From: Jakub Sitnicki jakub@cloudflare.com
stable inclusion from stable-v5.10.136 commit 042fb1c281f357d58308366b5e2ddd8e5f1ad384 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I5ZWNE
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=...
--------------------------------
commit 8f50f16ff39dd4e2d43d1548ca66925652f8aff7 upstream.
Add coverage to the verifier tests and tests for reading bpf_sock fields to ensure that 32-bit, 16-bit, and 8-bit loads from dst_port field are allowed only at intended offsets and produce expected values.
While 16-bit and 8-bit access to dst_port field is straight-forward, 32-bit wide loads need be allowed and produce a zero-padded 16-bit value for backward compatibility.
Signed-off-by: Jakub Sitnicki jakub@cloudflare.com Link: https://lore.kernel.org/r/20220130115518.213259-3-jakub@cloudflare.com Signed-off-by: Alexei Starovoitov ast@kernel.org [OP: backport to 5.10: adjusted context in sock_fields.c] Signed-off-by: Ovidiu Panait ovidiu.panait@windriver.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com Reviewed-by: Wei Li liwei391@huawei.com --- tools/include/uapi/linux/bpf.h | 3 +- .../selftests/bpf/prog_tests/sock_fields.c | 60 +++++++++----- .../selftests/bpf/progs/test_sock_fields.c | 41 ++++++++++ tools/testing/selftests/bpf/verifier/sock.c | 81 ++++++++++++++++++- 4 files changed, 162 insertions(+), 23 deletions(-)
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index a44cb5155825..9ff20fbad70e 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -4195,7 +4195,8 @@ struct bpf_sock { __u32 src_ip4; __u32 src_ip6[4]; __u32 src_port; /* host byte order */ - __u32 dst_port; /* network byte order */ + __be16 dst_port; /* network byte order */ + __u16 :16; /* zero padding */ __u32 dst_ip4; __u32 dst_ip6[4]; __u32 state; diff --git a/tools/testing/selftests/bpf/prog_tests/sock_fields.c b/tools/testing/selftests/bpf/prog_tests/sock_fields.c index af87118e748e..e8b5bf707cd4 100644 --- a/tools/testing/selftests/bpf/prog_tests/sock_fields.c +++ b/tools/testing/selftests/bpf/prog_tests/sock_fields.c @@ -1,9 +1,11 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright (c) 2019 Facebook */
+#define _GNU_SOURCE #include <netinet/in.h> #include <arpa/inet.h> #include <unistd.h> +#include <sched.h> #include <stdlib.h> #include <string.h> #include <errno.h> @@ -21,6 +23,7 @@ enum bpf_linum_array_idx { EGRESS_LINUM_IDX, INGRESS_LINUM_IDX, + READ_SK_DST_PORT_LINUM_IDX, __NR_BPF_LINUM_ARRAY_IDX, };
@@ -43,8 +46,16 @@ static __u64 child_cg_id; static int linum_map_fd; static __u32 duration;
-static __u32 egress_linum_idx = EGRESS_LINUM_IDX; -static __u32 ingress_linum_idx = INGRESS_LINUM_IDX; +static bool create_netns(void) +{ + if (!ASSERT_OK(unshare(CLONE_NEWNET), "create netns")) + return false; + + if (!ASSERT_OK(system("ip link set dev lo up"), "bring up lo")) + return false; + + return true; +}
static void print_sk(const struct bpf_sock *sk, const char *prefix) { @@ -92,19 +103,24 @@ static void check_result(void) { struct bpf_tcp_sock srv_tp, cli_tp, listen_tp; struct bpf_sock srv_sk, cli_sk, listen_sk; - __u32 ingress_linum, egress_linum; + __u32 idx, ingress_linum, egress_linum, linum; int err;
- err = bpf_map_lookup_elem(linum_map_fd, &egress_linum_idx, - &egress_linum); + idx = EGRESS_LINUM_IDX; + err = bpf_map_lookup_elem(linum_map_fd, &idx, &egress_linum); CHECK(err == -1, "bpf_map_lookup_elem(linum_map_fd)", "err:%d errno:%d\n", err, errno);
