drivers: net: Modify WireGuard for backward compat

Change-Id: I1c8e130a514a7b0329f8df8099cc84f4cc8d5822

19 files changed, 4192 insertions, 0 deletions

diff --git a/drivers/net/wireguard/compat/Makefile.include b/drivers/net/wireguard/compat/Makefile.include
new file mode 100644
index 000000000000..513dba444a37
--- /dev/null
+++ b/drivers/net/wireguard/compat/Makefile.include
@@ -0,0 +1,102 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+
+kbuild-dir := $(if $(filter /%,$(src)),$(src),$(srctree)/$(src))
+
+ccflags-y += -include $(kbuild-dir)/compat/compat.h
+asflags-y += -include $(kbuild-dir)/compat/compat-asm.h
+
+ifeq ($(wildcard $(srctree)/include/linux/ptr_ring.h),)
+ccflags-y += -I$(kbuild-dir)/compat/ptr_ring/include
+endif
+
+ifeq ($(wildcard $(srctree)/include/linux/siphash.h),)
+ccflags-y += -I$(kbuild-dir)/compat/siphash/include
+wireguard-y += compat/siphash/siphash.o
+endif
+
+ifeq ($(wildcard $(srctree)/include/net/dst_cache.h),)
+ccflags-y += -I$(kbuild-dir)/compat/dst_cache/include
+wireguard-y += compat/dst_cache/dst_cache.o
+endif
+
+ifeq ($(wildcard $(srctree)/arch/x86/include/asm/intel-family.h)$(CONFIG_X86),y)
+ccflags-y += -I$(kbuild-dir)/compat/intel-family-x86/include
+endif
+
+ifeq ($(wildcard $(srctree)/arch/x86/include/asm/fpu/api.h)$(CONFIG_X86),y)
+ccflags-y += -I$(kbuild-dir)/compat/fpu-x86/include
+endif
+
+ifeq ($(wildcard $(srctree)/arch/$(SRCARCH)/include/asm/simd.h)$(shell grep -s -F "generic-y += simd.h" "$(srctree)/arch/$(SRCARCH)/Makefile" "$(srctree)/arch/$(SRCARCH)/Makefile"),)
+ccflags-y += -I$(kbuild-dir)/compat/simd-asm/include
+endif
+
+ifeq ($(wildcard $(srctree)/include/linux/simd.h),)
+ccflags-y += -I$(kbuild-dir)/compat/simd/include
+endif
+
+ifeq ($(wildcard $(srctree)/include/net/udp_tunnel.h),)
+ccflags-y += -I$(kbuild-dir)/compat/udp_tunnel/include
+wireguard-y += compat/udp_tunnel/udp_tunnel.o
+endif
+
+ifeq ($(shell grep -s -F "int crypto_memneq" "$(srctree)/include/crypto/algapi.h"),)
+ccflags-y += -include $(kbuild-dir)/compat/memneq/include.h
+wireguard-y += compat/memneq/memneq.o
+endif
+
+ifeq ($(shell grep -s -F "addr_gen_mode" "$(srctree)/include/linux/ipv6.h"),)
+ccflags-y += -DCOMPAT_CANNOT_USE_DEV_CNF
+endif
+
+ifdef CONFIG_HZ
+ifeq ($(wildcard $(CURDIR)/include/generated/timeconst.h),)
+ccflags-y += $(shell bash -c '((a=$(CONFIG_HZ), b=1000000)); while ((b > 0)); do ((t=b, b=a%b, a=t)); done; echo "-DHZ_TO_USEC_NUM=$$((1000000/a)) -DHZ_TO_USEC_DEN=$$(($(CONFIG_HZ)/a))";')
+endif
+endif
+
+ifeq ($(wildcard $(srctree)/arch/arm/include/asm/neon.h)$(CONFIG_ARM),y)
+ccflags-y += -I$(kbuild-dir)/compat/neon-arm/include
+endif
+ifeq ($(wildcard $(srctree)/arch/arm64/include/asm/neon.h)$(CONFIG_ARM64),y)
+ccflags-y += -I$(kbuild-dir)/compat/neon-arm/include
+endif
+
+ifeq ($(CONFIG_X86_64),y)
+	ifeq ($(ssse3_instr),)
+		ssse3_instr := $(call as-instr,pshufb %xmm0$(comma)%xmm0,-DCONFIG_AS_SSSE3=1)
+		ccflags-y += $(ssse3_instr)
+		asflags-y += $(ssse3_instr)
+	endif
+	ifeq ($(avx_instr),)
+		avx_instr := $(call as-instr,vxorps %ymm0$(comma)%ymm1$(comma)%ymm2,-DCONFIG_AS_AVX=1)
+		ccflags-y += $(avx_instr)
+		asflags-y += $(avx_instr)
+	endif
+	ifeq ($(avx2_instr),)
+		avx2_instr := $(call as-instr,vpbroadcastb %xmm0$(comma)%ymm1,-DCONFIG_AS_AVX2=1)
+		ccflags-y += $(avx2_instr)
+		asflags-y += $(avx2_instr)
+	endif
+	ifeq ($(avx512_instr),)
+		avx512_instr := $(call as-instr,vpmovm2b %k1$(comma)%zmm5,-DCONFIG_AS_AVX512=1)
+		ccflags-y += $(avx512_instr)
+		asflags-y += $(avx512_instr)
+	endif
+	ifeq ($(bmi2_instr),)
+		bmi2_instr :=$(call as-instr,mulx %rax$(comma)%rax$(comma)%rax,-DCONFIG_AS_BMI2=1)
+		ccflags-y += $(bmi2_instr)
+		asflags-y += $(bmi2_instr)
+	endif
+	ifeq ($(adx_instr),)
+		adx_instr :=$(call as-instr,adcx %rax$(comma)%rax,-DCONFIG_AS_ADX=1)
+		ccflags-y += $(adx_instr)
+		asflags-y += $(adx_instr)
+	endif
+endif
+
+ifneq ($(shell grep -s -F "\#define LINUX_PACKAGE_ID \" Debian " "$(CURDIR)/include/generated/package.h"),)
+ccflags-y += -DISDEBIAN
+endif
diff --git a/drivers/net/wireguard/compat/checksum/checksum_partial_compat.h b/drivers/net/wireguard/compat/checksum/checksum_partial_compat.h
new file mode 100644
index 000000000000..3dfe5397a94f
--- /dev/null
+++ b/drivers/net/wireguard/compat/checksum/checksum_partial_compat.h
@@ -0,0 +1,208 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include <net/route.h>
+#include <net/esp.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <net/ip6_checksum.h>
+
+#define IP6_MF          0x0001
+#define IP6_OFFSET      0xFFF8
+static inline int skb_maybe_pull_tail(struct sk_buff *skb, unsigned int len, unsigned int max)
+{
+	if (skb_headlen(skb) >= len)
+		return 0;
+	if (max > skb->len)
+		max = skb->len;
+	if (__pskb_pull_tail(skb, max - skb_headlen(skb)) == NULL)
+		return -ENOMEM;
+	if (skb_headlen(skb) < len)
+		return -EPROTO;
+	return 0;
+}
+#define MAX_IP_HDR_LEN 128
+static inline int skb_checksum_setup_ip(struct sk_buff *skb, bool recalculate)
+{
+	unsigned int off;
+	bool fragment;
+	int err;
+	fragment = false;
+	err = skb_maybe_pull_tail(skb, sizeof(struct iphdr), MAX_IP_HDR_LEN);
+	if (err < 0)
+		goto out;
+	if (ip_hdr(skb)->frag_off & htons(IP_OFFSET | IP_MF))
+		fragment = true;
+	off = ip_hdrlen(skb);
+	err = -EPROTO;
+	if (fragment)
+		goto out;
+	switch (ip_hdr(skb)->protocol) {
+	case IPPROTO_TCP:
+		err = skb_maybe_pull_tail(skb,
+					  off + sizeof(struct tcphdr),
+					  MAX_IP_HDR_LEN);
+		if (err < 0)
+			goto out;
+
+		if (!skb_partial_csum_set(skb, off,
+					  offsetof(struct tcphdr, check))) {
+			err = -EPROTO;
+			goto out;
+		}
+
+		if (recalculate)
+			tcp_hdr(skb)->check =
+				~csum_tcpudp_magic(ip_hdr(skb)->saddr,
+						   ip_hdr(skb)->daddr,
+						   skb->len - off,
+						   IPPROTO_TCP, 0);
+		break;
+	case IPPROTO_UDP:
+		err = skb_maybe_pull_tail(skb,
+					  off + sizeof(struct udphdr),
+					  MAX_IP_HDR_LEN);
+		if (err < 0)
+			goto out;
+
+		if (!skb_partial_csum_set(skb, off,
+					  offsetof(struct udphdr, check))) {
+			err = -EPROTO;
+			goto out;
+		}
+
+		if (recalculate)
+			udp_hdr(skb)->check =
+				~csum_tcpudp_magic(ip_hdr(skb)->saddr,
+						   ip_hdr(skb)->daddr,
+						   skb->len - off,
+						   IPPROTO_UDP, 0);
+		break;
+	default:
+		goto out;
+	}
+	err = 0;
+out:
+	return err;
+}
+#define MAX_IPV6_HDR_LEN 256
+#define OPT_HDR(type, skb, off) \
+	(type *)(skb_network_header(skb) + (off))
+static inline int skb_checksum_setup_ipv6(struct sk_buff *skb, bool recalculate)
+{
+	int err;
+	u8 nexthdr;
+	unsigned int off;
+	unsigned int len;
+	bool fragment;
+	bool done;
+	fragment = false;
+	done = false;
+	off = sizeof(struct ipv6hdr);
+	err = skb_maybe_pull_tail(skb, off, MAX_IPV6_HDR_LEN);
+	if (err < 0)
+		goto out;
+	nexthdr = ipv6_hdr(skb)->nexthdr;
+	len = sizeof(struct ipv6hdr) + ntohs(ipv6_hdr(skb)->payload_len);
+	while (off <= len && !done) {
+		switch (nexthdr) {
+		case IPPROTO_DSTOPTS:
+		case IPPROTO_HOPOPTS:
+		case IPPROTO_ROUTING: {
+			struct ipv6_opt_hdr *hp;
+
+			err = skb_maybe_pull_tail(skb, off + sizeof(struct ipv6_opt_hdr), MAX_IPV6_HDR_LEN);
+			if (err < 0)
+				goto out;
+			hp = OPT_HDR(struct ipv6_opt_hdr, skb, off);
+			nexthdr = hp->nexthdr;
+			off += ipv6_optlen(hp);
+			break;
+		}
+		case IPPROTO_FRAGMENT: {
+			struct frag_hdr *hp;
+			err = skb_maybe_pull_tail(skb, off + sizeof(struct frag_hdr), MAX_IPV6_HDR_LEN);
+			if (err < 0)
+				goto out;
+			hp = OPT_HDR(struct frag_hdr, skb, off);
+			if (hp->frag_off & htons(IP6_OFFSET | IP6_MF))
+				fragment = true;
+			nexthdr = hp->nexthdr;
+			off += sizeof(struct frag_hdr);
+			break;
+		}
+		default:
+			done = true;
+			break;
+		}
+	}
+	err = -EPROTO;
+	if (!done || fragment)
+		goto out;
+	switch (nexthdr) {
+		case IPPROTO_TCP:
+			err = skb_maybe_pull_tail(skb,
+						  off + sizeof(struct tcphdr),
+						  MAX_IPV6_HDR_LEN);
+			if (err < 0)
+				goto out;
+
+			if (!skb_partial_csum_set(skb, off,
+						  offsetof(struct tcphdr, check))) {
+				err = -EPROTO;
+				goto out;
+			}
+
+			if (recalculate)
+				tcp_hdr(skb)->check =
+					~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
+							 &ipv6_hdr(skb)->daddr,
+							 skb->len - off,
+							 IPPROTO_TCP, 0);
+			break;
+		case IPPROTO_UDP:
+			err = skb_maybe_pull_tail(skb,
+						  off + sizeof(struct udphdr),
+						  MAX_IPV6_HDR_LEN);
+			if (err < 0)
+				goto out;
+
+			if (!skb_partial_csum_set(skb, off,
+						  offsetof(struct udphdr, check))) {
+				err = -EPROTO;
+				goto out;
+			}
+
+			if (recalculate)
+				udp_hdr(skb)->check =
+					~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
+							 &ipv6_hdr(skb)->daddr,
+							 skb->len - off,
+							 IPPROTO_UDP, 0);
+			break;
+		default:
+			goto out;
+	}
+	err = 0;
+out:
+	return err;
+}
+static inline int skb_checksum_setup(struct sk_buff *skb, bool recalculate)
+{
+	int err;
+	switch (skb->protocol) {
+	case htons(ETH_P_IP):
+		err = skb_checksum_setup_ip(skb, recalculate);
+		break;
+
+	case htons(ETH_P_IPV6):
+		err = skb_checksum_setup_ipv6(skb, recalculate);
+		break;
+	default:
+		err = -EPROTO;
+		break;
+	}
+	return err;
+}
diff --git a/drivers/net/wireguard/compat/compat-asm.h b/drivers/net/wireguard/compat/compat-asm.h
new file mode 100644
index 000000000000..4e427e50e9c6
--- /dev/null
+++ b/drivers/net/wireguard/compat/compat-asm.h
@@ -0,0 +1,78 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_COMPATASM_H
+#define _WG_COMPATASM_H
+
+#include <linux/linkage.h>
+#include <linux/kconfig.h>
+#include <linux/version.h>
+
+/* PaX compatibility */
+#if defined(RAP_PLUGIN)
+#undef ENTRY
+#define ENTRY RAP_ENTRY
+#endif
+
+#if defined(__LINUX_ARM_ARCH__) && LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0)
+	.irp	c,,eq,ne,cs,cc,mi,pl,vs,vc,hi,ls,ge,lt,gt,le,hs,lo
+	.macro	ret\c, reg
+#if __LINUX_ARM_ARCH__ < 6
+	mov\c	pc, \reg
+#else
+	.ifeqs	"\reg", "lr"
+	bx\c	\reg
+	.else
+	mov\c	pc, \reg
+	.endif
+#endif
+	.endm
+	.endr
+#endif
+
+#if defined(__LINUX_ARM_ARCH__) && LINUX_VERSION_CODE < KERNEL_VERSION(3, 15, 0)
+#include <asm/assembler.h>
+#define lspush push
+#define lspull pull
+#undef push
+#undef pull
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 4, 76)
+#define SYM_FUNC_START ENTRY
+#define SYM_FUNC_END ENDPROC
+#endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 5, 0)
+#define blake2s_compress_ssse3 zinc_blake2s_compress_ssse3
+#define blake2s_compress_avx512 zinc_blake2s_compress_avx512
+#define poly1305_init_arm zinc_poly1305_init_arm
+#define poly1305_blocks_arm zinc_poly1305_blocks_arm
+#define poly1305_emit_arm zinc_poly1305_emit_arm
+#define poly1305_blocks_neon zinc_poly1305_blocks_neon
+#define poly1305_emit_neon zinc_poly1305_emit_neon
+#define poly1305_init_mips zinc_poly1305_init_mips
+#define poly1305_blocks_mips zinc_poly1305_blocks_mips
+#define poly1305_emit_mips zinc_poly1305_emit_mips
+#define poly1305_init_x86_64 zinc_poly1305_init_x86_64
+#define poly1305_blocks_x86_64 zinc_poly1305_blocks_x86_64
+#define poly1305_emit_x86_64 zinc_poly1305_emit_x86_64
+#define poly1305_emit_avx zinc_poly1305_emit_avx
+#define poly1305_blocks_avx zinc_poly1305_blocks_avx
+#define poly1305_blocks_avx2 zinc_poly1305_blocks_avx2
+#define poly1305_blocks_avx512 zinc_poly1305_blocks_avx512
+#define curve25519_neon zinc_curve25519_neon
+#define hchacha20_ssse3 zinc_hchacha20_ssse3
+#define chacha20_ssse3 zinc_chacha20_ssse3
+#define chacha20_avx2 zinc_chacha20_avx2
+#define chacha20_avx512 zinc_chacha20_avx512
+#define chacha20_avx512vl zinc_chacha20_avx512vl
+#define chacha20_mips zinc_chacha20_mips
+#define chacha20_arm zinc_chacha20_arm
+#define hchacha20_arm zinc_hchacha20_arm
+#define chacha20_neon zinc_chacha20_neon
+#endif
+
+#endif /* _WG_COMPATASM_H */
diff --git a/drivers/net/wireguard/compat/compat.h b/drivers/net/wireguard/compat/compat.h
new file mode 100644
index 000000000000..9f6e725db531
--- /dev/null
+++ b/drivers/net/wireguard/compat/compat.h
@@ -0,0 +1,1118 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_COMPAT_H
+#define _WG_COMPAT_H
+
+#include <linux/kconfig.h>
+#include <linux/version.h>
+#include <linux/types.h>
+#include <generated/utsrelease.h>
+
+#ifdef RHEL_MAJOR
+#if RHEL_MAJOR == 7
+#define ISRHEL7
+#if RHEL_MINOR == 8
+#define ISCENTOS7
+#endif
+#elif RHEL_MAJOR == 8
+#define ISRHEL8
+#if RHEL_MINOR == 2
+#define ISCENTOS8
+#endif
+#endif
+#endif
+#ifdef UTS_UBUNTU_RELEASE_ABI
+#if LINUX_VERSION_CODE == KERNEL_VERSION(3, 13, 11)
+#define ISUBUNTU1404
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 5, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 4, 0)
+#define ISUBUNTU1604
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 16, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 15, 0)
+#define ISUBUNTU1804
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(5, 0, 0)
+#define ISUBUNTU1904
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(5, 4, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(5, 3, 0)
+#define ISUBUNTU1910
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 10, 0)
+#error "WireGuard requires Linux >= 3.10"
+#endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 6, 0)
+#error "WireGuard has been merged into Linux >= 5.6 and therefore this compatibility module is no longer required."
