From patchwork Thu Sep 3 21:00:14 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 261555 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH, MAILING_LIST_MULTI, SIGNED_OFF_BY, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 23DF1C433E2 for ; Thu, 3 Sep 2020 21:00:46 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id E64602067C for ; Thu, 3 Sep 2020 21:00:45 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="eBwxJcWg" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729446AbgICVAo (ORCPT ); Thu, 3 Sep 2020 17:00:44 -0400 Received: from hqnvemgate24.nvidia.com ([216.228.121.143]:19835 "EHLO hqnvemgate24.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729348AbgICVAn (ORCPT ); Thu, 3 Sep 2020 17:00:43 -0400 Received: from hqpgpgate101.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate24.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Thu, 03 Sep 2020 13:58:33 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate101.nvidia.com (PGP Universal service); Thu, 03 Sep 2020 14:00:42 -0700 X-PGP-Universal: processed; by hqpgpgate101.nvidia.com on Thu, 03 Sep 2020 14:00:42 -0700 Received: from sx1.mtl.com (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Sep 2020 21:00:36 +0000 From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski CC: , Maxim Mikityanskiy , "Saeed Mahameed" Subject: [net-next 02/10] net/mlx5e: Refactor xmit functions Date: Thu, 3 Sep 2020 14:00:14 -0700 Message-ID: <20200903210022.22774-3-saeedm@nvidia.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200903210022.22774-1-saeedm@nvidia.com> References: <20200903210022.22774-1-saeedm@nvidia.com> MIME-Version: 1.0 X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL101.nvidia.com (172.20.187.10) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1599166713; bh=hd6XBGCYFZChOwxI4uOUKLCeTle+gst5PyqNCfzV38o=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-Originating-IP:X-ClientProxiedBy; b=eBwxJcWgGiQyr/JCUXTnhidjEqod0B5LDA76THYoZgU59IUkjUbvYfn4kTOncWpv7 3t/jiMhQ+YyVGXY+MRriokHyVo4Frwksc3g7CKAOecka+ZF/iQX4wOMhicPBbLSahO j9Qrk7a+AgsBEb+Yje9khx4E2TRTgYBwm9csuZc+rT/GJaxo6t+4cmHDws0vrpyO+E D+f4KwfCBIU1gZAJZiEyHiiYsQQI5P9C86W+d3rBQnRH/T1fuM0DJKt/5btfqvZS0G oDyeX2XW18cbscStrL8oOtc/Xts+3b7I4qGFy3MUEsFJn2tFQW8eWkZmwyMY2797jQ LaijIoP1m8cmg== Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Maxim Mikityanskiy A huge function mlx5e_sq_xmit was split into several to achieve multiple goals: 1. Reuse the code in IPoIB. 2. Better intergrate with TLS, IPSEC, GENEVE and checksum offloads. Now it's possible to reserve space in the WQ before running eseg-based offloads, so: 2.1. It's not needed to copy cseg and eseg after mlx5e_fill_sq_frag_edge anymore. 2.2. mlx5e_txqsq_get_next_pi will be used instead of the legacy mlx5e_fill_sq_frag_edge for better code maintainability and reuse. 3. Prepare for the upcoming TX MPWQE for SKBs. It will intervene after mlx5e_sq_calc_wqe_attr to check if it's possible to use MPWQE, and the code flow will split into two paths: MPWQE and non-MPWQE. Two high-level functions are provided to send packets: * mlx5e_xmit is called by the networking stack, runs offloads and sends the packet. In one of the following patches, MPWQE support will be added to this flow. * mlx5e_sq_xmit_simple is called by the TLS offload, runs only the checksum offload and sends the packet. This change has no performance impact in TCP single stream test and XDP_TX single stream test. UDP pktgen (burst 32), single stream: Packet rate: 17.55 Mpps -> 19.23 Mpps Instructions per packet: 420 -> 360 Cycles per packet: 165 -> 142 CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (x86_64) NIC: Mellanox ConnectX-6 Dx To get this performance gain, manual optimizations of function inlining were performed. It's important to have mlx5e_sq_xmit_wqe inline, otherwise the packet rate will be 1 Mpps less in UDP pktgen test. __always_inline is required, because gcc uninlines it when it's called from two places (mlx5e_xmit and mlx5e_sq_xmit_simple). Signed-off-by: Maxim Mikityanskiy Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 63 +-- .../mellanox/mlx5/core/en_accel/en_accel.h | 5 + .../mellanox/mlx5/core/en_accel/tls_rxtx.c | 6 +- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 391 ++++++++++-------- 4 files changed, 243 insertions(+), 222 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 9334c9c3e208..d4ee22789ab0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -41,8 +41,6 @@ void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq); u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev); netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev); -void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more); bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget); void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq); @@ -188,23 +186,6 @@ static inline u16 mlx5e_icosq_get_next_pi(struct mlx5e_icosq *sq, u16 size) return pi; } -static inline void -mlx5e_fill_sq_frag_edge(struct mlx5e_txqsq *sq, struct mlx5_wq_cyc *wq, - u16 pi, u16 nnops) -{ - struct mlx5e_tx_wqe_info *edge_wi, *wi = &sq->db.wqe_info[pi]; - - edge_wi = wi + nnops; - - /* fill sq frag edge with nops to avoid wqe wrapping two pages */ - for (; wi < edge_wi; wi++) { - memset(wi, 0, sizeof(*wi)); - wi->num_wqebbs = 1; - mlx5e_post_nop(wq, sq->sqn, &sq->pc); - } - sq->stats->nop += nnops; -} - static inline void mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc, void __iomem *uar_map, struct mlx5_wqe_ctrl_seg *ctrl) @@ -223,29 +204,6 @@ mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc, void __iomem *uar_map, mlx5_write64((__be32 *)ctrl, uar_map); } -static inline bool mlx5e_transport_inline_tx_wqe(struct mlx5_wqe_ctrl_seg *cseg) -{ - return cseg && !!cseg->tis_tir_num; -} - -static inline u8 -mlx5e_tx_wqe_inline_mode(struct mlx5e_txqsq *sq, struct mlx5_wqe_ctrl_seg *cseg, - struct sk_buff *skb) -{ - u8 mode; - - if (mlx5e_transport_inline_tx_wqe(cseg)) - return MLX5_INLINE_MODE_TCP_UDP; - - mode = sq->min_inline_mode; - - if (skb_vlan_tag_present(skb) && - test_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state)) - mode = max_t(u8, MLX5_INLINE_MODE_L2, mode); - - return mode; -} - static inline void mlx5e_cq_arm(struct mlx5e_cq *cq) { struct mlx5_core_cq *mcq; @@ -286,6 +244,27 @@ mlx5e_tx_dma_unmap(struct device *pdev, struct mlx5e_sq_dma *dma) } } +static inline void mlx5e_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, + struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg) +{ + if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { + eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM; + if (skb->encapsulation) { + eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM | + MLX5_ETH_WQE_L4_INNER_CSUM; + sq->stats->csum_partial_inner++; + } else { + eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM; + sq->stats->csum_partial++; + } + } else { + sq->stats->csum_none++; + } +} + +void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more); + static inline void mlx5e_rqwq_reset(struct mlx5e_rq *rq) { if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h index 110476bdeffb..23d4ef5ab9c5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h @@ -145,6 +145,11 @@ static inline bool mlx5e_accel_tx_finish(struct mlx5e_priv *priv, } #endif +#if IS_ENABLED(CONFIG_GENEVE) + if (skb->encapsulation) + mlx5e_tx_tunnel_accel(skb, &wqe->eth); +#endif + return true; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c index b0c31d49ff8d..c36560b3e93d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c @@ -189,12 +189,10 @@ static bool mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, struct mlx5e_tls *tls) { u32 tcp_seq = ntohl(tcp_hdr(skb)->seq); - struct mlx5e_tx_wqe *wqe; struct sync_info info; struct sk_buff *nskb; int linear_len = 0; int headln; - u16 pi; int i; sq->stats->tls_ooo++; @@ -246,9 +244,7 @@ static bool mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, sq->stats->tls_resync_bytes += nskb->len; mlx5e_tls_complete_sync_skb(skb, nskb, tcp_seq, headln, cpu_to_be64(info.rcd_sn)); - pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); - wqe = MLX5E_TX_FETCH_WQE(sq, pi); - mlx5e_sq_xmit(sq, nskb, wqe, pi, true); + mlx5e_sq_xmit_simple(sq, nskb, true); return true; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index