tc_mqprio_qopt *mqprio) { - struct mlx5e_tir *tir; - void *tirc; - int inlen; - int err = 0; - u32 *in; - int ix; + struct mlx5e_params new_params; + u8 tc = mqprio->num_tc; + int err; - inlen = MLX5_ST_SZ_BYTES(create_tir_in); - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) - return -ENOMEM; + mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS; - for (ix = 0; ix < priv->max_nch; ix++) { - memset(in, 0, inlen); - tir = &tirs[ix]; - tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); - mlx5e_build_direct_tir_ctx(priv, tir->rqt.rqtn, tirc); - err = mlx5e_create_tir(priv->mdev, tir, in, inlen); - if (unlikely(err)) - goto err_destroy_ch_tirs; - } + if (tc && tc != MLX5E_MAX_NUM_TC) + return -EINVAL; - goto out; + new_params = priv->channels.params; + mlx5e_params_mqprio_dcb_set(&new_params, tc ? tc : 1); -err_destroy_ch_tirs: - mlx5_core_warn(priv->mdev, "create tirs failed, %d\n", err); - for (ix--; ix >= 0; ix--) - mlx5e_destroy_tir(priv->mdev, &tirs[ix]); + err = mlx5e_safe_switch_params(priv, &new_params, + mlx5e_num_channels_changed_ctx, NULL, true); -out: - kvfree(in); + if (!err && priv->mqprio_rl) { + mlx5e_mqprio_rl_cleanup(priv->mqprio_rl); + mlx5e_mqprio_rl_free(priv->mqprio_rl); + priv->mqprio_rl = NULL; + } + priv->max_opened_tc = max_t(u8, priv->max_opened_tc, + mlx5e_get_dcb_num_tc(&priv->channels.params)); return err; } -void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc) +static int mlx5e_mqprio_channel_validate(struct mlx5e_priv *priv, + struct tc_mqprio_qopt_offload *mqprio) { + struct net_device *netdev = priv->netdev; + struct mlx5e_ptp *ptp_channel; + int agg_count = 0; int i; - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); + ptp_channel = priv->channels.ptp; + if (ptp_channel && test_bit(MLX5E_PTP_STATE_TX, ptp_channel->state)) { + netdev_err(netdev, + "Cannot activate MQPRIO mode channel since it conflicts with TX port TS\n"); + return -EINVAL; + } - if (!inner_ttc || !mlx5e_tunnel_inner_ft_supported(priv->mdev)) - return; + if (mqprio->qopt.offset[0] != 0 || mqprio->qopt.num_tc < 1 || + mqprio->qopt.num_tc > MLX5E_MAX_NUM_MQPRIO_CH_TC) + return -EINVAL; - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5e_destroy_tir(priv->mdev, &priv->inner_indir_tir[i]); -} + for (i = 0; i < mqprio->qopt.num_tc; i++) { + if (!mqprio->qopt.count[i]) { + netdev_err(netdev, "Zero size for queue-group (%d) is not supported\n", i); + return -EINVAL; + } + if (mqprio->min_rate[i]) { + netdev_err(netdev, "Min tx rate is not supported\n"); + return -EINVAL; + } -void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv, struct mlx5e_tir *tirs) -{ - int i; + if (mqprio->max_rate[i]) { + int err; - for (i = 0; i < priv->max_nch; i++) - mlx5e_destroy_tir(priv->mdev, &tirs[i]); -} + err = mlx5e_qos_bytes_rate_check(priv->mdev, mqprio->max_rate[i]); + if (err) + return err; + } -static int mlx5e_modify_channels_scatter_fcs(struct mlx5e_channels *chs, bool enable) -{ - int err = 0; - int i; + if (mqprio->qopt.offset[i] != agg_count) { + netdev_err(netdev, "Discontinuous queues config is not supported\n"); + return -EINVAL; + } + agg_count += mqprio->qopt.count[i]; + } - for (i = 0; i < chs->num; i++) { - err = mlx5e_modify_rq_scatter_fcs(&chs->c[i]->rq, enable); - if (err) - return err; + if (priv->channels.params.num_channels != agg_count) { + netdev_err(netdev, "Num of queues (%d) does not match available (%d)\n", + agg_count, priv->channels.params.num_channels); + return -EINVAL; } return 0; } -static int mlx5e_modify_channels_vsd(struct mlx5e_channels *chs, bool vsd) +static bool mlx5e_mqprio_rate_limit(u8 num_tc, u64 max_rate[]) { - int err = 0; - int i; - - for (i = 0; i < chs->num; i++) { - err = mlx5e_modify_rq_vsd(&chs->c[i]->rq, vsd); - if (err) - return err; - } + int tc; - return 0; + for (tc = 0; tc < num_tc; tc++) + if (max_rate[tc]) + return true; + return false; } -static int mlx5e_setup_tc_mqprio(struct mlx5e_priv *priv, - struct tc_mqprio_qopt *mqprio) +static struct mlx5e_mqprio_rl *mlx5e_mqprio_rl_create(struct mlx5_core_dev *mdev, + u8 num_tc, u64 max_rate[]) { - struct mlx5e_channels new_channels = {}; - u8 tc = mqprio->num_tc; - int err = 0; + struct mlx5e_mqprio_rl *rl; + int err; - mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS; + if (!mlx5e_mqprio_rate_limit(num_tc, max_rate)) + return NULL; - if (tc && tc != MLX5E_MAX_NUM_TC) - return -EINVAL; + rl = mlx5e_mqprio_rl_alloc(); + if (!rl) + return ERR_PTR(-ENOMEM); - mutex_lock(&priv->state_lock); + err = mlx5e_mqprio_rl_init(rl, mdev, num_tc, max_rate); + if (err) { + mlx5e_mqprio_rl_free(rl); + return ERR_PTR(err); + } - new_channels.params = priv->channels.params; - new_channels.params.num_tc = tc ? tc : 1; + return rl; +} - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { - priv->channels.params = new_channels.params; - goto out; - } +static int mlx5e_setup_tc_mqprio_channel(struct mlx5e_priv *priv, + struct tc_mqprio_qopt_offload *mqprio) +{ + mlx5e_fp_preactivate preactivate; + struct mlx5e_params new_params; + struct mlx5e_mqprio_rl *rl; + bool nch_changed; + int err; - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + err = mlx5e_mqprio_channel_validate(priv, mqprio); if (err) - goto out; + return err; - priv->max_opened_tc = max_t(u8, priv->max_opened_tc, - new_channels.params.num_tc); -out: - mutex_unlock(&priv->state_lock); - return err; -} + rl = mlx5e_mqprio_rl_create(priv->mdev, mqprio->qopt.num_tc, mqprio->max_rate); + if (IS_ERR(rl)) + return PTR_ERR(rl); -#ifdef CONFIG_MLX5_ESWITCH -static int mlx5e_setup_tc_cls_flower(struct mlx5e_priv *priv, - struct flow_cls_offload *cls_flower, - unsigned long flags) -{ - switch (cls_flower->command) { - case FLOW_CLS_REPLACE: - return mlx5e_configure_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_DESTROY: - return mlx5e_delete_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_STATS: - return mlx5e_stats_flower(priv->netdev, priv, cls_flower, - flags); - default: - return -EOPNOTSUPP; + new_params = priv->channels.params; + mlx5e_params_mqprio_channel_set(&new_params, mqprio, rl); + + nch_changed = mlx5e_get_dcb_num_tc(&priv->channels.params) > 1; + preactivate = nch_changed ? mlx5e_num_channels_changed_ctx : + mlx5e_update_netdev_queues_ctx; + err = mlx5e_safe_switch_params(priv, &new_params, preactivate, NULL, true); + if (err) { + if (rl) { + mlx5e_mqprio_rl_cleanup(rl); + mlx5e_mqprio_rl_free(rl); + } + return err; + } + + if (priv->mqprio_rl) { + mlx5e_mqprio_rl_cleanup(priv->mqprio_rl); + mlx5e_mqprio_rl_free(priv->mqprio_rl); } + priv->mqprio_rl = rl; + + return 0; } -static int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, - void *cb_priv) +static int mlx5e_setup_tc_mqprio(struct mlx5e_priv *priv, + struct tc_mqprio_qopt_offload *mqprio) { - unsigned long flags = MLX5_TC_FLAG(INGRESS) | MLX5_TC_FLAG(NIC_OFFLOAD); - struct mlx5e_priv *priv = cb_priv; + /* MQPRIO is another toplevel qdisc that can't be attached + * simultaneously with the offloaded HTB. + */ + if (WARN_ON(mlx5e_selq_is_htb_enabled(&priv->selq))) + return -EINVAL; - switch (type) { - case TC_SETUP_CLSFLOWER: - return mlx5e_setup_tc_cls_flower(priv, type_data, flags); + switch (mqprio->mode) { + case TC_MQPRIO_MODE_DCB: + return mlx5e_setup_tc_mqprio_dcb(priv, &mqprio->qopt); + case TC_MQPRIO_MODE_CHANNEL: + return mlx5e_setup_tc_mqprio_channel(priv, mqprio); default: return -EOPNOTSUPP; } } -#endif static LIST_HEAD(mlx5e_block_cb_list); @@ -3483,9 +3584,17 @@ static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, void *type_data) { struct mlx5e_priv *priv = netdev_priv(dev); + bool tc_unbind = false; + int err; + + if (type == TC_SETUP_BLOCK && + ((struct flow_block_offload *)type_data)->command == FLOW_BLOCK_UNBIND) + tc_unbind = true; + + if (!netif_device_present(dev) && !tc_unbind) + return -ENODEV; switch (type) { -#ifdef CONFIG_MLX5_ESWITCH case TC_SETUP_BLOCK: { struct flow_block_offload *f = type_data; @@ -3495,9 +3604,16 @@ static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, mlx5e_setup_tc_block_cb, priv, priv, true); } -#endif case TC_SETUP_QDISC_MQPRIO: - return mlx5e_setup_tc_mqprio(priv, type_data); + mutex_lock(&priv->state_lock); + err = mlx5e_setup_tc_mqprio(priv, type_data); + mutex_unlock(&priv->state_lock); + return err; + case TC_SETUP_QDISC_HTB: + mutex_lock(&priv->state_lock); + err = mlx5e_htb_setup_tc(priv, type_data); + mutex_unlock(&priv->state_lock); + return err; default: return -EOPNOTSUPP; } @@ -3507,14 +3623,15 @@ void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s) { int i; - for (i = 0; i < priv->max_nch; i++) { - struct mlx5e_channel_stats *channel_stats = &priv->channel_stats[i]; + for (i = 0; i < priv->stats_nch; i++) { + struct mlx5e_channel_stats *channel_stats = priv->channel_stats[i]; struct mlx5e_rq_stats *xskrq_stats = &channel_stats->xskrq; struct mlx5e_rq_stats *rq_stats = &channel_stats->rq; int j; s->rx_packets += rq_stats->packets + xskrq_stats->packets; s->rx_bytes += rq_stats->bytes + xskrq_stats->bytes; + s->multicast += rq_stats->mcast_packets + xskrq_stats->mcast_packets; for (j = 0; j < priv->max_opened_tc; j++) { struct mlx5e_sq_stats *sq_stats = &channel_stats->sq[j]; @@ -3524,25 +3641,55 @@ void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s) s->tx_dropped += sq_stats->dropped; } } + if (priv->tx_ptp_opened) { + for (i = 0; i < priv->max_opened_tc; i++) { + struct mlx5e_sq_stats *sq_stats = &priv->ptp_stats.sq[i]; + + s->tx_packets += sq_stats->packets; + s->tx_bytes += sq_stats->bytes; + s->tx_dropped += sq_stats->dropped; + } + } + if (priv->rx_ptp_opened) { + struct mlx5e_rq_stats *rq_stats = &priv->ptp_stats.rq; + + s->rx_packets += rq_stats->packets; + s->rx_bytes += rq_stats->bytes; + s->multicast += rq_stats->mcast_packets; + } } void mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) { struct mlx5e_priv *priv = netdev_priv(dev); - struct mlx5e_vport_stats *vstats = &priv->stats.vport; struct mlx5e_pport_stats *pstats = &priv->stats.pport; - if (!mlx5e_monitor_counter_supported(priv)) { + if (!netif_device_present(dev)) + return; + + /* In switchdev mode, monitor counters doesn't monitor + * rx/tx stats of 802_3. The update stats mechanism + * should keep the 802_3 layout counters updated + */ + if (!mlx5e_monitor_counter_supported(priv) || + mlx5e_is_uplink_rep(priv)) { /* update HW stats in background for next time */ mlx5e_queue_update_stats(priv); } if (mlx5e_is_uplink_rep(priv)) { + struct mlx5e_vport_stats *vstats = &priv->stats.vport; + stats->rx_packets = PPORT_802_3_GET(pstats, a_frames_received_ok); stats->rx_bytes = PPORT_802_3_GET(pstats, a_octets_received_ok); stats->tx_packets = PPORT_802_3_GET(pstats, a_frames_transmitted_ok); stats->tx_bytes = PPORT_802_3_GET(pstats, a_octets_transmitted_ok); + + /* vport multicast also counts packets that are dropped due to steering + * or rx out of buffer + */ + stats->multicast = VPORT_COUNTER_GET(vstats, received_eth_multicast.packets); } else { mlx5e_fold_sw_stats64(priv, stats); } @@ -3552,7 +3699,8 @@ mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) stats->rx_length_errors = PPORT_802_3_GET(pstats, a_in_range_length_errors) + PPORT_802_3_GET(pstats, a_out_of_range_length_field) + - PPORT_802_3_GET(pstats, a_frame_too_long_errors); + PPORT_802_3_GET(pstats, a_frame_too_long_errors) + + VNIC_ENV_GET(&priv->stats.vnic, eth_wqe_too_small); stats->rx_crc_errors = PPORT_802_3_GET(pstats, a_frame_check_sequence_errors); stats->rx_frame_errors = PPORT_802_3_GET(pstats, a_alignment_errors); @@ -3560,19 +3708,21 @@ mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) stats->rx_errors = stats->rx_length_errors + stats->rx_crc_errors + stats->rx_frame_errors; stats->tx_errors = stats->tx_aborted_errors + stats->tx_carrier_errors; +} - /* vport multicast also counts packets that are dropped due to steering - * or rx out of buffer - */ - stats->multicast = - VPORT_COUNTER_GET(vstats, received_eth_multicast.packets); +static void mlx5e_nic_set_rx_mode(struct mlx5e_priv *priv) +{ + if (mlx5e_is_uplink_rep(priv)) + return; /* no rx mode for uplink rep */ + + queue_work(priv->wq, &priv->set_rx_mode_work); } static void mlx5e_set_rx_mode(struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); - queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_nic_set_rx_mode(priv); } static int mlx5e_set_mac(struct net_device *netdev, void *addr) @@ -3584,10 +3734,10 @@ static int mlx5e_set_mac(struct net_device *netdev, void *addr) return -EADDRNOTAVAIL; netif_addr_lock_bh(netdev); - ether_addr_copy(netdev->dev_addr, saddr->sa_data); + eth_hw_addr_set(netdev, saddr->sa_data); netif_addr_unlock_bh(netdev); - queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_nic_set_rx_mode(priv); return 0; } @@ -3606,45 +3756,62 @@ static int set_feature_lro(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5e_channels new_channels = {}; - struct mlx5e_params *old_params; + struct mlx5e_params *cur_params; + struct mlx5e_params new_params; + bool reset = true; int err = 0; - bool reset; mutex_lock(&priv->state_lock); - if (enable && priv->xsk.refcnt) { - netdev_warn(netdev, "LRO is incompatible with AF_XDP (%hu XSKs are active)\n", - priv->xsk.refcnt); - err = -EINVAL; - goto out; - } + cur_params = &priv->channels.params; + new_params = *cur_params; - old_params = &priv->channels.params; - if (enable && !MLX5E_GET_PFLAG(old_params, MLX5E_PFLAG_RX_STRIDING_RQ)) { - netdev_warn(netdev, "can't set LRO with legacy RQ\n"); - err = -EINVAL; + if (enable) + new_params.packet_merge.type = MLX5E_PACKET_MERGE_LRO; + else if (new_params.packet_merge.type == MLX5E_PACKET_MERGE_LRO) + new_params.packet_merge.type = MLX5E_PACKET_MERGE_NONE; + else goto out; - } - reset = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (!(cur_params->packet_merge.type == MLX5E_PACKET_MERGE_SHAMPO && + new_params.packet_merge.type == MLX5E_PACKET_MERGE_LRO)) { + if (cur_params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { + if (mlx5e_rx_mpwqe_is_linear_skb(mdev, cur_params, NULL) == + mlx5e_rx_mpwqe_is_linear_skb(mdev, &new_params, NULL)) + reset = false; + } + } - new_channels.params = *old_params; - new_channels.params.lro_en = enable; + err = mlx5e_safe_switch_params(priv, &new_params, + mlx5e_modify_tirs_packet_merge_ctx, NULL, reset); +out: + mutex_unlock(&priv->state_lock); + return err; +} - if (old_params->rq_wq_type != MLX5_WQ_TYPE_CYCLIC) { - if (mlx5e_rx_mpwqe_is_linear_skb(mdev, old_params, NULL) == - mlx5e_rx_mpwqe_is_linear_skb(mdev, &new_channels.params, NULL)) - reset = false; - } +static int set_feature_hw_gro(struct net_device *netdev, bool enable) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_params new_params; + bool reset = true; + int err = 0; - if (!reset) { - *old_params = new_channels.params; - err = mlx5e_modify_tirs_lro(priv); + mutex_lock(&priv->state_lock); + new_params = priv->channels.params; + + if (enable) { + new_params.packet_merge.type = MLX5E_PACKET_MERGE_SHAMPO; + new_params.packet_merge.shampo.match_criteria_type = + MLX5_RQC_SHAMPO_MATCH_CRITERIA_TYPE_EXTENDED; + new_params.packet_merge.shampo.alignment_granularity = + MLX5_RQC_SHAMPO_NO_MATCH_ALIGNMENT_GRANULARITY_STRIDE; + } else if (new_params.packet_merge.type == MLX5E_PACKET_MERGE_SHAMPO) { + new_params.packet_merge.type = MLX5E_PACKET_MERGE_NONE; + } else { goto out; } - err = mlx5e_safe_switch_channels(priv, &new_channels, mlx5e_modify_tirs_lro); + err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, reset); out: mutex_unlock(&priv->state_lock); return err; @@ -3655,27 +3822,39 @@ static int set_feature_cvlan_filter(struct net_device *netdev, bool enable) struct mlx5e_priv *priv = netdev_priv(netdev); if (enable) - mlx5e_enable_cvlan_filter(priv); + mlx5e_enable_cvlan_filter(priv->fs, + !!(priv->netdev->flags & IFF_PROMISC)); else - mlx5e_disable_cvlan_filter(priv); + mlx5e_disable_cvlan_filter(priv->fs, + !!(priv->netdev->flags & IFF_PROMISC)); return 0; } -#ifdef CONFIG_MLX5_ESWITCH -static int set_feature_tc_num_filters(struct net_device *netdev, bool enable) +static int set_feature_hw_tc(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); + int err = 0; - if (!enable && mlx5e_tc_num_filters(priv, MLX5_TC_FLAG(NIC_OFFLOAD))) { +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + int tc_flag = mlx5e_is_uplink_rep(priv) ? MLX5_TC_FLAG(ESW_OFFLOAD) : + MLX5_TC_FLAG(NIC_OFFLOAD); + if (!enable && mlx5e_tc_num_filters(priv, tc_flag)) { netdev_err(netdev, "Active offloaded tc filters, can't turn hw_tc_offload off\n"); return -EINVAL; } +#endif - return 0; + mutex_lock(&priv->state_lock); + if (!enable && mlx5e_selq_is_htb_enabled(&priv->selq)) { + netdev_err(netdev, "Active HTB offload, can't turn hw_tc_offload off\n"); + err = -EINVAL; + } + mutex_unlock(&priv->state_lock); + + return err; } -#endif static int set_feature_rx_all(struct net_device *netdev, bool enable) { @@ -3685,20 +3864,67 @@ static int set_feature_rx_all(struct net_device *netdev, bool enable) return mlx5_set_port_fcs(mdev, !enable); } +static int mlx5e_set_rx_port_ts(struct mlx5_core_dev *mdev, bool enable) +{ + u32 in[MLX5_ST_SZ_DW(pcmr_reg)] = {}; + bool supported, curr_state; + int err; + + if (!MLX5_CAP_GEN(mdev, ports_check)) + return 0; + + err = mlx5_query_ports_check(mdev, in, sizeof(in)); + if (err) + return err; + + supported = MLX5_GET(pcmr_reg, in, rx_ts_over_crc_cap); + curr_state = MLX5_GET(pcmr_reg, in, rx_ts_over_crc); + + if (!supported || enable == curr_state) + return 0; + + MLX5_SET(pcmr_reg, in, local_port, 1); + MLX5_SET(pcmr_reg, in, rx_ts_over_crc, enable); + + return mlx5_set_ports_check(mdev, in, sizeof(in)); +} + static int set_feature_rx_fcs(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_channels *chs = &priv->channels; + struct mlx5_core_dev *mdev = priv->mdev; int err; mutex_lock(&priv->state_lock); - priv->channels.params.scatter_fcs_en = enable; - err = mlx5e_modify_channels_scatter_fcs(&priv->channels, enable); - if (err) - priv->channels.params.scatter_fcs_en = !enable; + if (enable) { + err = mlx5e_set_rx_port_ts(mdev, false); + if (err) + goto out; - mutex_unlock(&priv->state_lock); + chs->params.scatter_fcs_en = true; + err = mlx5e_modify_channels_scatter_fcs(chs, true); + if (err) { + chs->params.scatter_fcs_en = false; + mlx5e_set_rx_port_ts(mdev, true); + } + } else { + chs->params.scatter_fcs_en = false; + err = mlx5e_modify_channels_scatter_fcs(chs, false); + if (err) { + chs->params.scatter_fcs_en = true; + goto out; + } + err = mlx5e_set_rx_port_ts(mdev, true); + if (err) { + mlx5_core_warn(mdev, "Failed to set RX port timestamp %d\n", err); + err = 0; + } + } +out: + mutex_unlock(&priv->state_lock); return err; } @@ -3709,20 +3935,45 @@ static int set_feature_rx_vlan(struct net_device *netdev, bool enable) mutex_lock(&priv->state_lock); + mlx5e_fs_set_vlan_strip_disable(priv->fs, !enable); priv->channels.params.vlan_strip_disable = !enable; + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) goto unlock; err = mlx5e_modify_channels_vsd(&priv->channels, !enable); - if (err) + if (err) { + mlx5e_fs_set_vlan_strip_disable(priv->fs, enable); priv->channels.params.vlan_strip_disable = enable; - + } unlock: mutex_unlock(&priv->state_lock); return err; } +int mlx5e_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_flow_steering *fs = priv->fs; + + if (mlx5e_is_uplink_rep(priv)) + return 0; /* no vlan table for uplink rep */ + + return mlx5e_fs_vlan_rx_add_vid(fs, dev, proto, vid); +} + +int mlx5e_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_flow_steering *fs = priv->fs; + + if (mlx5e_is_uplink_rep(priv)) + return 0; /* no vlan table for uplink rep */ + + return mlx5e_fs_vlan_rx_kill_vid(fs, dev, proto, vid); +} + #ifdef CONFIG_MLX5_EN_ARFS static int set_feature_arfs(struct net_device *netdev, bool enable) { @@ -3730,9 +3981,9 @@ static int set_feature_arfs(struct net_device *netdev, bool enable) int err; if (enable) - err = mlx5e_arfs_enable(priv); + err = mlx5e_arfs_enable(priv->fs); else - err = mlx5e_arfs_disable(priv); + err = mlx5e_arfs_disable(priv->fs); return err; } @@ -3740,12 +3991,11 @@ static int set_feature_arfs(struct net_device *netdev, bool enable) static int mlx5e_handle_feature(struct net_device *netdev, netdev_features_t *features, - netdev_features_t wanted_features, netdev_features_t feature, mlx5e_feature_handler feature_handler) { - netdev_features_t changes = wanted_features ^ netdev->features; - bool enable = !!(wanted_features & feature); + netdev_features_t changes = *features ^ netdev->features; + bool enable = !!(*features & feature); int err; if (!