bpf-fixes

-----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEE+soXsSLHKoYyzcli6rmadz2vbToFAmf6sD8ACgkQ6rmadz2v
 bTq86w//bbg2S1ZhSXXQvgRSbxfecvJ0r6XGDOaMsKxPXcqpbaMoSCYx2D8puO+b
 xm0vc+5qXlzuTHq9I8flDKrWdA+/sHxLQhXjcBA796vaY6IgJEnapf3kENyzZ3Vp
 agpNPlZe9FLaANDRivTFPVgzVjr07/3eL7VKItASksb/3yjBSa+vrIJVfGF1krQT
 slxTMzVMzB+p0MdKVjmeGn5EodWXp8TdVzQBPb8vnCn7U1h1HULSh4j1+nZ/Z1yr
 zC4/pVPmdDJe1H8ghBGm4f0nY+EwXPtZiVbXnYS2FhgjvthRKFYIyxN9F6kg7AD7
 NG0T6xw/QYNfPTR40PSiV/WHhH5qa2zRVtlepVU7tqqmsyRXi+0Eq/MfJyiuNzgN
 WWmJec0O/Ax4r2Xs/QgX3mFlRnLNi5gmc7fuOARmayAlqElZ9QdB2x6ebW5Fk4Qx
 9oyQACpcu6/oUKgeMSo52MDa82wUPPxpC6qdsefmQYaAcOKM5MD4SNd+eEnfX03E
 RAaItTW9az57a2BL9C/ejJO/SwY4Er+O8B3PO7GaKiURMSZa5nVlY+2QB2fJy6TA
 7IvSYjFD5E4risMbZgPFCqWkQ0yHbY7zEn/tbcNC5AFZoKv70jELPQTLPXq7UPLe
 BuKoL9VJyeXF7E1MQqQH33q3tfcwlIL++piCNHvTQoPadEba2dM=
 =Mezb
 -----END PGP SIGNATURE-----

Merge tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf

Pull bpf fixes from Alexei Starovoitov:

 - Followup fixes for resilient spinlock (Kumar Kartikeya Dwivedi):
     - Make res_spin_lock test less verbose, since it was spamming BPF
       CI on failure, and make the check for AA deadlock stronger
     - Fix rebasing mistake and use architecture provided
       res_smp_cond_load_acquire
     - Convert BPF maps (queue_stack and ringbuf) to resilient spinlock
       to address long standing syzbot reports

 - Make sure that classic BPF load instruction from SKF_[NET|LL]_OFF
   offsets works when skb is fragmeneted (Willem de Bruijn)

* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  bpf: Convert ringbuf map to rqspinlock
  bpf: Convert queue_stack map to rqspinlock
  bpf: Use architecture provided res_smp_cond_load_acquire
  selftests/bpf: Make res_spin_lock AA test condition stronger
  selftests/net: test sk_filter support for SKF_NET_OFF on frags
  bpf: support SKF_NET_OFF and SKF_LL_OFF on skb frags
  selftests/bpf: Make res_spin_lock test less verbose

arch/arm64/include/asm/rqspinlock.h

@@ -86,7 +86,7 @@
 
 #endif
 
-#define res_smp_cond_load_acquire_timewait(v, c) smp_cond_load_acquire_timewait(v, c, 0, 1)
+#define res_smp_cond_load_acquire(v, c) smp_cond_load_acquire_timewait(v, c, 0, 1)
 
 #include <asm-generic/rqspinlock.h>
 

kernel/bpf/queue_stack_maps.c

@@ -9,13 +9,14 @@
 #include <linux/slab.h>
 #include <linux/btf_ids.h>
 #include "percpu_freelist.h"
+#include <asm/rqspinlock.h>
 
 #define QUEUE_STACK_CREATE_FLAG_MASK \
 	(BPF_F_NUMA_NODE | BPF_F_ACCESS_MASK)
 
 struct bpf_queue_stack {
 	struct bpf_map map;
-	raw_spinlock_t lock;
+	rqspinlock_t lock;
 	u32 head, tail;
 	u32 size; /* max_entries + 1 */
 
