Merge tag 'net-next-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from Jakub Kicinski:
 "Core:

   - Enable memcg accounting for various networking objects.

  BPF:

   - Introduce bpf timers.

   - Add perf link and opaque bpf_cookie which the program can read out
     again, to be used in libbpf-based USDT library.

   - Add bpf_task_pt_regs() helper to access user space pt_regs in
     kprobes, to help user space stack unwinding.

   - Add support for UNIX sockets for BPF sockmap.

   - Extend BPF iterator support for UNIX domain sockets.

   - Allow BPF TCP congestion control progs and bpf iterators to call
     bpf_setsockopt(), e.g. to switch to another congestion control
     algorithm.

  Protocols:

   - Support IOAM Pre-allocated Trace with IPv6.

   - Support Management Component Transport Protocol.

   - bridge: multicast: add vlan support.

   - netfilter: add hooks for the SRv6 lightweight tunnel driver.

   - tcp:
       - enable mid-stream window clamping (by user space or BPF)
       - allow data-less, empty-cookie SYN with TFO_SERVER_COOKIE_NOT_REQD
       - more accurate DSACK processing for RACK-TLP

   - mptcp:
       - add full mesh path manager option
       - add partial support for MP_FAIL
       - improve use of backup subflows
       - optimize option processing

   - af_unix: add OOB notification support.

   - ipv6: add IFLA_INET6_RA_MTU to expose MTU value advertised by the
     router.

   - mac80211: Target Wake Time support in AP mode.

   - can: j1939: extend UAPI to notify about RX status.

  Driver APIs:

   - Add page frag support in page pool API.

   - Many improvements to the DSA (distributed switch) APIs.

   - ethtool: extend IRQ coalesce uAPI with timer reset modes.

   - devlink: control which auxiliary devices are created.

   - Support CAN PHYs via the generic PHY subsystem.

   - Proper cross-chip support for tag_8021q.

   - Allow TX forwarding for the software bridge data path to be
     offloaded to capable devices.

  Drivers:

   - veth: more flexible channels number configuration.

   - openvswitch: introduce per-cpu upcall dispatch.

   - Add internet mix (IMIX) mode to pktgen.

   - Transparently handle XDP operations in the bonding driver.

   - Add LiteETH network driver.

   - Renesas (ravb):
       - support Gigabit Ethernet IP

   - NXP Ethernet switch (sja1105):
       - fast aging support
       - support for "H" switch topologies
       - traffic termination for ports under VLAN-aware bridge

   - Intel 1G Ethernet
       - support getcrosststamp() with PCIe PTM (Precision Time
         Measurement) for better time sync
       - support Credit-Based Shaper (CBS) offload, enabling HW traffic
         prioritization and bandwidth reservation

   - Broadcom Ethernet (bnxt)
       - support pulse-per-second output
       - support larger Rx rings

   - Mellanox Ethernet (mlx5)
       - support ethtool RSS contexts and MQPRIO channel mode
       - support LAG offload with bridging
       - support devlink rate limit API
       - support packet sampling on tunnels

   - Huawei Ethernet (hns3):
       - basic devlink support
       - add extended IRQ coalescing support
       - report extended link state

   - Netronome Ethernet (nfp):
       - add conntrack offload support

   - Broadcom WiFi (brcmfmac):
       - add WPA3 Personal with FT to supported cipher suites
       - support 43752 SDIO device

   - Intel WiFi (iwlwifi):
       - support scanning hidden 6GHz networks
       - support for a new hardware family (Bz)

   - Xen pv driver:
       - harden netfront against malicious backends

   - Qualcomm mobile
       - ipa: refactor power management and enable automatic suspend
       - mhi: move MBIM to WWAN subsystem interfaces

  Refactor:

   - Ambient BPF run context and cgroup storage cleanup.

   - Compat rework for ndo_ioctl.

  Old code removal:

   - prism54 remove the obsoleted driver, deprecated by the p54 driver.

   - wan: remove sbni/granch driver"

* tag 'net-next-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1715 commits)
  net: Add depends on OF_NET for LiteX's LiteETH
  ipv6: seg6: remove duplicated include
  net: hns3: remove unnecessary spaces
  net: hns3: add some required spaces
  net: hns3: clean up a type mismatch warning
  net: hns3: refine function hns3_set_default_feature()
  ipv6: remove duplicated 'net/lwtunnel.h' include
  net: w5100: check return value after calling platform_get_resource()
  net/mlxbf_gige: Make use of devm_platform_ioremap_resourcexxx()
  net: mdio: mscc-miim: Make use of the helper function devm_platform_ioremap_resource()
  net: mdio-ipq4019: Make use of devm_platform_ioremap_resource()
  fou: remove sparse errors
  ipv4: fix endianness issue in inet_rtm_getroute_build_skb()
  octeontx2-af: Set proper errorcode for IPv4 checksum errors
  octeontx2-af: Fix static code analyzer reported issues
  octeontx2-af: Fix mailbox errors in nix_rss_flowkey_cfg
  octeontx2-af: Fix loop in free and unmap counter
  af_unix: fix potential NULL deref in unix_dgram_connect()
  dpaa2-eth: Replace strlcpy with strscpy
  octeontx2-af: Use NDC TX for transmit packet data
  ...

