[PATCH bpf] bpf: selftest: Fix flaky tcp_hdr_options test when adding addr to lo

[PATCH bpf] bpf: selftest: Fix flaky tcp_hdr_options test when adding addr to lo

[Martin KaFai Lau]

The tcp_hdr_options test adds a "::eB9F" addr to the lo dev.
However, this non loopback address will have a race on ipv6 dad
which may lead to EADDRNOTAVAIL error from time to time.

Even nodad is used in the iproute2 command, there is still a race in
when the route will be added.  This will then lead to ENETUNREACH from
time to time.

To avoid the above, this patch uses the default loopback address "::1"
to do the test.

Fixes: ad2f8eb0095e ("bpf: selftests: Tcp header options")
Reported-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
---
 .../bpf/prog_tests/tcp_hdr_options.c          | 26 +------------------
 .../bpf/progs/test_misc_tcp_hdr_options.c     |  2 +-
 2 files changed, 2 insertions(+), 26 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c b/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
index c86e67214a9e..c85174cdcb77 100644
--- a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
+++ b/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c
@@ -15,7 +15,7 @@
 #include "test_tcp_hdr_options.skel.h"
 #include "test_misc_tcp_hdr_options.skel.h"
 
-#define LO_ADDR6 "::eB9F"
+#define LO_ADDR6 "::1"
 #define CG_NAME "/tcpbpf-hdr-opt-test"
 
 struct bpf_test_option exp_passive_estab_in;
@@ -40,27 +40,6 @@ struct sk_fds {
 	int active_lport;
 };
 
-static int add_lo_addr(void)
-{
-	char ip_addr_cmd[256];
-	int cmdlen;
-
-	cmdlen = snprintf(ip_addr_cmd, sizeof(ip_addr_cmd),
-			  "ip -6 addr add %s/128 dev lo scope host",
-			  LO_ADDR6);
-
-	if (CHECK(cmdlen >= sizeof(ip_addr_cmd), "compile ip cmd",
-		  "failed to add host addr %s to lo. ip cmdlen is too long\n",
-		  LO_ADDR6))
-		return -1;
-
-	if (CHECK(system(ip_addr_cmd), "run ip cmd",
-		  "failed to add host addr %s to lo\n", LO_ADDR6))
-		return -1;
-
-	return 0;
-}
-
 static int create_netns(void)
 {
 	if (CHECK(unshare(CLONE_NEWNET), "create netns",
@@ -72,9 +51,6 @@ static int create_netns(void)
 		  "failed to bring lo link up\n"))
 		return -1;
 
-	if (add_lo_addr())
-		return -1;
-
 	return 0;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c b/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c
index 72ec0178f653..6077a025092c 100644
--- a/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c
+++ b/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c
@@ -16,7 +16,7 @@
 #define BPF_PROG_TEST_TCP_HDR_OPTIONS
 #include "test_tcp_hdr_options.h"
 
-__u16 last_addr16_n = __bpf_htons(0xeB9F);
+__u16 last_addr16_n = __bpf_htons(1);
 __u16 active_lport_n = 0;
 __u16 active_lport_h = 0;
 __u16 passive_lport_n = 0;
-- 
2.24.1

Re: [PATCH bpf] bpf: selftest: Fix flaky tcp_hdr_options test when adding addr to lo

[Andrii Nakryiko]

On Tue, Oct 13, 2020 at 4:13 AM Martin KaFai Lau <kafai@fb.com> wrote:
>
> The tcp_hdr_options test adds a "::eB9F" addr to the lo dev.
> However, this non loopback address will have a race on ipv6 dad
> which may lead to EADDRNOTAVAIL error from time to time.
>
> Even nodad is used in the iproute2 command, there is still a race in
> when the route will be added.  This will then lead to ENETUNREACH from
> time to time.
>
> To avoid the above, this patch uses the default loopback address "::1"
> to do the test.
>
> Fixes: ad2f8eb0095e ("bpf: selftests: Tcp header options")
> Reported-by: Andrii Nakryiko <andriin@fb.com>
> Signed-off-by: Martin KaFai Lau <kafai@fb.com>
> ---

Less shelling out is always good :)

Acked-by: Andrii Nakryiko <andrii@kernel.org>

>  .../bpf/prog_tests/tcp_hdr_options.c          | 26 +------------------
>  .../bpf/progs/test_misc_tcp_hdr_options.c     |  2 +-
>  2 files changed, 2 insertions(+), 26 deletions(-)
>

[...]