e15aa53ff83e..f967bc0573c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -144,23 +144,6 @@ static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs) memcpy(&vhdr->h_vlan_encapsulated_proto, skb->data + cpy1_sz, cpy2_sz); } -static inline void -mlx5e_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg) -{ - if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { - eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM; - if (skb->encapsulation) { - eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM | - MLX5_ETH_WQE_L4_INNER_CSUM; - sq->stats->csum_partial_inner++; - } else { - eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM; - sq->stats->csum_partial++; - } - } else - sq->stats->csum_none++; -} - static inline u16 mlx5e_tx_get_gso_ihs(struct mlx5e_txqsq *sq, struct sk_buff *skb) { @@ -232,22 +215,121 @@ mlx5e_txwqe_build_dsegs(struct mlx5e_txqsq *sq, struct sk_buff *skb, return -ENOMEM; } +struct mlx5e_tx_attr { + u32 num_bytes; + u16 headlen; + u16 ihs; + __be16 mss; + u8 opcode; +}; + +struct mlx5e_tx_wqe_attr { + u16 ds_cnt; + u16 ds_cnt_inl; + u8 num_wqebbs; +}; + +static inline u8 +mlx5e_tx_wqe_inline_mode(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_accel_tx_state *accel) +{ + u8 mode; + +#ifdef CONFIG_MLX5_EN_TLS + if (accel && accel->tls.tls_tisn) + return MLX5_INLINE_MODE_TCP_UDP; +#endif + + mode = sq->min_inline_mode; + + if (skb_vlan_tag_present(skb) && + test_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state)) + mode = max_t(u8, MLX5_INLINE_MODE_L2, mode); + + return mode; +} + +static inline void mlx5e_sq_xmit_prepare(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_accel_tx_state *accel, + struct mlx5e_tx_attr *attr) +{ + struct mlx5e_sq_stats *stats = sq->stats; + + if (skb_is_gso(skb)) { + u16 ihs = mlx5e_tx_get_gso_ihs(sq, skb); + + *attr = (struct mlx5e_tx_attr) { + .opcode = MLX5_OPCODE_LSO, + .mss = cpu_to_be16(skb_shinfo(skb)->gso_size), + .ihs = ihs, + .num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs, + .headlen = skb_headlen(skb) - ihs, + }; + + stats->packets += skb_shinfo(skb)->gso_segs; + } else { + u8 mode = mlx5e_tx_wqe_inline_mode(sq, skb, accel); + u16 ihs = mlx5e_calc_min_inline(mode, skb); + + *attr = (struct mlx5e_tx_attr) { + .opcode = MLX5_OPCODE_SEND, + .mss = cpu_to_be16(0), + .ihs = ihs, + .num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN), + .headlen = skb_headlen(skb) - ihs, + }; + + stats->packets++; + } + + stats->bytes += attr->num_bytes; +} + +static inline void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, + const struct mlx5e_tx_attr *attr, + struct mlx5e_tx_wqe_attr *wqe_attr) +{ + u16 ds_cnt = sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS; + u16 ds_cnt_inl = 0; + + ds_cnt += !!attr->headlen + skb_shinfo(skb)->nr_frags; + + if (attr->ihs) { + u16 inl = attr->ihs - INL_HDR_START_SZ; + + if (skb_vlan_tag_present(skb)) + inl += VLAN_HLEN; + + ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); + ds_cnt += ds_cnt_inl; + } + + *wqe_attr = (struct mlx5e_tx_wqe_attr) { + .ds_cnt = ds_cnt, + .ds_cnt_inl = ds_cnt_inl, + .num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS), + }; +} + static inline void mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, - u8 opcode, u16 ds_cnt, u8 num_wqebbs, u32 num_bytes, u8 num_dma, + const struct mlx5e_tx_attr *attr, + const struct mlx5e_tx_wqe_attr *wqe_attr, u8 num_dma, struct mlx5e_tx_wqe_info *wi, struct mlx5_wqe_ctrl_seg *cseg, bool xmit_more) { struct mlx5_wq_cyc *wq = &sq->wq; bool send_doorbell; - wi->num_bytes = num_bytes; - wi->num_dma = num_dma; - wi->num_wqebbs = num_wqebbs; - wi->skb = skb; + *wi = (struct mlx5e_tx_wqe_info) { + .skb = skb, + .num_bytes = attr->num_bytes, + .num_dma = num_dma, + .num_wqebbs = wqe_attr->num_wqebbs, + }; - cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode); - cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt); + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | attr->opcode); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | wqe_attr->ds_cnt); if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; @@ -258,105 +340,44 @@ mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, sq->stats->stopped++; } - send_doorbell = __netdev_tx_sent_queue(sq->txq, num_bytes, - xmit_more); + send_doorbell = __netdev_tx_sent_queue(sq->txq, attr->num_bytes, xmit_more); if (send_doorbell) mlx5e_notify_hw(wq, sq->pc, sq->uar_map, cseg); } -void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) +static __always_inline void +mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, + const struct mlx5e_tx_attr *attr, const struct mlx5e_tx_wqe_attr *wqe_attr, + struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) { - struct mlx5_wq_cyc *wq = &sq->wq; struct mlx5_wqe_ctrl_seg *cseg; struct mlx5_wqe_eth_seg *eseg; struct mlx5_wqe_data_seg *dseg; struct mlx5e_tx_wqe_info *wi; struct mlx5e_sq_stats *stats = sq->stats; - u16 headlen, ihs, contig_wqebbs_room; - u16 ds_cnt, ds_cnt_inl = 0; - u8 num_wqebbs, opcode; - u32 num_bytes; int num_dma; - __be16 mss; - /* Calc ihs and ds cnt, no writes to wqe yet */ - ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; - if (skb_is_gso(skb)) { - opcode = MLX5_OPCODE_LSO; - mss = cpu_to_be16(skb_shinfo(skb)->gso_size); - ihs = mlx5e_tx_get_gso_ihs(sq, skb); - num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs; - stats->packets += skb_shinfo(skb)->gso_segs; - } else { - u8 mode = mlx5e_tx_wqe_inline_mode(sq, &wqe->ctrl, skb); - - opcode = MLX5_OPCODE_SEND; - mss = 0; - ihs = mlx5e_calc_min_inline(mode, skb); - num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN); - stats->packets++; - } - - stats->bytes += num_bytes; stats->xmit_more += xmit_more; - headlen = skb->len - ihs - skb->data_len; - ds_cnt += !!headlen; - ds_cnt += skb_shinfo(skb)->nr_frags; - - if (ihs) { - u16 inl = ihs + !!skb_vlan_tag_present(skb) * VLAN_HLEN - INL_HDR_START_SZ; - - ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); - ds_cnt += ds_cnt_inl; - } - - num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - if (unlikely(contig_wqebbs_room < num_wqebbs)) { -#ifdef CONFIG_MLX5_EN_IPSEC - struct mlx5_wqe_eth_seg cur_eth = wqe->eth; -#endif -#ifdef CONFIG_MLX5_EN_TLS - struct mlx5_wqe_ctrl_seg cur_ctrl = wqe->ctrl; -#endif - mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - wqe = MLX5E_TX_FETCH_WQE(sq, pi); -#ifdef CONFIG_MLX5_EN_IPSEC - wqe->eth = cur_eth; -#endif -#ifdef CONFIG_MLX5_EN_TLS - wqe->ctrl = cur_ctrl; -#endif - } - /* fill wqe */ wi = &sq->db.wqe_info[pi]; cseg = &wqe->ctrl; eseg = &wqe->eth; dseg = wqe->data; -#if IS_ENABLED(CONFIG_GENEVE) - if (skb->encapsulation) - mlx5e_tx_tunnel_accel(skb, eseg); -#endif - mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); - - eseg->mss = mss; + eseg->mss = attr->mss; - if (ihs) { + if (attr->ihs) { if (skb_vlan_tag_present(skb)) { - eseg->inline_hdr.sz = cpu_to_be16(ihs + VLAN_HLEN); - mlx5e_insert_vlan(eseg->inline_hdr.start, skb, ihs); + eseg->inline_hdr.sz = cpu_to_be16(attr->ihs + VLAN_HLEN); + mlx5e_insert_vlan(eseg->inline_hdr.start, skb, attr->ihs); stats->added_vlan_packets++; } else { - eseg->inline_hdr.sz = cpu_to_be16(ihs); - memcpy(eseg->inline_hdr.start, skb->data, ihs); + eseg->inline_hdr.sz = cpu_to_be16(attr->ihs); + memcpy(eseg->inline_hdr.start, skb->data, attr->ihs); } - dseg += ds_cnt_inl; + dseg += wqe_attr->ds_cnt_inl; } else if (skb_vlan_tag_present(skb)) { eseg->insert.type = cpu_to_be16(MLX5_ETH_WQE_INSERT_VLAN); if (skb->vlan_proto == cpu_to_be16(ETH_P_8021AD)) @@ -365,12 +386,12 @@ void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, stats->added_vlan_packets++; } - num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + ihs, headlen, dseg); + num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + attr->ihs, + attr->headlen, dseg); if (unlikely(num_dma < 0)) goto err_drop; - mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, - num_dma, wi, cseg, xmit_more); + mlx5e_txwqe_complete(sq, skb, attr, wqe_attr, num_dma, wi, cseg, xmit_more); return; @@ -383,6 +404,8 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5e_accel_tx_state accel = {}; + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; struct mlx5e_tx_wqe *wqe; struct mlx5e_txqsq *sq; u16 pi; @@ -393,19 +416,64 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) if (unlikely(!mlx5e_accel_tx_begin(dev, sq, skb, &accel))) goto out; - pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + mlx5e_sq_xmit_prepare(sq, skb, &accel, &attr); + mlx5e_sq_calc_wqe_attr(skb, &attr, &wqe_attr); + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); wqe = MLX5E_TX_FETCH_WQE(sq, pi); /* May update the WQE, but may not post other WQEs. */ if (unlikely(!mlx5e_accel_tx_finish(priv, sq, skb, wqe, &accel))) goto out; - mlx5e_sq_xmit(sq, skb, wqe, pi, netdev_xmit_more()); + mlx5e_txwqe_build_eseg_csum(sq, skb, &wqe->eth); + mlx5e_sq_xmit_wqe(sq, skb, &attr, &wqe_attr, wqe, pi, netdev_xmit_more()); out: return NETDEV_TX_OK; } +void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more) +{ + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; + struct mlx5e_tx_wqe *wqe; + u16 pi; + + mlx5e_sq_xmit_prepare(sq, skb, NULL, &attr); + mlx5e_sq_calc_wqe_attr(skb, &attr, &wqe_attr); + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + mlx5e_txwqe_build_eseg_csum(sq, skb, &wqe->eth); + mlx5e_sq_xmit_wqe(sq, skb, &attr, &wqe_attr, wqe, pi, xmit_more); +} + +static inline void mlx5e_tx_wi_dma_unmap(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) +{ + int i; + + for (i = 0; i < wi->num_dma; i++) { + struct mlx5e_sq_dma *dma = mlx5e_dma_get(sq, (*dma_fifo_cc)++); + + mlx5e_tx_dma_unmap(sq->pdev, dma); + } +} + +static inline void mlx5e_consume_skb(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_cqe64 *cqe, int napi_budget) +{ + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { + struct skb_shared_hwtstamps hwts = {}; + u64 ts = get_cqe_ts(cqe); + + hwts.hwtstamp = mlx5_timecounter_cyc2time(sq->clock, ts); + skb_tstamp_tx(skb, &hwts); + } + + napi_consume_skb(skb, napi_budget); +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_sq_stats *stats; @@ -452,7 +520,6 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) do { struct sk_buff *skb; - int j; last_wqe = (sqcc == wqe_counter); @@ -460,33 +527,18 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) wi = &sq->db.wqe_info[ci]; skb = wi->skb; + sqcc += wi->num_wqebbs; + if (unlikely(!skb)) { mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); - sqcc += wi->num_wqebbs; continue; } - if (unlikely(skb_shinfo(skb)->tx_flags & - SKBTX_HW_TSTAMP)) { - struct skb_shared_hwtstamps hwts = {}; - - hwts.hwtstamp = - mlx5_timecounter_cyc2time(sq->clock, - get_cqe_ts(cqe)); - skb_tstamp_tx(skb, &hwts); - } - - for (j = 0; j < wi->num_dma; j++) { - struct mlx5e_sq_dma *dma = - mlx5e_dma_get(sq, dma_fifo_cc++); - - mlx5e_tx_dma_unmap(sq->pdev, dma); - } + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_consume_skb(sq, wi->skb, cqe, napi_budget); npkts++; nbytes += wi->num_bytes; - sqcc += wi->num_wqebbs; - napi_consume_skb(skb, napi_budget); } while (!last_wqe); if (unlikely(get_cqe_opcode(cqe) == MLX5_CQE_REQ_ERR)) { @@ -531,7 +583,6 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) u32 dma_fifo_cc, nbytes = 0; u16 ci, sqcc, npkts = 0; struct sk_buff *skb; - int i; sqcc = sq->cc; dma_fifo_cc = sq->dma_fifo_cc; @@ -541,23 +592,18 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) wi = &sq->db.wqe_info[ci]; skb = wi->skb; + sqcc += wi->num_wqebbs; + if (!skb) { mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); - sqcc += wi->num_wqebbs; continue; } - for (i = 0; i < wi->num_dma; i++) { - struct mlx5e_sq_dma *dma = - mlx5e_dma_get(sq, dma_fifo_cc++); - - mlx5e_tx_dma_unmap(sq->pdev, dma); - } - + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); dev_kfree_skb_any(skb); + npkts++; nbytes += wi->num_bytes; - sqcc += wi->num_wqebbs; } sq->dma_fifo_cc = dma_fifo_cc; @@ -576,9 +622,34 @@ mlx5i_txwqe_build_datagram(struct mlx5_av *av, u32 dqpn, u32 dqkey, dseg->av.key.qkey.qkey = cpu_to_be32(dqkey); } +static void mlx5i_sq_calc_wqe_attr(struct sk_buff *skb, + const struct mlx5e_tx_attr *attr, + struct mlx5e_tx_wqe_attr *wqe_attr) +{ + u16 ds_cnt = sizeof(struct mlx5i_tx_wqe) / MLX5_SEND_WQE_DS; + u16 ds_cnt_inl = 0; + + ds_cnt += !!attr->headlen + skb_shinfo(skb)->nr_frags; + + if (attr->ihs) { + u16 inl = attr->ihs - INL_HDR_START_SZ; + + ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); + ds_cnt += ds_cnt_inl; + } + + *wqe_attr = (struct mlx5e_tx_wqe_attr) { + .ds_cnt = ds_cnt, + .ds_cnt_inl = ds_cnt_inl, + .num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS), + }; +} + void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5_av *av, u32 dqpn, u32 dqkey, bool xmit_more) { + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; struct mlx5i_tx_wqe *wqe; struct mlx5_wqe_datagram_seg *datagram; @@ -588,47 +659,17 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5e_tx_wqe_info *wi; struct mlx5e_sq_stats *stats = sq->stats; - u16 ds_cnt, ds_cnt_inl = 0; - u8 num_wqebbs, opcode; - u16 headlen, ihs, pi; - u32 num_bytes; int num_dma; - __be16 mss; + u16 pi; - /* Calc ihs and ds cnt, no writes to wqe yet */ - ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; - if (skb_is_gso(skb)) { - opcode = MLX5_OPCODE_LSO; - mss = cpu_to_be16(skb_shinfo(skb)->gso_size); - ihs = mlx5e_tx_get_gso_ihs(sq, skb); - num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs; - stats->packets += skb_shinfo(skb)->gso_segs; - } else { - u8 mode = mlx5e_tx_wqe_inline_mode(sq, NULL, skb); + mlx5e_sq_xmit_prepare(sq, skb, NULL, &attr); + mlx5i_sq_calc_wqe_attr(skb, &attr, &wqe_attr); - opcode = MLX5_OPCODE_SEND; - mss = 0; - ihs = mlx5e_calc_min_inline(mode, skb); - num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN); - stats->packets++; - } + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); + wqe = MLX5I_SQ_FETCH_WQE(sq, pi); - stats->bytes += num_bytes; stats->xmit_more += xmit_more; - headlen = skb->len - ihs - skb->data_len; - ds_cnt += !!headlen; - ds_cnt += skb_shinfo(skb)->nr_frags; - - if (ihs) { - ds_cnt_inl = DIV_ROUND_UP(ihs - INL_HDR_START_SZ, MLX5_SEND_WQE_DS); - ds_cnt += ds_cnt_inl; - } - - num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS); - pi = mlx5e_txqsq_get_next_pi(sq, num_wqebbs); - wqe = MLX5I_SQ_FETCH_WQE(sq, pi); - /* fill wqe */ wi = &sq->db.wqe_info[pi]; cseg = &wqe->ctrl; @@ -640,20 +681,20 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); - eseg->mss = mss; + eseg->mss = attr.mss; - if (ihs) { - memcpy(eseg->inline_hdr.start, skb->data, ihs); - eseg->inline_hdr.sz = cpu_to_be16(ihs); - dseg += ds_cnt_inl; + if (attr.ihs) { + memcpy(eseg->inline_hdr.start, skb->data, attr.ihs); + eseg->inline_hdr.sz = cpu_to_be16(attr.ihs); + dseg += wqe_attr.ds_cnt_inl; } - num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + ihs, headlen, dseg); + num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + attr.ihs, + attr.headlen, dseg); if (unlikely(num_dma < 0)) goto err_drop; - mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, - num_dma, wi, cseg, xmit_more); + mlx5e_txwqe_complete(sq, skb, &attr, &wqe_attr, num_dma, wi, cseg, xmit_more); return; From patchwork Thu Sep 3 21:00:17 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 261554 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH, MAILING_LIST_MULTI, SIGNED_OFF_BY, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8D863C433E2 for ; Thu, 3 Sep 2020 21:00:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 5F8C920684 for ; Thu, 3 Sep 2020 21:00:58 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="M2Vb3RZM" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728556AbgICVA6 (ORCPT ); Thu, 3 Sep 2020 17:00:58 -0400 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:2775 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729348AbgICVAt (ORCPT ); Thu, 3 Sep 2020 17:00:49 -0400 Received: from hqpgpgate102.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Thu, 03 Sep 2020 14:00:00 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate102.nvidia.com (PGP Universal service); Thu, 03 Sep 2020 14:00:47 -0700 X-PGP-Universal: processed; by hqpgpgate102.nvidia.com on Thu, 03 Sep 2020 14:00:47 -0700 Received: from sx1.mtl.com (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Sep 2020 21:00:37 +0000 From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski CC: , Maxim Mikityanskiy , "Saeed Mahameed" Subject: [net-next 05/10] net/mlx5e: Move the TLS resync check out of the function Date: Thu, 3 Sep 2020 14:00:17 -0700 Message-ID: <20200903210022.22774-6-saeedm@nvidia.