+#endif
+
+#if defined(ISRHEL7)
+#include <linux/skbuff.h>
+#define headers_end headers_start
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(3, 18, 0)
+#define headers_start data
+#define headers_end data
+#endif
+
+#include <linux/cache.h>
+#include <linux/init.h>
+#ifndef __ro_after_init
+#define __ro_after_init __read_mostly
+#endif
+
+#include <linux/compiler.h>
+#ifndef READ_ONCE
+#define READ_ONCE ACCESS_ONCE
+#endif
+#ifndef WRITE_ONCE
+#ifdef ACCESS_ONCE_RW
+#define WRITE_ONCE(p, v) (ACCESS_ONCE_RW(p) = (v))
+#else
+#define WRITE_ONCE(p, v) (ACCESS_ONCE(p) = (v))
+#endif
+#endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include "udp_tunnel/udp_tunnel_partial_compat.h"
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0) && !defined(DEBUG) && defined(net_dbg_ratelimited)
+#undef net_dbg_ratelimited
+#define net_dbg_ratelimited(fmt, ...) do { if (0) no_printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__); } while (0)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0)
+#include <linux/rcupdate.h>
+#ifndef RCU_LOCKDEP_WARN
+#define RCU_LOCKDEP_WARN(cond, message) rcu_lockdep_assert(!(cond), message)
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0) && !defined(ISRHEL7)
+#define ipv6_dst_lookup(a, b, c, d) ipv6_dst_lookup(b, c, d)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 16, 83)
+#define ipv6_dst_lookup_flow(a, b, c, d) ipv6_dst_lookup_flow(b, c, d)
+#elif (LINUX_VERSION_CODE < KERNEL_VERSION(5, 4, 5) && LINUX_VERSION_CODE >= KERNEL_VERSION(5, 4, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(5, 3, 18) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 20, 0) && !defined(ISUBUNTU1904)) || (!defined(ISRHEL8) && !defined(ISDEBIAN) && !defined(ISUBUNTU1804) && LINUX_VERSION_CODE < KERNEL_VERSION(4, 19, 119) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 15, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 14, 181) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 224) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 5, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 4, 224) && !defined(ISUBUNTU1604) && (!defined(ISRHEL7) || defined(ISCENTOS7)))
+#define ipv6_dst_lookup_flow(a, b, c, d) ipv6_dst_lookup(a, b, &dst, c) + (void *)0 ?: dst
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0) && IS_ENABLED(CONFIG_IPV6) && !defined(ISRHEL7)
+#include <net/ipv6.h>
+struct ipv6_stub_type {
+	void *udpv6_encap_enable;
+	int (*ipv6_dst_lookup)(struct sock *sk, struct dst_entry **dst, struct flowi6 *fl6);
+};
+static const struct ipv6_stub_type ipv6_stub_impl = {
+	.udpv6_encap_enable = (void *)1,
+	.ipv6_dst_lookup = ip6_dst_lookup
+};
+static const struct ipv6_stub_type *ipv6_stub = &ipv6_stub_impl;
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 8, 0) && IS_ENABLED(CONFIG_IPV6) && !defined(ISRHEL7)
+#include <net/addrconf.h>
+static inline bool ipv6_mod_enabled(void)
+{
+	return ipv6_stub != NULL && ipv6_stub->udpv6_encap_enable != NULL;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 11, 0) && !defined(ISRHEL7)
+#include <linux/skbuff.h>
+static inline void skb_reset_tc(struct sk_buff *skb)
+{
+#ifdef CONFIG_NET_CLS_ACT
+	skb->tc_verd = 0;
+#endif
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 11, 0)
+#include <linux/random.h>
+#include <linux/siphash.h>
+static inline u32 __compat_get_random_u32(void)
+{
+	static siphash_key_t key;
+	static u32 counter = 0;
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 4, 0)
+	static bool has_seeded = false;
+	if (unlikely(!has_seeded)) {
+		get_random_bytes(&key, sizeof(key));
+		has_seeded = true;
+	}
+#else
+	get_random_once(&key, sizeof(key));
+#endif
+	return siphash_2u32(counter++, get_random_int(), &key);
+}
+#define get_random_u32 __compat_get_random_u32
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 18, 0) && !defined(ISRHEL7)
+static inline void netif_keep_dst(struct net_device *dev)
+{
+	dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+}
+#define COMPAT_CANNOT_USE_CSUM_LEVEL
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0) && !defined(ISRHEL7)
+#include <linux/netdevice.h>
+#ifndef netdev_alloc_pcpu_stats
+#define pcpu_sw_netstats pcpu_tstats
+#endif
+#ifndef netdev_alloc_pcpu_stats
+#define netdev_alloc_pcpu_stats alloc_percpu
+#endif
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(3, 15, 0) && !defined(ISRHEL7)
+#include <linux/netdevice.h>
+#ifndef netdev_alloc_pcpu_stats
+#define netdev_alloc_pcpu_stats(type)					\
+({									\
+	typeof(type) __percpu *pcpu_stats = alloc_percpu(type);		\
+	if (pcpu_stats)	{						\
+		int __cpu;						\
+		for_each_possible_cpu(__cpu) {				\
+			typeof(type) *stat;				\
+			stat = per_cpu_ptr(pcpu_stats, __cpu);		\
+			u64_stats_init(&stat->syncp);			\
+		}							\
+	}								\
+	pcpu_stats;							\
+})
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0) && !defined(ISRHEL7)
+#include "checksum/checksum_partial_compat.h"
+static inline void *__compat_pskb_put(struct sk_buff *skb, struct sk_buff *tail, int len)
+{
+	if (tail != skb) {
+		skb->data_len += len;
+		skb->len += len;
+	}
+	return skb_put(tail, len);
+}
+#define pskb_put __compat_pskb_put
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 11, 0) && !defined(ISRHEL7)
+#include <net/xfrm.h>
+static inline void skb_scrub_packet(struct sk_buff *skb, bool xnet)
+{
+#ifdef CONFIG_CAVIUM_OCTEON_IPFWD_OFFLOAD
+	memset(&skb->cvm_info, 0, sizeof(skb->cvm_info));
+	skb->cvm_reserved = 0;
+#endif
+	skb->tstamp.tv64 = 0;
+	skb->pkt_type = PACKET_HOST;
+	skb->skb_iif = 0;
+	skb_dst_drop(skb);
+	secpath_reset(skb);
+	nf_reset(skb);
+	nf_reset_trace(skb);
+	if (!xnet)
+		return;
+	skb_orphan(skb);
+	skb->mark = 0;
+}
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)
+#define skb_scrub_packet(a, b) skb_scrub_packet(a)
+#endif
+
+#if ((LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 13, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 63) || defined(ISUBUNTU1404)) && !defined(ISRHEL7)
+#include <linux/random.h>
+static inline u32 __compat_prandom_u32_max(u32 ep_ro)
+{
+	return (u32)(((u64)prandom_u32() * ep_ro) >> 32);
+}
+#define prandom_u32_max __compat_prandom_u32_max
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0)
+#include <linux/kernel.h>
+#ifndef U8_MAX
+#define U8_MAX ((u8)~0U)
+#endif
+#ifndef S8_MAX
+#define S8_MAX ((s8)(U8_MAX >> 1))
+#endif
+#ifndef S8_MIN
+#define S8_MIN ((s8)(-S8_MAX - 1))
+#endif
+#ifndef U16_MAX
+#define U16_MAX ((u16)~0U)
+#endif
+#ifndef S16_MAX
+#define S16_MAX ((s16)(U16_MAX >> 1))
+#endif
+#ifndef S16_MIN
+#define S16_MIN ((s16)(-S16_MAX - 1))
+#endif
+#ifndef U32_MAX
+#define U32_MAX ((u32)~0U)
+#endif
+#ifndef S32_MAX
+#define S32_MAX ((s32)(U32_MAX >> 1))
+#endif
+#ifndef S32_MIN
+#define S32_MIN ((s32)(-S32_MAX - 1))
+#endif
+#ifndef U64_MAX
+#define U64_MAX ((u64)~0ULL)
+#endif
+#ifndef S64_MAX
+#define S64_MAX ((s64)(U64_MAX >> 1))
+#endif
+#ifndef S64_MIN
+#define S64_MIN ((s64)(-S64_MAX - 1))
+#endif
+#endif
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 3) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 35) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 15, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 24) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 13, 0) && !defined(ISUBUNTU1404)) || (LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 33) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(3, 10, 60) && !defined(ISRHEL7))
+static inline void memzero_explicit(void *s, size_t count)
+{
+	memset(s, 0, count);
+	barrier();
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0) && !defined(ISRHEL7)
+static const struct in6_addr __compat_in6addr_any = IN6ADDR_ANY_INIT;
+#define in6addr_any __compat_in6addr_any
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 13, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 2, 0)
+#include <linux/completion.h>
+#include <linux/random.h>
+#include <linux/errno.h>
+struct rng_initializer {
+	struct completion done;
+	struct random_ready_callback cb;
+};
+static inline void rng_initialized_callback(struct random_ready_callback *cb)
+{
+	complete(&container_of(cb, struct rng_initializer, cb)->done);
+}
+static inline int wait_for_random_bytes(void)
+{
+	static bool rng_is_initialized = false;
+	int ret;
+	if (unlikely(!rng_is_initialized)) {
+		struct rng_initializer rng = {
+			.done = COMPLETION_INITIALIZER(rng.done),
+			.cb = { .owner = THIS_MODULE, .func = rng_initialized_callback }
+		};
+		ret = add_random_ready_callback(&rng.cb);
+		if (!ret) {
+			ret = wait_for_completion_interruptible(&rng.done);
+			if (ret) {
+				del_random_ready_callback(&rng.cb);
+				return ret;
+			}
+		} else if (ret != -EALREADY)
+			return ret;
+		rng_is_initialized = true;
+	}
+	return 0;
+}
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 2, 0)
+/* This is a disaster. Without this API, we really have no way of
+ * knowing if it's initialized. We just return that it has and hope
+ * for the best... */
+static inline int wait_for_random_bytes(void)
+{
+	return 0;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 19, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 2, 0) && !defined(ISRHEL8)
+#include <linux/random.h>
+#include <linux/slab.h>
+struct rng_is_initialized_callback {
+	struct random_ready_callback cb;
+	atomic_t *rng_state;
+};
+static inline void rng_is_initialized_callback(struct random_ready_callback *cb)
+{
+	struct rng_is_initialized_callback *rdy = container_of(cb, struct rng_is_initialized_callback, cb);
+	atomic_set(rdy->rng_state, 2);
+	kfree(rdy);
+}
+static inline bool rng_is_initialized(void)
+{
+	static atomic_t rng_state = ATOMIC_INIT(0);
+
+	if (atomic_read(&rng_state) == 2)
+		return true;
+
+	if (atomic_cmpxchg(&rng_state, 0, 1) == 0) {
+		int ret;
+		struct rng_is_initialized_callback *rdy = kmalloc(sizeof(*rdy), GFP_ATOMIC);
+		if (!rdy) {
+			atomic_set(&rng_state, 0);
+			return false;
+		}
+		rdy->cb.owner = THIS_MODULE;
+		rdy->cb.func = rng_is_initialized_callback;
+		rdy->rng_state = &rng_state;
+		ret = add_random_ready_callback(&rdy->cb);
+		if (ret)
+			kfree(rdy);
+		if (ret == -EALREADY) {
+			atomic_set(&rng_state, 2);
+			return true;
+		} else if (ret)
+			atomic_set(&rng_state, 0);
+		return false;
+	}
+	return false;
+}
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 2, 0)
+/* This is a disaster. Without this API, we really have no way of
+ * knowing if it's initialized. We just return that it has and hope
+ * for the best... */
+static inline bool rng_is_initialized(void)
+{
+	return true;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 13, 0)
+static inline int get_random_bytes_wait(void *buf, int nbytes)
+{
+	int ret = wait_for_random_bytes();
+	if (unlikely(ret))
+		return ret;
+	get_random_bytes(buf, nbytes);
+	return 0;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 11, 0) && !defined(ISRHEL7)
+#define system_power_efficient_wq system_unbound_wq
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 3, 0)
+#include <linux/ktime.h>
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0)
+#include <linux/hrtimer.h>
+#ifndef ktime_get_real_ts64
+#define timespec64 timespec
+#define ktime_get_real_ts64 ktime_get_real_ts
+#endif
+#else
+#include <linux/timekeeping.h>
+#endif
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 11, 0)
+static inline u64 __compat_jiffies64_to_nsecs(u64 j)
+{
+#if !(NSEC_PER_SEC % HZ)
+	return (NSEC_PER_SEC / HZ) * j;
+#else
+	return div_u64(j * HZ_TO_USEC_NUM, HZ_TO_USEC_DEN) * 1000;
+#endif
+}
+#define jiffies64_to_nsecs __compat_jiffies64_to_nsecs
+#endif
+static inline u64 ktime_get_coarse_boottime_ns(void)
+{
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0)
+	return ktime_to_ns(ktime_get_boottime());
+#elif (LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 12) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 20, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(4, 19, 53)
+	return ktime_to_ns(ktime_mono_to_any(ns_to_ktime(jiffies64_to_nsecs(get_jiffies_64())), TK_OFFS_BOOT));
+#else
+	return ktime_to_ns(ktime_get_coarse_boottime());
+#endif
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0)
+#include <linux/inetdevice.h>
+static inline __be32 __compat_confirm_addr_indev(struct in_device *in_dev, __be32 dst,  __be32 local, int scope)
+{
+	int same = 0;
+	__be32 addr = 0;
+	for_ifa(in_dev) {
+		if (!addr && (local == ifa->ifa_local || !local) && ifa->ifa_scope <= scope) {
+			addr = ifa->ifa_local;
+			if (same)
+				break;
+		}
+		if (!same) {
+			same = (!local || inet_ifa_match(local, ifa)) && (!dst || inet_ifa_match(dst, ifa));
+			if (same && addr) {
+				if (local || !dst)
+					break;
+				if (inet_ifa_match(addr, ifa))
+					break;
+				if (ifa->ifa_scope <= scope) {
+					addr = ifa->ifa_local;
+					break;
+				}
+				same = 0;
+			}
+		}
+	} endfor_ifa(in_dev);
+	return same ? addr : 0;
+}
+static inline __be32 __compat_inet_confirm_addr(struct net *net, struct in_device *in_dev, __be32 dst, __be32 local, int scope)
+{
+	__be32 addr = 0;
+	struct net_device *dev;
+	if (in_dev)
+		return __compat_confirm_addr_indev(in_dev, dst, local, scope);
+	rcu_read_lock();
+	for_each_netdev_rcu(net, dev) {
+		in_dev = __in_dev_get_rcu(dev);
+		if (in_dev) {
+			addr = __compat_confirm_addr_indev(in_dev, dst, local, scope);
+			if (addr)
+				break;
+		}
+	}
+	rcu_read_unlock();
+	return addr;
+}
+#define inet_confirm_addr __compat_inet_confirm_addr
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 12, 0)
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+static inline void *__compat_kvmalloc(size_t size, gfp_t flags)
+{
+	gfp_t kmalloc_flags = flags;
+	void *ret;
+	if (size > PAGE_SIZE) {
+		kmalloc_flags |= __GFP_NOWARN;
+		if (!(kmalloc_flags & __GFP_REPEAT) || (size <= PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER))
+			kmalloc_flags |= __GFP_NORETRY;
+	}
+	ret = kmalloc(size, kmalloc_flags);
+	if (ret || size <= PAGE_SIZE)
+		return ret;
+	return __vmalloc(size, flags, PAGE_KERNEL);
+}
+static inline void *__compat_kvzalloc(size_t size, gfp_t flags)
+{
+	return __compat_kvmalloc(size, flags | __GFP_ZERO);
+}
+#define kvmalloc __compat_kvmalloc
+#define kvzalloc __compat_kvzalloc
+#endif
+
+#if ((LINUX_VERSION_CODE < KERNEL_VERSION(3, 15, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 13, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 41)) && !defined(ISUBUNTU1404)
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+static inline void __compat_kvfree(const void *addr)
+{
+	if (is_vmalloc_addr(addr))
+		vfree(addr);
+	else
+		kfree(addr);
+}
+#define kvfree __compat_kvfree
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 11, 9)
+#include <linux/netdevice.h>
+#define priv_destructor destructor
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 13, 0)
+#define wg_newlink(a,b,c,d,e) wg_newlink(a,b,c,d)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 12, 0)
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#define nlmsg_parse(a, b, c, d, e, f) nlmsg_parse(a, b, c, d, e)
+#define nla_parse_nested(a, b, c, d, e) nla_parse_nested(a, b, c, d)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 10, 0) && !defined(ISRHEL7)
+static inline struct nlattr **genl_family_attrbuf(const struct genl_family *family)
+{
+	return family->attrbuf;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0)
+#define PTR_ERR_OR_ZERO(p) PTR_RET(p)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 7, 0)
+#include <net/netlink.h>
+#define nla_put_u64_64bit(a, b, c, d) nla_put_u64(a, b, c)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0)
+#include <net/genetlink.h>
+#ifndef GENL_UNS_ADMIN_PERM
+#define GENL_UNS_ADMIN_PERM GENL_ADMIN_PERM
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 10, 0) && !defined(ISRHEL7)
+#include <net/genetlink.h>
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 13, 0)
+#define genl_register_family(a) genl_register_family_with_ops(a, genl_ops, ARRAY_SIZE(genl_ops))
+#define COMPAT_CANNOT_USE_CONST_GENL_OPS
+#else
+#define genl_register_family(a) genl_register_family_with_ops(a, genl_ops)
+#endif
+#define COMPAT_CANNOT_USE_GENL_NOPS
+#endif
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(4, 14, 2) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 14, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 13, 16) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 65) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 5, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 4, 101) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 19, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(3, 18, 84)
+#define __COMPAT_NETLINK_DUMP_BLOCK { \
+	int ret; \
+	skb->end -= nlmsg_total_size(sizeof(int)); \
+	ret = wg_get_device_dump_real(skb, cb); \
+	skb->end += nlmsg_total_size(sizeof(int)); \
+	return ret; \
+}
+#define __COMPAT_NETLINK_DUMP_OVERRIDE
+#else
+#define __COMPAT_NETLINK_DUMP_BLOCK return wg_get_device_dump_real(skb, cb);
+#endif
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 8) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 15, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(4, 14, 25) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 10, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 87)
+#define wg_get_device_dump(a, b) wg_get_device_dump_real(a, b); \
+static int wg_get_device_dump(a, b) { \
+	struct wg_device *wg = (struct wg_device *)cb->args[0]; \
+	if (!wg) { \
+		int ret = wg_get_device_start(cb); \
+		if (ret) \
+			return ret; \
+	} \
+	__COMPAT_NETLINK_DUMP_BLOCK \
+} \
+static int wg_get_device_dump_real(a, b)
+#define COMPAT_CANNOT_USE_NETLINK_START
+#elif defined(__COMPAT_NETLINK_DUMP_OVERRIDE)
+#define wg_get_device_dump(a, b) wg_get_device_dump_real(a, b); \
+static int wg_get_device_dump(a, b) { \
+	__COMPAT_NETLINK_DUMP_BLOCK \
+} \
+static int wg_get_device_dump_real(a, b)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0)
+#define COMPAT_CANNOT_USE_IN6_DEV_GET
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0)
+#define COMPAT_CANNOT_USE_IFF_NO_QUEUE
+#endif
+
+#if defined(CONFIG_X86_64) && LINUX_VERSION_CODE < KERNEL_VERSION(4, 4, 0)
+#include <asm/user.h>
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 2, 0)
+#include <asm/xsave.h>
+#include <asm/xcr.h>
+static inline int cpu_has_xfeatures(u64 xfeatures_needed, const char **feature_name)
+{
+	return boot_cpu_has(X86_FEATURE_XSAVE) && xgetbv(XCR_XFEATURE_ENABLED_MASK) & xfeatures_needed;
+}
+#endif
+#ifndef XFEATURE_MASK_YMM
+#define XFEATURE_MASK_YMM XSTATE_YMM
+#endif
+#ifndef XFEATURE_MASK_SSE
+#define XFEATURE_MASK_SSE XSTATE_SSE
+#endif
+#ifndef XSTATE_AVX512
+#define XSTATE_AVX512 (XSTATE_OPMASK | XSTATE_ZMM_Hi256 | XSTATE_Hi16_ZMM)
+#endif
+#ifndef XFEATURE_MASK_AVX512
+#define XFEATURE_MASK_AVX512 XSTATE_AVX512
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0) && defined(CONFIG_X86_64)
+/* This is incredibly dumb and reckless, but as it turns out, there's
+ * not really hardware Linux runs properly on that supports F but not BW
+ * and VL, so in practice this isn't so bad. Plus, this is compat layer,
+ * so the bar remains fairly low.