@@ -78,7 +79,7 @@ static struct bpf_map *queue_stack_map_alloc(union bpf_attr *attr)
 
 	qs->size = size;
 
-	raw_spin_lock_init(&qs->lock);
+	raw_res_spin_lock_init(&qs->lock);
 
 	return &qs->map;
 }
@@ -98,12 +99,8 @@ static long __queue_map_get(struct bpf_map *map, void *value, bool delete)
 	int err = 0;
 	void *ptr;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_empty(qs)) {
 		memset(value, 0, qs->map.value_size);
@@ -120,7 +117,7 @@ static long __queue_map_get(struct bpf_map *map, void *value, bool delete)
 	}
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, flags);
 	return err;
 }
 
@@ -133,12 +130,8 @@ static long __stack_map_get(struct bpf_map *map, void *value, bool delete)
 	void *ptr;
 	u32 index;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_empty(qs)) {
 		memset(value, 0, qs->map.value_size);
@@ -157,7 +150,7 @@ static long __stack_map_get(struct bpf_map *map, void *value, bool delete)
 		qs->head = index;
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, flags);
 	return err;
 }
 
@@ -203,12 +196,8 @@ static long queue_stack_map_push_elem(struct bpf_map *map, void *value,
 	if (flags & BPF_NOEXIST || flags > BPF_EXIST)
 		return -EINVAL;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, irq_flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, irq_flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, irq_flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_full(qs)) {
 		if (!replace) {
@@ -227,7 +216,7 @@ static long queue_stack_map_push_elem(struct bpf_map *map, void *value,
 		qs->head = 0;
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, irq_flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, irq_flags);
 	return err;
 }
 

kernel/bpf/ringbuf.c

@@ -11,6 +11,7 @@
 #include <linux/kmemleak.h>
 #include <uapi/linux/btf.h>
 #include <linux/btf_ids.h>
+#include <asm/rqspinlock.h>
 
 #define RINGBUF_CREATE_FLAG_MASK (BPF_F_NUMA_NODE)
 
@@ -29,7 +30,7 @@ struct bpf_ringbuf {
 	u64 mask;
 	struct page **pages;
 	int nr_pages;
-	raw_spinlock_t spinlock ____cacheline_aligned_in_smp;
+	rqspinlock_t spinlock ____cacheline_aligned_in_smp;
 	/* For user-space producer ring buffers, an atomic_t busy bit is used
 	 * to synchronize access to the ring buffers in the kernel, rather than
 	 * the spinlock that is used for kernel-producer ring buffers. This is
@@ -173,7 +174,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node)
 	if (!rb)
 		return NULL;
 
-	raw_spin_lock_init(&rb->spinlock);
+	raw_res_spin_lock_init(&rb->spinlock);
 	atomic_set(&rb->busy, 0);
 	init_waitqueue_head(&rb->waitq);
 	init_irq_work(&rb->work, bpf_ringbuf_notify);
@@ -416,12 +417,8 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 
 	cons_pos = smp_load_acquire(&rb->consumer_pos);
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&rb->spinlock, flags))
-			return NULL;
-	} else {
-		raw_spin_lock_irqsave(&rb->spinlock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&rb->spinlock, flags))
+		return NULL;
 
 	pend_pos = rb->pending_pos;
 	prod_pos = rb->producer_pos;
@@ -446,7 +443,7 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 	 */
 	if (new_prod_pos - cons_pos > rb->mask ||
 	    new_prod_pos - pend_pos > rb->mask) {
-		raw_spin_unlock_irqrestore(&rb->spinlock, flags);
+		raw_res_spin_unlock_irqrestore(&rb->spinlock, flags);
 		return NULL;
 	}
 
@@ -458,7 +455,7 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 	/* pairs with consumer's smp_load_acquire() */
 	smp_store_release(&rb->producer_pos, new_prod_pos);
 