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200903210022.22774-1-saeedm@nvidia.com> References: <20200903210022.22774-1-saeedm@nvidia.com> MIME-Version: 1.0 X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL101.nvidia.com (172.20.187.10) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1599166800; bh=jhqzFzm306dz05igvrLWIp302+lW8RdL515BaMtnknw=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-Originating-IP:X-ClientProxiedBy; b=M2Vb3RZM1u8dWT+QQU4qrZnuBHG43cNWX1/SlpIDQ+75a6xYxM4MdhYOj9QQ/r2q1 G80sGymGFEZXlKJMKuyW4yjPAQwPYjnwVo6d+EBDoejX/TD8fdEALTigh08bOMGOds RaYqWGkhuKU7K1yns5tKxI1Jtej6Ug6OjQ4Jge3YlFwXnCZ+jInZqTQN+kX7HlZqxm ecqEbNa+QeZj1KnlGn24QcGH8rwFb2JY4du1CMr+BSsq8B5s/wSrg/90Qwv7z+VHTZ 37oD8Hj8HU2dqZOzM2lf/DWkyM2+Y07FvboFNdwvTR41sS/AQJvl/G5k2FdOxVyA/e eOB/+TfvbL2Og== Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Maxim Mikityanskiy Before this patch, mlx5e_ktls_tx_handle_resync_dump_comp checked for resync_dump_frag_page. It happened for all WQEs without an SKB, including padding WQEs, and required a function call. Normally, padding WQEs happen more often than TLS resyncs. Take this check out of the function and put it to an inline function to save a call on all padding WQEs. Signed-off-by: Maxim Mikityanskiy Signed-off-by: Saeed Mahameed --- .../ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c | 3 --- .../mellanox/mlx5/core/en_accel/ktls_txrx.h | 14 +++++++++++--- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 4 ++-- 3 files changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c index f4861545b236..b140e13fdcc8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c @@ -345,9 +345,6 @@ void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_sq_stats *stats; struct mlx5e_sq_dma *dma; - if (!wi->resync_dump_frag_page) - return; - dma = mlx5e_dma_get(sq, (*dma_fifo_cc)++); stats = sq->stats; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h index ff4c740af10b..fcfb156cf09d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h @@ -29,11 +29,19 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc); +static inline void +mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) +{ + if (unlikely(wi->resync_dump_frag_page)) + mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, dma_fifo_cc); +} #else static inline void -mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, - struct mlx5e_tx_wqe_info *wi, - u32 *dma_fifo_cc) +mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) { } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 46bdbbbfaf65..869b3313dabf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -530,7 +530,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) sqcc += wi->num_wqebbs; if (unlikely(!skb)) { - mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); + mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); continue; } @@ -595,7 +595,7 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) sqcc += wi->num_wqebbs; if (!skb) { - mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); + mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); continue; } From patchwork Thu Sep 3 21:00:18 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 261551 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH, MAILING_LIST_MULTI, SIGNED_OFF_BY, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 829BFC07548 for ; Thu, 3 Sep 2020 21:01:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 4DD13206A5 for ; Thu, 3 Sep 2020 21:01:20 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="Nplp0TDJ" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729517AbgICVBR (ORCPT ); Thu, 3 Sep 2020 17:01:17 -0400 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:2839 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729450AbgICVAz (ORCPT ); Thu, 3 Sep 2020 17:00:55 -0400 Received: from hqpgpgate102.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Thu, 03 Sep 2020 14:00:00 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate102.nvidia.com (PGP Universal service); Thu, 03 Sep 2020 14:00:47 -0700 X-PGP-Universal: processed; by hqpgpgate102.nvidia.com on Thu, 03 Sep 2020 14:00:47 -0700 Received: from sx1.mtl.com (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Sep 2020 21:00:38 +0000 From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski CC: , Maxim Mikityanskiy , "Tariq Toukan" , Saeed Mahameed Subject: [net-next 06/10] net/mlx5e: Support multiple SKBs in a TX WQE Date: Thu, 3 Sep 2020 14:00:18 -0700 Message-ID: <20200903210022.22774-7-saeedm@nvidia.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200903210022.22774-1-saeedm@nvidia.com> References: <20200903210022.22774-1-saeedm@nvidia.com> MIME-Version: 1.0 X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL101.nvidia.com (172.20.187.10) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1599166800; bh=DwVlOGh5MkvPalG7WO/D20TJ0VLotVDh99N0S6UUrOs=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-Originating-IP:X-ClientProxiedBy; b=Nplp0TDJJQAcmEw9dJIxpodckd0RWtHK0Bb3IRGIqACKZEYMOKB7nVjX5oOT7FjRN oqvZ67Znc3I2NpFRU3YRWvYTEbpJI5Zyx76sjUeU3oUTXFw7OwMidx2IIoDRL8Qecx KNnXD4cDyblM+HnfQJPFqP3TD5h1qQ05heNxOMWvGZrBvwodjUwy07yn7w+H3Up/3b zjTGiu3jHhN9SCqMEL1UKdz1DIRBka8D2RXK0p1GQaLkk83SkxgoWw3T0ypHbtejyt THLzvl0/RLnqLiGLkfF+2ffUASuDI9egnX87a0EweWHrQ4WreziXRdDrYrpJu61qPo t9FMDmdr/sKZg== Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Maxim Mikityanskiy TX MPWQE support for SKBs is coming in one of the following patches, and a single MPWQE can send multiple SKBs. This commit prepares the TX path code to handle such cases: 1. An additional FIFO for SKBs is added, just like the FIFO for DMA chunks. 2. struct mlx5e_tx_wqe_info will contain num_fifo_pkts. If a given WQE contains only one packet, num_fifo_pkts will be zero, and the SKB will be stored in mlx5e_tx_wqe_info, as usual. If num_fifo_pkts > 0, the SKB pointer will be NULL, and the SKBs will be stored in the FIFO. This change has no performance impact in TCP single stream test and XDP_TX single stream test. UDP pktgen (burst 32), single stream: Packet rate: 19.23 Mpps -> 19.12 Mpps Instructions per packet: 360 -> 354 Cycles per packet: 142 -> 140 CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (x86_64) NIC: Mellanox ConnectX-6 Dx Signed-off-by: Maxim Mikityanskiy Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++ .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 18 +++++ .../mellanox/mlx5/core/en_accel/ktls_txrx.h | 10 ++- .../net/ethernet/mellanox/mlx5/core/en_main.c | 7 +- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 71 ++++++++++++++----- 5 files changed, 89 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 4f33658da25a..6ab60074fca9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -317,11 +317,13 @@ struct mlx5e_txqsq { /* dirtied @completion */ u16 cc; + u16 skb_fifo_cc; u32 dma_fifo_cc; struct dim dim; /* Adaptive Moderation */ /* dirtied @xmit */ u16 pc ____cacheline_aligned_in_smp; + u16 skb_fifo_pc; u32 dma_fifo_pc; struct mlx5e_cq cq; @@ -329,9 +331,11 @@ struct mlx5e_txqsq { /* read only */ struct mlx5_wq_cyc wq; u32 dma_fifo_mask; + u16 skb_fifo_mask; struct mlx5e_sq_stats *stats; struct { struct mlx5e_sq_dma *dma_fifo; + struct sk_buff **skb_fifo; struct mlx5e_tx_wqe_info *wqe_info; } db; void __iomem *uar_map; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 155b89998891..7baac2971758 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -105,6 +105,7 @@ struct mlx5e_tx_wqe_info { u32 num_bytes; u8 num_wqebbs; u8 num_dma; + u8 num_fifo_pkts; #ifdef CONFIG_MLX5_EN_TLS struct page *resync_dump_frag_page; #endif @@ -231,6 +232,23 @@ mlx5e_dma_push(struct mlx5e_txqsq *sq, dma_addr_t addr, u32 size, dma->type = map_type; } +static inline struct sk_buff **mlx5e_skb_fifo_get(struct mlx5e_txqsq *sq, u16 i) +{ + return &sq->db.skb_fifo[i & sq->skb_fifo_mask]; +} + +static inline void mlx5e_skb_fifo_push(struct mlx5e_txqsq *sq, struct sk_buff *skb) +{ + struct sk_buff **skb_item = mlx5e_skb_fifo_get(sq, sq->skb_fifo_pc++); + + *skb_item = skb; +} + +static inline struct sk_buff *mlx5e_skb_fifo_pop(struct mlx5e_txqsq *sq) +{ + return *mlx5e_skb_fifo_get(sq, sq->skb_fifo_cc++); +} + static inline void mlx5e_tx_dma_unmap(struct device *pdev, struct mlx5e_sq_dma *dma) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h index fcfb156cf09d..7521c9be735b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h @@ -29,20 +29,24 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc); -static