| Message ID | 20240509-upstream-bpf-next-20240506-mptcp-subflow-test-v2-2-4048c2948665@kernel.org (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | selftests/bpf: new MPTCP subflow subtest | expand |
Hello, On 09/05/2024 17:49, Matthieu Baerts (NGI0) wrote: > From: Geliang Tang <tanggeliang@kylinos.cn> > > This patch adds a subtest named test_subflow to load and verify the newly > added mptcp subflow example in test_mptcp. Add a helper endpoint_init() > to add a new subflow endpoint. Add another helper ss_search() to verify the > fwmark and congestion values set by mptcp_subflow prog using setsockopts. > > Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/76 > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn> > Reviewed-by: Mat Martineau <martineau@kernel.org> > Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> > --- > tools/testing/selftests/bpf/prog_tests/mptcp.c | 109 +++++++++++++++++++++++++ > 1 file changed, 109 insertions(+) > > diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c > index 274d2e033e39..6039b0ff3801 100644 > --- a/tools/testing/selftests/bpf/prog_tests/mptcp.c > +++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c (...) > @@ -340,10 +344,115 @@ static void test_mptcpify(void) > close(cgroup_fd); > } > > +static int endpoint_init(char *flags) > +{ > + SYS(fail, "ip -net %s link add veth1 type veth peer name veth2", NS_TEST); > + SYS(fail, "ip -net %s addr add %s/24 dev veth1", NS_TEST, ADDR_1); > + SYS(fail, "ip -net %s link set dev veth1 up", NS_TEST); > + SYS(fail, "ip -net %s addr add %s/24 dev veth2", NS_TEST, ADDR_2); > + SYS(fail, "ip -net %s link set dev veth2 up", NS_TEST); > + SYS(fail, "ip -net %s mptcp endpoint add %s %s", NS_TEST, ADDR_2, flags); I just noticed that this command is failing on the BPF CI: https://github.com/kernel-patches/bpf/actions/runs/9020020315?pr=7009 Is it possible that an old version of IPRoute2 is installed? 'ip mptcp' is supported since v5.8.0 (from 2020). It looks like Ubuntu Focal 20.04 is being used, which has the v5.5.0. Do we then need to find another way to set the MPTCP endpoints? Cheers, Matt
On Thu, May 9, 2024 at 11:31 AM Matthieu Baerts <matttbe@kernel.org> wrote: > > Hello, > > On 09/05/2024 17:49, Matthieu Baerts (NGI0) wrote: > > From: Geliang Tang <tanggeliang@kylinos.cn> > > > > This patch adds a subtest named test_subflow to load and verify the newly > > added mptcp subflow example in test_mptcp. Add a helper endpoint_init() > > to add a new subflow endpoint. Add another helper ss_search() to verify the > > fwmark and congestion values set by mptcp_subflow prog using setsockopts. > > > > Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/76 > > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn> > > Reviewed-by: Mat Martineau <martineau@kernel.org> > > Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> > > --- > > tools/testing/selftests/bpf/prog_tests/mptcp.c | 109 +++++++++++++++++++++++++ > > 1 file changed, 109 insertions(+) > > > > diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c > > index 274d2e033e39..6039b0ff3801 100644 > > --- a/tools/testing/selftests/bpf/prog_tests/mptcp.c > > +++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c > > (...) > > > @@ -340,10 +344,115 @@ static void test_mptcpify(void) > > close(cgroup_fd); > > } > > > > +static int endpoint_init(char *flags) > > +{ > > + SYS(fail, "ip -net %s link add veth1 type veth peer name veth2", NS_TEST); > > + SYS(fail, "ip -net %s addr add %s/24 dev veth1", NS_TEST, ADDR_1); > > + SYS(fail, "ip -net %s link set dev veth1 up", NS_TEST); > > + SYS(fail, "ip -net %s addr add %s/24 dev veth2", NS_TEST, ADDR_2); > > + SYS(fail, "ip -net %s link set dev veth2 up", NS_TEST); > > + SYS(fail, "ip -net %s mptcp endpoint add %s %s", NS_TEST, ADDR_2, flags); > > I just noticed that this command is failing on the BPF CI: > > https://github.com/kernel-patches/bpf/actions/runs/9020020315?pr=7009 > > Is it possible that an old version of IPRoute2 is installed? > 'ip mptcp' is supported since v5.8.0 (from 2020). > > It looks like Ubuntu Focal 20.04 is being used, which has the v5.5.0. Do > we then need to find another way to set the MPTCP endpoints? Manu, any idea?