- err = bpf_map_lookup_elem(linum_map_fd, &ingress_linum_idx, - &ingress_linum); + idx = INGRESS_LINUM_IDX; + err = bpf_map_lookup_elem(linum_map_fd, &idx, &ingress_linum); CHECK(err == -1, "bpf_map_lookup_elem(linum_map_fd)", "err:%d errno:%d\n", err, errno);
+ idx = READ_SK_DST_PORT_LINUM_IDX; + err = bpf_map_lookup_elem(linum_map_fd, &idx, &linum); + ASSERT_OK(err, "bpf_map_lookup_elem(linum_map_fd, READ_SK_DST_PORT_IDX)"); + ASSERT_EQ(linum, 0, "failure in read_sk_dst_port on line"); + memcpy(&srv_sk, &skel->bss->srv_sk, sizeof(srv_sk)); memcpy(&srv_tp, &skel->bss->srv_tp, sizeof(srv_tp)); memcpy(&cli_sk, &skel->bss->cli_sk, sizeof(cli_sk)); @@ -263,7 +279,7 @@ static void test(void) char buf[DATA_LEN];
/* Prepare listen_fd */ - listen_fd = start_server(AF_INET6, SOCK_STREAM, "::1", 0, 0); + listen_fd = start_server(AF_INET6, SOCK_STREAM, "::1", 0xcafe, 0); /* start_server() has logged the error details */ if (CHECK_FAIL(listen_fd == -1)) goto done; @@ -331,8 +347,12 @@ static void test(void)
void test_sock_fields(void) { - struct bpf_link *egress_link = NULL, *ingress_link = NULL; int parent_cg_fd = -1, child_cg_fd = -1; + struct bpf_link *link; + + /* Use a dedicated netns to have a fixed listen port */ + if (!create_netns()) + return;
/* Create a cgroup, get fd, and join it */ parent_cg_fd = test__join_cgroup(PARENT_CGROUP); @@ -353,17 +373,20 @@ void test_sock_fields(void) if (CHECK(!skel, "test_sock_fields__open_and_load", "failed\n")) goto done;
- egress_link = bpf_program__attach_cgroup(skel->progs.egress_read_sock_fields, - child_cg_fd); - if (CHECK(IS_ERR(egress_link), "attach_cgroup(egress)", "err:%ld\n", - PTR_ERR(egress_link))) + link = bpf_program__attach_cgroup(skel->progs.egress_read_sock_fields, child_cg_fd); + if (!ASSERT_OK_PTR(link, "attach_cgroup(egress_read_sock_fields)")) + goto done; + skel->links.egress_read_sock_fields = link; + + link = bpf_program__attach_cgroup(skel->progs.ingress_read_sock_fields, child_cg_fd); + if (!ASSERT_OK_PTR(link, "attach_cgroup(ingress_read_sock_fields)")) goto done; + skel->links.ingress_read_sock_fields = link;
- ingress_link = bpf_program__attach_cgroup(skel->progs.ingress_read_sock_fields, - child_cg_fd); - if (CHECK(IS_ERR(ingress_link), "attach_cgroup(ingress)", "err:%ld\n", - PTR_ERR(ingress_link))) + link = bpf_program__attach_cgroup(skel->progs.read_sk_dst_port, child_cg_fd); + if (!ASSERT_OK_PTR(link, "attach_cgroup(read_sk_dst_port")) goto done; + skel->links.read_sk_dst_port = link;
linum_map_fd = bpf_map__fd(skel->maps.linum_map); sk_pkt_out_cnt_fd = bpf_map__fd(skel->maps.sk_pkt_out_cnt); @@ -372,8 +395,7 @@ void test_sock_fields(void) test();
done: - bpf_link__destroy(egress_link); - bpf_link__destroy(ingress_link); + test_sock_fields__detach(skel); test_sock_fields__destroy(skel); if (child_cg_fd != -1) close(child_cg_fd); diff --git a/tools/testing/selftests/bpf/progs/test_sock_fields.c b/tools/testing/selftests/bpf/progs/test_sock_fields.c index 7967348b11af..3e2e3ee51cc9 100644 --- a/tools/testing/selftests/bpf/progs/test_sock_fields.c +++ b/tools/testing/selftests/bpf/progs/test_sock_fields.c @@ -12,6 +12,7 @@ enum bpf_linum_array_idx { EGRESS_LINUM_IDX, INGRESS_LINUM_IDX, + READ_SK_DST_PORT_LINUM_IDX, __NR_BPF_LINUM_ARRAY_IDX, };
@@ -250,4 +251,44 @@ int ingress_read_sock_fields(struct __sk_buff *skb) return CG_OK; }
