From: Vladimir Oltean <vladimir.oltean@nxp.com> To: netdev@vger.kernel.org, Jakub Kicinski <kuba@kernel.org>, "David S. Miller" <davem@davemloft.net> Cc: Andrew Lunn <andrew@lunn.ch>, Florian Fainelli <f.fainelli@gmail.com>, Vivien Didelot <vivien.didelot@gmail.com>, Jiri Pirko <jiri@resnulli.us>, Ido Schimmel <idosch@idosch.org>, Tobias Waldekranz <tobias@waldekranz.com>, Roopa Prabhu <roopa@nvidia.com>, Nikolay Aleksandrov <nikolay@nvidia.com>, Stephen Hemminger <stephen@networkplumber.org>, bridge@lists.linux-foundation.org, Alexander Duyck <alexander.duyck@gmail.com> Subject: [RFC PATCH v2 net-next 10/10] net: dsa: tag_dsa: offload the bridge forwarding process Date: Sat, 3 Jul 2021 14:57:05 +0300 [thread overview] Message-ID: <20210703115705.1034112-11-vladimir.oltean@nxp.com> (raw) In-Reply-To: <20210703115705.1034112-1-vladimir.oltean@nxp.com> From: Tobias Waldekranz <tobias@waldekranz.com> Allow the DSA tagger to generate FORWARD frames for offloaded skbs sent from a bridge that we offload, allowing the switch to handle any frame replication that may be required. This also means that source address learning takes place on packets sent from the CPU, meaning that return traffic no longer needs to be flooded as unknown unicast. Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> --- net/dsa/dsa_priv.h | 11 +++++++++ net/dsa/tag_dsa.c | 60 +++++++++++++++++++++++++++++++++++++++------- 2 files changed, 63 insertions(+), 8 deletions(-) diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index c577338b5bb7..c070157cd967 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -389,6 +389,17 @@ static inline struct sk_buff *dsa_untag_bridge_pvid(struct sk_buff *skb) return skb; } +static inline struct net_device * +dsa_slave_get_sb_dev(const struct net_device *dev, struct sk_buff *skb) +{ + u16 queue_mapping = skb_get_queue_mapping(skb); + struct netdev_queue *txq; + + txq = netdev_get_tx_queue(dev, queue_mapping); + + return txq->sb_dev; +} + /* switch.c */ int dsa_switch_register_notifier(struct dsa_switch *ds); void dsa_switch_unregister_notifier(struct dsa_switch *ds); diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c index a822355afc90..9151ed141b3e 100644 --- a/net/dsa/tag_dsa.c +++ b/net/dsa/tag_dsa.c @@ -125,8 +125,49 @@ enum dsa_code { static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, u8 extra) { + struct net_device *sb_dev = dsa_slave_get_sb_dev(dev, skb); struct dsa_port *dp = dsa_slave_to_port(dev); + u8 tag_dev, tag_port; + enum dsa_cmd cmd; u8 *dsa_header; + u16 pvid = 0; + int err; + + if (sb_dev) { + /* Don't bother finding the accel_priv corresponding with this + * subordinate device, we know it's the bridge becase we can't + * offload anything else, so just search for it under the port, + * we know it's the same. + */ + struct dsa_bridge_fwd_accel_priv *accel_priv = dp->accel_priv; + struct dsa_switch_tree *dst = dp->ds->dst; + + cmd = DSA_CMD_FORWARD; + + /* When offloading forwarding for a bridge, inject FORWARD + * packets on behalf of a virtual switch device with an index + * past the physical switches. + */ + tag_dev = dst->last_switch + 1 + accel_priv->bridge_num; + tag_port = 0; + + /* If we are offloading forwarding for a VLAN-unaware bridge, + * inject packets to hardware using the bridge's pvid, since + * that's where the packets ingressed from. + */ + if (!br_vlan_enabled(sb_dev)) { + /* Safe because __dev_queue_xmit() runs under + * rcu_read_lock_bh() + */ + err = br_vlan_get_pvid_rcu(sb_dev, &pvid); + if (err) + return