(changes & feature)) @@ -3753,35 +4003,35 @@ static int mlx5e_handle_feature(struct net_device *netdev, err = feature_handler(netdev, enable); if (err) { + MLX5E_SET_FEATURE(features, feature, !enable); netdev_err(netdev, "%s feature %pNF failed, err %d\n", enable ? "Enable" : "Disable", &feature, err); return err; } - MLX5E_SET_FEATURE(features, feature, enable); return 0; } int mlx5e_set_features(struct net_device *netdev, netdev_features_t features) { - netdev_features_t oper_features = netdev->features; + netdev_features_t oper_features = features; int err = 0; #define MLX5E_HANDLE_FEATURE(feature, handler) \ - mlx5e_handle_feature(netdev, &oper_features, features, feature, handler) + mlx5e_handle_feature(netdev, &oper_features, feature, handler) err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_GRO_HW, set_feature_hw_gro); err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER, set_feature_cvlan_filter); -#ifdef CONFIG_MLX5_ESWITCH - err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_tc_num_filters); -#endif + err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_hw_tc); err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all); err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs); err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan); #ifdef CONFIG_MLX5_EN_ARFS err |= MLX5E_HANDLE_FEATURE(NETIF_F_NTUPLE, set_feature_arfs); #endif + err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TLS_RX, mlx5e_ktls_set_feature_rx); if (err) { netdev->features = oper_features; @@ -3791,15 +4041,40 @@ int mlx5e_set_features(struct net_device *netdev, netdev_features_t features) return 0; } +static netdev_features_t mlx5e_fix_uplink_rep_features(struct net_device *netdev, + netdev_features_t features) +{ + features &= ~NETIF_F_HW_TLS_RX; + if (netdev->features & NETIF_F_HW_TLS_RX) + netdev_warn(netdev, "Disabling hw_tls_rx, not supported in switchdev mode\n"); + + features &= ~NETIF_F_HW_TLS_TX; + if (netdev->features & NETIF_F_HW_TLS_TX) + netdev_warn(netdev, "Disabling hw_tls_tx, not supported in switchdev mode\n"); + + features &= ~NETIF_F_NTUPLE; + if (netdev->features & NETIF_F_NTUPLE) + netdev_warn(netdev, "Disabling ntuple, not supported in switchdev mode\n"); + + features &= ~NETIF_F_GRO_HW; + if (netdev->features & NETIF_F_GRO_HW) + netdev_warn(netdev, "Disabling HW_GRO, not supported in switchdev mode\n"); + + return features; +} + static netdev_features_t mlx5e_fix_features(struct net_device *netdev, netdev_features_t features) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_vlan_table *vlan; struct mlx5e_params *params; + vlan = mlx5e_fs_get_vlan(priv->fs); mutex_lock(&priv->state_lock); params = &priv->channels.params; - if (!bitmap_empty(priv->fs.vlan.active_svlans, VLAN_N_VID)) { + if (!vlan || + !bitmap_empty(mlx5e_vlan_get_active_svlans(vlan), VLAN_N_VID)) { /* HW strips the outer C-tag header, this is a problem * for S-tag traffic. */ @@ -3807,19 +4082,56 @@ static netdev_features_t mlx5e_fix_features(struct net_device *netdev, if (!params->vlan_strip_disable) netdev_warn(netdev, "Dropping C-tag vlan stripping offload due to S-tag vlan\n"); } + if (!MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ)) { if (features & NETIF_F_LRO) { netdev_warn(netdev, "Disabling LRO, not supported in legacy RQ\n"); features &= ~NETIF_F_LRO; } + if (features & NETIF_F_GRO_HW) { + netdev_warn(netdev, "Disabling HW-GRO, not supported in legacy RQ\n"); + features &= ~NETIF_F_GRO_HW; + } + } + + if (params->xdp_prog) { + if (features & NETIF_F_LRO) { + netdev_warn(netdev, "LRO is incompatible with XDP\n"); + features &= ~NETIF_F_LRO; + } + if (features & NETIF_F_GRO_HW) { + netdev_warn(netdev, "HW GRO is incompatible with XDP\n"); + features &= ~NETIF_F_GRO_HW; + } + } + + if (priv->xsk.refcnt) { + if (features & NETIF_F_LRO) { + netdev_warn(netdev, "LRO is incompatible with AF_XDP (%u XSKs are active)\n", + priv->xsk.refcnt); + features &= ~NETIF_F_LRO; + } + if (features & NETIF_F_GRO_HW) { + netdev_warn(netdev, "HW GRO is incompatible with AF_XDP (%u XSKs are active)\n", + priv->xsk.refcnt); + features &= ~NETIF_F_GRO_HW; + } } if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) { features &= ~NETIF_F_RXHASH; if (netdev->features & NETIF_F_RXHASH) netdev_warn(netdev, "Disabling rxhash, not supported when CQE compress is active\n"); + + if (features & NETIF_F_GRO_HW) { + netdev_warn(netdev, "Disabling HW-GRO, not supported when CQE compress is active\n"); + features &= ~NETIF_F_GRO_HW; + } } + if (mlx5e_is_uplink_rep(priv)) + features = mlx5e_fix_uplink_rep_features(netdev, features); + mutex_unlock(&priv->state_lock); return features; @@ -3833,13 +4145,14 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev, u16 ix; for (ix = 0; ix < chs->params.num_channels; ix++) { - struct xdp_umem *umem = mlx5e_xsk_get_umem(&chs->params, chs->params.xsk, ix); + struct xsk_buff_pool *xsk_pool = + mlx5e_xsk_get_pool(&chs->params, chs->params.xsk, ix); struct mlx5e_xsk_param xsk; - if (!umem) + if (!xsk_pool) continue; - mlx5e_build_xsk_param(umem, &xsk); + mlx5e_build_xsk_param(xsk_pool, &xsk); if (!mlx5e_validate_xsk_param(new_params, &xsk, mdev)) { u32 hr = mlx5e_get_linear_rq_headroom(new_params, &xsk); @@ -3850,10 +4163,10 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev, * 2. Size of SKBs allocated on XDP_PASS <= PAGE_SIZE. */ max_mtu_frame = MLX5E_HW2SW_MTU(new_params, xsk.chunk_size - hr); - max_mtu_page = mlx5e_xdp_max_mtu(new_params, &xsk); + max_mtu_page = MLX5E_HW2SW_MTU(new_params, SKB_MAX_HEAD(0)); max_mtu = min(max_mtu_frame, max_mtu_page); - netdev_err(netdev, "MTU %d is too big for an XSK running on channel %hu. Try MTU <= %d\n", + netdev_err(netdev, "MTU %d is too big for an XSK running on channel %u. Try MTU <= %d\n", new_params->sw_mtu, ix, max_mtu); return false; } @@ -3862,81 +4175,145 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev, return true; } +static bool mlx5e_params_validate_xdp(struct net_device *netdev, + struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + bool is_linear; + + /* No XSK params: AF_XDP can't be enabled yet at the point of setting + * the XDP program. + */ + is_linear = mlx5e_rx_is_linear_skb(mdev, params, NULL); + + if (!is_linear && params->rq_wq_type != MLX5_WQ_TYPE_CYCLIC) { + netdev_warn(netdev, "XDP is not allowed with striding RQ and MTU(%d) > %d\n", + params->sw_mtu, + mlx5e_xdp_max_mtu(params, NULL)); + return false; + } + if (!is_linear && !params->xdp_prog->aux->xdp_has_frags) { + netdev_warn(netdev, "MTU(%d) > %d, too big for an XDP program not aware of multi buffer\n", + params->sw_mtu, + mlx5e_xdp_max_mtu(params, NULL)); + return false; + } + + return true; +} + int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, - change_hw_mtu_cb set_mtu_cb) + mlx5e_fp_preactivate preactivate) { struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5e_channels new_channels = {}; + struct mlx5e_params new_params; struct mlx5e_params *params; + bool reset = true; int err = 0; - bool reset; mutex_lock(&priv->state_lock); params = &priv->channels.params; - reset = !params->lro_en; - reset = reset && test_bit(MLX5E_STATE_OPENED, &priv->state); - - new_channels.params = *params; - new_channels.params.sw_mtu = new_mtu; + new_params = *params; + new_params.sw_mtu = new_mtu; + err = mlx5e_validate_params(priv->mdev, &new_params); + if (err) + goto out; - if (params->xdp_prog && - !mlx5e_rx_is_linear_skb(&new_channels.params, NULL)) { - netdev_err(netdev, "MTU(%d) > %d is not allowed while XDP enabled\n", - new_mtu, mlx5e_xdp_max_mtu(params, NULL)); + if (new_params.xdp_prog && !mlx5e_params_validate_xdp(netdev, priv->mdev, + &new_params)) { err = -EINVAL; goto out; } if (priv->xsk.refcnt && !mlx5e_xsk_validate_mtu(netdev, &priv->channels, - &new_channels.params, priv->mdev)) { + &new_params, priv->mdev)) { err = -EINVAL; goto out; } - if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { - bool is_linear = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev, - &new_channels.params, - NULL); - u8 ppw_old = mlx5e_mpwqe_log_pkts_per_wqe(params, NULL); - u8 ppw_new = mlx5e_mpwqe_log_pkts_per_wqe(&new_channels.params, NULL); - - /* If XSK is active, XSK RQs are linear. */ - is_linear |= priv->xsk.refcnt; - - /* Always reset in linear mode - hw_mtu is used in data path. */ - reset = reset && (is_linear || (ppw_old != ppw_new)); - } - - if (!reset) { - params->sw_mtu = new_mtu; - if (set_mtu_cb) - set_mtu_cb(priv); - netdev->mtu = params->sw_mtu; - goto out; + if (params->packet_merge.type == MLX5E_PACKET_MERGE_LRO) + reset = false; + + if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ && + params->packet_merge.type != MLX5E_PACKET_MERGE_SHAMPO) { + bool is_linear_old = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev, params, NULL); + bool is_linear_new = mlx5e_rx_mpwqe_is_linear_skb(priv->mdev, + &new_params, NULL); + u8 sz_old = mlx5e_mpwqe_get_log_rq_size(priv->mdev, params, NULL); + u8 sz_new = mlx5e_mpwqe_get_log_rq_size(priv->mdev, &new_params, NULL); + + /* Always reset in linear mode - hw_mtu is used in data path. + * Check that the mode was non-linear and didn't change. + * If XSK is active, XSK RQs are linear. + * Reset if the RQ size changed, even if it's non-linear. + */ + if (!is_linear_old && !is_linear_new && !priv->xsk.refcnt && + sz_old == sz_new) + reset = false; } - err = mlx5e_safe_switch_channels(priv, &new_channels, set_mtu_cb); - if (err) - goto out; - - netdev->mtu = new_channels.params.sw_mtu; + err = mlx5e_safe_switch_params(priv, &new_params, preactivate, NULL, reset); out: + netdev->mtu = params->sw_mtu; mutex_unlock(&priv->state_lock); return err; } static int mlx5e_change_nic_mtu(struct net_device *netdev, int new_mtu) { - return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu); + return mlx5e_change_mtu(netdev, new_mtu, mlx5e_set_dev_port_mtu_ctx); +} + +int mlx5e_ptp_rx_manage_fs_ctx(struct mlx5e_priv *priv, void *ctx) +{ + bool set = *(bool *)ctx; + + return mlx5e_ptp_rx_manage_fs(priv, set); +} + +static int mlx5e_hwstamp_config_no_ptp_rx(struct mlx5e_priv *priv, bool rx_filter) +{ + bool rx_cqe_compress_def = priv->channels.params.rx_cqe_compress_def; + int err; + + if (!rx_filter) + /* Reset CQE compression to Admin default */ + return mlx5e_modify_rx_cqe_compression_locked(priv, rx_cqe_compress_def, false); + + if (!MLX5E_GET_PFLAG(&priv->channels.params, MLX5E_PFLAG_RX_CQE_COMPRESS)) + return 0; + + /* Disable CQE compression */ + netdev_warn(priv->netdev, "Disabling RX cqe compression\n"); + err = mlx5e_modify_rx_cqe_compression_locked(priv, false, true); + if (err) + netdev_err(priv->netdev, "Failed disabling cqe compression err=%d\n", err); + + return err; +} + +static int mlx5e_hwstamp_config_ptp_rx(struct mlx5e_priv *priv, bool ptp_rx) +{ + struct mlx5e_params new_params; + + if (ptp_rx == priv->channels.params.ptp_rx) + return 0; + + new_params = priv->channels.params; + new_params.ptp_rx = ptp_rx; + return mlx5e_safe_switch_params(priv, &new_params, mlx5e_ptp_rx_manage_fs_ctx, + &new_params.ptp_rx, true); } int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) { struct hwtstamp_config config; + bool rx_cqe_compress_def; + bool ptp_rx; int err; if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz) || @@ -3956,11 +4333,12 @@ int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) } mutex_lock(&priv->state_lock); + rx_cqe_compress_def = priv->channels.params.rx_cqe_compress_def; + /* RX HW timestamp */ switch (config.rx_filter) { case HWTSTAMP_FILTER_NONE: - /* Reset CQE compression to Admin default */ - mlx5e_modify_rx_cqe_compression_locked(priv, priv->channels.params.rx_cqe_compress_def); + ptp_rx = false; break; case HWTSTAMP_FILTER_ALL: case HWTSTAMP_FILTER_SOME: @@ -3977,22 +4355,25 @@ int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - /* Disable CQE compression */ - if (MLX5E_GET_PFLAG(&priv->channels.params, MLX5E_PFLAG_RX_CQE_COMPRESS)) - netdev_warn(priv->netdev, "Disabling RX cqe compression\n"); - err = mlx5e_modify_rx_cqe_compression_locked(priv, false); - if (err) { - netdev_err(priv->netdev, "Failed disabling cqe compression err=%d\n", err); - mutex_unlock(&priv->state_lock); - return err; - } config.rx_filter = HWTSTAMP_FILTER_ALL; + /* ptp_rx is set if both HW TS is set and CQE + * compression is