+static __noinline bool sk_dst_port__load_word(struct bpf_sock *sk) +{ + __u32 *word = (__u32 *)&sk->dst_port; + return word[0] == bpf_htonl(0xcafe0000); +} + +static __noinline bool sk_dst_port__load_half(struct bpf_sock *sk) +{ + __u16 *half = (__u16 *)&sk->dst_port; + return half[0] == bpf_htons(0xcafe); +} + +static __noinline bool sk_dst_port__load_byte(struct bpf_sock *sk) +{ + __u8 *byte = (__u8 *)&sk->dst_port; + return byte[0] == 0xca && byte[1] == 0xfe; +} + +SEC("cgroup_skb/egress") +int read_sk_dst_port(struct __sk_buff *skb) +{ + __u32 linum, linum_idx; + struct bpf_sock *sk; + + linum_idx = READ_SK_DST_PORT_LINUM_IDX; + + sk = skb->sk; + if (!sk) + RET_LOG(); + + if (!sk_dst_port__load_word(sk)) + RET_LOG(); + if (!sk_dst_port__load_half(sk)) + RET_LOG(); + if (!sk_dst_port__load_byte(sk)) + RET_LOG(); + + return CG_OK; +} + char _license[] SEC("license") = "GPL"; diff --git a/tools/testing/selftests/bpf/verifier/sock.c b/tools/testing/selftests/bpf/verifier/sock.c index ce13ece08d51..8c224eac93df 100644 --- a/tools/testing/selftests/bpf/verifier/sock.c +++ b/tools/testing/selftests/bpf/verifier/sock.c @@ -121,7 +121,25 @@ .result = ACCEPT, }, { - "sk_fullsock(skb->sk): sk->dst_port [narrow load]", + "sk_fullsock(skb->sk): sk->dst_port [word load] (backward compatibility)", + .insns = { + BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), + BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_EMIT_CALL(BPF_FUNC_sk_fullsock), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port)), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .prog_type = BPF_PROG_TYPE_CGROUP_SKB, + .result = ACCEPT, +}, +{ + "sk_fullsock(skb->sk): sk->dst_port [half load]", .insns = { BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), @@ -139,7 +157,64 @@ .result = ACCEPT, }, { - "sk_fullsock(skb->sk): sk->dst_port [load 2nd byte]", + "sk_fullsock(skb->sk): sk->dst_port [half load] (invalid)", + .insns = { + BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), + BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_EMIT_CALL(BPF_FUNC_sk_fullsock), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .prog_type = BPF_PROG_TYPE_CGROUP_SKB, + .result = REJECT, + .errstr = "invalid sock access", +}, +{ + "sk_fullsock(skb->sk): sk->dst_port [byte load]", + .insns = { + BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), + BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_EMIT_CALL(BPF_FUNC_sk_fullsock), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_0, offsetof(struct bpf_sock, dst_port)), + BPF_LDX_MEM(BPF_B, BPF_REG_2, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 1), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .prog_type = BPF_PROG_TYPE_CGROUP_SKB, + .result = ACCEPT, +}, +{ + "sk_fullsock(skb->sk): sk->dst_port [byte load] (invalid)", + .insns = { + BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), + BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_EMIT_CALL(BPF_FUNC_sk_fullsock), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .prog_type = BPF_PROG_TYPE_CGROUP_SKB, + .result = REJECT, + .errstr = "invalid sock access", +}, +{ + "sk_fullsock(skb->sk): past sk->dst_port [half load] (invalid)", .insns = { BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)), BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2), @@ -149,7 +224,7 @@ BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 2), BPF_MOV64_IMM(BPF_REG_0, 0), BPF_EXIT_INSN(), - BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_0, offsetof(struct bpf_sock, dst_port) + 1), + BPF_LDX_MEM(BPF_H, BPF_REG_0, BPF_REG_0, offsetofend(struct bpf_sock, dst_port)), BPF_MOV64_IMM(BPF_REG_0, 0), BPF_EXIT_INSN(), },