+ */
+#include <asm/cpufeature.h>
+#ifndef X86_FEATURE_AVX512BW
+#define X86_FEATURE_AVX512BW X86_FEATURE_AVX512F
+#endif
+#ifndef X86_FEATURE_AVX512VL
+#define X86_FEATURE_AVX512VL X86_FEATURE_AVX512F
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 11, 0)
+struct __compat_dummy_container { char dev; };
+#define netdev_notifier_info net_device *)data); __attribute((unused)) char __compat_dummy_variable = ((struct __compat_dummy_container
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 14, 0)
+#define timer_setup(a, b, c) setup_timer(a, ((void (*)(unsigned long))b), ((unsigned long)a))
+#define from_timer(var, callback_timer, timer_fieldname) container_of(callback_timer, typeof(*var), timer_fieldname)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 3)
+#define COMPAT_CANNOT_USE_AVX512
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 0)
+#include <net/genetlink.h>
+#define genl_dump_check_consistent(a, b) genl_dump_check_consistent(a, b, &genl_family)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 13, 0) && !defined(ISRHEL7)
+static inline void *skb_put_data(struct sk_buff *skb, const void *data, unsigned int len)
+{
+	void *tmp = skb_put(skb, len);
+	memcpy(tmp, data, len);
+	return tmp;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 19, 0) && !defined(ISRHEL7)
+#define napi_complete_done(n, work_done) napi_complete(n)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 5, 0)
+#include <linux/netdevice.h>
+/* NAPI_STATE_SCHED gets set by netif_napi_add anyway, so this is safe.
+ * Also, kernels without NAPI_STATE_NO_BUSY_POLL don't have a call to
+ * napi_hash_add inside of netif_napi_add.
+ */
+#define NAPI_STATE_NO_BUSY_POLL NAPI_STATE_SCHED
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0)
+#include <linux/atomic.h>
+#ifndef atomic_read_acquire
+#define atomic_read_acquire(v) ({ int __compat_p1 = atomic_read(v); smp_rmb(); __compat_p1; })
+#endif
+#ifndef atomic_set_release
+#define atomic_set_release(v, i) ({ smp_wmb(); atomic_set(v, i); })
+#endif
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0)
+#include <linux/atomic.h>
+#ifndef atomic_read_acquire
+#define atomic_read_acquire(v) smp_load_acquire(&(v)->counter)
+#endif
+#ifndef atomic_set_release
+#define atomic_set_release(v, i) smp_store_release(&(v)->counter, (i))
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 17, 0)
+static inline void le32_to_cpu_array(u32 *buf, unsigned int words)
+{
+	while (words--) {
+		__le32_to_cpus(buf);
+		buf++;
+	}
+}
+static inline void cpu_to_le32_array(u32 *buf, unsigned int words)
+{
+	while (words--) {
+		__cpu_to_le32s(buf);
+		buf++;
+	}
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 14, 0)
+#include <crypto/algapi.h>
+static inline void crypto_xor_cpy(u8 *dst, const u8 *src1, const u8 *src2,
+				  unsigned int size)
+{
+	if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) &&
+	    __builtin_constant_p(size) &&
+	    (size % sizeof(unsigned long)) == 0) {
+		unsigned long *d = (unsigned long *)dst;
+		unsigned long *s1 = (unsigned long *)src1;
+		unsigned long *s2 = (unsigned long *)src2;
+
+		while (size > 0) {
+			*d++ = *s1++ ^ *s2++;
+			size -= sizeof(unsigned long);
+		}
+	} else {
+		if (unlikely(dst != src1))
+			memmove(dst, src1, size);
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 11, 0)
+		crypto_xor(dst, src2, size);
+#else
+		__crypto_xor(dst, src2, size);
+#endif
+	}
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0)
+#define read_cpuid_part() read_cpuid_part_number()
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0) && !defined(ISRHEL7)
+#define hlist_add_behind(a, b) hlist_add_after(b, a)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 0, 0)
+#define totalram_pages() totalram_pages
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 18, 0)
+struct __kernel_timespec {
+	int64_t tv_sec, tv_nsec;
+};
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 0)
+#include <linux/time64.h>
+#ifdef __kernel_timespec
+#undef __kernel_timespec
+struct __kernel_timespec {
+	int64_t tv_sec, tv_nsec;
+};
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 12, 0)
+#include <linux/kernel.h>
+#ifndef ALIGN_DOWN
+#define ALIGN_DOWN(x, a) __ALIGN_KERNEL((x) - ((a) - 1), (a))
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 0) && !defined(ISRHEL8)
+#include <linux/skbuff.h>
+#define skb_probe_transport_header(a) skb_probe_transport_header(a, 0)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 0) && !defined(ISRHEL7)
+#define ignore_df local_df
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 0) && !defined(ISRHEL8)
+/* Note that all intentional uses of the non-_bh variety need to explicitly
+ * undef these, conditionalized on COMPAT_CANNOT_DEPRECIATE_BH_RCU.
+ */
+#include <linux/rcupdate.h>
+static __always_inline void old_synchronize_rcu(void)
+{
+	synchronize_rcu();
+}
+static __always_inline void old_call_rcu(void *a, void *b)
+{
+	call_rcu(a, b);
+}
+static __always_inline void old_rcu_barrier(void)
+{
+	rcu_barrier();
+}
+#ifdef synchronize_rcu
+#undef synchronize_rcu
+#endif
+#ifdef call_rcu
+#undef call_rcu
+#endif
+#ifdef rcu_barrier
+#undef rcu_barrier
+#endif
+#define synchronize_rcu synchronize_rcu_bh
+#define call_rcu call_rcu_bh
+#define rcu_barrier rcu_barrier_bh
+#define COMPAT_CANNOT_DEPRECIATE_BH_RCU
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 19, 10) && !defined(ISRHEL8)
+static inline void skb_mark_not_on_list(struct sk_buff *skb)
+{
+	skb->next = NULL;
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 20, 0) && !defined(ISRHEL8)
+#define NLA_POLICY_EXACT_LEN(_len) { .type = NLA_UNSPEC, .len = _len }
+#endif
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 2, 0) && !defined(ISRHEL8)
+#define NLA_POLICY_MIN_LEN(_len) { .type = NLA_UNSPEC, .len = _len }
+#define COMPAT_CANNOT_INDIVIDUAL_NETLINK_OPS_POLICY
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 2, 0) && defined(__aarch64__)
+#define cpu_have_named_feature(name) (elf_hwcap & (HWCAP_ ## name))
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 1, 0)
+#include <linux/stddef.h>
+#ifndef offsetofend
+#define offsetofend(TYPE, MEMBER) (offsetof(TYPE, MEMBER) + sizeof(((TYPE *)0)->MEMBER))
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 0)
+#define genl_dumpit_info(cb) ({ \
+	struct { struct nlattr **attrs; } *a = (void *)((u8 *)cb->args + offsetofend(struct dump_ctx, next_allowedip)); \
+	BUILD_BUG_ON(sizeof(cb->args) < offsetofend(struct dump_ctx, next_allowedip) + sizeof(*a)); \
+	a->attrs = genl_family_attrbuf(&genl_family); \
+	if (nlmsg_parse(cb->nlh, GENL_HDRLEN + genl_family.hdrsize, a->attrs, genl_family.maxattr, device_policy, NULL) < 0) \
+		memset(a->attrs, 0, (genl_family.maxattr + 1) * sizeof(struct nlattr *)); \
+	a; \
+})
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 6, 0)
+#include <linux/skbuff.h>
+#ifndef skb_list_walk_safe
+#define skb_list_walk_safe(first, skb, next)                                   \
+	for ((skb) = (first), (next) = (skb) ? (skb)->next : NULL; (skb);      \
+	     (skb) = (next), (next) = (skb) ? (skb)->next : NULL)
+#endif
+#endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 5, 0)
+#define blake2s_init zinc_blake2s_init
+#define blake2s_init_key zinc_blake2s_init_key
+#define blake2s_update zinc_blake2s_update
+#define blake2s_final zinc_blake2s_final
+#define blake2s_hmac zinc_blake2s_hmac
+#define chacha20 zinc_chacha20
+#define hchacha20 zinc_hchacha20
+#define chacha20poly1305_encrypt zinc_chacha20poly1305_encrypt
+#define chacha20poly1305_encrypt_sg_inplace zinc_chacha20poly1305_encrypt_sg_inplace
+#define chacha20poly1305_decrypt zinc_chacha20poly1305_decrypt
+#define chacha20poly1305_decrypt_sg_inplace zinc_chacha20poly1305_decrypt_sg_inplace
+#define xchacha20poly1305_encrypt zinc_xchacha20poly1305_encrypt
+#define xchacha20poly1305_decrypt zinc_xchacha20poly1305_decrypt
+#define curve25519 zinc_curve25519
+#define curve25519_generate_secret zinc_curve25519_generate_secret
+#define curve25519_generate_public zinc_curve25519_generate_public
+#define poly1305_init zinc_poly1305_init
+#define poly1305_update zinc_poly1305_update
+#define poly1305_final zinc_poly1305_final
+#define blake2s_compress_ssse3 zinc_blake2s_compress_ssse3
+#define blake2s_compress_avx512 zinc_blake2s_compress_avx512
+#define poly1305_init_arm zinc_poly1305_init_arm
+#define poly1305_blocks_arm zinc_poly1305_blocks_arm
+#define poly1305_emit_arm zinc_poly1305_emit_arm
+#define poly1305_blocks_neon zinc_poly1305_blocks_neon
+#define poly1305_emit_neon zinc_poly1305_emit_neon
+#define poly1305_init_mips zinc_poly1305_init_mips
+#define poly1305_blocks_mips zinc_poly1305_blocks_mips
+#define poly1305_emit_mips zinc_poly1305_emit_mips
+#define poly1305_init_x86_64 zinc_poly1305_init_x86_64
+#define poly1305_blocks_x86_64 zinc_poly1305_blocks_x86_64
+#define poly1305_emit_x86_64 zinc_poly1305_emit_x86_64
+#define poly1305_emit_avx zinc_poly1305_emit_avx
+#define poly1305_blocks_avx zinc_poly1305_blocks_avx
+#define poly1305_blocks_avx2 zinc_poly1305_blocks_avx2
+#define poly1305_blocks_avx512 zinc_poly1305_blocks_avx512
+#define curve25519_neon zinc_curve25519_neon
+#define hchacha20_ssse3 zinc_hchacha20_ssse3
+#define chacha20_ssse3 zinc_chacha20_ssse3
+#define chacha20_avx2 zinc_chacha20_avx2
+#define chacha20_avx512 zinc_chacha20_avx512
+#define chacha20_avx512vl zinc_chacha20_avx512vl
+#define chacha20_mips zinc_chacha20_mips
+#define chacha20_arm zinc_chacha20_arm
+#define hchacha20_arm zinc_hchacha20_arm
+#define chacha20_neon zinc_chacha20_neon
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 19, 0) && !defined(ISRHEL7)
+#include <linux/skbuff.h>
+static inline int skb_ensure_writable(struct sk_buff *skb, int write_len)
+{
+	if (!pskb_may_pull(skb, write_len))
+		return -ENOMEM;
+
+	if (!skb_cloned(skb) || skb_clone_writable(skb, write_len))
+		return 0;
+
+	return pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 6, 0)
+#if IS_ENABLED(CONFIG_NF_NAT)
+#include <linux/ip.h>
+#include <linux/icmpv6.h>
+#include <net/ipv6.h>
+#include <net/icmp.h>
+#include <net/netfilter/nf_conntrack.h>
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 1, 0) && !defined(ISRHEL8)
+#include <net/netfilter/nf_nat_core.h>
+#endif
+static inline void __compat_icmp_ndo_send(struct sk_buff *skb_in, int type, int code, __be32 info)
+{
+	struct sk_buff *cloned_skb = NULL;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+	__be32 orig_ip;
+
+	ct = nf_ct_get(skb_in, &ctinfo);
+	if (!ct || !(ct->status & IPS_SRC_NAT)) {
+		icmp_send(skb_in, type, code, info);
+		return;
+	}
+
+	if (skb_shared(skb_in))
+		skb_in = cloned_skb = skb_clone(skb_in, GFP_ATOMIC);
+
+	if (unlikely(!skb_in || skb_network_header(skb_in) < skb_in->head ||
+	    (skb_network_header(skb_in) + sizeof(struct iphdr)) >
+	    skb_tail_pointer(skb_in) || skb_ensure_writable(skb_in,
+	    skb_network_offset(skb_in) + sizeof(struct iphdr))))
+		goto out;
+
+	orig_ip = ip_hdr(skb_in)->saddr;
+	ip_hdr(skb_in)->saddr = ct->tuplehash[0].tuple.src.u3.ip;
+	icmp_send(skb_in, type, code, info);
+	ip_hdr(skb_in)->saddr = orig_ip;
+out:
+	consume_skb(cloned_skb);
+}
+static inline void __compat_icmpv6_ndo_send(struct sk_buff *skb_in, u8 type, u8 code, __u32 info)
+{
+	struct sk_buff *cloned_skb = NULL;
+	enum ip_conntrack_info ctinfo;
+	struct in6_addr orig_ip;
+	struct nf_conn *ct;
+
+	ct = nf_ct_get(skb_in, &ctinfo);
+	if (!ct || !(ct->status & IPS_SRC_NAT)) {
+		icmpv6_send(skb_in, type, code, info);
+		return;
+	}
+
+	if (skb_shared(skb_in))
+		skb_in = cloned_skb = skb_clone(skb_in, GFP_ATOMIC);
+
+	if (unlikely(!skb_in || skb_network_header(skb_in) < skb_in->head ||
+	    (skb_network_header(skb_in) + sizeof(struct ipv6hdr)) >
+	    skb_tail_pointer(skb_in) || skb_ensure_writable(skb_in,
+	    skb_network_offset(skb_in) + sizeof(struct ipv6hdr))))
+		goto out;
+
+	orig_ip = ipv6_hdr(skb_in)->saddr;
+	ipv6_hdr(skb_in)->saddr = ct->tuplehash[0].tuple.src.u3.in6;
+	icmpv6_send(skb_in, type, code, info);
+	ipv6_hdr(skb_in)->saddr = orig_ip;
+out:
+	consume_skb(cloned_skb);
+}
+#else
+#define __compat_icmp_ndo_send icmp_send
+#define __compat_icmpv6_ndo_send icmpv6_send
+#endif
+#define icmp_ndo_send __compat_icmp_ndo_send
+#define icmpv6_ndo_send __compat_icmpv6_ndo_send
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 10, 0)
+#define COMPAT_CANNOT_USE_MAX_MTU
+#endif
+
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 14) && LINUX_VERSION_CODE >= KERNEL_VERSION(5, 5, 0)) || (LINUX_VERSION_CODE < KERNEL_VERSION(5, 4, 29) && !defined(ISUBUNTU1910) && !defined(ISUBUNTU1904) && (!defined(ISRHEL8) || defined(ISCENTOS8)))
+#include <linux/skbuff.h>
+#include <net/sch_generic.h>
+static inline void skb_reset_redirect(struct sk_buff *skb)
+{
+#ifdef CONFIG_NET_SCHED
+	skb_reset_tc(skb);
+#endif
+}
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0) && !defined(ISRHEL7)
+#define skb_get_hash skb_get_rxhash
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 15, 0) && !defined(ISRHEL7)
+#define hash rxhash
+#define l4_hash l4_rxhash
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 17, 0) && !defined(ISRHEL7)
+#define sw_hash ignore_df = 0; skb->nf_trace = skb->ooo_okay
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 3, 0)
+#define pre_exit exit
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 8, 0)
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+static inline __be16 ip_tunnel_parse_protocol(const struct sk_buff *skb)
+{
+	if (skb_network_header(skb) >= skb->head &&
+	    (skb_network_header(skb) + sizeof(struct iphdr)) <= skb_tail_pointer(skb) &&
+	    ip_hdr(skb)->version == 4)
+		return htons(ETH_P_IP);
+	if (skb_network_header(skb) >= skb->head &&
+	    (skb_network_header(skb) + sizeof(struct ipv6hdr)) <= skb_tail_pointer(skb) &&
+	    ipv6_hdr(skb)->version == 6)
+		return htons(ETH_P_IPV6);
+	return 0;
+}