set + */ + ptp_rx = rx_cqe_compress_def; break; default: - mutex_unlock(&priv->state_lock); - return -ERANGE; + err = -ERANGE; + goto err_unlock; } + if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) + err = mlx5e_hwstamp_config_no_ptp_rx(priv, + config.rx_filter != HWTSTAMP_FILTER_NONE); + else + err = mlx5e_hwstamp_config_ptp_rx(priv, ptp_rx); + if (err) + goto err_unlock; + memcpy(&priv->tstamp, &config, sizeof(config)); mutex_unlock(&priv->state_lock); @@ -4001,6 +4382,9 @@ int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; +err_unlock: + mutex_unlock(&priv->state_lock); + return err; } int mlx5e_hwstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr) @@ -4103,6 +4487,9 @@ static int mlx5e_set_vf_link_state(struct net_device *dev, int vf, struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5_core_dev *mdev = priv->mdev; + if (mlx5e_is_uplink_rep(priv)) + return -EOPNOTSUPP; + return mlx5_eswitch_set_vport_state(mdev->priv.eswitch, vf + 1, mlx5_ifla_link2vport(link_state)); } @@ -4114,6 +4501,9 @@ int mlx5e_get_vf_config(struct net_device *dev, struct mlx5_core_dev *mdev = priv->mdev; int err; + if (!netif_device_present(dev)) + return -EOPNOTSUPP; + err = mlx5_eswitch_get_vport_config(mdev->priv.eswitch, vf + 1, ivi); if (err) return err; @@ -4130,83 +4520,61 @@ int mlx5e_get_vf_stats(struct net_device *dev, return mlx5_eswitch_get_vport_stats(mdev->priv.eswitch, vf + 1, vf_stats); } -#endif - -struct mlx5e_vxlan_work { - struct work_struct work; - struct mlx5e_priv *priv; - u16 port; -}; -static void mlx5e_vxlan_add_work(struct work_struct *work) +static bool +mlx5e_has_offload_stats(const struct net_device *dev, int attr_id) { - struct mlx5e_vxlan_work *vxlan_work = - container_of(work, struct mlx5e_vxlan_work, work); - struct mlx5e_priv *priv = vxlan_work->priv; - u16 port = vxlan_work->port; - - mutex_lock(&priv->state_lock); - mlx5_vxlan_add_port(priv->mdev->vxlan, port); - mutex_unlock(&priv->state_lock); + struct mlx5e_priv *priv = netdev_priv(dev); - kfree(vxlan_work); -} + if (!netif_device_present(dev)) + return false; -static void mlx5e_vxlan_del_work(struct work_struct *work) -{ - struct mlx5e_vxlan_work *vxlan_work = - container_of(work, struct mlx5e_vxlan_work, work); - struct mlx5e_priv *priv = vxlan_work->priv; - u16 port = vxlan_work->port; + if (!mlx5e_is_uplink_rep(priv)) + return false; - mutex_lock(&priv->state_lock); - mlx5_vxlan_del_port(priv->mdev->vxlan, port); - mutex_unlock(&priv->state_lock); - kfree(vxlan_work); + return mlx5e_rep_has_offload_stats(dev, attr_id); } -static void mlx5e_vxlan_queue_work(struct mlx5e_priv *priv, u16 port, int add) +static int +mlx5e_get_offload_stats(int attr_id, const struct net_device *dev, + void *sp) { - struct mlx5e_vxlan_work *vxlan_work; - - vxlan_work = kmalloc(sizeof(*vxlan_work), GFP_ATOMIC); - if (!vxlan_work) - return; + struct mlx5e_priv *priv = netdev_priv(dev); - if (add) - INIT_WORK(&vxlan_work->work, mlx5e_vxlan_add_work); - else - INIT_WORK(&vxlan_work->work, mlx5e_vxlan_del_work); + if (!mlx5e_is_uplink_rep(priv)) + return -EOPNOTSUPP; - vxlan_work->priv = priv; - vxlan_work->port = port; - queue_work(priv->wq, &vxlan_work->work); + return mlx5e_rep_get_offload_stats(attr_id, dev, sp); } +#endif -void mlx5e_add_vxlan_port(struct net_device *netdev, struct udp_tunnel_info *ti) +static bool mlx5e_tunnel_proto_supported_tx(struct mlx5_core_dev *mdev, u8 proto_type) { - struct mlx5e_priv *priv = netdev_priv(netdev); - - if (ti->type != UDP_TUNNEL_TYPE_VXLAN) - return; - - if (!mlx5_vxlan_allowed(priv->mdev->vxlan)) - return; - - mlx5e_vxlan_queue_work(priv, be16_to_cpu(ti->port), 1); + switch (proto_type) { + case IPPROTO_GRE: + return MLX5_CAP_ETH(mdev, tunnel_stateless_gre); + case IPPROTO_IPIP: + case IPPROTO_IPV6: + return (MLX5_CAP_ETH(mdev, tunnel_stateless_ip_over_ip) || + MLX5_CAP_ETH(mdev, tunnel_stateless_ip_over_ip_tx)); + default: + return false; + } } -void mlx5e_del_vxlan_port(struct net_device *netdev, struct udp_tunnel_info *ti) +static bool mlx5e_gre_tunnel_inner_proto_offload_supported(struct mlx5_core_dev *mdev, + struct sk_buff *skb) { - struct mlx5e_priv *priv = netdev_priv(netdev); - - if (ti->type != UDP_TUNNEL_TYPE_VXLAN) - return; - - if (!mlx5_vxlan_allowed(priv->mdev->vxlan)) - return; - - mlx5e_vxlan_queue_work(priv, be16_to_cpu(ti->port), 0); + switch (skb->inner_protocol) { + case htons(ETH_P_IP): + case htons(ETH_P_IPV6): + case htons(ETH_P_TEB): + return true; + case htons(ETH_P_MPLS_UC): + case htons(ETH_P_MPLS_MC): + return MLX5_CAP_ETH(mdev, tunnel_stateless_mpls_over_gre); + } + return false; } static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv, @@ -4231,10 +4599,12 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv, switch (proto) { case IPPROTO_GRE: - return features; + if (mlx5e_gre_tunnel_inner_proto_offload_supported(priv->mdev, skb)) + return features; + break; case IPPROTO_IPIP: case IPPROTO_IPV6: - if (mlx5e_tunnel_proto_supported(priv->mdev, IPPROTO_IPIP)) + if (mlx5e_tunnel_proto_supported_tx(priv->mdev, IPPROTO_IPIP)) return features; break; case IPPROTO_UDP: @@ -4250,6 +4620,11 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv, if (port == GENEVE_UDP_PORT && mlx5_geneve_tx_allowed(priv->mdev)) return features; #endif + break; +#ifdef CONFIG_MLX5_EN_IPSEC + case IPPROTO_ESP: + return mlx5e_ipsec_feature_check(skb, features); +#endif } out: @@ -4266,11 +4641,6 @@ netdev_features_t mlx5e_features_check(struct sk_buff *skb, features = vlan_features_check(skb, features); features = vxlan_features_check(skb, features); -#ifdef CONFIG_MLX5_EN_IPSEC - if (mlx5e_ipsec_feature_check(skb, netdev, features)) - return features; -#endif - /* Validate if the tunneled packet is being offloaded by HW */ if (skb->encapsulation && (features & NETIF_F_CSUM_MASK || features & NETIF_F_GSO_MASK)) @@ -4283,8 +4653,7 @@ static void mlx5e_tx_timeout_work(struct work_struct *work) { struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, tx_timeout_work); - bool report_failed = false; - int err; + struct net_device *netdev = priv->netdev; int i; rtnl_lock(); @@ -4293,27 +4662,19 @@ static void mlx5e_tx_timeout_work(struct work_struct *work) if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) goto unlock; - for (i = 0; i < priv->channels.num * priv->channels.params.num_tc; i++) { + for (i = 0; i < netdev->real_num_tx_queues; i++) { struct netdev_queue *dev_queue = - netdev_get_tx_queue(priv->netdev, i); + netdev_get_tx_queue(netdev, i); struct mlx5e_txqsq *sq = priv->txq2sq[i]; if (!netif_xmit_stopped(dev_queue)) continue; if (mlx5e_reporter_tx_timeout(sq)) - report_failed = true; + /* break if tried to reopened channels */ + break; } - if (!report_failed) - goto unlock; - - err = mlx5e_safe_reopen_channels(priv); - if (err) - netdev_err(priv->netdev, - "mlx5e_safe_reopen_channels failed recovering from a tx_timeout, err(%d).\n", - err); - unlock: mutex_unlock(&priv->state_lock); rtnl_unlock(); @@ -4330,40 +4691,39 @@ static void mlx5e_tx_timeout(struct net_device *dev, unsigned int txqueue) static int mlx5e_xdp_allowed(struct mlx5e_priv *priv, struct bpf_prog *prog) { struct net_device *netdev = priv->netdev; - struct mlx5e_channels new_channels = {}; + struct mlx5e_params new_params; - if (priv->channels.params.lro_en) { - netdev_warn(netdev, "can't set XDP while LRO is on, disable LRO first\n"); + if (priv->channels.params.packet_merge.type != MLX5E_PACKET_MERGE_NONE) { + netdev_warn(netdev, "can't set XDP while HW-GRO/LRO is on, disable them first\n"); return -EINVAL; } - if (MLX5_IPSEC_DEV(priv->mdev)) { - netdev_warn(netdev, "can't set XDP with IPSec offload\n"); - return -EINVAL; - } - - new_channels.params = priv->channels.params; - new_channels.params.xdp_prog = prog; + new_params = priv->channels.params; + new_params.xdp_prog = prog; - /* No XSK params: AF_XDP can't be enabled yet at the point of setting - * the XDP program. - */ - if (!mlx5e_rx_is_linear_skb(&new_channels.params, NULL)) { - netdev_warn(netdev, "XDP is not allowed with MTU(%d) > %d\n", - new_channels.params.sw_mtu, - mlx5e_xdp_max_mtu(&new_channels.params, NULL)); + if (!mlx5e_params_validate_xdp(netdev, priv->mdev, &new_params)) return -EINVAL; - } return 0; } +static void mlx5e_rq_replace_xdp_prog(struct mlx5e_rq *rq, struct bpf_prog *prog) +{ + struct bpf_prog *old_prog; + + old_prog = rcu_replace_pointer(rq->xdp_prog, prog, + lockdep_is_held(&rq->priv->state_lock)); + if (old_prog) + bpf_prog_put(old_prog); +} + static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_params new_params; struct bpf_prog *old_prog; - bool reset, was_opened; int err = 0; + bool reset; int i; mutex_lock(&priv->state_lock); @@ -4374,91 +4734,59 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog) goto unlock; } - was_opened = test_bit(MLX5E_STATE_OPENED, &priv->state); /* no need for full reset when exchanging programs */ reset = (!priv->channels.params.xdp_prog || !prog); - if (was_opened && !reset) - /* num_channels is invariant here, so we can take the - * batched reference right upfront. - */ - bpf_prog_add(prog, priv->channels.num); - - if (was_opened && reset) { - struct mlx5e_channels new_channels = {}; + new_params = priv->channels.params; + new_params.xdp_prog = prog; - new_channels.params = priv->channels.params; - new_channels.params.xdp_prog = prog; - mlx5e_set_rq_type(priv->mdev, &new_channels.params); - old_prog = priv->channels.params.xdp_prog; - - err = mlx5e_safe_switch_channels(priv, &new_channels, NULL); + /* XDP affects striding RQ parameters. Block XDP if striding RQ won't be + * supported with the new parameters: if PAGE_SIZE is bigger than + * MLX5_MPWQE_LOG_STRIDE_SZ_MAX, striding RQ can't be used, even though + * the MTU is small enough for the linear mode, because XDP uses strides + * of PAGE_SIZE on regular RQs. + */ + if (reset && MLX5E_GET_PFLAG(&new_params, MLX5E_PFLAG_RX_STRIDING_RQ)) { + /* Checking for regular RQs here; XSK RQs were checked on XSK bind. */ + err = mlx5e_mpwrq_validate_regular(priv->mdev, &new_params); if (err) goto unlock; - } else { - /* exchange programs, extra prog reference we got from caller - * as long as we don't fail from this point onwards. - */ - old_prog = xchg(&priv->channels.params.xdp_prog, prog); } + old_prog = priv->channels.params.xdp_prog; + + err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, reset); + if (err) + goto unlock; + if (old_prog) bpf_prog_put(old_prog); - if (!was_opened && reset) /* change RQ type according to priv->xdp_prog */ - mlx5e_set_rq_type(priv->mdev, &priv->channels.params); - - if (!was_opened || reset) + if (!test_bit(MLX5E_STATE_OPENED, &priv->state) || reset) goto unlock; /* exchanging programs w/o reset, we update ref counts on behalf * of the channels RQs here. */ + bpf_prog_add(prog, priv->channels.num); for (i = 0; i < priv->channels.num; i++) { struct mlx5e_channel *c = priv->channels.c[i]; - bool xsk_open = test_bit(MLX5E_CHANNEL_STATE_XSK, c->state); - clear_bit(MLX5E_RQ_STATE_ENABLED, &c->rq.state); - if (xsk_open) - clear_bit(MLX5E_RQ_STATE_ENABLED, &c->xskrq.state); - napi_synchronize(&c->napi); - /* prevent mlx5e_poll_rx_cq from accessing rq->xdp_prog */ - - old_prog = xchg(&c->rq.xdp_prog, prog); - if (old_prog) - bpf_prog_put(old_prog); - - if (xsk_open) { - old_prog = xchg(&c->xskrq.xdp_prog, prog); - if (old_prog) - bpf_prog_put(old_prog); + mlx5e_rq_replace_xdp_prog(&c->rq, prog); + if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) { + bpf_prog_inc(prog); + mlx5e_rq_replace_xdp_prog(&c->xskrq, prog); } - - set_bit(MLX5E_RQ_STATE_ENABLED, &c->rq.state); - if (xsk_open) - set_bit(MLX5E_RQ_STATE_ENABLED, &c->xskrq.state); - /* napi_schedule in case we have missed anything */ - napi_schedule(&c->napi); } unlock: mutex_unlock(&priv->state_lock); - return err; -} - -static u32 mlx5e_xdp_query(struct net_device *dev) -{ - struct mlx5e_priv *priv = netdev_priv(dev); - const struct bpf_prog *xdp_prog; - u32 prog_id = 0; - mutex_lock(&priv->state_lock); - xdp_prog = priv->channels.params.xdp_prog; - if (xdp_prog) - prog_id = xdp_prog->aux->id; - mutex_unlock(&priv->state_lock); + /* Need to fix some features. */ + if (!err) + netdev_update_features(netdev); - return prog_id; + return err; } static int mlx5e_xdp(struct net_device *dev, struct netdev_bpf *xdp) @@ -4466,11 +4794,8 @@ static int mlx5e_xdp(struct net_device *dev, struct netdev_bpf *xdp) switch (xdp->command) { case XDP_SETUP_PROG: return mlx5e_xdp_set(dev, xdp->prog); - case XDP_QUERY_PROG: - xdp->prog_id = mlx5e_xdp_query(dev); - return 0; - case XDP_SETUP_XSK_UMEM: - return mlx5e_xsk_setup_umem(dev, xdp->xsk.umem, + case XDP_SETUP_XSK_POOL: + return mlx5e_xsk_setup_pool(dev, xdp->xsk.pool, xdp->xsk.queue_id); default: return -EINVAL; @@ -4546,10 +4871,8 @@ const struct net_device_ops mlx5e_netdev_ops = { .ndo_set_features = mlx5e_set_features, .ndo_fix_features = mlx5e_fix_features, .ndo_change_mtu = mlx5e_change_nic_mtu, - .ndo_do_ioctl = mlx5e_ioctl, + .ndo_eth_ioctl = mlx5e_ioctl, .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, - .ndo_udp_tunnel_add = mlx5e_add_vxlan_port, - .ndo_udp_tunnel_del = mlx5e_del_vxlan_port, .ndo_features_check = mlx5e_features_check, .ndo_tx_timeout = mlx5e_tx_timeout, .ndo_bpf = mlx5e_xdp, @@ -4571,122 +4894,12 @@ const struct net_device_ops mlx5e_netdev_ops = { .ndo_get_vf_config = mlx5e_get_vf_config, .ndo_set_vf_link_state = mlx5e_set_vf_link_state, .ndo_get_vf_stats = mlx5e_get_vf_stats, + .ndo_has_offload_stats = mlx5e_has_offload_stats, + .ndo_get_offload_stats = mlx5e_get_offload_stats, #endif + .ndo_get_devlink_port = mlx5e_get_devlink_port, }; -static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) -{ - if (MLX5_CAP_GEN(mdev, port_type) != MLX5_CAP_PORT_TYPE_ETH) - return -EOPNOTSUPP; - if (!MLX5_CAP_GEN(mdev, eth_net_offloads) || - !MLX5_CAP_GEN(mdev, nic_flow_table) || - !MLX5_CAP_ETH(mdev, csum_cap) || - !MLX5_CAP_ETH(mdev, max_lso_cap) || - !MLX5_CAP_ETH(mdev, vlan_cap) || - !MLX5_CAP_ETH(mdev, rss_ind_tbl_cap) || - MLX5_CAP_FLOWTABLE(mdev, - flow_table_properties_nic_receive.max_ft_level) - < 3) { - mlx5_core_warn(mdev, - "Not creating net device, some required device capabilities are missing\n"); - return -EOPNOTSUPP; - } - if (!MLX5_CAP_ETH(mdev, self_lb_en_modifiable)) - mlx5_core_warn(mdev, "Self loop back prevention is not supported\n"); - if (!MLX5_CAP_GEN(mdev, cq_moderation)) - mlx5_core_warn(mdev, "CQ moderation is not supported\n"); - - return 0; -} - -void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len, - int num_channels) -{ - int i; - - for (i = 0; i < len; i++) - indirection_rqt[i] = i % num_channels; -} - -static bool slow_pci_heuristic(struct mlx5_core_dev *mdev) -{ - u32 link_speed = 0; - u32 pci_bw = 0; - - mlx5e_port_max_linkspeed(mdev, &link_speed); - pci_bw = pcie_bandwidth_available(mdev->pdev, NULL, NULL, NULL); - mlx5_core_dbg_once(mdev, "Max link speed = %d, PCI BW = %d\n", - link_speed, pci_bw); - -#define MLX5E_SLOW_PCI_RATIO (2) - - return link_speed && pci_bw && - link_speed > MLX5E_SLOW_PCI_RATIO * pci_bw; -} - -static struct dim_cq_moder mlx5e_get_def_tx_moderation(u8 cq_period_mode) -{ - struct dim_cq_moder moder; - - moder.cq_period_mode = cq_period_mode; - moder.pkts = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; - moder.usec = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC; - if (cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) - moder.usec = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC_FROM_CQE; - - return moder; -} - -static struct dim_cq_moder mlx5e_get_def_rx_moderation(u8 cq_period_mode) -{ - struct dim_cq_moder moder; - - moder.cq_period_mode = cq_period_mode; - moder.pkts = MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; - moder.usec = MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; - if (cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) - moder.usec = MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; - - return moder; -} - -static u8 mlx5_to_net_dim_cq_period_mode(u8 cq_period_mode) -{ - return cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE ? - DIM_CQ_PERIOD_MODE_START_FROM_CQE : - DIM_CQ_PERIOD_MODE_START_FROM_EQE; -} - -void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) -{ - if (params->tx_dim_enabled) { - u8 dim_period_mode = mlx5_to_net_dim_cq_period_mode(cq_period_mode); - - params->tx_cq_moderation = net_dim_get_def_tx_moderation(dim_period_mode); - } else { - params->tx_cq_moderation = mlx5e_get_def_tx_moderation(cq_period_mode); - } - - MLX5E_SET_PFLAG(params, MLX5E_PFLAG_TX_CQE_BASED_MODER, - params->tx_cq_moderation.cq_period_mode == - MLX5_CQ_PERIOD_MODE_START_FROM_CQE); -} - -void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) -{ - if (params->rx_dim_enabled) { - u8 dim_period_mode = mlx5_to_net_dim_cq_period_mode(cq_period_mode); - - params->rx_cq_moderation = net_dim_get_def_rx_moderation(dim_period_mode); - } else { - params->rx_cq_moderation = mlx5e_get_def_rx_moderation(cq_period_mode); - } - - MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_CQE_BASED_MODER, - params->rx_cq_moderation.cq_period_mode == - MLX5_CQ_PERIOD_MODE_START_FROM_CQE); -} - static u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout) { int i; @@ -4699,46 +4912,9 @@ static u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeo return MLX5_CAP_ETH(mdev, lro_timer_supported_periods[i]); } -void mlx5e_build_rq_params(struct mlx5_core_dev *mdev, - struct mlx5e_params *params) -{ - /* Prefer Striding RQ, unless any of the following holds: - * - Striding RQ configuration is not possible/supported. - * - Slow PCI heuristic. - * - Legacy RQ would use linear SKB while Striding RQ would use non-linear. - * - * No XSK params: checking the availability of striding RQ in general. - */ - if (!slow_pci_heuristic(mdev) && - mlx5e_striding_rq_possible(mdev, params) && - (mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL) || - !mlx5e_rx_is_linear_skb(params, NULL))) - MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ, true); - mlx5e_set_rq_type(mdev, params); - mlx5e_init_rq_type_params(mdev, params); -} - -void mlx5e_build_rss_params(struct mlx5e_rss_params *rss_params, - u16 num_channels) -{ - enum mlx5e_traffic_types tt; - - rss_params->hfunc = ETH_RSS_HASH_TOP; - netdev_rss_key_fill(rss_params->toeplitz_hash_key, - sizeof(rss_params->toeplitz_hash_key)); - mlx5e_build_default_indir_rqt(rss_params->indirection_rqt, - MLX5E_INDIR_RQT_SIZE, num_channels); - for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) - rss_params->rx_hash_fields[tt] = - tirc_default_config[tt].rx_hash_fields; -} - -void mlx5e_build_nic_params(struct mlx5e_priv *priv, - struct mlx5e_xsk *xsk, - struct mlx5e_rss_params *rss_params, - struct mlx5e_params *params, - u16 mtu) +void mlx5e_build_nic_params(struct mlx5e_priv *priv, struct mlx5e_xsk *xsk, u16 mtu) { + struct mlx5e_params *params = &priv->channels.params; struct mlx5_core_dev *mdev = priv->mdev; u8 rx_cq_period_mode; @@ -4746,16 +4922,16 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, params->hard_mtu = MLX5E_ETH_HARD_MTU; params->num_channels = min_t(unsigned int, MLX5E_MAX_NUM_CHANNELS / 2, priv->max_nch); - params->num_tc = 1; + mlx5e_params_mqprio_reset(params); /* SQ */ params->log_sq_size = is_kdump_kernel() ? MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE : MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE; + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_SKB_TX_MPWQE, mlx5e_tx_mpwqe_supported(mdev)); /* XDP SQ */ - MLX5E_SET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE, - MLX5_CAP_ETH(mdev, enhanced_multi_pkt_send_wqe)); + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE, mlx5e_tx_mpwqe_supported(mdev)); /* set CQE compression */ params->rx_cqe_compress_def = false; @@ -4769,15 +4945,7 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, /* RQ */ mlx5e_build_rq_params(mdev, params); - /* HW LRO */ - - /* TODO: && MLX5_CAP_ETH(mdev, lro_cap) */ - if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { - /* No XSK params: checking the availability of striding RQ in general. */ - if (!mlx5e_rx_mpwqe_is_linear_skb(mdev, params, NULL)) - params->lro_en = !slow_pci_heuristic(mdev); - } - params->lro_timeout = mlx5e_choose_lro_timeout(mdev, MLX5E_DEFAULT_LRO_TIMEOUT); + params->packet_merge.timeout = mlx5e_choose_lro_timeout(mdev, MLX5E_DEFAULT_LRO_TIMEOUT); /* CQ moderation params */ rx_cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? @@ -4791,25 +4959,75 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, /* TX inline */ mlx5_query_min_inline(mdev, ¶ms->tx_min_inline_mode); - /* RSS */ - mlx5e_build_rss_params(rss_params, params->num_channels); - params->tunneled_offload_en = - mlx5e_tunnel_inner_ft_supported(mdev); + params->tunneled_offload_en = mlx5_tunnel_inner_ft_supported(mdev); /* AF_XDP */ params->xsk = xsk; + + /* Do not update netdev->features directly in here + * on mlx5e_attach_netdev() we will call mlx5e_update_features() + * To update netdev->features please modify mlx5e_fix_features() + */ } static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + u8 addr[ETH_ALEN]; - mlx5_query_mac_address(priv->mdev, netdev->dev_addr); - if (is_zero_ether_addr(netdev->dev_addr) && + mlx5_query_mac_address(priv->mdev, addr); + if (is_zero_ether_addr(addr) && !MLX5_CAP_GEN(priv->mdev, vport_group_manager)) { eth_hw_addr_random(netdev); mlx5_core_info(priv->mdev, "Assigned random MAC address %pM\n", netdev->dev_addr); + return; } + + eth_hw_addr_set(netdev, addr); +} + +static int mlx5e_vxlan_set_port(struct net_device *netdev, unsigned int table, + unsigned int entry, struct udp_tunnel_info *ti) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + return mlx5_vxlan_add_port(priv->mdev->vxlan, ntohs(ti->port)); +} + +static int mlx5e_vxlan_unset_port(struct net_device *netdev, unsigned int table, + unsigned int entry, struct udp_tunnel_info *ti) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + return mlx5_vxlan_del_port(priv->mdev->vxlan, ntohs(ti->port)); +} + +void mlx5e_vxlan_set_netdev_info(struct mlx5e_priv *priv) +{ + if (!mlx5_vxlan_allowed(priv->mdev->vxlan)) + return; + + priv->nic_info.set_port = mlx5e_vxlan_set_port; + priv->nic_info.unset_port = mlx5e_vxlan_unset_port; + priv->nic_info.flags = UDP_TUNNEL_NIC_INFO_MAY_SLEEP | + UDP_TUNNEL_NIC_INFO_STATIC_IANA_VXLAN; + priv->nic_info.tables[0].tunnel_types = UDP_TUNNEL_TYPE_VXLAN; + /* Don't count the space hard-coded to the IANA port */ + priv->nic_info.tables[0].n_entries = + mlx5_vxlan_max_udp_ports(priv->mdev) - 1; + + priv->netdev->udp_tunnel_nic_info = &priv->nic_info; +} + +static bool mlx5e_tunnel_any_tx_proto_supported(struct mlx5_core_dev *mdev) +{ + int tt; + + for (tt = 0; tt < MLX5_NUM_TUNNEL_TT; tt++) { + if (mlx5e_tunnel_proto_supported_tx(mdev, mlx5_get_proto_by_tunnel_type(tt))) + return true; + } + return (mlx5_vxlan_allowed(mdev->vxlan) || mlx5_geneve_tx_allowed(mdev)); } static void mlx5e_build_nic_netdev(struct net_device *netdev) @@ -4823,10 +5041,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->netdev_ops = &mlx5e_netdev_ops; -#ifdef CONFIG_MLX5_CORE_EN_DCB - if (MLX5_CAP_GEN(mdev, vport_group_manager) && MLX5_CAP_GEN(mdev, qos)) - netdev->dcbnl_ops = &mlx5e_dcbnl_ops; -#endif + mlx5e_dcbnl_build_netdev(netdev); netdev->watchdog_timeo = 15 * HZ; @@ -4839,6 +5054,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->vlan_features |= NETIF_F_TSO6; netdev->vlan_features |= NETIF_F_RXCSUM; netdev->vlan_features |= NETIF_F_RXHASH; + netdev->vlan_features |= NETIF_F_GSO_PARTIAL; netdev->mpls_features |= NETIF_F_SG; netdev->mpls_features |= NETIF_F_HW_CSUM; @@ -4848,8 +5064,16 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_TX; netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_RX; + /* Tunneled LRO is not supported in the driver, and the same RQs are + * shared between inner and outer TIRs, so the driver can't disable LRO + * for inner TIRs while having it enabled for outer TIRs. Due to this, + * block LRO altogether if the firmware declares tunneled LRO support. + */ if (!!MLX5_CAP_ETH(mdev, lro_cap) && - mlx5e_check_fragmented_striding_rq_cap(mdev)) + !MLX5_CAP_ETH(mdev, tunnel_lro_vxlan) && + !MLX5_CAP_ETH(mdev, tunnel_lro_gre) && + mlx5e_check_fragmented_striding_rq_cap(mdev, PAGE_SHIFT, + MLX5E_MPWRQ_UMR_MODE_ALIGNED)) netdev->vlan_features |= NETIF_F_LRO; netdev->hw_features = netdev->vlan_features; @@ -4858,8 +5082,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER; netdev->hw_features |= NETIF_F_HW_VLAN_STAG_TX; - if (mlx5_vxlan_allowed(mdev->vxlan) || mlx5_geneve_tx_allowed(mdev) || - mlx5e_any_tunnel_proto_supported(mdev)) { + if (mlx5e_tunnel_any_tx_proto_supported(mdev)) { netdev->hw_enc_features |= NETIF_F_HW_CSUM; netdev->hw_enc_features |= NETIF_F_TSO; netdev->hw_enc_features |= NETIF_F_TSO6; @@ -4876,7 +5099,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) NETIF_F_GSO_UDP_TUNNEL_CSUM; } - if (mlx5e_tunnel_proto_supported(mdev, IPPROTO_GRE)) { + if (mlx5e_tunnel_proto_supported_tx(mdev, IPPROTO_GRE)) { netdev->hw_features |= NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM; netdev->hw_enc_features |= NETIF_F_GSO_GRE | @@ -4885,7 +5108,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) NETIF_F_GSO_GRE_CSUM; } - if (mlx5e_tunnel_proto_supported(mdev, IPPROTO_IPIP)) { + if (mlx5e_tunnel_proto_supported_tx(mdev, IPPROTO_IPIP)) { netdev->hw_features |= NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_IPXIP6; netdev->hw_enc_features |= NETIF_F_GSO_IPXIP4 | @@ -4894,7 +5117,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) NETIF_F_GSO_IPXIP6; } - netdev->hw_features |= NETIF_F_GSO_PARTIAL; netdev->gso_partial_features |= NETIF_F_GSO_UDP_L4; netdev->hw_features |= NETIF_F_GSO_UDP_L4; netdev->features |= NETIF_F_GSO_UDP_L4; @@ -4907,26 +5129,24 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) if (MLX5_CAP_ETH(mdev, scatter_fcs)) netdev->hw_features |= NETIF_F_RXFCS; + if (mlx5_qos_is_supported(mdev)) + netdev->hw_features |= NETIF_F_HW_TC; + netdev->features = netdev->hw_features; - if (!priv->channels.params.lro_en) - netdev->features &= ~NETIF_F_LRO; + /* Defaults */ if (fcs_enabled) netdev->features &= ~NETIF_F_RXALL; - - if (!priv->channels.params.scatter_fcs_en) - netdev->features &= ~NETIF_F_RXFCS; - - /* prefere CQE compression over rxhash */ - if (MLX5E_GET_PFLAG(&priv->channels.params, MLX5E_PFLAG_RX_CQE_COMPRESS)) - netdev->features &= ~NETIF_F_RXHASH; + netdev->features &= ~NETIF_F_LRO; + netdev->features &= ~NETIF_F_GRO_HW; + netdev->features &= ~NETIF_F_RXFCS; #define FT_CAP(f) MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_receive.f) if (FT_CAP(flow_modify_en) && FT_CAP(modify_root) && FT_CAP(identified_miss_table_mode) && FT_CAP(flow_table_modify)) { -#ifdef CONFIG_MLX5_ESWITCH +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) netdev->hw_features |= NETIF_F_HW_TC; #endif #ifdef CONFIG_MLX5_EN_ARFS @@ -4939,81 +5159,102 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->priv_flags |= IFF_UNICAST_FLT; + netif_set_tso_max_size(netdev, GSO_MAX_SIZE); mlx5e_set_netdev_dev_addr(netdev); + mlx5e_macsec_build_netdev(priv); mlx5e_ipsec_build_netdev(priv); - mlx5e_tls_build_netdev(priv); + mlx5e_ktls_build_netdev(priv); } void mlx5e_create_q_counters(struct mlx5e_priv *priv) { + u32 out[MLX5_ST_SZ_DW(alloc_q_counter_out)] = {}; + u32 in[MLX5_ST_SZ_DW(alloc_q_counter_in)] = {}; struct mlx5_core_dev *mdev = priv->mdev; int err; - err = mlx5_core_alloc_q_counter(mdev, &priv->q_counter); - if (err) { - mlx5_core_warn(mdev, "alloc queue counter failed, %d\n", err); - priv->q_counter = 0; - } + MLX5_SET(alloc_q_counter_in, in, opcode, MLX5_CMD_OP_ALLOC_Q_COUNTER); + err = mlx5_cmd_exec_inout(mdev, alloc_q_counter, in, out); + if (!err) + priv->q_counter = + MLX5_GET(alloc_q_counter_out, out, counter_set_id); - err = mlx5_core_alloc_q_counter(mdev, &priv->drop_rq_q_counter); - if (err) { - mlx5_core_warn(mdev, "alloc drop RQ counter failed, %d\n", err); - priv->drop_rq_q_counter = 0; - } + err = mlx5_cmd_exec_inout(mdev, alloc_q_counter, in, out); + if (!err) + priv->drop_rq_q_counter = + MLX5_GET(alloc_q_counter_out, out, counter_set_id); } void mlx5e_destroy_q_counters(struct mlx5e_priv *priv) { - if (priv->q_counter) - mlx5_core_dealloc_q_counter(priv->mdev, priv->q_counter); + u32 in[MLX5_ST_SZ_DW(dealloc_q_counter_in)] = {}; + + MLX5_SET(dealloc_q_counter_in, in, opcode, + MLX5_CMD_OP_DEALLOC_Q_COUNTER); + if (priv->q_counter) { + MLX5_SET(dealloc_q_counter_in, in, counter_set_id, + priv->q_counter); + mlx5_cmd_exec_in(priv->mdev, dealloc_q_counter, in); + } - if (priv->drop_rq_q_counter) - mlx5_core_dealloc_q_counter(priv->mdev, priv->drop_rq_q_counter); + if (priv->drop_rq_q_counter) { + MLX5_SET(dealloc_q_counter_in, in, counter_set_id, + priv->drop_rq_q_counter); + mlx5_cmd_exec_in(priv->mdev, dealloc_q_counter, in); + } } static int mlx5e_nic_init(struct mlx5_core_dev *mdev, - struct net_device *netdev, - const struct mlx5e_profile *profile, - void *ppriv) + struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5e_rss_params *rss = &priv->rss_params; + struct mlx5e_flow_steering *fs; int err; - err = mlx5e_netdev_init(netdev, priv, mdev, profile, ppriv); - if (err) - return err; - - mlx5e_build_nic_params(priv, &priv->xsk, rss, &priv->channels.params, - netdev->mtu); + mlx5e_build_nic_params(priv, &priv->xsk, netdev->mtu); + mlx5e_vxlan_set_netdev_info(priv); mlx5e_timestamp_init(priv); + fs = mlx5e_fs_init(priv->profile, mdev, + !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); + if (!fs) { + err = -ENOMEM; + mlx5_core_err(mdev, "FS initialization failed, %d\n", err); + return err; + } + priv->fs = fs; + err = mlx5e_ipsec_init(priv); if (err) mlx5_core_err(mdev, "IPSec initialization failed, %d\n", err); - err = mlx5e_tls_init(priv); + + err = mlx5e_ktls_init(priv); if (err) mlx5_core_err(mdev, "TLS initialization failed, %d\n", err); - mlx5e_build_nic_netdev(netdev); - mlx5e_health_create_reporters(priv); + mlx5e_health_create_reporters(priv); return 0; } static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) { mlx5e_health_destroy_reporters(priv); - mlx5e_tls_cleanup(priv); + mlx5e_ktls_cleanup(priv); mlx5e_ipsec_cleanup(priv); - mlx5e_netdev_cleanup(priv->netdev, priv); + mlx5e_fs_cleanup(priv->fs); } static int mlx5e_init_nic_rx(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; + enum mlx5e_rx_res_features features; int err; + priv->rx_res = mlx5e_rx_res_alloc(); + if (!priv->rx_res) + return -ENOMEM; + mlx5e_create_q_counters(priv); err = mlx5e_open_drop_rq(priv, &priv->drop_rq); @@ -5022,75 +5263,81 @@ static int mlx5e_init_nic_rx(struct mlx5e_priv *priv) goto err_destroy_q_counters; } - err = mlx5e_create_indirect_rqt(priv); + features = MLX5E_RX_RES_FEATURE_PTP; + if (priv->channels.params.tunneled_offload_en) + features |= MLX5E_RX_RES_FEATURE_INNER_FT; + err = mlx5e_rx_res_init(priv->rx_res, priv->mdev, features, + priv->max_nch, priv->drop_rq.rqn, + &priv->channels.params.packet_merge, + priv->channels.params.num_channels); if (err) goto err_close_drop_rq; - err = mlx5e_create_direct_rqts(priv, priv->direct_tir); - if (err) - goto err_destroy_indirect_rqts; - - err = mlx5e_create_indirect_tirs(priv, true); - if (err) - goto err_destroy_direct_rqts; - - err = mlx5e_create_direct_tirs(priv, priv->direct_tir); - if (err) - goto err_destroy_indirect_tirs; - - err = mlx5e_create_direct_rqts(priv, priv->xsk_tir); - if (unlikely(err)) - goto err_destroy_direct_tirs; - - err = mlx5e_create_direct_tirs(priv, priv->xsk_tir); - if (unlikely(err)) - goto err_destroy_xsk_rqts; - - err = mlx5e_create_flow_steering(priv); + err = mlx5e_create_flow_steering(priv->fs, priv->rx_res, priv->profile, + priv->netdev); if (err) { mlx5_core_warn(mdev, "create flow steering failed, %d\n", err); - goto err_destroy_xsk_tirs; + goto err_destroy_rx_res; } err = mlx5e_tc_nic_init(priv); if (err) goto err_destroy_flow_steering; + err = mlx5e_accel_init_rx(priv); + if (err) + goto err_tc_nic_cleanup; + +#ifdef CONFIG_MLX5_EN_ARFS + priv->netdev->rx_cpu_rmap = mlx5_eq_table_get_rmap(priv->mdev); +#endif + return 0; +err_tc_nic_cleanup: + mlx5e_tc_nic_cleanup(priv); err_destroy_flow_steering: - mlx5e_destroy_flow_steering(priv); -err_destroy_xsk_tirs: - mlx5e_destroy_direct_tirs(priv, priv->xsk_tir); -err_destroy_xsk_rqts: - mlx5e_destroy_direct_rqts(priv, priv->xsk_tir); -err_destroy_direct_tirs: - mlx5e_destroy_direct_tirs(priv, priv->direct_tir); -err_destroy_indirect_tirs: - mlx5e_destroy_indirect_tirs(priv, true); -err_destroy_direct_rqts: - mlx5e_destroy_direct_rqts(priv, priv->direct_tir); -err_destroy_indirect_rqts: - mlx5e_destroy_rqt(priv, &priv->indir_rqt); + mlx5e_destroy_flow_steering(priv->fs, !!