inline void +static inline bool mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc) { - if (unlikely(wi->resync_dump_frag_page)) + if (unlikely(wi->resync_dump_frag_page)) { mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, dma_fifo_cc); + return true; + } + return false; } #else -static inline void +static inline bool mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc) { + return false; } #endif /* CONFIG_MLX5_EN_TLS */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 26834625556d..b413aa168e4e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1040,6 +1040,7 @@ static void mlx5e_free_icosq(struct mlx5e_icosq *sq) static void mlx5e_free_txqsq_db(struct mlx5e_txqsq *sq) { kvfree(sq->db.wqe_info); + kvfree(sq->db.skb_fifo); kvfree(sq->db.dma_fifo); } @@ -1051,15 +1052,19 @@ 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 = kvzalloc_node(array_size(df_sz, + sizeof(*sq->db.skb_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 || !sq->db.wqe_info) { mlx5e_free_txqsq_db(sq); return -ENOMEM; } sq->dma_fifo_mask = df_sz - 1; + sq->skb_fifo_mask = df_sz - 1; return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 869b3313dabf..9ced350150b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -326,6 +326,7 @@ mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, .num_bytes = attr->num_bytes, .num_dma = num_dma, .num_wqebbs = wqe_attr->num_wqebbs, + .num_fifo_pkts = 0, }; cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | attr->opcode); @@ -474,6 +475,20 @@ static inline void mlx5e_consume_skb(struct mlx5e_txqsq *sq, struct sk_buff *skb napi_consume_skb(skb, napi_budget); } +static inline void mlx5e_tx_wi_consume_fifo_skbs(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + struct mlx5_cqe64 *cqe, + int napi_budget) +{ + int i; + + for (i = 0; i < wi->num_fifo_pkts; i++) { + struct sk_buff *skb = mlx5e_skb_fifo_pop(sq); + + mlx5e_consume_skb(sq, skb, cqe, napi_budget); + } +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_sq_stats *stats; @@ -519,26 +534,33 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) wqe_counter = be16_to_cpu(cqe->wqe_counter); do { - struct sk_buff *skb; - last_wqe = (sqcc == wqe_counter); ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.wqe_info[ci]; - skb = wi->skb; sqcc += wi->num_wqebbs; - if (unlikely(!skb)) { - mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); + if (likely(wi->skb)) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_consume_skb(sq, wi->skb, cqe, napi_budget); + + npkts++; + nbytes += wi->num_bytes; continue; } - mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); - mlx5e_consume_skb(sq, wi->skb, cqe, napi_budget); + if (unlikely(mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, + &dma_fifo_cc))) + continue; - npkts++; - nbytes += wi->num_bytes; + if (wi->num_fifo_pkts) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_tx_wi_consume_fifo_skbs(sq, wi, cqe, napi_budget); + + npkts += wi->num_fifo_pkts; + nbytes += wi->num_bytes; + } } while (!last_wqe); if (unlikely(get_cqe_opcode(cqe) == MLX5_CQE_REQ_ERR)) { @@ -577,12 +599,19 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) return (i == MLX5E_TX_CQ_POLL_BUDGET); } +static void mlx5e_tx_wi_kfree_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi) +{ + int i; + + for (i = 0; i < wi->num_fifo_pkts; i++) + dev_kfree_skb_any(mlx5e_skb_fifo_pop(sq)); +} + void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) { struct mlx5e_tx_wqe_info *wi; u32 dma_fifo_cc, nbytes = 0; u16 ci, sqcc, npkts = 0; - struct sk_buff *skb; sqcc = sq->cc; dma_fifo_cc = sq->dma_fifo_cc; @@ -590,20 +619,28 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) while (sqcc != sq->pc) { ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.wqe_info[ci]; - skb = wi->skb; sqcc += wi->num_wqebbs; - if (!skb) { - mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); + if (likely(wi->skb)) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + dev_kfree_skb_any(wi->skb); + + npkts++; + nbytes += wi->num_bytes; continue; } - mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); - dev_kfree_skb_any(skb); + if (unlikely(mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc))) + continue; - npkts++; - nbytes += wi->num_bytes; + if (wi->num_fifo_pkts) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_tx_wi_kfree_fifo_skbs(sq, wi); + + npkts += wi->num_fifo_pkts; + nbytes += wi->num_bytes; + } } sq->dma_fifo_cc = dma_fifo_cc; From patchwork Thu Sep 3 21:00:20 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 261552 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH, MAILING_LIST_MULTI, SIGNED_OFF_BY, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id A3C94C433E2 for ; Thu, 3 Sep 2020 21:01:12 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 7665520684 for ; Thu, 3 Sep 2020 21:01:12 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="A9ur9tbj" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729495AbgICVBJ (ORCPT ); Thu, 3 Sep 2020 17:01:09 -0400 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:2838 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728115AbgICVAz (ORCPT ); Thu, 3 Sep 2020 17:00:55 -0400 Received: from hqpgpgate102.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Thu, 03 Sep 2020 14:00:00 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate102.nvidia.com (PGP Universal service); Thu, 03 Sep 2020 14:00:47 -0700 X-PGP-Universal: processed; by hqpgpgate102.nvidia.com on Thu, 03 Sep 2020 14:00:47 -0700 Received: from sx1.mtl.com (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Sep 2020 21:00:39 +0000 From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski CC: , Maxim Mikityanskiy , "Tariq Toukan" , Saeed Mahameed Subject: [net-next 08/10] net/mlx5e: Rename xmit-related structs to generalize them Date: Thu, 3 Sep 2020 14:00:20 -0700 Message-ID: <20200903210022.22774-9-saeedm@nvidia.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200903210022.22774-1-saeedm@nvidia.com> References: <20200903210022.22774-1-saeedm@nvidia.com> MIME-Version: 1.0 X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL101.nvidia.com (172.20.187.10) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1599166800; bh=dQLkGRhn95EGUjzjPAeoniyGO+1RaspqoAOUDwqymYE=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-Originating-IP:X-ClientProxiedBy; b=A9ur9tbjuQ4+Kc5BPxRARqQ6rtPQYCUZK79NOT9XbMw9j5/GKXQA2tz2Rso8/NYf2 NJ/V5Oao1Av+Tavyqk+dvNyV4QHTKyNW6/vsbBxHm8vz71TlQoNoMfsyVAj9cXP81T pXAL6wACQQKoRGUTCOaSD4+8Hrgibsex4k9P3nIgej7amF2X+CLvLaV7WZsl1z0xOp 6swdkM3BvWG6NnIVAmQoSDZheuFSGLCOydtXfESLcxoPs4qqUHiAGTL4paQE8M8kJz QSblPjgXAj34+riFj6fDl6//+PxqDpfESksZyb/tUSKayREODuIC7qp+o5LHa9xRBT KYArd7578XBNQ== Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Maxim Mikityanskiy As preparation for the upcoming TX MPWQE support for SKBs, rename struct mlx5e_xdp_mpwqe to mlx5e_tx_mpwqe and move it above struct mlx5e_txqsq. This structure will be reused in the regular SQ and in the regular TX data path. Also rename mlx5e_xdp_xmit_data to mlx5e_xmit_data - it will be used in the upcoming TX MPWQE flow. Signed-off-by: Maxim Mikityanskiy Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 22 +++++++++---------- .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 2 +- .