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 1, 0) || defined(ISRHEL8)
+static const struct header_ops ip_tunnel_header_ops = { .parse_protocol = ip_tunnel_parse_protocol };
+#else
+#define header_ops hard_header_len
+#define ip_tunnel_header_ops *(char *)0 - (char *)0
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 9, 0)
+#define kfree_sensitive(a) kzfree(a)
+#endif
+
+#if defined(ISUBUNTU1604) || defined(ISRHEL7)
+#include <linux/siphash.h>
+#ifndef _WG_LINUX_SIPHASH_H
+#define hsiphash_1u32 siphash_1u32
+#define hsiphash_2u32 siphash_2u32
+#define hsiphash_3u32 siphash_3u32
+#define hsiphash_key_t siphash_key_t
+#endif
+#endif
+
+#ifdef CONFIG_VE
+#include <linux/netdev_features.h>
+#ifdef NETIF_F_VIRTUAL
+#undef NETIF_F_LLTX
+#define NETIF_F_LLTX (__NETIF_F(LLTX) | __NETIF_F(VIRTUAL))
+#endif
+#endif
+
+/* https://github.com/ClangBuiltLinux/linux/issues/7 */
+#if defined( __clang__) && (!defined(CONFIG_CLANG_VERSION) || CONFIG_CLANG_VERSION < 80000)
+#include <linux/bug.h>
+#undef BUILD_BUG_ON
+#define BUILD_BUG_ON(x)
+#endif
+
+/* PaX compatibility */
+#ifdef CONSTIFY_PLUGIN
+#include <linux/cache.h>
+#undef __read_mostly
+#define __read_mostly
+#endif
+#if (defined(RAP_PLUGIN) || defined(CONFIG_CFI_CLANG)) && LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 0)
+#include <linux/timer.h>
+#define wg_expired_retransmit_handshake(a) wg_expired_retransmit_handshake(unsigned long timer)
+#define wg_expired_send_keepalive(a) wg_expired_send_keepalive(unsigned long timer)
+#define wg_expired_new_handshake(a) wg_expired_new_handshake(unsigned long timer)
+#define wg_expired_zero_key_material(a) wg_expired_zero_key_material(unsigned long timer)
+#define wg_expired_send_persistent_keepalive(a) wg_expired_send_persistent_keepalive(unsigned long timer)
+#undef timer_setup
+#define timer_setup(a, b, c) setup_timer(a, ((void (*)(unsigned long))b), ((unsigned long)a))
+#undef from_timer
+#define from_timer(var, callback_timer, timer_fieldname) container_of((struct timer_list *)callback_timer, typeof(*var), timer_fieldname)
+#endif
+
+#endif /* _WG_COMPAT_H */
diff --git a/drivers/net/wireguard/compat/dst_cache/dst_cache.c b/drivers/net/wireguard/compat/dst_cache/dst_cache.c
new file mode 100644
index 000000000000..7ec22f768a8f
--- /dev/null
+++ b/drivers/net/wireguard/compat/dst_cache/dst_cache.c
@@ -0,0 +1,175 @@
+/*
+ * net/core/dst_cache.c - dst entry cache
+ *
+ * Copyright (c) 2016 Paolo Abeni <pabeni@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <net/dst_cache.h>
+#include <net/route.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ip6_fib.h>
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(4, 2, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)) || LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 50)
+static inline u32 rt6_get_cookie(const struct rt6_info *rt)
+{
+	if ((unlikely(rt->dst.flags & DST_NOCACHE) && rt->dst.from))
+		rt = (struct rt6_info *)(rt->dst.from);
+
+	return rt->rt6i_node ? rt->rt6i_node->fn_sernum : 0;
+}
+#endif
+#endif
+#include <uapi/linux/in.h>
+
+struct dst_cache_pcpu {
+	unsigned long refresh_ts;
+	struct dst_entry *dst;
+	u32 cookie;
+	union {
+		struct in_addr in_saddr;
+		struct in6_addr in6_saddr;
+	};
+};
+
+static void dst_cache_per_cpu_dst_set(struct dst_cache_pcpu *dst_cache,
+				      struct dst_entry *dst, u32 cookie)
+{
+	dst_release(dst_cache->dst);
+	if (dst)
+		dst_hold(dst);
+
+	dst_cache->cookie = cookie;
+	dst_cache->dst = dst;
+}
+
+static struct dst_entry *dst_cache_per_cpu_get(struct dst_cache *dst_cache,
+					       struct dst_cache_pcpu *idst)
+{
+	struct dst_entry *dst;
+
+	dst = idst->dst;
+	if (!dst)
+		goto fail;
+
+	/* the cache already hold a dst reference; it can't go away */
+	dst_hold(dst);
+
+	if (unlikely(!time_after(idst->refresh_ts, dst_cache->reset_ts) ||
+		     (dst->obsolete && !dst->ops->check(dst, idst->cookie)))) {
+		dst_cache_per_cpu_dst_set(idst, NULL, 0);
+		dst_release(dst);
+		goto fail;
+	}
+	return dst;
+
+fail:
+	idst->refresh_ts = jiffies;
+	return NULL;
+}
+
+struct dst_entry *dst_cache_get(struct dst_cache *dst_cache)
+{
+	if (!dst_cache->cache)
+		return NULL;
+
+	return dst_cache_per_cpu_get(dst_cache, this_cpu_ptr(dst_cache->cache));
+}
+
+struct rtable *dst_cache_get_ip4(struct dst_cache *dst_cache, __be32 *saddr)
+{
+	struct dst_cache_pcpu *idst;
+	struct dst_entry *dst;
+
+	if (!dst_cache->cache)
+		return NULL;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst = dst_cache_per_cpu_get(dst_cache, idst);
+	if (!dst)
+		return NULL;
+
+	*saddr = idst->in_saddr.s_addr;
+	return container_of(dst, struct rtable, dst);
+}
+
+void dst_cache_set_ip4(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       __be32 saddr)
+{
+	struct dst_cache_pcpu *idst;
+
+	if (!dst_cache->cache)
+		return;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst_cache_per_cpu_dst_set(idst, dst, 0);
+	idst->in_saddr.s_addr = saddr;
+}
+
+#if IS_ENABLED(CONFIG_IPV6)
+void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       const struct in6_addr *addr)
+{
+	struct dst_cache_pcpu *idst;
+
+	if (!dst_cache->cache)
+		return;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst_cache_per_cpu_dst_set(this_cpu_ptr(dst_cache->cache), dst,
+				  rt6_get_cookie((struct rt6_info *)dst));
+	idst->in6_saddr = *addr;
+}
+
+struct dst_entry *dst_cache_get_ip6(struct dst_cache *dst_cache,
+				    struct in6_addr *saddr)
+{
+	struct dst_cache_pcpu *idst;
+	struct dst_entry *dst;
+
+	if (!dst_cache->cache)
+		return NULL;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst = dst_cache_per_cpu_get(dst_cache, idst);
+	if (!dst)
+		return NULL;
+
+	*saddr = idst->in6_saddr;
+	return dst;
+}
+#endif
+
+int dst_cache_init(struct dst_cache *dst_cache, gfp_t gfp)
+{
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 18, 0)
+	BUG_ON(gfp & GFP_ATOMIC);
+	dst_cache->cache = alloc_percpu(struct dst_cache_pcpu);
+#else
+	dst_cache->cache = alloc_percpu_gfp(struct dst_cache_pcpu,
+					    gfp | __GFP_ZERO);
+#endif
+	if (!dst_cache->cache)
+		return -ENOMEM;
+
+	dst_cache_reset(dst_cache);
+	return 0;
+}
+
+void dst_cache_destroy(struct dst_cache *dst_cache)
+{
+	int i;
+
+	if (!dst_cache->cache)
+		return;
+
+	for_each_possible_cpu(i)
+		dst_release(per_cpu_ptr(dst_cache->cache, i)->dst);
+
+	free_percpu(dst_cache->cache);
+}
diff --git a/drivers/net/wireguard/compat/dst_cache/include/net/dst_cache.h b/drivers/net/wireguard/compat/dst_cache/include/net/dst_cache.h
new file mode 100644
index 000000000000..48021c0d6be1
--- /dev/null
+++ b/drivers/net/wireguard/compat/dst_cache/include/net/dst_cache.h
@@ -0,0 +1,97 @@
+#ifndef _WG_NET_DST_CACHE_H
+#define _WG_NET_DST_CACHE_H
+
+#include <linux/jiffies.h>
+#include <net/dst.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ip6_fib.h>
+#endif
+
+struct dst_cache {
+	struct dst_cache_pcpu __percpu *cache;
+	unsigned long reset_ts;
+};
+
+/**
+ *	dst_cache_get - perform cache lookup
+ *	@dst_cache: the cache
+ *
+ *	The caller should use dst_cache_get_ip4() if it need to retrieve the
+ *	source address to be used when xmitting to the cached dst.
+ *	local BH must be disabled.
+ */
+struct dst_entry *dst_cache_get(struct dst_cache *dst_cache);
+
+/**
+ *	dst_cache_get_ip4 - perform cache lookup and fetch ipv4 source address
+ *	@dst_cache: the cache
+ *	@saddr: return value for the retrieved source address
+ *
+ *	local BH must be disabled.
+ */
+struct rtable *dst_cache_get_ip4(struct dst_cache *dst_cache, __be32 *saddr);
+
+/**
+ *	dst_cache_set_ip4 - store the ipv4 dst into the cache
+ *	@dst_cache: the cache
+ *	@dst: the entry to be cached
+ *	@saddr: the source address to be stored inside the cache
+ *
+ *	local BH must be disabled.
+ */
+void dst_cache_set_ip4(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       __be32 saddr);
+
+#if IS_ENABLED(CONFIG_IPV6)
+
+/**
+ *	dst_cache_set_ip6 - store the ipv6 dst into the cache
+ *	@dst_cache: the cache
+ *	@dst: the entry to be cached
+ *	@saddr: the source address to be stored inside the cache
+ *
+ *	local BH must be disabled.
+ */
+void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       const struct in6_addr *addr);
+
+/**
+ *	dst_cache_get_ip6 - perform cache lookup and fetch ipv6 source address
+ *	@dst_cache: the cache
+ *	@saddr: return value for the retrieved source address
+ *
+ *	local BH must be disabled.
+ */
+struct dst_entry *dst_cache_get_ip6(struct dst_cache *dst_cache,
+				    struct in6_addr *saddr);
+#endif
+
+/**
+ *	dst_cache_reset - invalidate the cache contents
+ *	@dst_cache: the cache
+ *
+ *	This do not free the cached dst to avoid races and contentions.
+ *	the dst will be freed on later cache lookup.
+ */
+static inline void dst_cache_reset(struct dst_cache *dst_cache)
+{
+	dst_cache->reset_ts = jiffies;
+}
+
+/**
+ *	dst_cache_init - initialize the cache, allocating the required storage
+ *	@dst_cache: the cache
+ *	@gfp: allocation flags
+ */
+int dst_cache_init(struct dst_cache *dst_cache, gfp_t gfp);
+
+/**
+ *	dst_cache_destroy - empty the cache and free the allocated storage
+ *	@dst_cache: the cache
+ *
+ *	No synchronization is enforced: it must be called only when the cache
+ *	is unused.
+ */
+void dst_cache_destroy(struct dst_cache *dst_cache);
+
+#endif /* _WG_NET_DST_CACHE_H */
diff --git a/drivers/net/wireguard/compat/fpu-x86/include/asm/fpu/api.h b/drivers/net/wireguard/compat/fpu-x86/include/asm/fpu/api.h
new file mode 100644
index 000000000000..f3f9117bcb88
--- /dev/null
+++ b/drivers/net/wireguard/compat/fpu-x86/include/asm/fpu/api.h
@@ -0,0 +1 @@
+#include <asm/i387.h>
diff --git a/drivers/net/wireguard/compat/intel-family-x86/include/asm/intel-family.h b/drivers/net/wireguard/compat/intel-family-x86/include/asm/intel-family.h
new file mode 100644
index 000000000000..35a6bc4da8ad
--- /dev/null
+++ b/drivers/net/wireguard/compat/intel-family-x86/include/asm/intel-family.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_INTEL_FAMILY_H
+#define _ASM_X86_INTEL_FAMILY_H
+
+/*
+ * "Big Core" Processors (Branded as Core, Xeon, etc...)
+ *
+ * The "_X" parts are generally the EP and EX Xeons, or the
+ * "Extreme" ones, like Broadwell-E.
+ *
+ * Things ending in "2" are usually because we have no better
+ * name for them.  There's no processor called "SILVERMONT2".
+ */
+
+#define INTEL_FAM6_CORE_YONAH		0x0E
+
+#define INTEL_FAM6_CORE2_MEROM		0x0F
+#define INTEL_FAM6_CORE2_MEROM_L	0x16
+#define INTEL_FAM6_CORE2_PENRYN		0x17
+#define INTEL_FAM6_CORE2_DUNNINGTON	0x1D
+
+#define INTEL_FAM6_NEHALEM		0x1E
+#define INTEL_FAM6_NEHALEM_G		0x1F /* Auburndale / Havendale */
+#define INTEL_FAM6_NEHALEM_EP		0x1A
+#define INTEL_FAM6_NEHALEM_EX		0x2E
+
+#define INTEL_FAM6_WESTMERE		0x25
+#define INTEL_FAM6_WESTMERE_EP		0x2C
+#define INTEL_FAM6_WESTMERE_EX		0x2F
+
+#define INTEL_FAM6_SANDYBRIDGE		0x2A
+#define INTEL_FAM6_SANDYBRIDGE_X	0x2D
+#define INTEL_FAM6_IVYBRIDGE		0x3A
+#define INTEL_FAM6_IVYBRIDGE_X		0x3E
+
+#define INTEL_FAM6_HASWELL_CORE		0x3C
+#define INTEL_FAM6_HASWELL_X		0x3F
+#define INTEL_FAM6_HASWELL_ULT		0x45
+#define INTEL_FAM6_HASWELL_GT3E		0x46
+
+#define INTEL_FAM6_BROADWELL_CORE	0x3D
+#define INTEL_FAM6_BROADWELL_GT3E	0x47
+#define INTEL_FAM6_BROADWELL_X		0x4F
+#define INTEL_FAM6_BROADWELL_XEON_D	0x56
+
+#define INTEL_FAM6_SKYLAKE_MOBILE	0x4E
+#define INTEL_FAM6_SKYLAKE_DESKTOP	0x5E
+#define INTEL_FAM6_SKYLAKE_X		0x55
+#define INTEL_FAM6_KABYLAKE_MOBILE	0x8E
+#define INTEL_FAM6_KABYLAKE_DESKTOP	0x9E
+
+/* "Small Core" Processors (Atom) */
+
+#define INTEL_FAM6_ATOM_PINEVIEW	0x1C
+#define INTEL_FAM6_ATOM_LINCROFT	0x26
+#define INTEL_FAM6_ATOM_PENWELL		0x27
+#define INTEL_FAM6_ATOM_CLOVERVIEW	0x35
+#define INTEL_FAM6_ATOM_CEDARVIEW	0x36
+#define INTEL_FAM6_ATOM_SILVERMONT1	0x37 /* BayTrail/BYT / Valleyview */
+#define INTEL_FAM6_ATOM_SILVERMONT2	0x4D /* Avaton/Rangely */
+#define INTEL_FAM6_ATOM_AIRMONT		0x4C /* CherryTrail / Braswell */
+#define INTEL_FAM6_ATOM_MERRIFIELD	0x4A /* Tangier */
+#define INTEL_FAM6_ATOM_MOOREFIELD	0x5A /* Anniedale */
+#define INTEL_FAM6_ATOM_GOLDMONT	0x5C
+#define INTEL_FAM6_ATOM_DENVERTON	0x5F /* Goldmont Microserver */
+#define INTEL_FAM6_ATOM_GEMINI_LAKE	0x7A
+
+/* Xeon Phi */
+
+#define INTEL_FAM6_XEON_PHI_KNL		0x57 /* Knights Landing */
+#define INTEL_FAM6_XEON_PHI_KNM		0x85 /* Knights Mill */
+
+#endif /* _ASM_X86_INTEL_FAMILY_H */
diff --git a/drivers/net/wireguard/compat/memneq/include.h b/drivers/net/wireguard/compat/memneq/include.h
new file mode 100644
index 000000000000..2d18acd9b6c8
--- /dev/null
+++ b/drivers/net/wireguard/compat/memneq/include.h
@@ -0,0 +1,5 @@
+extern noinline unsigned long __crypto_memneq(const void *a, const void *b, size_t size);
+static inline int crypto_memneq(const void *a, const void *b, size_t size)
+{
+	return __crypto_memneq(a, b, size) != 0UL ? 1 : 0;
+}
diff --git a/drivers/net/wireguard/compat/memneq/memneq.c b/drivers/net/wireguard/compat/memneq/memneq.c
new file mode 100644
index 000000000000..1c427d405537
--- /dev/null
+++ b/drivers/net/wireguard/compat/memneq/memneq.c
@@ -0,0 +1,170 @@
+/*
+ * Constant-time equality testing of memory regions.
+ *
+ * Authors:
+ *
+ *   James Yonan <james@openvpn.net>
+ *   Daniel Borkmann <dborkman@redhat.com>
+ *
+ * This file is provided under a dual BSD/GPLv2 license.  When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * Copyright(c) 2013 OpenVPN Technologies, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ * The full GNU General Public License is included in this distribution
+ * in the file called LICENSE.GPL.
+ *
+ * BSD LICENSE
+ *
+ * Copyright(c) 2013 OpenVPN Technologies, Inc. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *   * Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *   * Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in
+ *     the documentation and/or other materials provided with the
+ *     distribution.