-	raw_spin_unlock_irqrestore(&rb->spinlock, flags);
+	raw_res_spin_unlock_irqrestore(&rb->spinlock, flags);
 
 	return (void *)hdr + BPF_RINGBUF_HDR_SZ;
 }

kernel/bpf/rqspinlock.c

@@ -253,7 +253,7 @@ static noinline int check_timeout(rqspinlock_t *lock, u32 mask,
 	})
 #else
 #define RES_CHECK_TIMEOUT(ts, ret, mask)			      \
-	({ (ret) = check_timeout(&(ts)); })
+	({ (ret) = check_timeout((lock), (mask), &(ts)); })
 #endif
 
 /*

net/core/filter.c

@@ -218,24 +218,36 @@ BPF_CALL_3(bpf_skb_get_nlattr_nest, struct sk_buff *, skb, u32, a, u32, x)
 	return 0;
 }
 
+static int bpf_skb_load_helper_convert_offset(const struct sk_buff *skb, int offset)
+{
+	if (likely(offset >= 0))
+		return offset;
+
+	if (offset >= SKF_NET_OFF)
+		return offset - SKF_NET_OFF + skb_network_offset(skb);
+
+	if (offset >= SKF_LL_OFF && skb_mac_header_was_set(skb))
+		return offset - SKF_LL_OFF + skb_mac_offset(skb);
+
+	return INT_MIN;
+}
+
 BPF_CALL_4(bpf_skb_load_helper_8, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	u8 tmp, *ptr;
+	u8 tmp;
 	const int len = sizeof(tmp);
 
-	if (offset >= 0) {
-		if (headlen - offset >= len)
-			return *(u8 *)(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return tmp;
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return *(u8 *)ptr;
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return *(u8 *)(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return tmp;
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_8_no_cache, const struct sk_buff *, skb,
@@ -248,21 +260,19 @@ BPF_CALL_2(bpf_skb_load_helper_8_no_cache, const struct sk_buff *, skb,
 BPF_CALL_4(bpf_skb_load_helper_16, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	__be16 tmp, *ptr;
+	__be16 tmp;
 	const int len = sizeof(tmp);
 
-	if (offset >= 0) {
-		if (headlen - offset >= len)
-			return get_unaligned_be16(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return be16_to_cpu(tmp);
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return get_unaligned_be16(ptr);
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return get_unaligned_be16(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return be16_to_cpu(tmp);
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_16_no_cache, const struct sk_buff *, skb,
@@ -275,21 +285,19 @@ BPF_CALL_2(bpf_skb_load_helper_16_no_cache, const struct sk_buff *, skb,
 BPF_CALL_4(bpf_skb_load_helper_32, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	__be32 tmp, *ptr;
+	__be32 tmp;
 	const int len = sizeof(tmp);
 
-	if (likely(offset >= 0)) {
-		if (headlen - offset >= len)
-			return get_unaligned_be32(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return be32_to_cpu(tmp);
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return get_unaligned_be32(ptr);
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return get_unaligned_be32(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return be32_to_cpu(tmp);
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_32_no_cache, const struct sk_buff *, skb,

tools/testing/selftests/bpf/prog_tests/res_spin_lock.c

@@ -25,8 +25,11 @@ static void *spin_lock_thread(void *arg)
 
 	while (!READ_ONCE(skip)) {
 		err = bpf_prog_test_run_opts(prog_fd, &topts);
-		ASSERT_OK(err, "test_run");
-		ASSERT_OK(topts.retval, "test_run retval");
+		if (err || topts.retval) {
+			ASSERT_OK(err, "test_run");
+			ASSERT_OK(topts.retval, "test_run retval");
+			break;
+		}
 	}
 	pthread_exit(arg);
 }

tools/testing/selftests/bpf/progs/res_spin_lock.c

@@ -38,13 +38,14 @@ int res_spin_lock_test(struct __sk_buff *ctx)
 	r = bpf_res_spin_lock(&elem1->lock);
 	if (r)
 		return r;
-	if (!bpf_res_spin_lock(&elem2->lock)) {
+	r = bpf_res_spin_lock(&elem2->lock);
+	if (!r) {
 		bpf_res_spin_unlock(&elem2->lock);
 		bpf_res_spin_unlock(&elem1->lock);
 		return -1;
 	}
 	bpf_res_spin_unlock(&elem1->lock);
-	return 0;
+	return r != -EDEADLK;
 }
 