(priv->netdev->hw_features & NETIF_F_NTUPLE), + priv->profile); +err_destroy_rx_res: + mlx5e_rx_res_destroy(priv->rx_res); err_close_drop_rq: mlx5e_close_drop_rq(&priv->drop_rq); err_destroy_q_counters: mlx5e_destroy_q_counters(priv); + mlx5e_rx_res_free(priv->rx_res); + priv->rx_res = NULL; return err; } static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv) { + mlx5e_accel_cleanup_rx(priv); mlx5e_tc_nic_cleanup(priv); - mlx5e_destroy_flow_steering(priv); - mlx5e_destroy_direct_tirs(priv, priv->xsk_tir); - mlx5e_destroy_direct_rqts(priv, priv->xsk_tir); - mlx5e_destroy_direct_tirs(priv, priv->direct_tir); - mlx5e_destroy_indirect_tirs(priv, true); - mlx5e_destroy_direct_rqts(priv, priv->direct_tir); - mlx5e_destroy_rqt(priv, &priv->indir_rqt); + mlx5e_destroy_flow_steering(priv->fs, !!(priv->netdev->hw_features & NETIF_F_NTUPLE), + priv->profile); + mlx5e_rx_res_destroy(priv->rx_res); mlx5e_close_drop_rq(&priv->drop_rq); mlx5e_destroy_q_counters(priv); + mlx5e_rx_res_free(priv->rx_res); + priv->rx_res = NULL; +} + +static void mlx5e_set_mqprio_rl(struct mlx5e_priv *priv) +{ + struct mlx5e_params *params; + struct mlx5e_mqprio_rl *rl; + + params = &priv->channels.params; + if (params->mqprio.mode != TC_MQPRIO_MODE_CHANNEL) + return; + + rl = mlx5e_mqprio_rl_create(priv->mdev, params->mqprio.num_tc, + params->mqprio.channel.max_rate); + if (IS_ERR(rl)) + rl = NULL; + priv->mqprio_rl = rl; + mlx5e_mqprio_rl_update_params(params, rl); } static int mlx5e_init_nic_tx(struct mlx5e_priv *priv) @@ -5103,44 +5350,56 @@ static int mlx5e_init_nic_tx(struct mlx5e_priv *priv) return err; } -#ifdef CONFIG_MLX5_CORE_EN_DCB + err = mlx5e_accel_init_tx(priv); + if (err) + goto err_destroy_tises; + + mlx5e_set_mqprio_rl(priv); mlx5e_dcbnl_initialize(priv); -#endif return 0; + +err_destroy_tises: + mlx5e_destroy_tises(priv); + return err; } static void mlx5e_nic_enable(struct mlx5e_priv *priv) { struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; + int err; + + mlx5e_fs_init_l2_addr(priv->fs, netdev); - mlx5e_init_l2_addr(priv); + err = mlx5e_macsec_init(priv); + if (err) + mlx5_core_err(mdev, "MACsec initialization failed, %d\n", err); /* Marking the link as currently not needed by the Driver */ if (!netif_running(netdev)) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); + mlx5e_modify_admin_state(mdev, MLX5_PORT_DOWN); mlx5e_set_netdev_mtu_boundaries(priv); mlx5e_set_dev_port_mtu(priv); - mlx5_lag_add(mdev, netdev); + mlx5_lag_add_netdev(mdev, netdev); mlx5e_enable_async_events(priv); + mlx5e_enable_blocking_events(priv); if (mlx5e_monitor_counter_supported(priv)) mlx5e_monitor_counter_init(priv); mlx5e_hv_vhca_stats_create(priv); if (netdev->reg_state != NETREG_REGISTERED) return; -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_init_app(priv); -#endif - queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_nic_set_rx_mode(priv); rtnl_lock(); if (netif_running(netdev)) mlx5e_open(netdev); + udp_tunnel_nic_reset_ntf(priv->netdev); netif_device_attach(netdev); rtnl_unlock(); } @@ -5149,10 +5408,8 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; -#ifdef CONFIG_MLX5_CORE_EN_DCB if (priv->netdev->reg_state == NETREG_REGISTERED) mlx5e_dcbnl_delete_app(priv); -#endif rtnl_lock(); if (netif_running(priv->netdev)) @@ -5160,19 +5417,27 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) netif_device_detach(priv->netdev); rtnl_unlock(); - queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_nic_set_rx_mode(priv); mlx5e_hv_vhca_stats_destroy(priv); if (mlx5e_monitor_counter_supported(priv)) mlx5e_monitor_counter_cleanup(priv); + mlx5e_disable_blocking_events(priv); + if (priv->en_trap) { + mlx5e_deactivate_trap(priv); + mlx5e_close_trap(priv->en_trap); + priv->en_trap = NULL; + } mlx5e_disable_async_events(priv); - mlx5_lag_remove(mdev); + mlx5_lag_remove_netdev(mdev, priv->netdev); + mlx5_vxlan_reset_to_default(mdev->vxlan); + mlx5e_macsec_cleanup(priv); } int mlx5e_update_nic_rx(struct mlx5e_priv *priv) { - return mlx5e_refresh_tirs(priv, false); + return mlx5e_refresh_tirs(priv, false, false); } static const struct mlx5e_profile mlx5e_nic_profile = { @@ -5185,35 +5450,102 @@ static const struct mlx5e_profile mlx5e_nic_profile = { .enable = mlx5e_nic_enable, .disable = mlx5e_nic_disable, .update_rx = mlx5e_update_nic_rx, - .update_stats = mlx5e_update_ndo_stats, + .update_stats = mlx5e_stats_update_ndo_stats, .update_carrier = mlx5e_update_carrier, - .rx_handlers.handle_rx_cqe = mlx5e_handle_rx_cqe, - .rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq, + .rx_handlers = &mlx5e_rx_handlers_nic, .max_tc = MLX5E_MAX_NUM_TC, - .rq_groups = MLX5E_NUM_RQ_GROUPS(XSK), .stats_grps = mlx5e_nic_stats_grps, .stats_grps_num = mlx5e_nic_stats_grps_num, + .features = BIT(MLX5E_PROFILE_FEATURE_PTP_RX) | + BIT(MLX5E_PROFILE_FEATURE_PTP_TX) | + BIT(MLX5E_PROFILE_FEATURE_QOS_HTB) | + BIT(MLX5E_PROFILE_FEATURE_FS_VLAN) | + BIT(MLX5E_PROFILE_FEATURE_FS_TC), }; -/* mlx5e generic netdev management API (move to en_common.c) */ +static int mlx5e_profile_max_num_channels(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + int nch; + + nch = mlx5e_get_max_num_channels(mdev); + + if (profile->max_nch_limit) + nch = min_t(int, nch, profile->max_nch_limit(mdev)); + return nch; +} + +static unsigned int +mlx5e_calc_max_nch(struct mlx5_core_dev *mdev, struct net_device *netdev, + const struct mlx5e_profile *profile) -/* mlx5e_netdev_init/cleanup must be called from profile->init/cleanup callbacks */ -int mlx5e_netdev_init(struct net_device *netdev, - struct mlx5e_priv *priv, - struct mlx5_core_dev *mdev, - const struct mlx5e_profile *profile, - void *ppriv) { + unsigned int max_nch, tmp; + + /* core resources */ + max_nch = mlx5e_profile_max_num_channels(mdev, profile); + + /* netdev rx queues */ + max_nch = min_t(unsigned int, max_nch, netdev->num_rx_queues); + + /* netdev tx queues */ + tmp = netdev->num_tx_queues; + if (mlx5_qos_is_supported(mdev)) + tmp -= mlx5e_qos_max_leaf_nodes(mdev); + if (MLX5_CAP_GEN(mdev, ts_cqe_to_dest_cqn)) + tmp -= profile->max_tc; + tmp = tmp / profile->max_tc; + max_nch = min_t(unsigned int, max_nch, tmp); + + return max_nch; +} + +int mlx5e_get_pf_num_tirs(struct mlx5_core_dev *mdev) +{ + /* Indirect TIRS: 2 sets of TTCs (inner + outer steering) + * and 1 set of direct TIRS + */ + return 2 * MLX5E_NUM_INDIR_TIRS + + mlx5e_profile_max_num_channels(mdev, &mlx5e_nic_profile); +} + +void mlx5e_set_rx_mode_work(struct work_struct *work) +{ + struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, + set_rx_mode_work); + + return mlx5e_fs_set_rx_mode_work(priv->fs, priv->netdev); +} + +/* mlx5e generic netdev management API (move to en_common.c) */ +int mlx5e_priv_init(struct mlx5e_priv *priv, + const struct mlx5e_profile *profile, + struct net_device *netdev, + struct mlx5_core_dev *mdev) +{ + int nch, num_txqs, node; + int err; + + num_txqs = netdev->num_tx_queues; + nch = mlx5e_calc_max_nch(mdev, netdev, profile); + node = dev_to_node(mlx5_core_dma_dev(mdev)); + /* priv init */ priv->mdev = mdev; priv->netdev = netdev; - priv->profile = profile; - priv->ppriv = ppriv; priv->msglevel = MLX5E_MSG_LEVEL; - priv->max_nch = netdev->num_rx_queues / max_t(u8, profile->rq_groups, 1); + priv->max_nch = nch; priv->max_opened_tc = 1; + if (!alloc_cpumask_var(&priv->scratchpad.cpumask, GFP_KERNEL)) + return -ENOMEM; + mutex_init(&priv->state_lock); + + err = mlx5e_selq_init(&priv->selq, &priv->state_lock); + if (err) + goto err_free_cpumask; + INIT_WORK(&priv->update_carrier_work, mlx5e_update_carrier_work); INIT_WORK(&priv->set_rx_mode_work, mlx5e_set_rx_mode_work); INIT_WORK(&priv->tx_timeout_work, mlx5e_tx_timeout_work); @@ -5221,45 +5553,112 @@ int mlx5e_netdev_init(struct net_device *netdev, priv->wq = create_singlethread_workqueue("mlx5e"); if (!priv->wq) - return -ENOMEM; + goto err_free_selq; - /* netdev init */ - netif_carrier_off(netdev); + priv->txq2sq = kcalloc_node(num_txqs, sizeof(*priv->txq2sq), GFP_KERNEL, node); + if (!priv->txq2sq) + goto err_destroy_workqueue; -#ifdef CONFIG_MLX5_EN_ARFS - netdev->rx_cpu_rmap = mlx5_eq_table_get_rmap(mdev); -#endif + priv->tx_rates = kcalloc_node(num_txqs, sizeof(*priv->tx_rates), GFP_KERNEL, node); + if (!priv->tx_rates) + goto err_free_txq2sq; + + priv->channel_stats = + kcalloc_node(nch, sizeof(*priv->channel_stats), GFP_KERNEL, node); + if (!priv->channel_stats) + goto err_free_tx_rates; return 0; + +err_free_tx_rates: + kfree(priv->tx_rates); +err_free_txq2sq: + kfree(priv->txq2sq); +err_destroy_workqueue: + destroy_workqueue(priv->wq); +err_free_selq: + mlx5e_selq_cleanup(&priv->selq); +err_free_cpumask: + free_cpumask_var(priv->scratchpad.cpumask); + return -ENOMEM; } -void mlx5e_netdev_cleanup(struct net_device *netdev, struct mlx5e_priv *priv) +void mlx5e_priv_cleanup(struct mlx5e_priv *priv) { + int i; + + /* bail if change profile failed and also rollback failed */ + if (!priv->mdev) + return; + + for (i = 0; i < priv->stats_nch; i++) + kvfree(priv->channel_stats[i]); + kfree(priv->channel_stats); + kfree(priv->tx_rates); + kfree(priv->txq2sq); destroy_workqueue(priv->wq); + mutex_lock(&priv->state_lock); + mlx5e_selq_cleanup(&priv->selq); + mutex_unlock(&priv->state_lock); + free_cpumask_var(priv->scratchpad.cpumask); + + for (i = 0; i < priv->htb_max_qos_sqs; i++) + kfree(priv->htb_qos_sq_stats[i]); + kvfree(priv->htb_qos_sq_stats); + + memset(priv, 0, sizeof(*priv)); +} + +static unsigned int mlx5e_get_max_num_txqs(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + unsigned int nch, ptp_txqs, qos_txqs; + + nch = mlx5e_profile_max_num_channels(mdev, profile); + + ptp_txqs = MLX5_CAP_GEN(mdev, ts_cqe_to_dest_cqn) && + mlx5e_profile_feature_cap(profile, PTP_TX) ? + profile->max_tc : 0; + + qos_txqs = mlx5_qos_is_supported(mdev) && + mlx5e_profile_feature_cap(profile, QOS_HTB) ? + mlx5e_qos_max_leaf_nodes(mdev) : 0; + + return nch * profile->max_tc + ptp_txqs + qos_txqs; +} + +static unsigned int mlx5e_get_max_num_rxqs(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + return mlx5e_profile_max_num_channels(mdev, profile); } -struct net_device *mlx5e_create_netdev(struct mlx5_core_dev *mdev, - const struct mlx5e_profile *profile, - int nch, - void *ppriv) +struct net_device * +mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile) { struct net_device *netdev; + unsigned int txqs, rxqs; int err; - netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), - nch * profile->max_tc, - nch * profile->rq_groups); + txqs = mlx5e_get_max_num_txqs(mdev, profile); + rxqs = mlx5e_get_max_num_rxqs(mdev, profile); + + netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), txqs, rxqs); if (!netdev) { mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); return NULL; } - err = profile->init(mdev, netdev, profile, ppriv); + err = mlx5e_priv_init(netdev_priv(netdev), profile, netdev, mdev); if (err) { - mlx5_core_err(mdev, "failed to init mlx5e profile %d\n", err); + mlx5_core_err(mdev, "mlx5e_priv_init failed, err=%d\n", err); goto err_free_netdev; } + netif_carrier_off(netdev); + netif_tx_disable(netdev); + dev_net_set(netdev, mlx5_core_net(mdev)); + return netdev; err_free_netdev: @@ -5268,24 +5667,77 @@ err_free_netdev: return NULL; } +static void mlx5e_update_features(struct net_device *netdev) +{ + if (netdev->reg_state != NETREG_REGISTERED) + return; /* features will be updated on netdev registration */ + + rtnl_lock(); + netdev_update_features(netdev); + rtnl_unlock(); +} + +static void mlx5e_reset_channels(struct net_device *netdev) +{ + netdev_reset_tc(netdev); +} + int mlx5e_attach_netdev(struct mlx5e_priv *priv) { - const struct mlx5e_profile *profile; + const bool take_rtnl = priv->netdev->reg_state == NETREG_REGISTERED; + const struct mlx5e_profile *profile = priv->profile; int max_nch; int err; - profile = priv->profile; clear_bit(MLX5E_STATE_DESTROYING, &priv->state); + if (priv->fs) + mlx5e_fs_set_state_destroy(priv->fs, + !