NULL; + } + } else { + cmd = DSA_CMD_FROM_CPU; + tag_dev = dp->ds->index; + tag_port = dp->index; + } if (skb->protocol == htons(ETH_P_8021Q)) { if (extra) { @@ -134,10 +175,10 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, memmove(skb->data, skb->data + extra, 2 * ETH_ALEN); } - /* Construct tagged FROM_CPU DSA tag from 802.1Q tag. */ + /* Construct tagged DSA tag from 802.1Q tag. */ dsa_header = skb->data + 2 * ETH_ALEN + extra; - dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | 0x20 | dp->ds->index; - dsa_header[1] = dp->index << 3; + dsa_header[0] = (cmd << 6) | 0x20 | tag_dev; + dsa_header[1] = tag_port << 3; /* Move CFI field from byte 2 to byte 1. */ if (dsa_header[2] & 0x10) { @@ -148,12 +189,13 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, skb_push(skb, DSA_HLEN + extra); memmove(skb->data, skb->data + DSA_HLEN + extra, 2 * ETH_ALEN); - /* Construct untagged FROM_CPU DSA tag. */ + /* Construct untagged DSA tag. */ dsa_header = skb->data + 2 * ETH_ALEN + extra; - dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | dp->ds->index; - dsa_header[1] = dp->index << 3; - dsa_header[2] = 0x00; - dsa_header[3] = 0x00; + + dsa_header[0] = (cmd << 6) | tag_dev; + dsa_header[1] = tag_port << 3; + dsa_header[2] = pvid >> 8; + dsa_header[3] = pvid & 0xff; } return skb; @@ -304,6 +346,7 @@ static const struct dsa_device_ops dsa_netdev_ops = { .xmit = dsa_xmit, .rcv = dsa_rcv, .needed_headroom = DSA_HLEN, + .bridge_fwd_offload = true, }; DSA_TAG_DRIVER(dsa_netdev_ops); @@ -347,6 +390,7 @@ static const struct dsa_device_ops edsa_netdev_ops = { .xmit = edsa_xmit, .rcv = edsa_rcv, .needed_headroom = EDSA_HLEN, + .bridge_fwd_offload = true, }; DSA_TAG_DRIVER(edsa_netdev_ops); -- 2.25.1
WARNING: multiple messages have this Message-ID (diff)
From: Vladimir Oltean <vladimir.oltean@nxp.com> To: netdev@vger.kernel.org, Jakub Kicinski <kuba@kernel.org>, "David S. Miller" <davem@davemloft.net> Cc: Andrew Lunn <andrew@lunn.ch>, Florian Fainelli <f.fainelli@gmail.com>, Jiri Pirko <jiri@resnulli.us>, bridge@lists.linux-foundation.org, Alexander Duyck <alexander.duyck@gmail.com>, Ido Schimmel <idosch@idosch.org>, Nikolay Aleksandrov <nikolay@nvidia.com>, Roopa Prabhu <roopa@nvidia.com>, Vivien Didelot <vivien.didelot@gmail.com>, Tobias Waldekranz <tobias@waldekranz.com> Subject: [Bridge] [RFC PATCH v2 net-next 10/10] net: dsa: tag_dsa: offload the bridge forwarding process Date: Sat, 3 Jul 2021 14:57:05 +0300 [thread overview] Message-ID: <20210703115705.1034112-11-vladimir.oltean@nxp.com> (raw) In-Reply-To: <20210703115705.1034112-1-vladimir.oltean@nxp.com> From: Tobias Waldekranz <tobias@waldekranz.com> Allow the DSA tagger to generate FORWARD frames for offloaded skbs sent from a bridge that we offload, allowing the switch to handle any frame replication that may be required. This also means that source address learning takes place on packets sent from the CPU, meaning that return traffic no longer needs to be flooded as unknown unicast. Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> --- net/dsa/dsa_priv.h | 11 +++++++++ net/dsa/tag_dsa.c | 60 +++++++++++++++++++++++++++++++++++++++------- 2 files changed, 63 insertions(+), 8 deletions(-) diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index c577338b5bb7..c070157cd967 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -389,6 +389,17 @@ static inline struct sk_buff *dsa_untag_bridge_pvid(struct sk_buff *skb) return skb; } +static inline struct net_device * +dsa_slave_get_sb_dev(const struct net_device *dev, struct