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 16 +++++++------- .../net/ethernet/mellanox/mlx5/core/en/xdp.h | 10 ++++----- .../ethernet/mellanox/mlx5/core/en/xsk/tx.c | 2 +- 5 files changed, 26 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 6ab60074fca9..3511836f0f4a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -312,6 +312,14 @@ enum { MLX5E_SQ_STATE_PENDING_XSK_TX, }; +struct mlx5e_tx_mpwqe { + /* Current MPWQE session */ + struct mlx5e_tx_wqe *wqe; + u8 ds_count; + u8 pkt_count; + u8 inline_on; +}; + struct mlx5e_txqsq { /* data path */ @@ -402,7 +410,7 @@ struct mlx5e_xdp_info { }; }; -struct mlx5e_xdp_xmit_data { +struct mlx5e_xmit_data { dma_addr_t dma_addr; void *data; u32 len; @@ -415,18 +423,10 @@ struct mlx5e_xdp_info_fifo { u32 mask; }; -struct mlx5e_xdp_mpwqe { - /* Current MPWQE session */ - struct mlx5e_tx_wqe *wqe; - u8 ds_count; - u8 pkt_count; - u8 inline_on; -}; - struct mlx5e_xdpsq; typedef int (*mlx5e_fp_xmit_xdp_frame_check)(struct mlx5e_xdpsq *); typedef bool (*mlx5e_fp_xmit_xdp_frame)(struct mlx5e_xdpsq *, - struct mlx5e_xdp_xmit_data *, + struct mlx5e_xmit_data *, struct mlx5e_xdp_info *, int); @@ -441,7 +441,7 @@ struct mlx5e_xdpsq { u32 xdpi_fifo_pc ____cacheline_aligned_in_smp; u16 pc; struct mlx5_wqe_ctrl_seg *doorbell_cseg; - struct mlx5e_xdp_mpwqe mpwqe; + struct mlx5e_tx_mpwqe mpwqe; struct mlx5e_cq cq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 09cf4236439e..1ac4607fba08 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -298,7 +298,7 @@ static inline void mlx5e_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more); -static inline bool mlx5e_tx_mpwqe_is_full(struct mlx5e_xdp_mpwqe *session) +static inline bool mlx5e_tx_mpwqe_is_full(struct mlx5e_tx_mpwqe *session) { return session->ds_count == MLX5E_TX_MPW_MAX_NUM_DS; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 0edd4ebeb90c..adacc4f9a3bf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -59,7 +59,7 @@ static inline bool mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, struct mlx5e_dma_info *di, struct xdp_buff *xdp) { - struct mlx5e_xdp_xmit_data xdptxd; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; struct xdp_frame *xdpf; dma_addr_t dma_addr; @@ -194,7 +194,7 @@ static u16 mlx5e_xdpsq_get_next_pi(struct mlx5e_xdpsq *sq, u16 size) static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5e_xdpsq_stats *stats = sq->stats; struct mlx5e_tx_wqe *wqe; u16 pi; @@ -203,7 +203,7 @@ static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) wqe = MLX5E_TX_FETCH_WQE(sq, pi); net_prefetchw(session->wqe->data); - *session = (struct mlx5e_xdp_mpwqe) { + *session = (struct mlx5e_tx_mpwqe) { .wqe = wqe, .ds_count = MLX5E_TX_WQE_EMPTY_DS_COUNT, .pkt_count = 0, @@ -216,7 +216,7 @@ static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) void mlx5e_xdp_mpwqe_complete(struct mlx5e_xdpsq *sq) { struct mlx5_wq_cyc *wq = &sq->wq; - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5_wqe_ctrl_seg *cseg = &session->wqe->ctrl; u16 ds_count = session->ds_count; u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); @@ -261,10 +261,10 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq } INDIRECT_CALLABLE_SCOPE bool -mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *xdptxd, +mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5e_xdpsq_stats *stats = sq->stats; if (unlikely(xdptxd->len > sq->hw_mtu)) { @@ -308,7 +308,7 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check(struct mlx5e_xdpsq *sq) } INDIRECT_CALLABLE_SCOPE bool -mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *xdptxd, +mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result) { struct mlx5_wq_cyc *wq = &sq->wq; @@ -505,7 +505,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, for (i = 0; i < n; i++) { struct xdp_frame *xdpf = frames[i]; - struct mlx5e_xdp_xmit_data xdptxd; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; bool ret; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h index 0dc38acab5a8..4bd8af478a4a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h @@ -58,11 +58,11 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, u32 flags); INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result)); INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result)); INDIRECT_CALLABLE_DECLARE(int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq *sq)); @@ -123,7 +123,7 @@ static inline bool mlx5e_xdp_get_inline_state(struct mlx5e_xdpsq *sq, bool cur) return cur; } -static inline bool mlx5e_xdp_mpqwe_is_full(struct mlx5e_xdp_mpwqe *session) +static inline bool mlx5e_xdp_mpqwe_is_full(struct mlx5e_tx_mpwqe *session) { if (session->inline_on) return session->ds_count + MLX5E_XDP_INLINE_WQE_MAX_DS_CNT > @@ -138,10 +138,10 @@ struct mlx5e_xdp_wqe_info { static inline void mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdpsq_stats *stats) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5_wqe_data_seg *dseg = (struct mlx5_wqe_data_seg *)session->wqe + session->ds_count; u32 dma_len = xdptxd->len; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c index aa91cbdfe969..fb671a457129 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c @@ -67,8 +67,8 @@ static void mlx5e_xsk_tx_post_err(struct mlx5e_xdpsq *sq, bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) { struct xsk_buff_pool *pool = sq->xsk_pool; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; - struct mlx5e_xdp_xmit_data xdptxd; bool work_done = true; bool flush = false; From patchwork Thu Sep 3 21:00:22 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Saeed Mahameed X-Patchwork-Id: 261553 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH, MAILING_LIST_MULTI, SIGNED_OFF_BY, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 451BFC43461 for ; Thu, 3 Sep 2020 21:01:06 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 086A020684 for ; Thu, 3 Sep 2020 21:01:06 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="MDb1GEm6" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729478AbgICVBC (ORCPT ); Thu, 3 Sep 2020 17:01:02 -0400 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:2778 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729447AbgICVAt (ORCPT ); Thu, 3 Sep 2020 17:00:49 -0400 Received: from hqpgpgate102.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Thu, 03 Sep 2020 14:00:00 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate102.nvidia.com (PGP Universal service); Thu, 03 Sep 2020 14:00:47 -0700 X-PGP-Universal: processed; by hqpgpgate102.nvidia.com on Thu, 03 Sep 2020 14:00:47 -0700 Received: from sx1.mtl.com (10.124.1.5) by HQMAIL107.nvidia.com (172.20.187.13) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Thu, 3 Sep 2020 21:00:40 +0000 From: Saeed Mahameed To: "David S. Miller" , Jakub Kicinski CC: , Maxim Mikityanskiy , "Saeed Mahameed" Subject: [net-next 10/10] net/mlx5e: Enhanced TX MPWQE for SKBs Date: Thu, 3 Sep 2020 14:00:22 -0700 Message-ID: <20200903210022.22774-11-saeedm@nvidia.com> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200903210022.22774-1-saeedm@nvidia.com> References: <20200903210022.22774-1-saeedm@nvidia.com> MIME-Version: 1.0 X-Originating-IP: [10.124.1.5] X-ClientProxiedBy: HQMAIL101.nvidia.com (172.20.187.10) To HQMAIL107.nvidia.com (172.20.187.13) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1599166800; bh=bLtge1re3pU0EKeTVltgIuziyp8QtXp5FmIq+YYhmQM=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding: Content-Type:X-Originating-IP:X-ClientProxiedBy; b=MDb1GEm6Rw9liZbCyaLWUIUpQXKN0suzO3WCJXaWa7vv+dl7uZo/WujZuIA1DDCft ZKiG7zCKhl1E2ouGiOiAX2fldQpEihdXcPodSYFvjJr1yYsDsLNNnIQXtrv66PqKP2 bI0x/shYmvPf3he33PCB0d387JzM6WUgv3Scsh59YXLw8M4U/bbpXTtXNGj4R7Ca6C V+0p4gINbl9iRAeTqUnHK+A/wnLtSGxW0V00mNWzpC+YR5Vteyqg8C4mYX2w6yXS4h 0sKjiczBYMmNMDko46Ng94ur0UK/BWgEX3cNb0wWzv4RaTdkNrul86tUBtTNUSRi1H RxBqUA9Ktr4Nw== Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Maxim Mikityanskiy This commit adds support for Enhanced TX MPWQE feature in the regular (SKB) data path. A MPWQE (multi-packet work queue element) can serve multiple packets, reducing the PCI bandwidth on control traffic. Two new stats (tx*_mpwqe_blks and tx*_mpwqe_pkts) are added. The feature is on by default and controlled by the skb_tx_mpwqe private flag. In a MPWQE, eseg is shared among all packets, so eseg-based offloads (IPSEC, GENEVE, checksum) run on a separate eseg that is compared to the eseg of the current MPWQE session to decide if the new packet can be added to the same session. MPWQE is not compatible with certain offloads and features, such as TLS offload, TSO, nonlinear SKBs. If such incompatible features are in use, the driver gracefully falls back to non-MPWQE. This change has no performance impact in TCP single stream test and XDP_TX single stream test. UDP pktgen, 64-byte packets, single stream, MPWQE off: Packet rate: 19.12 Mpps -> 20.02 Mpps Instructions per packet: 354 -> 347 Cycles per packet: 140 -> 129 UDP pktgen, 64-byte packets, single stream, MPWQE on: Packet rate: 19.12 Mpps -> 20.67 Mpps Instructions per packet: 354 -> 335 Cycles per packet: 140 -> 124 Enabling MPWQE can reduce PCI bandwidth: PCI Gen2, pktgen at fixed rate of 36864000 pps on 24 CPU cores: Inbound PCI utilization with MPWQE off: 81.3% Inbound PCI utilization with MPWQE on: 59.3% PCI Gen3, pktgen at fixed rate of 56064005 pps on 24 CPU cores: Inbound PCI utilization with MPWQE off: 65.8% Inbound PCI utilization with MPWQE on: 49.2% Enabling MPWQE can also reduce CPU load, increasing the packet rate in case of CPU bottleneck: PCI Gen2, pktgen at full rate on 24 CPU cores: Packet rate with MPWQE off: 37.4 Mpps Packet rate with MPWQE on: 49.1 Mpps PCI Gen3, pktgen at full rate on 24 CPU cores: Packet rate with MPWQE off: 56.2 Mpps Packet rate with MPWQE on: 67.0 Mpps Burst size in all pktgen tests is 32. CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (x86_64) NIC: Mellanox ConnectX-6 Dx To avoid performance degradation when MPWQE is off, manual optimizations of function inlining were performed. It's especially important to have mlx5e_sq_xmit_mpwqe noinline, otherwise gcc inlines it automatically and bloats mlx5e_xmit, slowing it down. Signed-off-by: Maxim Mikityanskiy Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 + .