+ *   * Neither the name of OpenVPN Technologies nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <crypto/algapi.h>
+
+/* Make the optimizer believe the variable can be manipulated arbitrarily. */
+#define COMPILER_OPTIMIZER_HIDE_VAR(var) asm("" : "=r" (var) : "0" (var))
+
+#ifndef __HAVE_ARCH_CRYPTO_MEMNEQ
+
+/* Generic path for arbitrary size */
+static inline unsigned long
+__crypto_memneq_generic(const void *a, const void *b, size_t size)
+{
+	unsigned long neq = 0;
+
+#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
+	while (size >= sizeof(unsigned long)) {
+		neq |= *(unsigned long *)a ^ *(unsigned long *)b;
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		a += sizeof(unsigned long);
+		b += sizeof(unsigned long);
+		size -= sizeof(unsigned long);
+	}
+#endif /* CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS */
+	while (size > 0) {
+		neq |= *(unsigned char *)a ^ *(unsigned char *)b;
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		a += 1;
+		b += 1;
+		size -= 1;
+	}
+	return neq;
+}
+
+/* Loop-free fast-path for frequently used 16-byte size */
+static inline unsigned long __crypto_memneq_16(const void *a, const void *b)
+{
+	unsigned long neq = 0;
+
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+	if (sizeof(unsigned long) == 8) {
+		neq |= *(unsigned long *)(a)   ^ *(unsigned long *)(b);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned long *)(a+8) ^ *(unsigned long *)(b+8);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+	} else if (sizeof(unsigned int) == 4) {
+		neq |= *(unsigned int *)(a)    ^ *(unsigned int *)(b);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned int *)(a+4)  ^ *(unsigned int *)(b+4);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned int *)(a+8)  ^ *(unsigned int *)(b+8);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned int *)(a+12) ^ *(unsigned int *)(b+12);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+	} else
+#endif /* CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS */
+	{
+		neq |= *(unsigned char *)(a)    ^ *(unsigned char *)(b);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+1)  ^ *(unsigned char *)(b+1);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+2)  ^ *(unsigned char *)(b+2);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+3)  ^ *(unsigned char *)(b+3);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+4)  ^ *(unsigned char *)(b+4);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+5)  ^ *(unsigned char *)(b+5);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+6)  ^ *(unsigned char *)(b+6);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+7)  ^ *(unsigned char *)(b+7);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+8)  ^ *(unsigned char *)(b+8);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+9)  ^ *(unsigned char *)(b+9);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+10) ^ *(unsigned char *)(b+10);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+11) ^ *(unsigned char *)(b+11);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+12) ^ *(unsigned char *)(b+12);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+13) ^ *(unsigned char *)(b+13);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+14) ^ *(unsigned char *)(b+14);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+		neq |= *(unsigned char *)(a+15) ^ *(unsigned char *)(b+15);
+		COMPILER_OPTIMIZER_HIDE_VAR(neq);
+	}
+
+	return neq;
+}
+
+/* Compare two areas of memory without leaking timing information,
+ * and with special optimizations for common sizes.  Users should
+ * not call this function directly, but should instead use
+ * crypto_memneq defined in crypto/algapi.h.
+ */
+noinline unsigned long __crypto_memneq(const void *a, const void *b,
+				       size_t size)
+{
+	switch (size) {
+	case 16:
+		return __crypto_memneq_16(a, b);
+	default:
+		return __crypto_memneq_generic(a, b, size);
+	}
+}
+
+#endif /* __HAVE_ARCH_CRYPTO_MEMNEQ */
diff --git a/drivers/net/wireguard/compat/neon-arm/include/asm/neon.h b/drivers/net/wireguard/compat/neon-arm/include/asm/neon.h
new file mode 100644
index 000000000000..980d831e201a
--- /dev/null
+++ b/drivers/net/wireguard/compat/neon-arm/include/asm/neon.h
@@ -0,0 +1,7 @@
+#ifndef _ARCH_ARM_ASM_NEON
+#define _ARCH_ARM_ASM_NEON
+#define kernel_neon_begin() \
+	BUILD_BUG_ON_MSG(1, "This kernel does not support ARM NEON")
+#define kernel_neon_end() \
+	BUILD_BUG_ON_MSG(1, "This kernel does not support ARM NEON")
+#endif
diff --git a/drivers/net/wireguard/compat/ptr_ring/include/linux/ptr_ring.h b/drivers/net/wireguard/compat/ptr_ring/include/linux/ptr_ring.h
new file mode 100644
index 000000000000..9d514bac1388
--- /dev/null
+++ b/drivers/net/wireguard/compat/ptr_ring/include/linux/ptr_ring.h
@@ -0,0 +1,674 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ *	Definitions for the 'struct ptr_ring' datastructure.
+ *
+ *	Author:
+ *		Michael S. Tsirkin <mst@redhat.com>
+ *
+ *	Copyright (C) 2016 Red Hat, Inc.
+ *
+ *	This is a limited-size FIFO maintaining pointers in FIFO order, with
+ *	one CPU producing entries and another consuming entries from a FIFO.
+ *
+ *	This implementation tries to minimize cache-contention when there is a
+ *	single producer and a single consumer CPU.
+ */
+
+#ifndef _LINUX_PTR_RING_H
+#define _LINUX_PTR_RING_H 1
+
+#ifdef __KERNEL__
+#include <linux/spinlock.h>
+#include <linux/cache.h>
+#include <linux/types.h>
+#include <linux/compiler.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <asm/errno.h>
+#endif
+
+struct ptr_ring {
+	int producer ____cacheline_aligned_in_smp;
+	spinlock_t producer_lock;
+	int consumer_head ____cacheline_aligned_in_smp; /* next valid entry */
+	int consumer_tail; /* next entry to invalidate */
+	spinlock_t consumer_lock;
+	/* Shared consumer/producer data */
+	/* Read-only by both the producer and the consumer */
+	int size ____cacheline_aligned_in_smp; /* max entries in queue */
+	int batch; /* number of entries to consume in a batch */
+	void **queue;
+};
+
+/* Note: callers invoking this in a loop must use a compiler barrier,
+ * for example cpu_relax().
+ *
+ * NB: this is unlike __ptr_ring_empty in that callers must hold producer_lock:
+ * see e.g. ptr_ring_full.
+ */
+static inline bool __ptr_ring_full(struct ptr_ring *r)
+{
+	return r->queue[r->producer];
+}
+
+static inline bool ptr_ring_full(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock(&r->producer_lock);
+	ret = __ptr_ring_full(r);
+	spin_unlock(&r->producer_lock);
+
+	return ret;
+}
+
+static inline bool ptr_ring_full_irq(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock_irq(&r->producer_lock);
+	ret = __ptr_ring_full(r);
+	spin_unlock_irq(&r->producer_lock);
+
+	return ret;
+}
+
+static inline bool ptr_ring_full_any(struct ptr_ring *r)
+{
+	unsigned long flags;
+	bool ret;
+
+	spin_lock_irqsave(&r->producer_lock, flags);
+	ret = __ptr_ring_full(r);
+	spin_unlock_irqrestore(&r->producer_lock, flags);
+
+	return ret;
+}
+
+static inline bool ptr_ring_full_bh(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock_bh(&r->producer_lock);
+	ret = __ptr_ring_full(r);
+	spin_unlock_bh(&r->producer_lock);
+
+	return ret;
+}
+
+/* Note: callers invoking this in a loop must use a compiler barrier,
+ * for example cpu_relax(). Callers must hold producer_lock.
+ * Callers are responsible for making sure pointer that is being queued
+ * points to a valid data.
+ */
+static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr)
+{
+	if (unlikely(!r->size) || r->queue[r->producer])
+		return -ENOSPC;
+
+	/* Make sure the pointer we are storing points to a valid data. */
+	/* Pairs with smp_read_barrier_depends in __ptr_ring_consume. */
+	smp_wmb();
+
+	WRITE_ONCE(r->queue[r->producer++], ptr);
+	if (unlikely(r->producer >= r->size))
+		r->producer = 0;
+	return 0;
+}
+
+/*
+ * Note: resize (below) nests producer lock within consumer lock, so if you
+ * consume in interrupt or BH context, you must disable interrupts/BH when
+ * calling this.
+ */
+static inline int ptr_ring_produce(struct ptr_ring *r, void *ptr)
+{
+	int ret;
+
+	spin_lock(&r->producer_lock);
+	ret = __ptr_ring_produce(r, ptr);
+	spin_unlock(&r->producer_lock);
+
+	return ret;
+}
+
+static inline int ptr_ring_produce_irq(struct ptr_ring *r, void *ptr)
+{
+	int ret;
+
+	spin_lock_irq(&r->producer_lock);
+	ret = __ptr_ring_produce(r, ptr);
+	spin_unlock_irq(&r->producer_lock);
+
+	return ret;
+}
+
+static inline int ptr_ring_produce_any(struct ptr_ring *r, void *ptr)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&r->producer_lock, flags);
+	ret = __ptr_ring_produce(r, ptr);
+	spin_unlock_irqrestore(&r->producer_lock, flags);
+
+	return ret;
+}
+
+static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr)
+{
+	int ret;
+
+	spin_lock_bh(&r->producer_lock);
+	ret = __ptr_ring_produce(r, ptr);
+	spin_unlock_bh(&r->producer_lock);
+
+	return ret;
+}
+
+static inline void *__ptr_ring_peek(struct ptr_ring *r)
+{
+	if (likely(r->size))
+		return READ_ONCE(r->queue[r->consumer_head]);
+	return NULL;
+}
+
+/*
+ * Test ring empty status without taking any locks.
+ *
+ * NB: This is only safe to call if ring is never resized.
+ *
+ * However, if some other CPU consumes ring entries at the same time, the value
+ * returned is not guaranteed to be correct.
+ *
+ * In this case - to avoid incorrectly detecting the ring
+ * as empty - the CPU consuming the ring entries is responsible
+ * for either consuming all ring entries until the ring is empty,
+ * or synchronizing with some other CPU and causing it to
+ * re-test __ptr_ring_empty and/or consume the ring enteries
+ * after the synchronization point.
+ *
+ * Note: callers invoking this in a loop must use a compiler barrier,
+ * for example cpu_relax().
+ */
+static inline bool __ptr_ring_empty(struct ptr_ring *r)
+{
+	if (likely(r->size))
+		return !r->queue[READ_ONCE(r->consumer_head)];
+	return true;
+}
+
+static inline bool ptr_ring_empty(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock(&r->consumer_lock);
+	ret = __ptr_ring_empty(r);
+	spin_unlock(&r->consumer_lock);
+
+	return ret;
+}
+
+static inline bool ptr_ring_empty_irq(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock_irq(&r->consumer_lock);
+	ret = __ptr_ring_empty(r);
+	spin_unlock_irq(&r->consumer_lock);
+
+	return ret;
+}
+
+static inline bool ptr_ring_empty_any(struct ptr_ring *r)
+{
+	unsigned long flags;
+	bool ret;
+
+	spin_lock_irqsave(&r->consumer_lock, flags);
+	ret = __ptr_ring_empty(r);
+	spin_unlock_irqrestore(&r->consumer_lock, flags);
+
+	return ret;
+}
+
+static inline bool ptr_ring_empty_bh(struct ptr_ring *r)
+{
+	bool ret;
+
+	spin_lock_bh(&r->consumer_lock);
+	ret = __ptr_ring_empty(r);
+	spin_unlock_bh(&r->consumer_lock);
+
+	return ret;
+}
+
+/* Must only be called after __ptr_ring_peek returned !NULL */
+static inline void __ptr_ring_discard_one(struct ptr_ring *r)
+{
+	/* Fundamentally, what we want to do is update consumer
+	 * index and zero out the entry so producer can reuse it.
+	 * Doing it naively at each consume would be as simple as:
+	 *       consumer = r->consumer;
+	 *       r->queue[consumer++] = NULL;
+	 *       if (unlikely(consumer >= r->size))
+	 *               consumer = 0;
+	 *       r->consumer = consumer;
+	 * but that is suboptimal when the ring is full as producer is writing
+	 * out new entries in the same cache line.  Defer these updates until a
+	 * batch of entries has been consumed.
+	 */
+	/* Note: we must keep consumer_head valid at all times for __ptr_ring_empty
+	 * to work correctly.
+	 */
+	int consumer_head = r->consumer_head;
+	int head = consumer_head++;
+
+	/* Once we have processed enough entries invalidate them in
+	 * the ring all at once so producer can reuse their space in the ring.
+	 * We also do this when we reach end of the ring - not mandatory
+	 * but helps keep the implementation simple.
+	 */
+	if (unlikely(consumer_head - r->consumer_tail >= r->batch ||
+		     consumer_head >= r->size)) {
+		/* Zero out entries in the reverse order: this way we touch the
+		 * cache line that producer might currently be reading the last;
+		 * producer won't make progress and touch other cache lines
+		 * besides the first one until we write out all entries.
+		 */
+		while (likely(head >= r->consumer_tail))
+			r->queue[head--] = NULL;
+		r->consumer_tail = consumer_head;
+	}
+	if (unlikely(consumer_head >= r->size)) {
+		consumer_head = 0;
+		r->consumer_tail = 0;
+	}
+	/* matching READ_ONCE in __ptr_ring_empty for lockless tests */
+	WRITE_ONCE(r->consumer_head, consumer_head);
+}
+
+static inline void *__ptr_ring_consume(struct ptr_ring *r)
+{
+	void *ptr;
+
+	ptr = __ptr_ring_peek(r);
+	if (ptr)
+		__ptr_ring_discard_one(r);
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 0)
+	/* The READ_ONCE in __ptr_ring_peek doesn't imply a barrier on old kernels. */
+	smp_read_barrier_depends();
+#endif
+
+	return ptr;
+}
+
+static inline int __ptr_ring_consume_batched(struct ptr_ring *r,
+					     void **array, int n)
+{
+	void *ptr;
+	int i;
+
+	for (i = 0; i < n; i++) {
+		ptr = __ptr_ring_consume(r);
+		if (!ptr)
+			break;
+		array[i] = ptr;
+	}
+
+	return i;
+}
+
+/*
+ * Note: resize (below) nests producer lock within consumer lock, so if you
+ * call this in interrupt or BH context, you must disable interrupts/BH when
+ * producing.
+ */
+static inline void *ptr_ring_consume(struct ptr_ring *r)
+{
+	void *ptr;
+
+	spin_lock(&r->consumer_lock);
+	ptr = __ptr_ring_consume(r);
+	spin_unlock(&r->consumer_lock);
+
+	return ptr;
+}
+
+static inline void *ptr_ring_consume_irq(struct ptr_ring *r)
+{
+	void *ptr;
+
+	spin_lock_irq(&r->consumer_lock);
+	ptr = __ptr_ring_consume(r);
+	spin_unlock_irq(&r->consumer_lock);
+
+	return ptr;
+}
+
+static inline void *ptr_ring_consume_any(struct ptr_ring *r)
+{
+	unsigned long flags;
+	void *ptr;
+
+	spin_lock_irqsave(&r->consumer_lock, flags);
+	ptr = __ptr_ring_consume(r);
+	spin_unlock_irqrestore(&r->consumer_lock, flags);
+
+	return ptr;
+}
+
+static inline void *ptr_ring_consume_bh(struct ptr_ring *r)
+{
+	void *ptr;
+
+	spin_lock_bh(&r->consumer_lock);
+	ptr = __ptr_ring_consume(r);
+	spin_unlock_bh(&r->consumer_lock);
+
+	return ptr;
+}
+
+static inline int ptr_ring_consume_batched(struct ptr_ring *r,
+					   void **array, int n)
+{
+	int ret;
+
+	spin_lock(&r->consumer_lock);
+	ret = __ptr_ring_consume_batched(r, array, n);
+	spin_unlock(&r->consumer_lock);
+
+	return ret;
+}
+
+static inline int ptr_ring_consume_batched_irq(struct ptr_ring *r,
+					       void **array, int n)
+{
+	int ret;
+
+	spin_lock_irq(&r->consumer_lock);
+	ret = __ptr_ring_consume_batched(r, array, n);
+	spin_unlock_irq(&r->consumer_lock);
+
+	return ret;
+}
+
+static inline int ptr_ring_consume_batched_any(struct ptr_ring *r,
+					       void **array, int n)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&r->consumer_lock, flags);
+	ret = __ptr_ring_consume_batched(r, array, n);
+	spin_unlock_irqrestore(&r->consumer_lock, flags);
+
+	return ret;
+}
+
+static inline int ptr_ring_consume_batched_bh(struct ptr_ring *r,
+					      void **array, int n)
+{
+	int ret;
+
+	spin_lock_bh(&r->consumer_lock);
+	ret = __ptr_ring_consume_batched(r, array, n);
+	spin_unlock_bh(&r->consumer_lock);
+
+	return ret;
+}
+
+/* Cast to structure type and call a function without discarding from FIFO.
+ * Function must return a value.
+ * Callers must take consumer_lock.
+ */
+#define __PTR_RING_PEEK_CALL(r, f) ((f)(__ptr_ring_peek(r)))
+
+#define PTR_RING_PEEK_CALL(r, f) ({ \
+	typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \
+	\
+	spin_lock(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \
+	spin_unlock(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v; \
+})
+
+#define PTR_RING_PEEK_CALL_IRQ(r, f) ({ \
+	typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \
+	\
+	spin_lock_irq(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \
+	spin_unlock_irq(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v; \
+})
+
+#define PTR_RING_PEEK_CALL_BH(r, f) ({ \
+	typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \
+	\
+	spin_lock_bh(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \
+	spin_unlock_bh(&(r)->consumer_lock); \
+	__PTR_RING_PEEK_CALL_v; \
+})
+
+#define PTR_RING_PEEK_CALL_ANY(r, f) ({ \
+	typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \
+	unsigned long __PTR_RING_PEEK_CALL_f;\
+	\
+	spin_lock_irqsave(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \
+	__PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \
+	spin_unlock_irqrestore(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \
+	__PTR_RING_PEEK_CALL_v; \
+})
+
+/* Not all gfp_t flags (besides GFP_KERNEL) are allowed. See
+ * documentation for vmalloc for which of them are legal.
+ */
+static inline void **__ptr_ring_init_queue_alloc(unsigned int size, gfp_t gfp)
+{
+	if (size > KMALLOC_MAX_SIZE / sizeof(void *))
+		return NULL;
+	return kvmalloc(size * sizeof(void *), gfp | __GFP_ZERO);
+}
+
+static inline void __ptr_ring_set_size(struct ptr_ring *r, int size)
+{
+	r->size = size;
+	r->batch = SMP_CACHE_BYTES * 2 / sizeof(*(r->queue));
+	/* We need to set batch at least to 1 to make logic
+	 * in __ptr_ring_discard_one work correctly.
+	 * Batching too much (because ring is small) would cause a lot of
+	 * burstiness. Needs tuning, for now disable batching.
+	 */
+	if (r->batch > r->size / 2 || !r->batch)
+		r->batch = 1;
+}
+
+static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp)
+{
+	r->queue = __ptr_ring_init_queue_alloc(size, gfp);
+	if (!r->queue)
+		return -ENOMEM;
+
+	__ptr_ring_set_size(r, size);
+	r->producer = r->consumer_head = r->consumer_tail = 0;
+	spin_lock_init(&r->producer_lock);
+	spin_lock_init(&r->consumer_lock);
+
+	return 0;
+}
+
+/*
+ * Return entries into ring. Destroy entries that don't fit.
+ *
+ * Note: this is expected to be a rare slow path operation.
+ *
+ * Note: producer lock is nested within consumer lock, so if you
+ * resize you must make sure all uses nest correctly.
+ * In particular if you consume ring in interrupt or BH context, you must
+ * disable interrupts/BH when doing so.
+ */
+static inline void ptr_ring_unconsume(struct ptr_ring *r, void **batch, int n,
+				      void (*destroy)(void *))
+{
+	unsigned long flags;
+	int head;
+
+	spin_lock_irqsave(&r->consumer_lock, flags);
+	spin_lock(&r->producer_lock);
+
+	if (!r->size)
+		goto done;
+
+	/*
+	 * Clean out buffered entries (for simplicity). This way following code
+	 * can test entries for NULL and if not assume they are valid.