sk_buff *skb) +{ + u16 queue_mapping = skb_get_queue_mapping(skb); + struct netdev_queue *txq; + + txq = netdev_get_tx_queue(dev, queue_mapping); + + return txq->sb_dev; +} + /* switch.c */ int dsa_switch_register_notifier(struct dsa_switch *ds); void dsa_switch_unregister_notifier(struct dsa_switch *ds); diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c index a822355afc90..9151ed141b3e 100644 --- a/net/dsa/tag_dsa.c +++ b/net/dsa/tag_dsa.c @@ -125,8 +125,49 @@ enum dsa_code { static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, u8 extra) { + struct net_device *sb_dev = dsa_slave_get_sb_dev(dev, skb); struct dsa_port *dp = dsa_slave_to_port(dev); + u8 tag_dev, tag_port; + enum dsa_cmd cmd; u8 *dsa_header; + u16 pvid = 0; + int err; + + if (sb_dev) { + /* Don't bother finding the accel_priv corresponding with this + * subordinate device, we know it's the bridge becase we can't + * offload anything else, so just search for it under the port, + * we know it's the same. + */ + struct dsa_bridge_fwd_accel_priv *accel_priv = dp->accel_priv; + struct dsa_switch_tree *dst = dp->ds->dst; + + cmd = DSA_CMD_FORWARD; + + /* When offloading forwarding for a bridge, inject FORWARD + * packets on behalf of a virtual switch device with an index + * past the physical switches. + */ + tag_dev = dst->last_switch + 1 + accel_priv->bridge_num; + tag_port = 0; + + /* If we are offloading forwarding for a VLAN-unaware bridge, + * inject packets to hardware using the bridge's pvid, since + * that's where the packets ingressed from. + */ + if (!br_vlan_enabled(sb_dev)) { + /* Safe because __dev_queue_xmit() runs under + * rcu_read_lock_bh() + */ + err = br_vlan_get_pvid_rcu(sb_dev, &pvid); + if (err) + return NULL; + } + } else { + cmd = DSA_CMD_FROM_CPU; + tag_dev = dp->ds->index; + tag_port = dp->index; + } if (skb->protocol == htons(ETH_P_8021Q)) { if (extra) { @@ -134,10 +175,10 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, memmove(skb->data, skb->data + extra, 2 * ETH_ALEN); } - /* Construct tagged FROM_CPU DSA tag from 802.1Q tag. */ + /* Construct tagged DSA tag from 802.1Q tag. */ dsa_header = skb->data + 2 * ETH_ALEN + extra; - dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | 0x20 | dp->ds->index; - dsa_header[1] = dp->index << 3; + dsa_header[0] = (cmd << 6) | 0x20 | tag_dev; + dsa_header[1] = tag_port << 3; /* Move CFI field from byte 2 to byte 1. */ if (dsa_header[2] & 0x10) { @@ -148,12 +189,13 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev, skb_push(skb, DSA_HLEN + extra); memmove(skb->data, skb->data + DSA_HLEN + extra, 2 * ETH_ALEN); - /* Construct untagged FROM_CPU DSA tag. */ + /* Construct untagged DSA tag. */ dsa_header = skb->data + 2 * ETH_ALEN + extra; - dsa_header[0] = (DSA_CMD_FROM_CPU << 6) | dp->ds->index; - dsa_header[1] = dp->index << 3; - dsa_header[2] = 0x00; - dsa_header[3] = 0x00; + + dsa_header[0] = (cmd << 6) | tag_dev; + dsa_header[1] = tag_port << 3; + dsa_header[2] = pvid >> 8; + dsa_header[3] = pvid & 0xff; } return skb; @@ -304,6 +346,7 @@ static const struct dsa_device_ops dsa_netdev_ops = { .xmit = dsa_xmit, .rcv = dsa_rcv, .needed_headroom = DSA_HLEN, + .bridge_fwd_offload = true, }; DSA_TAG_DRIVER(dsa_netdev_ops); @@ -347,6 +390,7 @@ static const struct dsa_device_ops edsa_netdev_ops = { .xmit = edsa_xmit, .rcv = edsa_rcv, .needed_headroom = EDSA_HLEN, + .bridge_fwd_offload = true, }; DSA_TAG_DRIVER(edsa_netdev_ops); -- 2.25.1
