From: Jose Abreu <Jose.Abreu@synopsys.com> To: netdev@vger.kernel.org Cc: Joao Pinto <Joao.Pinto@synopsys.com>, Jose Abreu <Jose.Abreu@synopsys.com>, Giuseppe Cavallaro <peppe.cavallaro@st.com>, Alexandre Torgue <alexandre.torgue@st.com>, Jose Abreu <Jose.Abreu@synopsys.com>, "David S. Miller" <davem@davemloft.net>, Maxime Coquelin <mcoquelin.stm32@gmail.com>, linux-stm32@st-md-mailman.stormreply.com, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org Subject: [PATCH net-next 6/6] net: stmmac: Implement UDP Segmentation Offload Date: Mon, 11 Nov 2019 15:42:39 +0100 [thread overview] Message-ID: <7a62a0532edaec95312308d7a4ced7eae331bc6d.1573482992.git.Jose.Abreu@synopsys.com> (raw) In-Reply-To: <cover.1573482991.git.Jose.Abreu@synopsys.com> In-Reply-To: <cover.1573482991.git.Jose.Abreu@synopsys.com> Implement the UDP Segmentation Offload feature in stmmac. This is only available in GMAC4+ cores. Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com> --- Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com> Cc: Alexandre Torgue <alexandre.torgue@st.com> Cc: Jose Abreu <joabreu@synopsys.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: netdev@vger.kernel.org Cc: linux-stm32@st-md-mailman.stormreply.com Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 32 ++++++++++++++++------- 1 file changed, 23 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a2fac7772666..39b4efd521f9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -36,6 +36,7 @@ #endif /* CONFIG_DEBUG_FS */ #include <linux/net_tstamp.h> #include <linux/phylink.h> +#include <linux/udp.h> #include <net/pkt_cls.h> #include "stmmac_ptp.h" #include "stmmac.h" @@ -2916,9 +2917,9 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) u32 queue = skb_get_queue_mapping(skb); struct stmmac_tx_queue *tx_q; unsigned int first_entry; + u8 proto_hdr_len, hdr; int tmp_pay_len = 0; u32 pay_len, mss; - u8 proto_hdr_len; dma_addr_t des; bool has_vlan; int i; @@ -2926,7 +2927,13 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) tx_q = &priv->tx_queue[queue]; /* Compute header lengths */ - proto_hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) { + proto_hdr_len = skb_transport_offset(skb) + sizeof(struct udphdr); + hdr = sizeof(struct udphdr); + } else { + proto_hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + hdr = tcp_hdrlen(skb); + } /* Desc availability based on threshold should be enough safe */ if (unlikely(stmmac_tx_avail(priv, queue) < @@ -2956,8 +2963,8 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) } if (netif_msg_tx_queued(priv)) { - pr_info("%s: tcphdrlen %d, hdr_len %d, pay_len %d, mss %d\n", - __func__, tcp_hdrlen(skb), proto_hdr_len, pay_len, mss); + pr_info("%s: hdrlen %d, hdr_len %d, pay_len %d, mss %d\n", + __func__, hdr, proto_hdr_len, pay_len, mss); pr_info("\tskb->len %d, skb->data_len %d\n", skb->len, skb->data_len); } @@ -3071,7 +3078,7 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) proto_hdr_len, pay_len, 1, tx_q->tx_skbuff_dma[first_entry].last_segment, - tcp_hdrlen(skb) / 4, (skb->len - proto_hdr_len)); + hdr / 4, (skb->len - proto_hdr_len)); /* If context desc is used to change MSS */ if (mss_desc) { @@ -3130,6 +3137,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) int i, csum_insertion = 0, is_jumbo = 0; u32 queue = skb_get_queue_mapping(skb); int nfrags = skb_shinfo(skb)->nr_frags; + int gso = skb_shinfo(skb)->gso_type; struct dma_desc *desc, *first; struct stmmac_tx_queue *tx_q; unsigned int first_entry; @@ -3145,7 +3153,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) /* Manage oversized TCP frames for GMAC4 device */ if (skb_is_gso(skb) && priv->tso) { - if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) + if (gso & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) + return stmmac_tso_xmit(skb, dev); + if (priv->plat->has_gmac4 && (gso & SKB_GSO_UDP_L4)) return stmmac_tso_xmit(skb, dev); } @@ -4036,11 +4046,13 @@ static int stmmac_setup_tc(struct net_device *ndev, enum tc_setup_type type, static u16 stmmac_select_queue(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev) { - if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) { + int gso = skb_shinfo(skb)->gso_type; + + if (gso & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6 | SKB_GSO_UDP_L4)) { /* - * There is no way to determine the number of TSO + * There is no way to determine the number of TSO/USO * capable Queues. Let's use always the Queue 0 - * because if TSO is supported then at least this + * because if TSO/USO is supported then at least this * one will be capable. */ return 0; @@ -4555,6 +4567,8 @@ int stmmac_dvr_probe(struct device *device, if ((priv->plat->tso_en) && (priv->dma_cap.tsoen)) { ndev->hw_features |= NETIF_F_TSO | NETIF_F_TSO6; + if (priv->plat->has_gmac4) + ndev->hw_features |= NETIF_F_GSO_UDP_L4; priv->tso = true; dev_info(priv->device, "TSO feature enabled\n"); } -- 2.7.4
WARNING: multiple messages have this Message-ID (diff)
From: Jose Abreu <Jose.Abreu@synopsys.com> To: netdev@vger.kernel.org Cc: Jose Abreu <Jose.Abreu@synopsys.com>, Joao Pinto <Joao.Pinto@synopsys.com>, Alexandre Torgue <alexandre.torgue@st.com>, linux-kernel@vger.kernel.org, linux-stm32@st-md-mailman.stormreply.com, Maxime Coquelin <mcoquelin.stm32@gmail.com>, Giuseppe Cavallaro <peppe.cavallaro@st.com>, "David S. Miller" <davem@davemloft.net>, linux-arm-kernel@lists.infradead.org Subject: [PATCH net-next 6/6] net: stmmac: Implement UDP Segmentation Offload Date: Mon, 11 Nov 2019 15:42:39 +0100 [thread overview] Message-ID: <7a62a0532edaec95312308d7a4ced7eae331bc6d.1573482992.git.Jose.Abreu@synopsys.com> (raw) In-Reply-To: <cover.1573482991.git.Jose.Abreu@synopsys.com> In-Reply-To: <cover.1573482991.git.Jose.Abreu@synopsys.com> Implement the UDP Segmentation Offload feature in stmmac. This is only available in GMAC4+ cores. Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com> --- Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com> Cc: Alexandre Torgue <alexandre.torgue@st.com> Cc: Jose Abreu <joabreu@synopsys.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: netdev@vger.kernel.org Cc: linux-stm32@st-md-mailman.stormreply.com Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 32 ++++++++++++++++------- 1 file changed, 23 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a2fac7772666..39b4efd521f9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -36,6 +36,7 @@ #endif /* CONFIG_DEBUG_FS */ #include <linux/net_tstamp.h> #include <linux/phylink.h> +#include <linux/udp.h> #include <net/pkt_cls.h> #include "stmmac_ptp.h" #include "stmmac.h" @@ -2916,9 +2917,9 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) u32 queue = skb_get_queue_mapping(skb); struct stmmac_tx_queue *tx_q; unsigned int first_entry; + u8 proto_hdr_len, hdr; int tmp_pay_len = 0; u32 pay_len, mss; - u8 proto_hdr_len; dma_addr_t des; bool has_vlan; int i; @@ -2926,7 +2927,13 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) tx_q = &priv->tx_queue[queue]; /* Compute header lengths */ - proto_hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) { + proto_hdr_len = skb_transport_offset(skb) + sizeof(struct udphdr); + hdr = sizeof(struct udphdr); + } else { + proto_hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + hdr = tcp_hdrlen(skb); + } /* Desc availability based on threshold should be enough safe */ if (unlikely(stmmac_tx_avail(priv, queue) < @@ -2956,8 +2963,8 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) } if (netif_msg_tx_queued(priv)) { - pr_info("%s: tcphdrlen %d, hdr_len %d, pay_len %d, mss %d\n", - __func__, tcp_hdrlen(skb), proto_hdr_len, pay_len, mss); + pr_info("%s: hdrlen %d, hdr_len %d, pay_len %d, mss %d\n", + __func__, hdr, proto_hdr_len, pay_len, mss); pr_info("\tskb->len %d, skb->data_len %d\n", skb->len, skb->data_len); } @@ -3071,7 +3078,7 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) proto_hdr_len, pay_len, 1, tx_q->tx_skbuff_dma[first_entry].last_segment, - tcp_hdrlen(skb) / 4, (skb->len - proto_hdr_len)); + hdr / 4, (skb->len - proto_hdr_len)); /* If context desc is used to change MSS */ if (mss_desc) { @@ -3130,6 +3137,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) int i, csum_insertion = 0, is_jumbo = 0; u32 queue = skb_get_queue_mapping(skb); int nfrags = skb_shinfo(skb)->nr_frags; + int gso = skb_shinfo(skb)->gso_type; struct dma_desc *desc, *first; struct stmmac_tx_queue *tx_q; unsigned int first_entry; @@ -3145,7 +3153,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) /* Manage oversized TCP frames for GMAC4 device */ if (skb_is_gso(skb) && priv->tso) { - if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) + if (gso & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) + return stmmac_tso_xmit(skb, dev); + if (priv->plat->has_gmac4 && (gso & SKB_GSO_UDP_L4)) return stmmac_tso_xmit(skb, dev); } @@ -4036,11 +4046,13 @@ static int stmmac_setup_tc(struct net_device *ndev, enum tc_setup_type type, static u16 stmmac_select_queue(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev) { - if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) { + int gso = skb_shinfo(skb)->gso_type; + + if (gso & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6 | SKB_GSO_UDP_L4)) { /* - * There is no way to determine the number of TSO + * There is no way to determine the number of TSO/USO * capable Queues. Let's use always the Queue 0 - * because if TSO is supported then at least this + * because if TSO/USO is supported then at least this * one will be capable. */ return 0; @@ -4555,6 +4567,8 @@ int stmmac_dvr_probe(struct device *device, if ((priv->plat->tso_en) && (priv->dma_cap.tsoen)) { ndev->hw_features |= NETIF_F_TSO | NETIF_F_TSO6; + if (priv->plat->has_gmac4) + ndev->hw_features |= NETIF_F_GSO_UDP_L4; priv->tso = true; dev_info(priv->device, "TSO feature enabled\n"); } -- 2.7.4 _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-11-11 14:43 UTC|newest] Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-11-11 14:42 [PATCH net-next 0/6] net: stmmac: Improvements for -next Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` [PATCH net-next 1/6] net: stmmac: Fix sparse warning Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` [PATCH net-next 2/6] net: stmmac: gmac4+: Add Split Header support Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` [PATCH net-next 3/6] net: stmmac: xgmac: Add C45 PHY support in the MDIO callbacks Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` [PATCH net-next 4/6] net: stmmac: tc: Remove the speed dependency Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` [PATCH net-next 5/6] net: stmmac: Rework stmmac_rx() Jose Abreu 2019-11-11 14:42 ` Jose Abreu 2019-11-11 14:42 ` Jose Abreu [this message] 2019-11-11 14:42 ` [PATCH net-next 6/6] net: stmmac: Implement UDP Segmentation Offload Jose Abreu 2019-11-12 7:13 ` [PATCH net-next 0/6] net: stmmac: Improvements for -next David Miller 2019-11-12 7:13 ` David Miller
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=7a62a0532edaec95312308d7a4ced7eae331bc6d.1573482992.git.Jose.Abreu@synopsys.com \ --to=jose.abreu@synopsys.com \ --cc=Joao.Pinto@synopsys.com \ --cc=alexandre.torgue@st.com \ --cc=davem@davemloft.net \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-stm32@st-md-mailman.stormreply.com \ --cc=mcoquelin.stm32@gmail.com \ --cc=netdev@vger.kernel.org \ --cc=peppe.cavallaro@st.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.