+	 */
+	head = r->consumer_head - 1;
+	while (likely(head >= r->consumer_tail))
+		r->queue[head--] = NULL;
+	r->consumer_tail = r->consumer_head;
+
+	/*
+	 * Go over entries in batch, start moving head back and copy entries.
+	 * Stop when we run into previously unconsumed entries.
+	 */
+	while (n) {
+		head = r->consumer_head - 1;
+		if (head < 0)
+			head = r->size - 1;
+		if (r->queue[head]) {
+			/* This batch entry will have to be destroyed. */
+			goto done;
+		}
+		r->queue[head] = batch[--n];
+		r->consumer_tail = head;
+		/* matching READ_ONCE in __ptr_ring_empty for lockless tests */
+		WRITE_ONCE(r->consumer_head, head);
+	}
+
+done:
+	/* Destroy all entries left in the batch. */
+	while (n)
+		destroy(batch[--n]);
+	spin_unlock(&r->producer_lock);
+	spin_unlock_irqrestore(&r->consumer_lock, flags);
+}
+
+static inline void **__ptr_ring_swap_queue(struct ptr_ring *r, void **queue,
+					   int size, gfp_t gfp,
+					   void (*destroy)(void *))
+{
+	int producer = 0;
+	void **old;
+	void *ptr;
+
+	while ((ptr = __ptr_ring_consume(r)))
+		if (producer < size)
+			queue[producer++] = ptr;
+		else if (destroy)
+			destroy(ptr);
+
+	if (producer >= size)
+		producer = 0;
+	__ptr_ring_set_size(r, size);
+	r->producer = producer;
+	r->consumer_head = 0;
+	r->consumer_tail = 0;
+	old = r->queue;
+	r->queue = queue;
+
+	return old;
+}
+
+/*
+ * Note: producer lock is nested within consumer lock, so if you
+ * resize you must make sure all uses nest correctly.
+ * In particular if you consume ring in interrupt or BH context, you must
+ * disable interrupts/BH when doing so.
+ */
+static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp,
+				  void (*destroy)(void *))
+{
+	unsigned long flags;
+	void **queue = __ptr_ring_init_queue_alloc(size, gfp);
+	void **old;
+
+	if (!queue)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&(r)->consumer_lock, flags);
+	spin_lock(&(r)->producer_lock);
+
+	old = __ptr_ring_swap_queue(r, queue, size, gfp, destroy);
+
+	spin_unlock(&(r)->producer_lock);
+	spin_unlock_irqrestore(&(r)->consumer_lock, flags);
+
+	kvfree(old);
+
+	return 0;
+}
+
+/*
+ * Note: producer lock is nested within consumer lock, so if you
+ * resize you must make sure all uses nest correctly.
+ * In particular if you consume ring in interrupt or BH context, you must
+ * disable interrupts/BH when doing so.
+ */
+static inline int ptr_ring_resize_multiple(struct ptr_ring **rings,
+					   unsigned int nrings,
+					   int size,
+					   gfp_t gfp, void (*destroy)(void *))
+{
+	unsigned long flags;
+	void ***queues;
+	int i;
+
+	queues = kmalloc_array(nrings, sizeof(*queues), gfp);
+	if (!queues)
+		goto noqueues;
+
+	for (i = 0; i < nrings; ++i) {
+		queues[i] = __ptr_ring_init_queue_alloc(size, gfp);
+		if (!queues[i])
+			goto nomem;
+	}
+
+	for (i = 0; i < nrings; ++i) {
+		spin_lock_irqsave(&(rings[i])->consumer_lock, flags);
+		spin_lock(&(rings[i])->producer_lock);
+		queues[i] = __ptr_ring_swap_queue(rings[i], queues[i],
+						  size, gfp, destroy);
+		spin_unlock(&(rings[i])->producer_lock);
+		spin_unlock_irqrestore(&(rings[i])->consumer_lock, flags);
+	}
+
+	for (i = 0; i < nrings; ++i)
+		kvfree(queues[i]);
+
+	kfree(queues);
+
+	return 0;
+
+nomem:
+	while (--i >= 0)
+		kvfree(queues[i]);
+
+	kfree(queues);
+
+noqueues:
+	return -ENOMEM;
+}
+
+static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *))
+{
+	void *ptr;
+
+	if (destroy)
+		while ((ptr = ptr_ring_consume(r)))
+			destroy(ptr);
+	kvfree(r->queue);
+}
+
+#endif /* _LINUX_PTR_RING_H  */
diff --git a/drivers/net/wireguard/compat/simd-asm/include/asm/simd.h b/drivers/net/wireguard/compat/simd-asm/include/asm/simd.h
new file mode 100644
index 000000000000..a975b38b5578
--- /dev/null
+++ b/drivers/net/wireguard/compat/simd-asm/include/asm/simd.h
@@ -0,0 +1,21 @@
+#ifndef _COMPAT_ASM_SIMD_H
+#define _COMPAT_ASM_SIMD_H
+
+#if defined(CONFIG_X86_64)
+#include <asm/fpu/api.h>
+#endif
+
+static __must_check inline bool may_use_simd(void)
+{
+#if defined(CONFIG_X86_64)
+	return irq_fpu_usable();
+#elif defined(CONFIG_ARM64) && defined(CONFIG_KERNEL_MODE_NEON)
+	return true;
+#elif defined(CONFIG_ARM) && defined(CONFIG_KERNEL_MODE_NEON)
+	return !in_nmi() && !in_irq() && !in_serving_softirq();
+#else
+	return false;
+#endif
+}
+
+#endif
diff --git a/drivers/net/wireguard/compat/simd/include/linux/simd.h b/drivers/net/wireguard/compat/simd/include/linux/simd.h
new file mode 100644
index 000000000000..c75c72471e9e
--- /dev/null
+++ b/drivers/net/wireguard/compat/simd/include/linux/simd.h
@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _WG_SIMD_H
+#define _WG_SIMD_H
+
+#include <linux/sched.h>
+#include <asm/simd.h>
+#if defined(CONFIG_X86_64)
+#include <linux/version.h>
+#include <asm/fpu/api.h>
+#elif defined(CONFIG_KERNEL_MODE_NEON)
+#include <asm/neon.h>
+#endif
+
+typedef enum {
+	HAVE_NO_SIMD = 1 << 0,
+	HAVE_FULL_SIMD = 1 << 1,
+	HAVE_SIMD_IN_USE = 1 << 31
+} simd_context_t;
+
+#define DONT_USE_SIMD ((simd_context_t []){ HAVE_NO_SIMD })
+
+static inline void simd_get(simd_context_t *ctx)
+{
+	*ctx = !IS_ENABLED(CONFIG_PREEMPT_RT_BASE) && may_use_simd() ? HAVE_FULL_SIMD : HAVE_NO_SIMD;
+}
+
+static inline void simd_put(simd_context_t *ctx)
+{
+#if defined(CONFIG_X86_64)
+	if (*ctx & HAVE_SIMD_IN_USE)
+		kernel_fpu_end();
+#elif defined(CONFIG_KERNEL_MODE_NEON)
+	if (*ctx & HAVE_SIMD_IN_USE)
+		kernel_neon_end();
+#endif
+	*ctx = HAVE_NO_SIMD;
+}
+
+static inline bool simd_relax(simd_context_t *ctx)
+{
+#ifdef CONFIG_PREEMPT
+	if ((*ctx & HAVE_SIMD_IN_USE) && need_resched()) {
+		simd_put(ctx);
+		simd_get(ctx);
+		return true;
+	}
+#endif
+	return false;
+}
+
+static __must_check inline bool simd_use(simd_context_t *ctx)
+{
+	if (!(*ctx & HAVE_FULL_SIMD))
+		return false;
+	if (*ctx & HAVE_SIMD_IN_USE)
+		return true;
+#if defined(CONFIG_X86_64)
+	kernel_fpu_begin();
+#elif defined(CONFIG_KERNEL_MODE_NEON)
+	kernel_neon_begin();
+#endif
+	*ctx |= HAVE_SIMD_IN_USE;
+	return true;
+}
+
+#endif /* _WG_SIMD_H */
diff --git a/drivers/net/wireguard/compat/siphash/include/linux/siphash.h b/drivers/net/wireguard/compat/siphash/include/linux/siphash.h
new file mode 100644
index 000000000000..1e5e337d15bf
--- /dev/null
+++ b/drivers/net/wireguard/compat/siphash/include/linux/siphash.h
@@ -0,0 +1,140 @@
+/* Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ *
+ * This file is provided under a dual BSD/GPLv2 license.
+ *
+ * SipHash: a fast short-input PRF
+ * https://131002.net/siphash/
+ *
+ * This implementation is specifically for SipHash2-4 for a secure PRF
+ * and HalfSipHash1-3/SipHash1-3 for an insecure PRF only suitable for
+ * hashtables.
+ */
+
+#ifndef _WG_LINUX_SIPHASH_H
+#define _WG_LINUX_SIPHASH_H
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+
+#define SIPHASH_ALIGNMENT __alignof__(u64)
+typedef struct {
+	u64 key[2];
+} siphash_key_t;
+
+u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key);
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key);
+#endif
+
+u64 siphash_1u64(const u64 a, const siphash_key_t *key);
+u64 siphash_2u64(const u64 a, const u64 b, const siphash_key_t *key);
+u64 siphash_3u64(const u64 a, const u64 b, const u64 c,
+		 const siphash_key_t *key);
+u64 siphash_4u64(const u64 a, const u64 b, const u64 c, const u64 d,
+		 const siphash_key_t *key);
+u64 siphash_1u32(const u32 a, const siphash_key_t *key);
+u64 siphash_3u32(const u32 a, const u32 b, const u32 c,
+		 const siphash_key_t *key);
+
+static inline u64 siphash_2u32(const u32 a, const u32 b,
+			       const siphash_key_t *key)
+{
+	return siphash_1u64((u64)b << 32 | a, key);
+}
+static inline u64 siphash_4u32(const u32 a, const u32 b, const u32 c,
+			       const u32 d, const siphash_key_t *key)
+{
+	return siphash_2u64((u64)b << 32 | a, (u64)d << 32 | c, key);
+}
+
+
+static inline u64 ___siphash_aligned(const __le64 *data, size_t len,
+				     const siphash_key_t *key)
+{
+	if (__builtin_constant_p(len) && len == 4)
+		return siphash_1u32(le32_to_cpup((const __le32 *)data), key);
+	if (__builtin_constant_p(len) && len == 8)
+		return siphash_1u64(le64_to_cpu(data[0]), key);
+	if (__builtin_constant_p(len) && len == 16)
+		return siphash_2u64(le64_to_cpu(data[0]), le64_to_cpu(data[1]),
+				    key);
+	if (__builtin_constant_p(len) && len == 24)
+		return siphash_3u64(le64_to_cpu(data[0]), le64_to_cpu(data[1]),
+				    le64_to_cpu(data[2]), key);
+	if (__builtin_constant_p(len) && len == 32)
+		return siphash_4u64(le64_to_cpu(data[0]), le64_to_cpu(data[1]),
+				    le64_to_cpu(data[2]), le64_to_cpu(data[3]),
+				    key);
+	return __siphash_aligned(data, len, key);
+}
+
+/**
+ * siphash - compute 64-bit siphash PRF value
+ * @data: buffer to hash
+ * @size: size of @data
+ * @key: the siphash key
+ */
+static inline u64 siphash(const void *data, size_t len,
+			  const siphash_key_t *key)
+{
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+	if (!IS_ALIGNED((unsigned long)data, SIPHASH_ALIGNMENT))
+		return __siphash_unaligned(data, len, key);
+#endif
+	return ___siphash_aligned(data, len, key);
+}
+
+#define HSIPHASH_ALIGNMENT __alignof__(unsigned long)
+typedef struct {
+	unsigned long key[2];
+} hsiphash_key_t;
+
+u32 __hsiphash_aligned(const void *data, size_t len,
+		       const hsiphash_key_t *key);
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+u32 __hsiphash_unaligned(const void *data, size_t len,
+			 const hsiphash_key_t *key);
+#endif
+
+u32 hsiphash_1u32(const u32 a, const hsiphash_key_t *key);
+u32 hsiphash_2u32(const u32 a, const u32 b, const hsiphash_key_t *key);
+u32 hsiphash_3u32(const u32 a, const u32 b, const u32 c,
+		  const hsiphash_key_t *key);
+u32 hsiphash_4u32(const u32 a, const u32 b, const u32 c, const u32 d,
+		  const hsiphash_key_t *key);
+
+static inline u32 ___hsiphash_aligned(const __le32 *data, size_t len,
+				      const hsiphash_key_t *key)
+{
+	if (__builtin_constant_p(len) && len == 4)
+		return hsiphash_1u32(le32_to_cpu(data[0]), key);
+	if (__builtin_constant_p(len) && len == 8)
+		return hsiphash_2u32(le32_to_cpu(data[0]), le32_to_cpu(data[1]),
+				     key);
+	if (__builtin_constant_p(len) && len == 12)
+		return hsiphash_3u32(le32_to_cpu(data[0]), le32_to_cpu(data[1]),
+				     le32_to_cpu(data[2]), key);
+	if (__builtin_constant_p(len) && len == 16)
+		return hsiphash_4u32(le32_to_cpu(data[0]), le32_to_cpu(data[1]),
+				     le32_to_cpu(data[2]), le32_to_cpu(data[3]),
+				     key);
+	return __hsiphash_aligned(data, len, key);
+}
+
+/**
+ * hsiphash - compute 32-bit hsiphash PRF value
+ * @data: buffer to hash
+ * @size: size of @data
+ * @key: the hsiphash key
+ */
+static inline u32 hsiphash(const void *data, size_t len,
+			   const hsiphash_key_t *key)
+{
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+	if (!IS_ALIGNED((unsigned long)data, HSIPHASH_ALIGNMENT))
+		return __hsiphash_unaligned(data, len, key);
+#endif
+	return ___hsiphash_aligned(data, len, key);
+}
+
+#endif /* _WG_LINUX_SIPHASH_H */
diff --git a/drivers/net/wireguard/compat/siphash/siphash.c b/drivers/net/wireguard/compat/siphash/siphash.c
new file mode 100644
index 000000000000..58855328e6e0
--- /dev/null
+++ b/drivers/net/wireguard/compat/siphash/siphash.c
@@ -0,0 +1,539 @@
+/* Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ *
+ * This file is provided under a dual BSD/GPLv2 license.
+ *
+ * SipHash: a fast short-input PRF
+ * https://131002.net/siphash/
+ *
+ * This implementation is specifically for SipHash2-4 for a secure PRF
+ * and HalfSipHash1-3/SipHash1-3 for an insecure PRF only suitable for
+ * hashtables.
+ */
+
+#include <linux/siphash.h>
+#include <asm/unaligned.h>
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0)
+#ifdef __LITTLE_ENDIAN
+#define bytemask_from_count(cnt)	(~(~0ul << (cnt)*8))
+#else
+#define bytemask_from_count(cnt)	(~(~0ul >> (cnt)*8))
+#endif
+#endif
+
+#if defined(CONFIG_DCACHE_WORD_ACCESS) && BITS_PER_LONG == 64
+#include <linux/dcache.h>
+#include <asm/word-at-a-time.h>
+#endif
+
+#define SIPROUND \
+	do { \
+	v0 += v1; v1 = rol64(v1, 13); v1 ^= v0; v0 = rol64(v0, 32); \
+	v2 += v3; v3 = rol64(v3, 16); v3 ^= v2; \
+	v0 += v3; v3 = rol64(v3, 21); v3 ^= v0; \
+	v2 += v1; v1 = rol64(v1, 17); v1 ^= v2; v2 = rol64(v2, 32); \
+	} while (0)
+
+#define PREAMBLE(len) \
+	u64 v0 = 0x736f6d6570736575ULL; \
+	u64 v1 = 0x646f72616e646f6dULL; \
+	u64 v2 = 0x6c7967656e657261ULL; \
+	u64 v3 = 0x7465646279746573ULL; \
+	u64 b = ((u64)(len)) << 56; \
+	v3 ^= key->key[1]; \
+	v2 ^= key->key[0]; \
+	v1 ^= key->key[1]; \
+	v0 ^= key->key[0];
+
+#define POSTAMBLE \
+	v3 ^= b; \
+	SIPROUND; \
+	SIPROUND; \
+	v0 ^= b; \
+	v2 ^= 0xff; \
+	SIPROUND; \
+	SIPROUND; \
+	SIPROUND; \
+	SIPROUND; \
+	return (v0 ^ v1) ^ (v2 ^ v3);
+
+u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u64));
+	const u8 left = len & (sizeof(u64) - 1);
+	u64 m;
+	PREAMBLE(len)
+	for (; data != end; data += sizeof(u64)) {
+		m = le64_to_cpup(data);
+		v3 ^= m;
+		SIPROUND;
+		SIPROUND;
+		v0 ^= m;
+	}
+#if defined(CONFIG_DCACHE_WORD_ACCESS) && BITS_PER_LONG == 64
+	if (left)
+		b |= le64_to_cpu((__force __le64)(load_unaligned_zeropad(data) &
+						  bytemask_from_count(left)));
+#else
+	switch (left) {
+	case 7: b |= ((u64)end[6]) << 48;
+	case 6: b |= ((u64)end[5]) << 40;
+	case 5: b |= ((u64)end[4]) << 32;
+	case 4: b |= le32_to_cpup(data); break;
+	case 3: b |= ((u64)end[2]) << 16;
+	case 2: b |= le16_to_cpup(data); break;
+	case 1: b |= end[0];
+	}
+#endif
+	POSTAMBLE
+}
+
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u64));
+	const u8 left = len & (sizeof(u64) - 1);
+	u64 m;
+	PREAMBLE(len)
+	for (; data != end; data += sizeof(u64)) {
+		m = get_unaligned_le64(data);
+		v3 ^= m;
+		SIPROUND;
+		SIPROUND;
+		v0 ^= m;
+	}
+#if defined(CONFIG_DCACHE_WORD_ACCESS) && BITS_PER_LONG == 64
+	if (left)
+		b |= le64_to_cpu((__force __le64)(load_unaligned_zeropad(data) &
+						  bytemask_from_count(left)));
+#else
+	switch (left) {
+	case 7: b |= ((u64)end[6]) << 48;
+	case 6: b |= ((u64)end[5]) << 40;
+	case 5: b |= ((u64)end[4]) << 32;
+	case 4: b |= get_unaligned_le32(end); break;
+	case 3: b |= ((u64)end[2]) << 16;
+	case 2: b |= get_unaligned_le16(end); break;
+	case 1: b |= end[0];
+	}
+#endif
+	POSTAMBLE
+}
+#endif
+
+/**
+ * siphash_1u64 - compute 64-bit siphash PRF value of a u64
+ * @first: first u64
+ * @key: the siphash key
+ */
+u64 siphash_1u64(const u64 first, const siphash_key_t *key)
+{
+	PREAMBLE(8)
+	v3 ^= first;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= first;
+	POSTAMBLE
+}
+
+/**
+ * siphash_2u64 - compute 64-bit siphash PRF value of 2 u64
+ * @first: first u64
+ * @second: second u64
+ * @key: the siphash key
+ */
+u64 siphash_2u64(const u64 first, const u64 second, const siphash_key_t *key)
+{
+	PREAMBLE(16)
+	v3 ^= first;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= second;
+	POSTAMBLE
+}
+
+/**
+ * siphash_3u64 - compute 64-bit siphash PRF value of 3 u64
+ * @first: first u64
+ * @second: second u64
+ * @third: third u64
+ * @key: the siphash key
+ */
+u64 siphash_3u64(const u64 first, const u64 second, const u64 third,
+		 const siphash_key_t *key)
+{
+	PREAMBLE(24)
+	v3 ^= first;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= second;
+	v3 ^= third;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= third;
+	POSTAMBLE
+}
+
+/**
+ * siphash_4u64 - compute 64-bit siphash PRF value of 4 u64
+ * @first: first u64
+ * @second: second u64
+ * @third: third u64
+ * @forth: forth u64
+ * @key: the siphash key
+ */
+u64 siphash_4u64(const u64 first, const u64 second, const u64 third,
+		 const u64 forth, const siphash_key_t *key)
+{
+	PREAMBLE(32)
+	v3 ^= first;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= second;
+	v3 ^= third;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= third;
+	v3 ^= forth;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= forth;
+	POSTAMBLE
+}
+
+u64 siphash_1u32(const u32 first, const siphash_key_t *key)
+{
+	PREAMBLE(4)
+	b |= first;
+	POSTAMBLE
+}
+
+u64 siphash_3u32(const u32 first, const u32 second, const u32 third,
+		 const siphash_key_t *key)
+{
+	u64 combined = (u64)second << 32 | first;
+	PREAMBLE(12)
+	v3 ^= combined;
+	SIPROUND;
+	SIPROUND;
+	v0 ^= combined;
+	b |= third;
+	POSTAMBLE
+}
+
+#if BITS_PER_LONG == 64
+/* Note that on 64-bit, we make HalfSipHash1-3 actually be SipHash1-3, for
+ * performance reasons. On 32-bit, below, we actually implement HalfSipHash1-3.