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 1 + .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 1 + .../net/ethernet/mellanox/mlx5/core/en/xdp.h | 1 + .../mellanox/mlx5/core/en_accel/en_accel.h | 29 +-- .../mellanox/mlx5/core/en_accel/tls_rxtx.c | 2 + .../ethernet/mellanox/mlx5/core/en_ethtool.c | 15 +- .../net/ethernet/mellanox/mlx5/core/en_main.c | 11 ++ .../ethernet/mellanox/mlx5/core/en_stats.c | 6 + .../ethernet/mellanox/mlx5/core/en_stats.h | 4 + .../net/ethernet/mellanox/mlx5/core/en_tx.c | 184 +++++++++++++++++- 11 files changed, 240 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 3511836f0f4a..2abb0857ede0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -221,6 +221,7 @@ enum mlx5e_priv_flag { MLX5E_PFLAG_RX_STRIDING_RQ, MLX5E_PFLAG_RX_NO_CSUM_COMPLETE, MLX5E_PFLAG_XDP_TX_MPWQE, + MLX5E_PFLAG_SKB_TX_MPWQE, MLX5E_NUM_PFLAGS, /* Keep last */ }; @@ -304,6 +305,7 @@ struct mlx5e_sq_dma { enum { MLX5E_SQ_STATE_ENABLED, + MLX5E_SQ_STATE_MPWQE, MLX5E_SQ_STATE_RECOVERING, MLX5E_SQ_STATE_IPSEC, MLX5E_SQ_STATE_AM, @@ -315,6 +317,7 @@ enum { struct mlx5e_tx_mpwqe { /* Current MPWQE session */ struct mlx5e_tx_wqe *wqe; + u32 bytes_count; u8 ds_count; u8 pkt_count; u8 inline_on; @@ -333,6 +336,7 @@ struct mlx5e_txqsq { u16 pc ____cacheline_aligned_in_smp; u16 skb_fifo_pc; u32 dma_fifo_pc; + struct mlx5e_tx_mpwqe mpwqe; struct mlx5e_cq cq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 1ac4607fba08..749881987094 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -297,6 +297,7 @@ static inline void mlx5e_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, } void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more); +void mlx5e_tx_mpwqe_ensure_complete(struct mlx5e_txqsq *sq); static inline bool mlx5e_tx_mpwqe_is_full(struct mlx5e_tx_mpwqe *session) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index adacc4f9a3bf..307eb64889c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -205,6 +205,7 @@ static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) *session = (struct mlx5e_tx_mpwqe) { .wqe = wqe, + .bytes_count = 0, .ds_count = MLX5E_TX_WQE_EMPTY_DS_COUNT, .pkt_count = 0, .inline_on = mlx5e_xdp_get_inline_state(sq, session->inline_on), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h index 4bd8af478a4a..d487e5e37162 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h @@ -147,6 +147,7 @@ mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq, u32 dma_len = xdptxd->len; session->pkt_count++; + session->bytes_count += dma_len; if (session->inline_on && dma_len <= MLX5E_XDP_INLINE_WQE_SZ_THRSD) { struct mlx5_wqe_inline_seg *inline_dseg = diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h index 23d4ef5ab9c5..2ea1cdc1ca54 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h @@ -128,31 +128,38 @@ static inline bool mlx5e_accel_tx_begin(struct net_device *dev, return true; } -static inline bool mlx5e_accel_tx_finish(struct mlx5e_priv *priv, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, - struct mlx5e_accel_tx_state *state) -{ -#ifdef CONFIG_MLX5_EN_TLS - mlx5e_tls_handle_tx_wqe(sq, &wqe->ctrl, &state->tls); -#endif +/* Part of the eseg touched by TX offloads */ +#define MLX5E_ACCEL_ESEG_LEN offsetof(struct mlx5_wqe_eth_seg, mss) +static inline bool mlx5e_accel_tx_eseg(struct mlx5e_priv *priv, + struct mlx5e_txqsq *sq, + struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg) +{ #ifdef CONFIG_MLX5_EN_IPSEC if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state)) { - if (unlikely(!mlx5e_ipsec_handle_tx_skb(priv, &wqe->eth, skb))) + if (unlikely(!mlx5e_ipsec_handle_tx_skb(priv, eseg, skb))) return false; } #endif #if IS_ENABLED(CONFIG_GENEVE) if (skb->encapsulation) - mlx5e_tx_tunnel_accel(skb, &wqe->eth); + mlx5e_tx_tunnel_accel(skb, eseg); #endif return true; } +static inline void mlx5e_accel_tx_finish(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe *wqe, + struct mlx5e_accel_tx_state *state) +{ +#ifdef CONFIG_MLX5_EN_TLS + mlx5e_tls_handle_tx_wqe(sq, &wqe->ctrl, &state->tls); +#endif +} + static inline int mlx5e_accel_init_rx(struct mlx5e_priv *priv) { return mlx5e_ktls_init_rx(priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c index c36560b3e93d..6982b193ee8a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c @@ -270,6 +270,8 @@ bool mlx5e_tls_handle_tx_skb(struct net_device *netdev, struct mlx5e_txqsq *sq, if (!datalen) return true; + mlx5e_tx_mpwqe_ensure_complete(sq); + tls_ctx = tls_get_ctx(skb->sk); if (WARN_ON_ONCE(tls_ctx->netdev != netdev)) goto err_out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 5cb1e4839eb7..2c34bb57048c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -1901,7 +1901,7 @@ static int set_pflag_rx_no_csum_complete(struct net_device *netdev, bool enable) return 0; } -static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) +static int set_pflag_tx_mpwqe_common(struct net_device *netdev, u32 flag, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -1913,7 +1913,7 @@ static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) new_channels.params = priv->channels.params; - MLX5E_SET_PFLAG(&new_channels.params, MLX5E_PFLAG_XDP_TX_MPWQE, enable); + MLX5E_SET_PFLAG(&new_channels.params, flag, enable); if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { priv->channels.params = new_channels.params; @@ -1924,6 +1924,16 @@ static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) return err; } +static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) +{ + return set_pflag_tx_mpwqe_common(netdev, MLX5E_PFLAG_XDP_TX_MPWQE, enable); +} + +static int set_pflag_skb_tx_mpwqe(struct net_device *netdev, bool enable) +{ + return set_pflag_tx_mpwqe_common(netdev, MLX5E_PFLAG_SKB_TX_MPWQE, enable); +} + static const struct pflag_desc mlx5e_priv_flags[MLX5E_NUM_PFLAGS] = { { "rx_cqe_moder", set_pflag_rx_cqe_based_moder }, { "tx_cqe_moder", set_pflag_tx_cqe_based_moder }, @@ -1931,6 +1941,7 @@ static const struct pflag_desc mlx5e_priv_flags[MLX5E_NUM_PFLAGS] = { { "rx_striding_rq", set_pflag_rx_striding_rq }, { "rx_no_csum_complete", set_pflag_rx_no_csum_complete }, { "xdp_tx_mpwqe", set_pflag_xdp_tx_mpwqe }, + { "skb_tx_mpwqe", set_pflag_skb_tx_mpwqe }, }; static int mlx5e_handle_pflag(struct net_device *netdev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index b413aa168e4e..f8ad4a724a63 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1075,6 +1075,12 @@ static int mlx5e_calc_sq_stop_room(struct mlx5e_txqsq *sq, u8 log_sq_size) sq->stop_room = mlx5e_tls_get_stop_room(sq); sq->stop_room += mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); + if (test_bit(MLX5E_SQ_STATE_MPWQE, &sq->state)) + /* A MPWQE can take up to the maximum-sized WQE + all the normal + * stop room can be taken if a new packet breaks the active + * MPWQE session and allocates its WQEs right away. + */ + sq->stop_room += mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); if (WARN_ON(sq->stop_room >= sq_size)) { netdev_err(sq->channel->netdev, "Stop room %hu is bigger than the SQ size %d\n", @@ -1116,6 +1122,8 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state); if (mlx5_accel_is_tls_device(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_TLS, &sq->state); + if (param->is_mpw) + set_bit(MLX5E_SQ_STATE_MPWQE, &sq->state); err = mlx5e_calc_sq_stop_room(sq, params->log_sq_size); if (err) return err; @@ -2168,6 +2176,7 @@ static void mlx5e_build_sq_param(struct mlx5e_priv *priv, 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); + param->is_mpw = MLX5E_GET_PFLAG(params, MLX5E_PFLAG_SKB_TX_MPWQE); mlx5e_build_tx_cq_param(priv, params, ¶m->cqp); } @@ -4721,6 +4730,8 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, 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, + MLX5_CAP_ETH(mdev, enhanced_multi_pkt_send_wqe)); /* XDP SQ */ MLX5E_SET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index e3b2f59408e6..20d7815ffbf4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -98,6 +98,8 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_bytes) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_added_vlan_packets) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_nop) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_mpwqe_blks) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_mpwqe_pkts) }, #ifdef CONFIG_MLX5_EN_TLS { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_encrypted_packets) }, @@ -353,6 +355,8 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(sw) s->tx_tso_inner_bytes += sq_stats->tso_inner_bytes; s->tx_added_vlan_packets += sq_stats->added_vlan_packets; s->tx_nop += sq_stats->nop; + s->tx_mpwqe_blks += sq_stats->mpwqe_blks; + s->tx_mpwqe_pkts += sq_stats->mpwqe_pkts; s->tx_queue_stopped += sq_stats->stopped; s->tx_queue_wake += sq_stats->wake; s->tx_queue_dropped += sq_stats->dropped; @@ -1527,6 +1531,8 @@ static const struct counter_desc sq_stats_desc[] = { { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, csum_partial_inner) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, added_vlan_packets) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, nop) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, mpwqe_blks) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, mpwqe_pkts) }, #ifdef CONFIG_MLX5_EN_TLS { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_packets) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_bytes) }, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 2e1cca1923b9..fd198965ba82 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -117,6 +117,8 @@ struct mlx5e_sw_stats { u64 tx_tso_inner_bytes; u64 tx_added_vlan_packets; u64 tx_nop; + u64 tx_mpwqe_blks; + u64 tx_mpwqe_pkts; u64 rx_lro_packets; u64 rx_lro_bytes; u64 rx_ecn_mark; @@ -345,6 +347,8 @@ struct mlx5e_sq_stats { u64 csum_partial_inner; u64 added_vlan_packets; u64 nop; + u64 mpwqe_blks; + u64 mpwqe_pkts; #ifdef CONFIG_MLX5_EN_TLS u64 tls_encrypted_packets; u64 tls_encrypted_bytes; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 3b68c8333875..d8f1acca37f8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -412,6 +412,166 @@ mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, dev_kfree_skb_any(skb); } +static inline bool mlx5e_tx_skb_supports_mpwqe(struct sk_buff *skb, struct mlx5e_tx_attr *attr) +{ + return !skb_is_nonlinear(skb) && !skb_vlan_tag_present(skb) && !attr->ihs; +} + +static inline bool mlx5e_tx_mpwqe_same_eseg(struct mlx5e_txqsq *sq, struct mlx5_wqe_eth_seg *eseg) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + + /* Assumes the session is already running and has at least one packet. */ + return !memcmp(&session->wqe->eth, eseg, MLX5E_ACCEL_ESEG_LEN); +} + +static void mlx5e_tx_mpwqe_session_start(struct mlx5e_txqsq *sq, + struct mlx5_wqe_eth_seg *eseg) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_wqe *wqe; + u16 pi; + + pi = mlx5e_txqsq_get_next_pi(sq, MLX5E_TX_MPW_MAX_WQEBBS); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + prefetchw(wqe->data); + + *session = (struct mlx5e_tx_mpwqe) { + .wqe = wqe, + .bytes_count = 0, + .ds_count = MLX5E_TX_WQE_EMPTY_DS_COUNT, + .pkt_count = 0, + .inline_on = 0, + }; + + memcpy(&session->wqe->eth, eseg, MLX5E_ACCEL_ESEG_LEN); + + sq->stats->mpwqe_blks++; +} + +static inline bool mlx5e_tx_mpwqe_session_is_active(struct mlx5e_txqsq *sq) +{ + return sq->mpwqe.wqe; +} + +static inline void mlx5e_tx_mpwqe_add_dseg(struct mlx5e_txqsq *sq, struct mlx5e_xmit_data *txd) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + struct mlx5_wqe_data_seg *dseg; + + dseg = (struct mlx5_wqe_data_seg *)session->wqe + session->ds_count; + + session->pkt_count++; + session->bytes_count += txd->len; + + dseg->addr = cpu_to_be64(txd->dma_addr); + dseg->byte_count = cpu_to_be32(txd->len); + dseg->lkey = sq->mkey_be; + session->ds_count++; + + sq->stats->mpwqe_pkts++; +} + +static struct mlx5_wqe_ctrl_seg *mlx5e_tx_mpwqe_session_complete(struct mlx5e_txqsq *sq) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + u8 ds_count = session->ds_count; + struct mlx5_wqe_ctrl_seg *cseg; + struct mlx5e_tx_wqe_info *wi; + u16 pi; + + cseg = &session->wqe->ctrl; + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_ENHANCED_MPSW); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_count); + + pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + wi = &sq->db.wqe_info[pi]; + *wi = (struct mlx5e_tx_wqe_info) { + .skb = NULL, + .num_bytes = session->bytes_count, + .num_wqebbs = DIV_ROUND_UP(ds_count, MLX5_SEND_WQEBB_NUM_DS), + .num_dma = session->pkt_count, + .num_fifo_pkts = session->pkt_count, + }; + + sq->pc += wi->num_wqebbs; + + session->wqe = NULL; + + mlx5e_tx_check_stop(sq); + + return cseg; +} + +static noinline void +mlx5e_sq_xmit_mpwqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg, bool xmit_more) +{ + struct mlx5_wqe_ctrl_seg *cseg; + struct mlx5e_xmit_data txd; + + if (!mlx5e_tx_mpwqe_session_is_active(sq)) { + mlx5e_tx_mpwqe_session_start(sq, eseg); + } else if (!mlx5e_tx_mpwqe_same_eseg(sq, eseg)) { + mlx5e_tx_mpwqe_session_complete(sq); + mlx5e_tx_mpwqe_session_start(sq, eseg); + } + + sq->stats->xmit_more += xmit_more; + + txd.data = skb->data; + txd.len = skb->len; + + txd.dma_addr = dma_map_single(sq->pdev, txd.data, txd.len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(sq->pdev, txd.dma_addr))) + goto err_unmap; + mlx5e_dma_push(sq, txd.dma_addr, txd.len, MLX5E_DMA_MAP_SINGLE); + + mlx5e_skb_fifo_push(sq, skb); + + mlx5e_tx_mpwqe_add_dseg(sq, &txd); + + mlx5e_tx_skb_update_hwts_flags(skb); + + if (unlikely(mlx5e_tx_mpwqe_is_full(&sq->mpwqe))) { + /* Might stop the queue and affect the retval of __netdev_tx_sent_queue. */ + cseg = mlx5e_tx_mpwqe_session_complete(sq); + + if (__netdev_tx_sent_queue(sq->txq, txd.len, xmit_more)) + mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, cseg); + } else if (__netdev_tx_sent_queue(sq->txq, txd.len, xmit_more)) { + /* Might stop the queue, but we were asked to ring the doorbell anyway. */ + cseg = mlx5e_tx_mpwqe_session_complete(sq); + + mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, cseg); + } + + return; + +err_unmap: + mlx5e_dma_unmap_wqe_err(sq, 1); + sq->stats->dropped++; + dev_kfree_skb_any(skb); +} + +void mlx5e_tx_mpwqe_ensure_complete(struct mlx5e_txqsq *sq) +{ + /* Unlikely in non-MPWQE workloads; not important in MPWQE workloads. */ + if (unlikely(mlx5e_tx_mpwqe_session_is_active(sq))) + mlx5e_tx_mpwqe_session_complete(sq); +} + +static inline bool mlx5e_txwqe_build_eseg(struct mlx5e_priv *priv, struct mlx5e_txqsq *sq, + struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg) +{ + if (unlikely(!mlx5e_accel_tx_eseg(priv, sq, skb, eseg))) + return false; + + mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); + + return true; +} + netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -426,21 +586,35 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) /* May send SKBs and WQEs. */ if (unlikely(!mlx5e_accel_tx_begin(dev, sq, skb, &accel))) - goto out; + return NETDEV_TX_OK; mlx5e_sq_xmit_prepare(sq, skb, &accel, &attr); + + if (test_bit(MLX5E_SQ_STATE_MPWQE, &sq->state)) { + if (mlx5e_tx_skb_supports_mpwqe(skb, &attr)) { + struct mlx5_wqe_eth_seg eseg = {}; + + if (unlikely(!mlx5e_txwqe_build_eseg(priv, sq, skb, &eseg))) + return NETDEV_TX_OK; + + mlx5e_sq_xmit_mpwqe(sq, skb, &eseg, netdev_xmit_more()); + return NETDEV_TX_OK; + } + + mlx5e_tx_mpwqe_ensure_complete(sq); + } + mlx5e_sq_calc_wqe_attr(skb, &attr, &wqe_attr); pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); wqe = MLX5E_TX_FETCH_WQE(sq, pi); /* May update the WQE, but may not post other WQEs. */ - if (unlikely(!mlx5e_accel_tx_finish(priv, sq, skb, wqe, &accel))) - goto out; + mlx5e_accel_tx_finish(sq, wqe, &accel); + if (unlikely(!mlx5e_txwqe_build_eseg(priv, sq, skb, &wqe->eth))) + return NETDEV_TX_OK; - mlx5e_txwqe_build_eseg_csum(sq, skb, &wqe->eth); mlx5e_sq_xmit_wqe(sq, skb, &attr, &wqe_attr, wqe, pi, netdev_xmit_more()); -out: return NETDEV_TX_OK; }