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); + + /* Validate the max_wqe_size_sq capability. */ + if (WARN_ON_ONCE(mlx5e_get_max_sq_wqebbs(priv->mdev) < MLX5E_MAX_TX_WQEBBS)) { + mlx5_core_warn(priv->mdev, "MLX5E: Max SQ WQEBBs firmware capability: %u, needed %lu\n", + mlx5e_get_max_sq_wqebbs(priv->mdev), MLX5E_MAX_TX_WQEBBS); + return -EIO; + } /* max number of channels may have changed */ - max_nch = mlx5e_get_max_num_channels(priv->mdev); + max_nch = mlx5e_calc_max_nch(priv->mdev, priv->netdev, profile); if (priv->channels.params.num_channels > max_nch) { mlx5_core_warn(priv->mdev, "MLX5E: Reducing number of channels to %d\n", max_nch); + /* Reducing the number of channels - RXFH has to be reset, and + * mlx5e_num_channels_changed below will build the RQT. + */ + priv->netdev->priv_flags &= ~IFF_RXFH_CONFIGURED; priv->channels.params.num_channels = max_nch; - mlx5e_build_default_indir_rqt(priv->rss_params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, max_nch); + if (priv->channels.params.mqprio.mode == TC_MQPRIO_MODE_CHANNEL) { + mlx5_core_warn(priv->mdev, "MLX5E: Disabling MQPRIO channel mode\n"); + mlx5e_params_mqprio_reset(&priv->channels.params); + } + } + if (max_nch != priv->max_nch) { + mlx5_core_warn(priv->mdev, + "MLX5E: Updating max number of channels from %u to %u\n", + priv->max_nch, max_nch); + priv->max_nch = max_nch; } + /* 1. Set the real number of queues in the kernel the first time. + * 2. Set our default XPS cpumask. + * 3. Build the RQT. + * + * rtnl_lock is required by netif_set_real_num_*_queues in case the + * netdev has been registered by this point (if this function was called + * in the reload or resume flow). + */ + if (take_rtnl) + rtnl_lock(); + err = mlx5e_num_channels_changed(priv); + if (take_rtnl) + rtnl_unlock(); + if (err) + goto out; + err = profile->init_tx(priv); if (err) goto out; @@ -5297,12 +5749,20 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) if (profile->enable) profile->enable(priv); + mlx5e_update_features(priv->netdev); + return 0; err_cleanup_tx: profile->cleanup_tx(priv); out: + mlx5e_reset_channels(priv->netdev); + set_bit(MLX5E_STATE_DESTROYING, &priv->state); + if (priv->fs) + mlx5e_fs_set_state_destroy(priv->fs, + !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); + cancel_work_sync(&priv->update_stats_work); return err; } @@ -5311,6 +5771,9 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv) const struct mlx5e_profile *profile = priv->profile; set_bit(MLX5E_STATE_DESTROYING, &priv->state); + if (priv->fs) + mlx5e_fs_set_state_destroy(priv->fs, + !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); if (profile->disable) profile->disable(priv); @@ -5318,26 +5781,89 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv) profile->cleanup_rx(priv); profile->cleanup_tx(priv); + mlx5e_reset_channels(priv->netdev); cancel_work_sync(&priv->update_stats_work); } +static int +mlx5e_netdev_attach_profile(struct net_device *netdev, struct mlx5_core_dev *mdev, + const struct mlx5e_profile *new_profile, void *new_ppriv) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + err = mlx5e_priv_init(priv, new_profile, netdev, mdev); + if (err) { + mlx5_core_err(mdev, "mlx5e_priv_init failed, err=%d\n", err); + return err; + } + netif_carrier_off(netdev); + priv->profile = new_profile; + priv->ppriv = new_ppriv; + err = new_profile->init(priv->mdev, priv->netdev); + if (err) + goto priv_cleanup; + err = mlx5e_attach_netdev(priv); + if (err) + goto profile_cleanup; + return err; + +profile_cleanup: + new_profile->cleanup(priv); +priv_cleanup: + mlx5e_priv_cleanup(priv); + return err; +} + +int mlx5e_netdev_change_profile(struct mlx5e_priv *priv, + const struct mlx5e_profile *new_profile, void *new_ppriv) +{ + const struct mlx5e_profile *orig_profile = priv->profile; + struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = priv->mdev; + void *orig_ppriv = priv->ppriv; + int err, rollback_err; + + /* cleanup old profile */ + mlx5e_detach_netdev(priv); + priv->profile->cleanup(priv); + mlx5e_priv_cleanup(priv); + + err = mlx5e_netdev_attach_profile(netdev, mdev, new_profile, new_ppriv); + if (err) { /* roll back to original profile */ + netdev_warn(netdev, "%s: new profile init failed, %d\n", __func__, err); + goto rollback; + } + + return 0; + +rollback: + rollback_err = mlx5e_netdev_attach_profile(netdev, mdev, orig_profile, orig_ppriv); + if (rollback_err) + netdev_err(netdev, "%s: failed to rollback to orig profile, %d\n", + __func__, rollback_err); + return err; +} + +void mlx5e_netdev_attach_nic_profile(struct mlx5e_priv *priv) +{ + mlx5e_netdev_change_profile(priv, &mlx5e_nic_profile, NULL); +} + void mlx5e_destroy_netdev(struct mlx5e_priv *priv) { - const struct mlx5e_profile *profile = priv->profile; struct net_device *netdev = priv->netdev; - if (profile->cleanup) - profile->cleanup(priv); + mlx5e_priv_cleanup(priv); free_netdev(netdev); } -/* mlx5e_attach and mlx5e_detach scope should be only creating/destroying - * hardware contexts and to connect it to the current netdev. - */ -static int mlx5e_attach(struct mlx5_core_dev *mdev, void *vpriv) +static int mlx5e_resume(struct auxiliary_device *adev) { - struct mlx5e_priv *priv = vpriv; + struct mlx5_adev *edev = container_of(adev, struct mlx5_adev, adev); + struct mlx5e_priv *priv = auxiliary_get_drvdata(adev); struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = edev->mdev; int err; if (netif_device_present(netdev)) @@ -5356,111 +5882,132 @@ static int mlx5e_attach(struct mlx5_core_dev *mdev, void *vpriv) return 0; } -static void mlx5e_detach(struct mlx5_core_dev *mdev, void *vpriv) +static int mlx5e_suspend(struct auxiliary_device *adev, pm_message_t state) { - struct mlx5e_priv *priv = vpriv; + struct mlx5e_priv *priv = auxiliary_get_drvdata(adev); struct net_device *netdev = priv->netdev; - -#ifdef CONFIG_MLX5_ESWITCH - if (MLX5_ESWITCH_MANAGER(mdev) && vpriv == mdev) - return; -#endif + struct mlx5_core_dev *mdev = priv->mdev; if (!netif_device_present(netdev)) - return; + return -ENODEV; mlx5e_detach_netdev(priv); mlx5e_destroy_mdev_resources(mdev); + return 0; } -static void *mlx5e_add(struct mlx5_core_dev *mdev) +static int mlx5e_probe(struct auxiliary_device *adev, + const struct auxiliary_device_id *id) { + struct mlx5_adev *edev = container_of(adev, struct mlx5_adev, adev); + const struct mlx5e_profile *profile = &mlx5e_nic_profile; + struct mlx5_core_dev *mdev = edev->mdev; struct net_device *netdev; - void *priv; + pm_message_t state = {}; + struct mlx5e_priv *priv; int err; - int nch; - err = mlx5e_check_required_hca_cap(mdev); - if (err) - return NULL; - -#ifdef CONFIG_MLX5_ESWITCH - if (MLX5_ESWITCH_MANAGER(mdev) && - mlx5_eswitch_mode(mdev->priv.eswitch) == MLX5_ESWITCH_OFFLOADS) { - mlx5e_rep_register_vport_reps(mdev); - return mdev; - } -#endif - - nch = mlx5e_get_max_num_channels(mdev); - netdev = mlx5e_create_netdev(mdev, &mlx5e_nic_profile, nch, NULL); + netdev = mlx5e_create_netdev(mdev, profile); if (!netdev) { mlx5_core_err(mdev, "mlx5e_create_netdev failed\n"); - return NULL; + return -ENOMEM; } - dev_net_set(netdev, mlx5_core_net(mdev)); + mlx5e_build_nic_netdev(netdev); + priv = netdev_priv(netdev); + auxiliary_set_drvdata(adev, priv); + + priv->profile = profile; + priv->ppriv = NULL; - err = mlx5e_attach(mdev, priv); + err = mlx5e_devlink_port_register(priv); if (err) { - mlx5_core_err(mdev, "mlx5e_attach failed, %d\n", err); + mlx5_core_err(mdev, "mlx5e_devlink_port_register failed, %d\n", err); goto err_destroy_netdev; } + err = profile->init(mdev, netdev); + if (err) { + mlx5_core_err(mdev, "mlx5e_nic_profile init failed, %d\n", err); + goto err_devlink_cleanup; + } + + err = mlx5e_resume(adev); + if (err) { + mlx5_core_err(mdev, "mlx5e_resume failed, %d\n", err); + goto err_profile_cleanup; + } + err = register_netdev(netdev); if (err) { mlx5_core_err(mdev, "register_netdev failed, %d\n", err); - goto err_detach; + goto err_resume; } -#ifdef CONFIG_MLX5_CORE_EN_DCB + mlx5e_devlink_port_type_eth_set(priv); + mlx5e_dcbnl_init_app(priv); -#endif - return priv; + mlx5_uplink_netdev_set(mdev, netdev); + return 0; -err_detach: - mlx5e_detach(mdev, priv); +err_resume: + mlx5e_suspend(adev, state); +err_profile_cleanup: + profile->cleanup(priv); +err_devlink_cleanup: + mlx5e_devlink_port_unregister(priv); err_destroy_netdev: mlx5e_destroy_netdev(priv); - return NULL; + return err; } -static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) +static void mlx5e_remove(struct auxiliary_device *adev) { - struct mlx5e_priv *priv; + struct mlx5e_priv *priv = auxiliary_get_drvdata(adev); + pm_message_t state = {}; -#ifdef CONFIG_MLX5_ESWITCH - if (MLX5_ESWITCH_MANAGER(mdev) && vpriv == mdev) { - mlx5e_rep_unregister_vport_reps(mdev); - return; - } -#endif - priv = vpriv; -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_delete_app(priv); -#endif unregister_netdev(priv->netdev); - mlx5e_detach(mdev, vpriv); + mlx5e_suspend(adev, state); + priv->profile->cleanup(priv); + mlx5e_devlink_port_unregister(priv); mlx5e_destroy_netdev(priv); } -static struct mlx5_interface mlx5e_interface = { - .add = mlx5e_add, - .remove = mlx5e_remove, - .attach = mlx5e_attach, - .detach = mlx5e_detach, - .protocol = MLX5_INTERFACE_PROTOCOL_ETH, +static const struct auxiliary_device_id mlx5e_id_table[] = { + { .name = MLX5_ADEV_NAME ".eth", }, + {}, +}; + +MODULE_DEVICE_TABLE(auxiliary, mlx5e_id_table); + +static struct auxiliary_driver mlx5e_driver = { + .name = "eth", + .probe = mlx5e_probe, + .remove = mlx5e_remove, + .suspend = mlx5e_suspend, + .resume = mlx5e_resume, + .id_table = mlx5e_id_table, }; -void mlx5e_init(void) +int mlx5e_init(void) { - mlx5e_ipsec_build_inverse_table(); + int ret; + mlx5e_build_ptys2ethtool_map(); - mlx5_register_interface(&mlx5e_interface); + ret = auxiliary_driver_register(&mlx5e_driver); + if (ret) + return ret; + + ret = mlx5e_rep_init(); + if (ret) + auxiliary_driver_unregister(&mlx5e_driver); + return ret; } void mlx5e_cleanup(void) { - mlx5_unregister_interface(&mlx5e_interface); + mlx5e_rep_cleanup(); + auxiliary_driver_unregister(&mlx5e_driver); } |