 SEC("tc")
@@ -124,12 +125,15 @@ int res_spin_lock_test_held_lock_max(struct __sk_buff *ctx)
 	/* Trigger AA, after exhausting entries in the held lock table. This
 	 * time, only the timeout can save us, as AA detection won't succeed.
 	 */
-	if (!bpf_res_spin_lock(locks[34])) {
+	ret = bpf_res_spin_lock(locks[34]);
+	if (!ret) {
 		bpf_res_spin_unlock(locks[34]);
 		ret = 1;
 		goto end;
 	}
 
+	ret = ret != -ETIMEDOUT ? 2 : 0;
+
 end:
 	for (i = i - 1; i >= 0; i--)
 		bpf_res_spin_unlock(locks[i]);

tools/testing/selftests/net/.gitignore

@@ -39,6 +39,7 @@ scm_rights
 sk_bind_sendto_listen
 sk_connect_zero_addr
 sk_so_peek_off
+skf_net_off
 socket
 so_incoming_cpu
 so_netns_cookie

tools/testing/selftests/net/Makefile

@@ -106,6 +106,8 @@ TEST_PROGS += ipv6_route_update_soft_lockup.sh
 TEST_PROGS += busy_poll_test.sh
 TEST_GEN_PROGS += proc_net_pktgen
 TEST_PROGS += lwt_dst_cache_ref_loop.sh
+TEST_PROGS += skf_net_off.sh
+TEST_GEN_FILES += skf_net_off
 