+ */
+
+#define HSIPROUND SIPROUND
+#define HPREAMBLE(len) PREAMBLE(len)
+#define HPOSTAMBLE \
+	v3 ^= b; \
+	HSIPROUND; \
+	v0 ^= b; \
+	v2 ^= 0xff; \
+	HSIPROUND; \
+	HSIPROUND; \
+	HSIPROUND; \
+	return (v0 ^ v1) ^ (v2 ^ v3);
+
+u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u64));
+	const u8 left = len & (sizeof(u64) - 1);
+	u64 m;
+	HPREAMBLE(len)
+	for (; data != end; data += sizeof(u64)) {
+		m = le64_to_cpup(data);
+		v3 ^= m;
+		HSIPROUND;
+		v0 ^= m;
+	}
+#if defined(CONFIG_DCACHE_WORD_ACCESS) && BITS_PER_LONG == 64
+	if (left)
+		b |= le64_to_cpu((__force __le64)(load_unaligned_zeropad(data) &
+						  bytemask_from_count(left)));
+#else
+	switch (left) {
+	case 7: b |= ((u64)end[6]) << 48;
+	case 6: b |= ((u64)end[5]) << 40;
+	case 5: b |= ((u64)end[4]) << 32;
+	case 4: b |= le32_to_cpup(data); break;
+	case 3: b |= ((u64)end[2]) << 16;
+	case 2: b |= le16_to_cpup(data); break;
+	case 1: b |= end[0];
+	}
+#endif
+	HPOSTAMBLE
+}
+
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+u32 __hsiphash_unaligned(const void *data, size_t len,
+			 const hsiphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u64));
+	const u8 left = len & (sizeof(u64) - 1);
+	u64 m;
+	HPREAMBLE(len)
+	for (; data != end; data += sizeof(u64)) {
+		m = get_unaligned_le64(data);
+		v3 ^= m;
+		HSIPROUND;
+		v0 ^= m;
+	}
+#if defined(CONFIG_DCACHE_WORD_ACCESS) && BITS_PER_LONG == 64
+	if (left)
+		b |= le64_to_cpu((__force __le64)(load_unaligned_zeropad(data) &
+						  bytemask_from_count(left)));
+#else
+	switch (left) {
+	case 7: b |= ((u64)end[6]) << 48;
+	case 6: b |= ((u64)end[5]) << 40;
+	case 5: b |= ((u64)end[4]) << 32;
+	case 4: b |= get_unaligned_le32(end); break;
+	case 3: b |= ((u64)end[2]) << 16;
+	case 2: b |= get_unaligned_le16(end); break;
+	case 1: b |= end[0];
+	}
+#endif
+	HPOSTAMBLE
+}
+#endif
+
+/**
+ * hsiphash_1u32 - compute 64-bit hsiphash PRF value of a u32
+ * @first: first u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_1u32(const u32 first, const hsiphash_key_t *key)
+{
+	HPREAMBLE(4)
+	b |= first;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_2u32 - compute 32-bit hsiphash PRF value of 2 u32
+ * @first: first u32
+ * @second: second u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_2u32(const u32 first, const u32 second, const hsiphash_key_t *key)
+{
+	u64 combined = (u64)second << 32 | first;
+	HPREAMBLE(8)
+	v3 ^= combined;
+	HSIPROUND;
+	v0 ^= combined;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_3u32 - compute 32-bit hsiphash PRF value of 3 u32
+ * @first: first u32
+ * @second: second u32
+ * @third: third u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_3u32(const u32 first, const u32 second, const u32 third,
+		  const hsiphash_key_t *key)
+{
+	u64 combined = (u64)second << 32 | first;
+	HPREAMBLE(12)
+	v3 ^= combined;
+	HSIPROUND;
+	v0 ^= combined;
+	b |= third;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_4u32 - compute 32-bit hsiphash PRF value of 4 u32
+ * @first: first u32
+ * @second: second u32
+ * @third: third u32
+ * @forth: forth u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_4u32(const u32 first, const u32 second, const u32 third,
+		  const u32 forth, const hsiphash_key_t *key)
+{
+	u64 combined = (u64)second << 32 | first;
+	HPREAMBLE(16)
+	v3 ^= combined;
+	HSIPROUND;
+	v0 ^= combined;
+	combined = (u64)forth << 32 | third;
+	v3 ^= combined;
+	HSIPROUND;
+	v0 ^= combined;
+	HPOSTAMBLE
+}
+#else
+#define HSIPROUND \
+	do { \
+	v0 += v1; v1 = rol32(v1, 5); v1 ^= v0; v0 = rol32(v0, 16); \
+	v2 += v3; v3 = rol32(v3, 8); v3 ^= v2; \
+	v0 += v3; v3 = rol32(v3, 7); v3 ^= v0; \
+	v2 += v1; v1 = rol32(v1, 13); v1 ^= v2; v2 = rol32(v2, 16); \
+	} while (0)
+
+#define HPREAMBLE(len) \
+	u32 v0 = 0; \
+	u32 v1 = 0; \
+	u32 v2 = 0x6c796765U; \
+	u32 v3 = 0x74656462U; \
+	u32 b = ((u32)(len)) << 24; \
+	v3 ^= key->key[1]; \
+	v2 ^= key->key[0]; \
+	v1 ^= key->key[1]; \
+	v0 ^= key->key[0];
+
+#define HPOSTAMBLE \
+	v3 ^= b; \
+	HSIPROUND; \
+	v0 ^= b; \
+	v2 ^= 0xff; \
+	HSIPROUND; \
+	HSIPROUND; \
+	HSIPROUND; \
+	return v1 ^ v3;
+
+u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u32));
+	const u8 left = len & (sizeof(u32) - 1);
+	u32 m;
+	HPREAMBLE(len)
+	for (; data != end; data += sizeof(u32)) {
+		m = le32_to_cpup(data);
+		v3 ^= m;
+		HSIPROUND;
+		v0 ^= m;
+	}
+	switch (left) {
+	case 3: b |= ((u32)end[2]) << 16;
+	case 2: b |= le16_to_cpup(data); break;
+	case 1: b |= end[0];
+	}
+	HPOSTAMBLE
+}
+
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+u32 __hsiphash_unaligned(const void *data, size_t len,
+			 const hsiphash_key_t *key)
+{
+	const u8 *end = data + len - (len % sizeof(u32));
+	const u8 left = len & (sizeof(u32) - 1);
+	u32 m;
+	HPREAMBLE(len)
+	for (; data != end; data += sizeof(u32)) {
+		m = get_unaligned_le32(data);
+		v3 ^= m;
+		HSIPROUND;
+		v0 ^= m;
+	}
+	switch (left) {
+	case 3: b |= ((u32)end[2]) << 16;
+	case 2: b |= get_unaligned_le16(end); break;
+	case 1: b |= end[0];
+	}
+	HPOSTAMBLE
+}
+#endif
+
+/**
+ * hsiphash_1u32 - compute 32-bit hsiphash PRF value of a u32
+ * @first: first u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_1u32(const u32 first, const hsiphash_key_t *key)
+{
+	HPREAMBLE(4)
+	v3 ^= first;
+	HSIPROUND;
+	v0 ^= first;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_2u32 - compute 32-bit hsiphash PRF value of 2 u32
+ * @first: first u32
+ * @second: second u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_2u32(const u32 first, const u32 second, const hsiphash_key_t *key)
+{
+	HPREAMBLE(8)
+	v3 ^= first;
+	HSIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	HSIPROUND;
+	v0 ^= second;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_3u32 - compute 32-bit hsiphash PRF value of 3 u32
+ * @first: first u32
+ * @second: second u32
+ * @third: third u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_3u32(const u32 first, const u32 second, const u32 third,
+		  const hsiphash_key_t *key)
+{
+	HPREAMBLE(12)
+	v3 ^= first;
+	HSIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	HSIPROUND;
+	v0 ^= second;
+	v3 ^= third;
+	HSIPROUND;
+	v0 ^= third;
+	HPOSTAMBLE
+}
+
+/**
+ * hsiphash_4u32 - compute 32-bit hsiphash PRF value of 4 u32
+ * @first: first u32
+ * @second: second u32
+ * @third: third u32
+ * @forth: forth u32
+ * @key: the hsiphash key
+ */
+u32 hsiphash_4u32(const u32 first, const u32 second, const u32 third,
+		  const u32 forth, const hsiphash_key_t *key)
+{
+	HPREAMBLE(16)
+	v3 ^= first;
+	HSIPROUND;
+	v0 ^= first;
+	v3 ^= second;
+	HSIPROUND;
+	v0 ^= second;
+	v3 ^= third;
+	HSIPROUND;
+	v0 ^= third;
+	v3 ^= forth;
+	HSIPROUND;
+	v0 ^= forth;
+	HPOSTAMBLE
+}
+#endif
diff --git a/drivers/net/wireguard/compat/udp_tunnel/include/net/udp_tunnel.h b/drivers/net/wireguard/compat/udp_tunnel/include/net/udp_tunnel.h
new file mode 100644
index 000000000000..8999527d6952
--- /dev/null
+++ b/drivers/net/wireguard/compat/udp_tunnel/include/net/udp_tunnel.h
@@ -0,0 +1,94 @@
+#ifndef _WG_NET_UDP_TUNNEL_H
+#define _WG_NET_UDP_TUNNEL_H
+
+#include <net/ip_tunnels.h>
+#include <net/udp.h>
+
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ipv6.h>
+#include <net/addrconf.h>
+#endif
+
+struct udp_port_cfg {
+	u8			family;
+
+	/* Used only for kernel-created sockets */
+	union {
+		struct in_addr		local_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr		local_ip6;
+#endif
+	};
+
+	union {
+		struct in_addr		peer_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr		peer_ip6;
+#endif
+	};
+
+	__be16			local_udp_port;
+	__be16			peer_udp_port;
+	unsigned int		use_udp_checksums:1,
+				use_udp6_tx_checksums:1,
+				use_udp6_rx_checksums:1,
+				ipv6_v6only:1;
+};
+
+int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg,
+		     struct socket **sockp);
+
+#if IS_ENABLED(CONFIG_IPV6)
+int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
+		     struct socket **sockp);
+#else
+static inline int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
+				   struct socket **sockp)
+{
+	return 0;
+}
+#endif
+
+static inline int udp_sock_create(struct net *net,
+				  struct udp_port_cfg *cfg,
+				  struct socket **sockp)
+{
+	if (cfg->family == AF_INET)
+		return udp_sock_create4(net, cfg, sockp);
+
+	if (cfg->family == AF_INET6)
+		return udp_sock_create6(net, cfg, sockp);
+
+	return -EPFNOSUPPORT;
+}
+
+typedef int (*udp_tunnel_encap_rcv_t)(struct sock *sk, struct sk_buff *skb);
+
+struct udp_tunnel_sock_cfg {
+	void *sk_user_data;
+	__u8  encap_type;
+	udp_tunnel_encap_rcv_t encap_rcv;
+};
+
+/* Setup the given (UDP) sock to receive UDP encapsulated packets */
+void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
+			   struct udp_tunnel_sock_cfg *sock_cfg);
+
+/* Transmit the skb using UDP encapsulation. */
+void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb,
+			 __be32 src, __be32 dst, __u8 tos, __u8 ttl,
+			 __be16 df, __be16 src_port, __be16 dst_port,
+			 bool xnet, bool nocheck);
+
+#if IS_ENABLED(CONFIG_IPV6)
+int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
+			 struct sk_buff *skb,
+			 struct net_device *dev, struct in6_addr *saddr,
+			 struct in6_addr *daddr,
+			 __u8 prio, __u8 ttl, __be32 label,
+			 __be16 src_port, __be16 dst_port, bool nocheck);
+#endif
+
+void udp_tunnel_sock_release(struct socket *sock);
+
+#endif /* _WG_NET_UDP_TUNNEL_H */
diff --git a/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel.c b/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel.c
new file mode 100644
index 000000000000..9b8770ae7b3f
--- /dev/null
+++ b/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel.c
@@ -0,0 +1,394 @@
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/socket.h>
+#include <linux/udp.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <net/net_namespace.h>
+#include <net/inet_common.h>
+#include <net/udp.h>
+#include <net/udp_tunnel.h>
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0)
+#define __sk_user_data(sk) ((*((void __rcu **)&(sk)->sk_user_data)))
+#define rcu_dereference_sk_user_data(sk) rcu_dereference(__sk_user_data((sk)))
+#define rcu_assign_sk_user_data(sk, ptr) rcu_assign_pointer(__sk_user_data((sk)), ptr)
+#endif
+
+/* This is global so, uh, only one real call site... This is the kind of horrific hack you'd expect to see in compat code. */
+static udp_tunnel_encap_rcv_t encap_rcv = NULL;
+static void __compat_sk_data_ready(struct sock *sk
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 15, 0)
+			      ,int unused_vulnerable_length_param
+#endif
+			      )
+{
+	struct sk_buff *skb;
+	while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) {
+		skb_orphan(skb);
+		sk_mem_reclaim(sk);
+		encap_rcv(sk, skb);
+	}
+}
+
+int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg,
+		     struct socket **sockp)
+{
+	int err;
+	struct socket *sock = NULL;
+	struct sockaddr_in udp_addr;
+
+	err = __sock_create(net, AF_INET, SOCK_DGRAM, 0, &sock, 1);
+	if (err < 0)
+		goto error;
+
+	udp_addr.sin_family = AF_INET;
+	udp_addr.sin_addr = cfg->local_ip;
+	udp_addr.sin_port = cfg->local_udp_port;
+	err = kernel_bind(sock, (struct sockaddr *)&udp_addr,
+			  sizeof(udp_addr));
+	if (err < 0)
+		goto error;
+
+	if (cfg->peer_udp_port) {
+		udp_addr.sin_family = AF_INET;
+		udp_addr.sin_addr = cfg->peer_ip;
+		udp_addr.sin_port = cfg->peer_udp_port;
+		err = kernel_connect(sock, (struct sockaddr *)&udp_addr,
+				     sizeof(udp_addr), 0);
+		if (err < 0)
+			goto error;
+	}
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 0)
+	sock->sk->sk_no_check = !cfg->use_udp_checksums;
+#else
+	sock->sk->sk_no_check_tx = !cfg->use_udp_checksums;
+#endif
+
+	*sockp = sock;
+	return 0;
+
+error:
+	if (sock) {
+		kernel_sock_shutdown(sock, SHUT_RDWR);
+		sock_release(sock);
+	}
+	*sockp = NULL;
+	return err;
+}
+
+void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
+			   struct udp_tunnel_sock_cfg *cfg)
+{
+	inet_sk(sock->sk)->mc_loop = 0;
+	encap_rcv = cfg->encap_rcv;
+	rcu_assign_sk_user_data(sock->sk, cfg->sk_user_data);
+	/* We force the cast in this awful way, due to various Android kernels
+	 * backporting things stupidly. */
+	*(void **)&sock->sk->sk_data_ready = (void *)__compat_sk_data_ready;
+}
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 0)
+static inline __sum16 udp_v4_check(int len, __be32 saddr,
+				   __be32 daddr, __wsum base)
+{
+	return csum_tcpudp_magic(saddr, daddr, len, IPPROTO_UDP, base);
+}
+
+static void udp_set_csum(bool nocheck, struct sk_buff *skb,
+		  __be32 saddr, __be32 daddr, int len)
+{
+	struct udphdr *uh = udp_hdr(skb);
+
+	if (nocheck)
+		uh->check = 0;
+	else if (skb_is_gso(skb))
+		uh->check = ~udp_v4_check(len, saddr, daddr, 0);
+	else if (skb_dst(skb) && skb_dst(skb)->dev &&
+		 (skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) {
+
+		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
+
+		skb->ip_summed = CHECKSUM_PARTIAL;
+		skb->csum_start = skb_transport_header(skb) - skb->head;
+		skb->csum_offset = offsetof(struct udphdr, check);
+		uh->check = ~udp_v4_check(len, saddr, daddr, 0);
+	} else {
+		__wsum csum;
+
+		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
+
+		uh->check = 0;
+		csum = skb_checksum(skb, 0, len, 0);
+		uh->check = udp_v4_check(len, saddr, daddr, csum);
+		if (uh->check == 0)
+			uh->check = CSUM_MANGLED_0;
+
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+}
+
+#endif
+
+static void __compat_fake_destructor(struct sk_buff *skb)
+{
+}
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 11, 0)
+static void __compat_iptunnel_xmit(struct rtable *rt, struct sk_buff *skb,
+		  __be32 src, __be32 dst, __u8 proto,
+		  __u8 tos, __u8 ttl, __be16 df, bool xnet)
+{
+	struct iphdr *iph;
+	struct pcpu_tstats *tstats = this_cpu_ptr(skb->dev->tstats);
+
+	skb_scrub_packet(skb, xnet);
+
+	skb->rxhash = 0;
+	skb_dst_set(skb, &rt->dst);
+	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+
+	/* Push down and install the IP header. */
+	skb_push(skb, sizeof(struct iphdr));
+	skb_reset_network_header(skb);
+
+	iph = ip_hdr(skb);
+
+	iph->version	=	4;
+	iph->ihl	=	sizeof(struct iphdr) >> 2;
+	iph->frag_off	=	df;
+	iph->protocol	=	proto;
+	iph->tos	=	tos;
+	iph->daddr	=	dst;
+	iph->saddr	=	src;
+	iph->ttl	=	ttl;
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 10, 53)
+	__ip_select_ident(iph, &rt->dst, (skb_shinfo(skb)->gso_segs ?: 1) - 1);
+#else