1 files changed, 66 insertions, 29 deletions

diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c
index a822355afc90..77d0ce89ab77 100644
--- a/net/dsa/tag_dsa.c
+++ b/net/dsa/tag_dsa.c
@@ -126,18 +126,53 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 				   u8 extra)
 {
 	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 (skb->offload_fwd_mark) {
+		struct dsa_switch_tree *dst = dp->ds->dst;
+		struct net_device *br = dp->bridge_dev;
+
+		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 + dp->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(br)) {
+			/* Safe because __dev_queue_xmit() runs under
+			 * rcu_read_lock_bh()
+			 */
+			err = br_vlan_get_pvid_rcu(br, &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) {
 			skb_push(skb, extra);
-			memmove(skb->data, skb->data + extra, 2 * ETH_ALEN);
+			dsa_alloc_etype_header(skb, extra);
 		}
 
-		/* Construct tagged FROM_CPU 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;
+		/* Construct tagged DSA tag from 802.1Q tag. */
+		dsa_header = dsa_etype_header_pos_tx(skb) + extra;
+		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) {
@@ -146,14 +181,15 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 		}
 	} else {
 		skb_push(skb, DSA_HLEN + extra);
-		memmove(skb->data, skb->data + DSA_HLEN + extra, 2 * ETH_ALEN);
-
-		/* Construct untagged FROM_CPU 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_alloc_etype_header(skb, DSA_HLEN + extra);
+
+		/* Construct untagged DSA tag. */
+		dsa_header = dsa_etype_header_pos_tx(skb) + extra;
+
+		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;
@@ -162,20 +198,18 @@ static struct sk_buff *dsa_xmit_ll(struct sk_buff *skb, struct net_device *dev,
 static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 				  u8 extra)
 {
+	bool trap = false, trunk = false;
 	int source_device, source_port;
-	bool trunk = false;
 	enum dsa_code code;
 	enum dsa_cmd cmd;
 	u8 *dsa_header;
 
 	/* The ethertype field is part of the DSA header. */
-	dsa_header = skb->data - 2;
+	dsa_header = dsa_etype_header_pos_rx(skb);
 
 	cmd = dsa_header[0] >> 6;
 	switch (cmd) {
 	case DSA_CMD_FORWARD:
-		skb->offload_fwd_mark = 1;
-
 		trunk = !!(dsa_header[1] & 7);
 		break;
 
@@ -194,7 +228,6 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 			 * device (like a bridge) that forwarding has
 			 * already been done by hardware.
 			 */
-			skb->offload_fwd_mark = 1;
 			break;
 		case DSA_CODE_MGMT_TRAP:
 		case DSA_CODE_IGMP_MLD_TRAP:
@@ -202,6 +235,7 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 			/* Traps have, by definition, not been
 			 * forwarded by hardware, so don't mark them.
 			 */
+			trap = true;
 			break;
 		default:
 			/* Reserved code, this could be anything. Drop
@@ -235,6 +269,15 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 	if (!skb->dev)
 		return NULL;
 
+	/* When using LAG offload, skb->dev is not a DSA slave interface,
+	 * so we cannot call dsa_default_offload_fwd_mark and we need to
+	 * special-case it.
+	 */
+	if (trunk)
+		skb->offload_fwd_mark = true;
+	else if (!trap)
+		dsa_default_offload_fwd_mark(skb);
+
 	/* If the 'tagged' bit is set; convert the DSA tag to a 802.1Q
 	 * tag, and delete the ethertype (extra) if applicable. If the
 	 * 'tagged' bit is cleared; delete the DSA tag, and ethertype
@@ -269,14 +312,10 @@ static struct sk_buff *dsa_rcv_ll(struct sk_buff *skb, struct net_device *dev,
 		memcpy(dsa_header, new_header, DSA_HLEN);
 
 		if (extra)
-			memmove(skb->data - ETH_HLEN,
-				skb->data - ETH_HLEN - extra,
-				2 * ETH_ALEN);
+			dsa_strip_etype_header(skb, extra);
 	} else {
 		skb_pull_rcsum(skb, DSA_HLEN);
-		memmove(skb->data - ETH_HLEN,
-			skb->data - ETH_HLEN - DSA_HLEN - extra,
-			2 * ETH_ALEN);
+		dsa_strip_etype_header(skb, DSA_HLEN + extra);
 	}
 
 	return skb;
@@ -289,8 +328,7 @@ static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	return dsa_xmit_ll(skb, dev, 0);
 }
 
-static struct sk_buff *dsa_rcv(struct sk_buff *skb, struct net_device *dev,
-			       struct packet_type *pt)
+static struct sk_buff *dsa_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	if (unlikely(!pskb_may_pull(skb, DSA_HLEN)))
 		return NULL;
@@ -322,7 +360,7 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!skb)
 		return NULL;
 
-	edsa_header = skb->data + 2 * ETH_ALEN;
+	edsa_header = dsa_etype_header_pos_tx(skb);
 	edsa_header[0] = (ETH_P_EDSA >> 8) & 0xff;
 	edsa_header[1] = ETH_P_EDSA & 0xff;
 	edsa_header[2] = 0x00;
@@ -330,8 +368,7 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
 	return skb;
 }
 
-static struct sk_buff *edsa_rcv(struct sk_buff *skb, struct net_device *dev,
-				struct packet_type *pt)
+static struct sk_buff *edsa_rcv(struct sk_buff *skb, struct net_device *dev)
 {
 	if (unlikely(!pskb_may_pull(skb, EDSA_HLEN)))
 		return NULL;