 # YNL files, must be before "include ..lib.mk"
 YNL_GEN_FILES := busy_poller netlink-dumps

tools/testing/selftests/net/skf_net_off.c

@@ -0,0 +1,244 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Open a tun device.
+ *
+ * [modifications: use IFF_NAPI_FRAGS, add sk filter]
+ *
+ * Expects the device to have been configured previously, e.g.:
+ *   sudo ip tuntap add name tap1 mode tap
+ *   sudo ip link set tap1 up
+ *   sudo ip link set dev tap1 addr 02:00:00:00:00:01
+ *   sudo ip -6 addr add fdab::1 peer fdab::2 dev tap1 nodad
+ *
+ * And to avoid premature pskb_may_pull:
+ *
+ *   sudo ethtool -K tap1 gro off
+ *   sudo bash -c 'echo 0 > /proc/sys/net/ipv4/ip_early_demux'
+ */
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <error.h>
+#include <fcntl.h>
+#include <getopt.h>
+#include <linux/filter.h>
+#include <linux/if.h>
+#include <linux/if_packet.h>
+#include <linux/if_tun.h>
+#include <linux/ipv6.h>
+#include <netinet/if_ether.h>
+#include <netinet/in.h>
+#include <netinet/ip.h>
+#include <netinet/ip6.h>
+#include <netinet/udp.h>
+#include <poll.h>
+#include <signal.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/socket.h>
+#include <sys/poll.h>
+#include <sys/types.h>
+#include <sys/uio.h>
+#include <unistd.h>
+
+static bool cfg_do_filter;
+static bool cfg_do_frags;
+static int cfg_dst_port = 8000;
+static char *cfg_ifname;
+
+static int tun_open(const char *tun_name)
+{
+	struct ifreq ifr = {0};
+	int fd, ret;
+
+	fd = open("/dev/net/tun", O_RDWR);
+	if (fd == -1)
+		error(1, errno, "open /dev/net/tun");
+
+	ifr.ifr_flags = IFF_TAP;
+	if (cfg_do_frags)
+		ifr.ifr_flags |= IFF_NAPI | IFF_NAPI_FRAGS;
+
+	strncpy(ifr.ifr_name, tun_name, IFNAMSIZ - 1);
+
+	ret = ioctl(fd, TUNSETIFF, &ifr);
+	if (ret)
+		error(1, ret, "ioctl TUNSETIFF");
+
+	return fd;
+}
+
+static void sk_set_filter(int fd)
+{
+	const int offset_proto = offsetof(struct ip6_hdr, ip6_nxt);
+	const int offset_dport = sizeof(struct ip6_hdr) + offsetof(struct udphdr, dest);
+
+	/* Filter UDP packets with destination port cfg_dst_port */
+	struct sock_filter filter_code[] = {
+		BPF_STMT(BPF_LD  + BPF_B   + BPF_ABS, SKF_AD_OFF + SKF_AD_PKTTYPE),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, PACKET_HOST, 0, 4),
+		BPF_STMT(BPF_LD  + BPF_B   + BPF_ABS, SKF_NET_OFF + offset_proto),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, IPPROTO_UDP, 0, 2),
+		BPF_STMT(BPF_LD  + BPF_H   + BPF_ABS, SKF_NET_OFF + offset_dport),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, cfg_dst_port, 1, 0),
+		BPF_STMT(BPF_RET + BPF_K, 0),
+		BPF_STMT(BPF_RET + BPF_K, 0xFFFF),
+	};
+
+	struct sock_fprog filter = {
+		sizeof(filter_code) / sizeof(filter_code[0]),
+		filter_code,
+	};
+
+	if (setsockopt(fd, SOL_SOCKET, SO_ATTACH_FILTER, &filter, sizeof(filter)))
+		error(1, errno, "setsockopt attach filter");
+}
+
+static int raw_open(void)
+{
+	int fd;
+
+	fd = socket(PF_INET6, SOCK_RAW, IPPROTO_UDP);
+	if (fd == -1)
+		error(1, errno, "socket raw (udp)");
+
+	if (cfg_do_filter)
+		sk_set_filter(fd);
+
+	return fd;
+}
+
+static void tun_write(int fd)
+{
+	const char eth_src[] = { 0x02, 0x00, 0x00, 0x00, 0x00, 0x02 };
+	const char eth_dst[] = { 0x02, 0x00, 0x00, 0x00, 0x00, 0x01 };
+	struct tun_pi pi = {0};
+	struct ipv6hdr ip6h = {0};
+	struct udphdr uh = {0};
+	struct ethhdr eth = {0};
+	uint32_t payload;
+	struct iovec iov[5];
+	int ret;
+
+	pi.proto = htons(ETH_P_IPV6);
+
+	memcpy(eth.h_source, eth_src, sizeof(eth_src));