next prev parent reply other threads:[~2021-07-03 11:58 UTC|newest] Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-07-03 11:56 [RFC PATCH v2 net-next 00/10] Allow forwarding for the software bridge data path to be offloaded to capable devices Vladimir Oltean 2021-07-03 11:56 ` [Bridge] " Vladimir Oltean 2021-07-03 11:56 ` [RFC PATCH v2 net-next 01/10] net: dfwd: constrain existing users to macvlan subordinates Vladimir Oltean 2021-07-03 11:56 ` [Bridge] " Vladimir Oltean 2021-07-03 11:56 ` [RFC PATCH v2 net-next 02/10] net: bridge: disambiguate offload_fwd_mark Vladimir Oltean 2021-07-03 11:56 ` [Bridge] " Vladimir Oltean 2021-07-09 13:23 ` Grygorii Strashko 2021-07-03 11:56 ` [RFC PATCH v2 net-next 03/10] net: bridge: switchdev: recycle unused hwdoms Vladimir Oltean 2021-07-03 11:56 ` [Bridge] " Vladimir Oltean 2021-07-03 11:56 ` [RFC PATCH v2 net-next 04/10] net: bridge: switchdev: allow the data plane forwarding to be offloaded Vladimir Oltean 2021-07-03 11:56 ` [Bridge] " Vladimir Oltean 2021-07-09 13:16 ` Grygorii Strashko 2021-07-09 14:09 ` Vladimir Oltean 2021-07-09 14:09 ` [Bridge] " Vladimir Oltean 2021-07-12 12:28 ` Tobias Waldekranz 2021-07-12 12:28 ` [Bridge] " Tobias Waldekranz 2021-07-12 13:03 ` Vladimir Oltean 2021-07-12 13:03 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` [RFC PATCH v2 net-next 05/10] net: extract helpers for binding a subordinate device to TX queues Vladimir Oltean 2021-07-03 11:57 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` [RFC PATCH v2 net-next 06/10] net: allow ndo_select_queue to go beyond dev->num_real_tx_queues Vladimir Oltean 2021-07-03 11:57 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` [RFC PATCH v2 net-next 07/10] net: dsa: track the number of switches in a tree Vladimir Oltean 2021-07-03 11:57 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` [RFC PATCH v2 net-next 08/10] net: dsa: add support for bridge forwarding offload Vladimir Oltean 2021-07-03 11:57 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` [RFC PATCH v2 net-next 09/10] net: dsa: mv88e6xxx: map virtual bridges with forwarding offload in the PVT Vladimir Oltean 2021-07-03 11:57 ` [Bridge] " Vladimir Oltean 2021-07-03 11:57 ` Vladimir Oltean [this message] 2021-07-03 11:57 ` [Bridge] [RFC PATCH v2 net-next 10/10] net: dsa: tag_dsa: offload the bridge forwarding process Vladimir Oltean 2021-07-03 22:04 ` [RFC PATCH v2 net-next 00/10] Allow forwarding for the software bridge data path to be offloaded to capable devices Tobias Waldekranz 2021-07-03 22:04 ` [Bridge] " Tobias Waldekranz 2021-07-04 8:11 ` Vladimir Oltean 2021-07-04 8:11 ` [Bridge] " Vladimir Oltean 2021-07-05 8:09 ` Tobias Waldekranz 2021-07-05 8:09 ` [Bridge] " Tobias Waldekranz 2021-07-05 8:54 ` Vladimir Oltean 2021-07-05 8:54 ` [Bridge] " Vladimir Oltean 2021-07-05 4:20 ` DENG Qingfang 2021-07-05 4:20 ` [Bridge] " DENG Qingfang 2021-07-05 8:32 ` Tobias Waldekranz 2021-07-05 8:32 ` [Bridge] " Tobias Waldekranz 2021-07-05 9:57 ` Vladimir Oltean 2021-07-05 9:57 ` [Bridge] " Vladimir Oltean
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=20210703115705.1034112-11-vladimir.oltean@nxp.com \ --to=vladimir.oltean@nxp.com \ --cc=alexander.duyck@gmail.com \ --cc=andrew@lunn.ch \ --cc=bridge@lists.linux-foundation.org \ --cc=davem@davemloft.net \ --cc=f.fainelli@gmail.com \ --cc=idosch@idosch.org \ --cc=jiri@resnulli.us \ --cc=kuba@kernel.org \ --cc=netdev@vger.kernel.org \ --cc=nikolay@nvidia.com \ --cc=roopa@nvidia.com \ --cc=stephen@networkplumber.org \ --cc=tobias@waldekranz.com \ --cc=vivien.didelot@gmail.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.