+	__ip_select_ident(iph, skb_shinfo(skb)->gso_segs ?: 1);
+#endif
+
+	iptunnel_xmit(skb, skb->dev);
+	u64_stats_update_begin(&tstats->syncp);
+	tstats->tx_bytes -= 8;
+	u64_stats_update_end(&tstats->syncp);
+}
+#define iptunnel_xmit __compat_iptunnel_xmit
+#endif
+
+void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb,
+			 __be32 src, __be32 dst, __u8 tos, __u8 ttl,
+			 __be16 df, __be16 src_port, __be16 dst_port,
+			 bool xnet, bool nocheck)
+{
+	struct udphdr *uh;
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)
+	struct net_device *dev = skb->dev;
+	int ret;
+#endif
+
+	__skb_push(skb, sizeof(*uh));
+	skb_reset_transport_header(skb);
+	uh = udp_hdr(skb);
+
+	uh->dest = dst_port;
+	uh->source = src_port;
+	uh->len = htons(skb->len);
+
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+
+	udp_set_csum(nocheck, skb, src, dst, skb->len);
+
+	if (!skb->sk)
+		skb->sk = sk;
+	if (!skb->destructor)
+		skb->destructor = __compat_fake_destructor;
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)
+	ret =
+#endif
+	     iptunnel_xmit(
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 15, 0)
+			   sk,
+#endif
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 12, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)
+			   dev_net(dev),
+#endif
+			   rt, skb, src, dst, IPPROTO_UDP, tos, ttl, df
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 12, 0) || LINUX_VERSION_CODE < KERNEL_VERSION(3, 11, 0)
+			   , xnet
+#endif
+	     );
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 11, 0)
+	if (ret)
+		iptunnel_xmit_stats(ret - 8, &dev->stats, dev->tstats);
+#endif
+}
+
+void udp_tunnel_sock_release(struct socket *sock)
+{
+	rcu_assign_sk_user_data(sock->sk, NULL);
+	kernel_sock_shutdown(sock, SHUT_RDWR);
+	sock_release(sock);
+}
+
+#if IS_ENABLED(CONFIG_IPV6)
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/socket.h>
+#include <linux/udp.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/in6.h>
+#include <net/udp.h>
+#include <net/udp_tunnel.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/ip6_tunnel.h>
+#include <net/ip6_checksum.h>
+
+int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
+		     struct socket **sockp)
+{
+	struct sockaddr_in6 udp6_addr;
+	int err;
+	struct socket *sock = NULL;
+
+	err = __sock_create(net, AF_INET6, SOCK_DGRAM, 0, &sock, 1);
+	if (err < 0)
+		goto error;
+
+	if (cfg->ipv6_v6only) {
+		int val = 1;
+
+		err = kernel_setsockopt(sock, IPPROTO_IPV6, IPV6_V6ONLY,
+					(char *) &val, sizeof(val));
+		if (err < 0)
+			goto error;
+	}
+
+	udp6_addr.sin6_family = AF_INET6;
+	memcpy(&udp6_addr.sin6_addr, &cfg->local_ip6,
+	       sizeof(udp6_addr.sin6_addr));
+	udp6_addr.sin6_port = cfg->local_udp_port;
+	err = kernel_bind(sock, (struct sockaddr *)&udp6_addr,
+			  sizeof(udp6_addr));
+	if (err < 0)
+		goto error;
+
+	if (cfg->peer_udp_port) {
+		udp6_addr.sin6_family = AF_INET6;
+		memcpy(&udp6_addr.sin6_addr, &cfg->peer_ip6,
+		       sizeof(udp6_addr.sin6_addr));
+		udp6_addr.sin6_port = cfg->peer_udp_port;
+		err = kernel_connect(sock,
+				     (struct sockaddr *)&udp6_addr,
+				     sizeof(udp6_addr), 0);
+	}
+	if (err < 0)
+		goto error;
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 0)
+	sock->sk->sk_no_check = !cfg->use_udp_checksums;
+#else
+	udp_set_no_check6_tx(sock->sk, !cfg->use_udp6_tx_checksums);
+	udp_set_no_check6_rx(sock->sk, !cfg->use_udp6_rx_checksums);
+#endif
+
+	*sockp = sock;
+	return 0;
+
+error:
+	if (sock) {
+		kernel_sock_shutdown(sock, SHUT_RDWR);
+		sock_release(sock);
+	}
+	*sockp = NULL;
+	return err;
+}
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 16, 0)
+static inline __sum16 udp_v6_check(int len,
+		const struct in6_addr *saddr,
+		const struct in6_addr *daddr,
+		__wsum base)
+{
+	return csum_ipv6_magic(saddr, daddr, len, IPPROTO_UDP, base);
+}
+static void udp6_set_csum(bool nocheck, struct sk_buff *skb,
+		   const struct in6_addr *saddr,
+		   const struct in6_addr *daddr, int len)
+{
+	struct udphdr *uh = udp_hdr(skb);
+
+	if (nocheck)
+		uh->check = 0;
+	else if (skb_is_gso(skb))
+		uh->check = ~udp_v6_check(len, saddr, daddr, 0);
+	else if (skb_dst(skb) && skb_dst(skb)->dev &&
+		 (skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) {
+
+		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
+
+		skb->ip_summed = CHECKSUM_PARTIAL;
+		skb->csum_start = skb_transport_header(skb) - skb->head;
+		skb->csum_offset = offsetof(struct udphdr, check);
+		uh->check = ~udp_v6_check(len, saddr, daddr, 0);
+	} else {
+		__wsum csum;
+
+		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
+
+		uh->check = 0;
+		csum = skb_checksum(skb, 0, len, 0);
+		uh->check = udp_v6_check(len, saddr, daddr, csum);
+		if (uh->check == 0)
+			uh->check = CSUM_MANGLED_0;
+
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+}
+#endif
+
+int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
+			 struct sk_buff *skb,
+			 struct net_device *dev, struct in6_addr *saddr,
+			 struct in6_addr *daddr,
+			 __u8 prio, __u8 ttl, __be32 label,
+			 __be16 src_port, __be16 dst_port, bool nocheck)
+{
+	struct udphdr *uh;
+	struct ipv6hdr *ip6h;
+
+	__skb_push(skb, sizeof(*uh));
+	skb_reset_transport_header(skb);
+	uh = udp_hdr(skb);
+
+	uh->dest = dst_port;
+	uh->source = src_port;
+
+	uh->len = htons(skb->len);
+
+	skb_dst_set(skb, dst);
+
+	udp6_set_csum(nocheck, skb, saddr, daddr, skb->len);
+
+	__skb_push(skb, sizeof(*ip6h));
+	skb_reset_network_header(skb);
+	ip6h		  = ipv6_hdr(skb);
+	ip6_flow_hdr(ip6h, prio, label);
+	ip6h->payload_len = htons(skb->len);
+	ip6h->nexthdr     = IPPROTO_UDP;
+	ip6h->hop_limit   = ttl;
+	ip6h->daddr	  = *daddr;
+	ip6h->saddr	  = *saddr;
+
+	if (!skb->sk)
+		skb->sk = sk;
+	if (!skb->destructor)
+		skb->destructor = __compat_fake_destructor;
+
+	ip6tunnel_xmit(skb, dev);
+	return 0;
+}
+#endif
diff --git a/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel_partial_compat.h b/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel_partial_compat.h
new file mode 100644
index 000000000000..0605896e902f
--- /dev/null
+++ b/drivers/net/wireguard/compat/udp_tunnel/udp_tunnel_partial_compat.h
@@ -0,0 +1,226 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 18, 0)
+#define udp_sock_create4 udp_sock_create
+#define udp_sock_create6 udp_sock_create
+#include <linux/socket.h>
+#include <linux/if.h>
+#include <linux/in.h>
+#include <net/ip_tunnels.h>
+#include <net/udp.h>
+#include <net/inet_common.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <linux/in6.h>
+#include <net/ipv6.h>
+#include <net/addrconf.h>
+#include <net/ip6_checksum.h>
+#include <net/ip6_tunnel.h>
+#endif
+static inline void __compat_fake_destructor(struct sk_buff *skb)
+{
+}
+typedef int (*udp_tunnel_encap_rcv_t)(struct sock *sk, struct sk_buff *skb);
+struct udp_tunnel_sock_cfg {
+        void *sk_user_data;
+        __u8  encap_type;
+        udp_tunnel_encap_rcv_t encap_rcv;
+};
+/* This is global so, uh, only one real call site... This is the kind of horrific hack you'd expect to see in compat code. */
+static udp_tunnel_encap_rcv_t encap_rcv = NULL;
+static void __compat_sk_data_ready(struct sock *sk)
+{
+	struct sk_buff *skb;
+	while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) {
+		skb_orphan(skb);
+		sk_mem_reclaim(sk);
+		encap_rcv(sk, skb);
+	}
+}
+static inline void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
+                           struct udp_tunnel_sock_cfg *cfg)
+{
+	struct sock *sk = sock->sk;
+	inet_sk(sk)->mc_loop = 0;
+	encap_rcv = cfg->encap_rcv;
+	rcu_assign_sk_user_data(sk, cfg->sk_user_data);
+	sk->sk_data_ready = __compat_sk_data_ready;
+}
+static inline void udp_tunnel_sock_release(struct socket *sock)
+{
+	rcu_assign_sk_user_data(sock->sk, NULL);
+	kernel_sock_shutdown(sock, SHUT_RDWR);
+	sk_release_kernel(sock->sk);
+}
+static inline int udp_tunnel_xmit_skb(struct socket *sock, struct rtable *rt,
+                        struct sk_buff *skb, __be32 src, __be32 dst,
+                        __u8 tos, __u8 ttl, __be16 df, __be16 src_port,
+                        __be16 dst_port, bool xnet)
+{
+	struct udphdr *uh;
+	__skb_push(skb, sizeof(*uh));
+	skb_reset_transport_header(skb);
+	uh = udp_hdr(skb);
+	uh->dest = dst_port;
+	uh->source = src_port;
+	uh->len = htons(skb->len);
+	udp_set_csum(sock->sk->sk_no_check_tx, skb, src, dst, skb->len);
+	return iptunnel_xmit(sock->sk, rt, skb, src, dst, IPPROTO_UDP,
+			     tos, ttl, df, xnet);
+}
+#if IS_ENABLED(CONFIG_IPV6)
+static inline int udp_tunnel6_xmit_skb(struct socket *sock, struct dst_entry *dst,
+                         struct sk_buff *skb, struct net_device *dev,
+                         struct in6_addr *saddr, struct in6_addr *daddr,
+                         __u8 prio, __u8 ttl, __be16 src_port,
+                         __be16 dst_port)
+{
+	struct udphdr *uh;
+	struct ipv6hdr *ip6h;
+	struct sock *sk = sock->sk;
+	__skb_push(skb, sizeof(*uh));
+	skb_reset_transport_header(skb);
+	uh = udp_hdr(skb);
+	uh->dest = dst_port;
+	uh->source = src_port;
+	uh->len = htons(skb->len);
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED
+			    | IPSKB_REROUTED);
+	skb_dst_set(skb, dst);
+	udp6_set_csum(udp_get_no_check6_tx(sk), skb, &inet6_sk(sk)->saddr,
+	              &sk->sk_v6_daddr, skb->len);
+	__skb_push(skb, sizeof(*ip6h));
+	skb_reset_network_header(skb);
+	ip6h		  = ipv6_hdr(skb);
+	ip6_flow_hdr(ip6h, prio, htonl(0));
+	ip6h->payload_len = htons(skb->len);
+	ip6h->nexthdr     = IPPROTO_UDP;
+	ip6h->hop_limit   = ttl;
+	ip6h->daddr	  = *daddr;
+	ip6h->saddr	  = *saddr;
+	ip6tunnel_xmit(skb, dev);
+	return 0;
+}
+#endif
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 0, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <linux/udp.h>
+#include <linux/skbuff.h>
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+#define udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) do { struct net_device *dev__ = (c)->dev; int ret__; ret__ = udp_tunnel_xmit_skb((b)->sk_socket, a, c, d, e, f, g, h, i, j, k); if (ret__) iptunnel_xmit_stats(ret__ - 8, &dev__->stats, dev__->tstats); } while (0)
+#if IS_ENABLED(CONFIG_IPV6)
+#define udp_tunnel6_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) udp_tunnel6_xmit_skb((b)->sk_socket, a, c, d, e, f, g, h, j, k);
+#endif
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(4, 1, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 18, 0)
+static inline void __compat_fake_destructor(struct sk_buff *skb)
+{
+}
+#endif
+#define udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) do { struct net_device *dev__ = (c)->dev; int ret__; if (!(c)->destructor) (c)->destructor = __compat_fake_destructor; if (!(c)->sk) (c)->sk = (b); ret__ = udp_tunnel_xmit_skb(a, c, d, e, f, g, h, i, j, k, l); if (ret__) iptunnel_xmit_stats(ret__ - 8, &dev__->stats, dev__->tstats); } while (0)
+#if IS_ENABLED(CONFIG_IPV6)
+#define udp_tunnel6_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) do { if (!(c)->destructor) (c)->destructor = __compat_fake_destructor; if (!(c)->sk) (c)->sk = (b); udp_tunnel6_xmit_skb(a, c, d, e, f, g, h, j, k, l); } while(0)
+#endif
+#else
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+#define udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) do { struct net_device *dev__ = (c)->dev; int ret__ = udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l); if (ret__) iptunnel_xmit_stats(ret__ - 8, &dev__->stats, dev__->tstats); } while (0)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 5, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(4, 3, 0)
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+#define udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) do { struct net_device *dev__ = (c)->dev; int ret__ = udp_tunnel_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l); iptunnel_xmit_stats(ret__, &dev__->stats, dev__->tstats); } while (0)
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0) && IS_ENABLED(CONFIG_IPV6) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+#define udp_tunnel6_xmit_skb(a, b, c, d, e, f, g, h, i, j, k, l) udp_tunnel6_xmit_skb(a, b, c, d, e, f, g, h, j, k, l)
+#endif
+
+#endif
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 3, 0) && LINUX_VERSION_CODE >= KERNEL_VERSION(3, 17, 0)
+#include <linux/skbuff.h>
+#include <linux/if.h>
+#include <net/udp_tunnel.h>
+struct __compat_udp_port_cfg {
+	u8 family;
+	union {
+		struct in_addr local_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr local_ip6;
+#endif
+	};
+	union {
+		struct in_addr peer_ip;
+#if IS_ENABLED(CONFIG_IPV6)
+		struct in6_addr peer_ip6;
+#endif
+	};
+	__be16 local_udp_port;
+	__be16 peer_udp_port;
+	unsigned int use_udp_checksums:1, use_udp6_tx_checksums:1, use_udp6_rx_checksums:1, ipv6_v6only:1;
+};
+static inline int __maybe_unused __compat_udp_sock_create(struct net *net, struct __compat_udp_port_cfg *cfg, struct socket **sockp)
+{
+	struct udp_port_cfg old_cfg = {
+		.family = cfg->family,
+		.local_ip = cfg->local_ip,
+#if IS_ENABLED(CONFIG_IPV6)
+		.local_ip6 = cfg->local_ip6,
+#endif
+		.peer_ip = cfg->peer_ip,
+#if IS_ENABLED(CONFIG_IPV6)
+		.peer_ip6 = cfg->peer_ip6,
+#endif
+		.local_udp_port = cfg->local_udp_port,
+		.peer_udp_port = cfg->peer_udp_port,
+		.use_udp_checksums = cfg->use_udp_checksums,
+		.use_udp6_tx_checksums = cfg->use_udp6_tx_checksums,
+		.use_udp6_rx_checksums = cfg->use_udp6_rx_checksums
+	};
+	if (cfg->family == AF_INET)
+		return udp_sock_create4(net, &old_cfg, sockp);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	if (cfg->family == AF_INET6) {
+		int ret;
+		int old_bindv6only;
+		struct net *nobns;
+
+		if (cfg->ipv6_v6only) {
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 2, 0)
+			nobns = &init_net;
+#else
+			nobns = net;
+#endif
+			/* Since udp_port_cfg only learned of ipv6_v6only in 4.3, we do this horrible
+			 * hack here and set the sysctl variable temporarily to something that will
+			 * set the right option for us in sock_create. It's super racey! */
+			old_bindv6only = nobns->ipv6.sysctl.bindv6only;
+			nobns->ipv6.sysctl.bindv6only = 1;
+		}
+		ret = udp_sock_create6(net, &old_cfg, sockp);
+		if (cfg->ipv6_v6only)
+			nobns->ipv6.sysctl.bindv6only = old_bindv6only;
+		return ret;
+	}
+#endif
+	return -EPFNOSUPPORT;
+}
+#define udp_port_cfg __compat_udp_port_cfg
+#define udp_sock_create(a, b, c) __compat_udp_sock_create(a, b, c)
+#endif