+	memcpy(eth.h_dest, eth_dst, sizeof(eth_dst));
+	eth.h_proto = htons(ETH_P_IPV6);
+
+	ip6h.version = 6;
+	ip6h.payload_len = htons(sizeof(uh) + sizeof(uint32_t));
+	ip6h.nexthdr = IPPROTO_UDP;
+	ip6h.hop_limit = 8;
+	if (inet_pton(AF_INET6, "fdab::2", &ip6h.saddr) != 1)
+		error(1, errno, "inet_pton src");
+	if (inet_pton(AF_INET6, "fdab::1", &ip6h.daddr) != 1)
+		error(1, errno, "inet_pton src");
+
+	uh.source = htons(8000);
+	uh.dest = htons(cfg_dst_port);
+	uh.len = ip6h.payload_len;
+	uh.check = 0;
+
+	payload = htonl(0xABABABAB);		/* Covered in IPv6 length */
+
+	iov[0].iov_base = &pi;
+	iov[0].iov_len  = sizeof(pi);
+	iov[1].iov_base = &eth;
+	iov[1].iov_len  = sizeof(eth);
+	iov[2].iov_base = &ip6h;
+	iov[2].iov_len  = sizeof(ip6h);
+	iov[3].iov_base = &uh;
+	iov[3].iov_len  = sizeof(uh);
+	iov[4].iov_base = &payload;
+	iov[4].iov_len  = sizeof(payload);
+
+	ret = writev(fd, iov, sizeof(iov) / sizeof(iov[0]));
+	if (ret <= 0)
+		error(1, errno, "writev");
+}
+
+static void raw_read(int fd)
+{
+	struct timeval tv = { .tv_usec = 100 * 1000 };
+	struct msghdr msg = {0};
+	struct iovec iov[2];
+	struct udphdr uh;
+	uint32_t payload[2];
+	int ret;
+
+	if (setsockopt(fd, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(tv)))
+		error(1, errno, "setsockopt rcvtimeo udp");
+
+	iov[0].iov_base = &uh;
+	iov[0].iov_len = sizeof(uh);
+
+	iov[1].iov_base = payload;
+	iov[1].iov_len = sizeof(payload);
+
+	msg.msg_iov = iov;
+	msg.msg_iovlen = sizeof(iov) / sizeof(iov[0]);
+
+	ret = recvmsg(fd, &msg, 0);
+	if (ret <= 0)
+		error(1, errno, "read raw");
+	if (ret != sizeof(uh) + sizeof(payload[0]))
+		error(1, errno, "read raw: len=%d\n", ret);
+
+	fprintf(stderr, "raw recv: 0x%x\n", payload[0]);
+}
+
+static void parse_opts(int argc, char **argv)
+{
+	int c;
+
+	while ((c = getopt(argc, argv, "fFi:")) != -1) {
+		switch (c) {
+		case 'f':
+			cfg_do_filter = true;
+			printf("bpf filter enabled\n");
+			break;
+		case 'F':
+			cfg_do_frags = true;
+			printf("napi frags mode enabled\n");
+			break;
+		case 'i':
+			cfg_ifname = optarg;
+			break;
+		default:
+			error(1, 0, "unknown option %c", optopt);
+			break;
+		}
+	}
+
+	if (!cfg_ifname)
+		error(1, 0, "must specify tap interface name (-i)");
+}
+
+int main(int argc, char **argv)
+{
+	int fdt, fdr;
+
+	parse_opts(argc, argv);
+
+	fdr = raw_open();
+	fdt = tun_open(cfg_ifname);
+
+	tun_write(fdt);
+	raw_read(fdr);
+
+	if (close(fdt))
+		error(1, errno, "close tun");
+	if (close(fdr))
+		error(1, errno, "close udp");
+
+	fprintf(stderr, "OK\n");
+	return 0;
+}
+

tools/testing/selftests/net/skf_net_off.sh

@@ -0,0 +1,30 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+readonly NS="ns-$(mktemp -u XXXXXX)"
+
+cleanup() {
+	ip netns del $NS
+}
+
+ip netns add $NS
+trap cleanup EXIT
+
+ip -netns $NS link set lo up
+ip -netns $NS tuntap add name tap1 mode tap
+ip -netns $NS link set tap1 up
+ip -netns $NS link set dev tap1 addr 02:00:00:00:00:01
+ip -netns $NS -6 addr add fdab::1 peer fdab::2 dev tap1 nodad
+ip netns exec $NS ethtool -K tap1 gro off
+
+# disable early demux, else udp_v6_early_demux pulls udp header into linear
+ip netns exec $NS sysctl -w net.ipv4.ip_early_demux=0
+
+echo "no filter"
+ip netns exec $NS ./skf_net_off -i tap1
+
+echo "filter, linear skb (-f)"
+ip netns exec $NS ./skf_net_off -i tap1 -f
+
+echo "filter, fragmented skb (-f) (-F)"
+ip netns exec $NS ./skf_net_off -i tap1 -f -F