On 5/10/24, 7:06 AM, "Alexei Starovoitov" <alexei.starovoitov@gmail.com <mailto:alexei.starovoitov@gmail.com>> wrote: !-------------------------------------------------------------------| This Message Is From an External Sender |-------------------------------------------------------------------! On Thu, May 9, 2024 at 11:31 AM Matthieu Baerts <matttbe@kernel.org <mailto:matttbe@kernel.org>> wrote: > > Hello, > > On 09/05/2024 17:49, Matthieu Baerts (NGI0) wrote: > > From: Geliang Tang <tanggeliang@kylinos.cn <mailto:tanggeliang@kylinos.cn>> > > > > This patch adds a subtest named test_subflow to load and verify the newly > > added mptcp subflow example in test_mptcp. Add a helper endpoint_init() > > to add a new subflow endpoint. Add another helper ss_search() to verify the > > fwmark and congestion values set by mptcp_subflow prog using setsockopts. > > > > Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/76 <https://github.com/multipath-tcp/mptcp_net-next/issues/76> > > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn <mailto:tanggeliang@kylinos.cn>> > > Reviewed-by: Mat Martineau <martineau@kernel.org <mailto:martineau@kernel.org>> > > Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org <mailto:matttbe@kernel.org>> > > --- > > tools/testing/selftests/bpf/prog_tests/mptcp.c | 109 +++++++++++++++++++++++++ > > 1 file changed, 109 insertions(+) > > > > diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c > > index 274d2e033e39..6039b0ff3801 100644 > > --- a/tools/testing/selftests/bpf/prog_tests/mptcp.c > > +++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c > > (...) > > > @@ -340,10 +344,115 @@ static void test_mptcpify(void) > > close(cgroup_fd); > > } > > > > +static int endpoint_init(char *flags) > > +{ > > + SYS(fail, "ip -net %s link add veth1 type veth peer name veth2", NS_TEST); > > + SYS(fail, "ip -net %s addr add %s/24 dev veth1", NS_TEST, ADDR_1); > > + SYS(fail, "ip -net %s link set dev veth1 up", NS_TEST); > > + SYS(fail, "ip -net %s addr add %s/24 dev veth2", NS_TEST, ADDR_2); > > + SYS(fail, "ip -net %s link set dev veth2 up", NS_TEST); > > + SYS(fail, "ip -net %s mptcp endpoint add %s %s", NS_TEST, ADDR_2, flags); > > I just noticed that this command is failing on the BPF CI: > > https://github.com/kernel-patches/bpf/actions/runs/9020020315?pr=7009 <https://github.com/kernel-patches/bpf/actions/runs/9020020315?pr=7009> > > Is it possible that an old version of IPRoute2 is installed? > 'ip mptcp' is supported since v5.8.0 (from 2020). > > It looks like Ubuntu Focal 20.04 is being used, which has the v5.5.0. Do > we then need to find another way to set the MPTCP endpoints? Manu, any idea? // retrying plain text format.... please outlook, please, make this happen! Indeed, this is running Ubuntu 20.04. I am planning to eventually update to 22.04 (which has iproute 5.15), But I don’t have a good ETA to give other than this is not going to be in the coming month. It does not seem that iproute provides an easy way to check its version, and the version returned by the container is: root@081a02e57175:/actions-runner# ip --json -V ip utility, iproute2-ss200127 root@081a02e57175:/actions-runner# dpkg -l iproute2 Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad) ||/ Name Version Architecture Description +++-==============-==============-============-==================================== ii iproute2 5.5.0-1ubuntu1 amd64 networking and traffic control tools How complicated is it to set MPTCP directly using netlink following https://github.com/iproute2/iproute2/blob/397383a30c3b0e3ff551042b6654898a0872b83e/ip/ipmptcp.c#L573 ? Seems tools/testing/selftests/bpf/netlink_helpers.c could be useful to deal with the netlink part. And the former link to find out how to format the message? Manu
diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c index 274d2e033e39..6039b0ff3801 100644 --- a/tools/testing/selftests/bpf/prog_tests/mptcp.c +++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c @@ -9,8 +9,12 @@ #include "network_helpers.h" #include "mptcp_sock.skel.h" #include "mptcpify.skel.h" +#include "mptcp_subflow.skel.h" #define NS_TEST "mptcp_ns" +#define ADDR_1 "10.0.1.1" +#define ADDR_2 "10.0.1.2" +#define PORT_1 10001 #ifndef IPPROTO_MPTCP #define IPPROTO_MPTCP 262 @@ -340,10 +344,115 @@ static void test_mptcpify(void) close(cgroup_fd); } +static int endpoint_init(char *flags) +{ + SYS(fail, "ip -net %s link add veth1 type veth peer name veth2", NS_TEST); + SYS(fail, "ip -net %s addr add %s/24 dev veth1", NS_TEST, ADDR_1); + SYS(fail, "ip -net %s link set dev veth1 up", NS_TEST); + SYS(fail, "ip -net %s addr add %s/24 dev veth2", NS_TEST, ADDR_2); + SYS(fail, "ip -net %s link set dev veth2 up", NS_TEST); + SYS(fail, "ip -net %s mptcp endpoint add %s %s", NS_TEST, ADDR_2, flags); + + return 0; +fail: + return -1; +} + +static int _ss_search(char *src, char *dst, char *port, char *keyword) +{ + char cmd[128]; + int n; + + n = snprintf(cmd, sizeof(cmd), + "ip netns exec %s ss -Menita src %s dst %s %s %d | grep -q '%s'", + NS_TEST, src, dst, port, PORT_1, keyword); + if (n < 0 || n >= sizeof(cmd)) + return -1; + + return system(cmd); +} + +static int ss_search(char *src, char *keyword) +{ + return _ss_search(src, ADDR_1, "dport", keyword); +} + +static void run_subflow(char *new) +{ + int server_fd, client_fd, err; + char cc[TCP_CA_NAME_MAX]; + socklen_t len = sizeof(cc); + + server_fd = start_mptcp_server(AF_INET, ADDR_1, PORT_1, 0); + if (!ASSERT_GE(server_fd, 0, "start_mptcp_server")) + return; + + client_fd = connect_to_fd(server_fd, 0); + if (!ASSERT_GE(client_fd, 0, "connect to fd")) + goto fail; + + err = getsockopt(server_fd, SOL_TCP, TCP_CONGESTION, cc, &len); + if (!ASSERT_OK(err, "getsockopt(srv_fd, TCP_CONGESTION)")) + goto fail; + + send_byte(client_fd); + + ASSERT_OK(ss_search(ADDR_1, "fwmark:0x1"), "ss_search fwmark:0x1"); + ASSERT_OK(ss_search(ADDR_2, "fwmark:0x2"), "ss_search fwmark:0x2"); + ASSERT_OK(ss_search(ADDR_1, new), "ss_search new cc"); + ASSERT_OK(ss_search(ADDR_2, cc), "ss_search default cc"); + + close(client_fd); +fail: + close(server_fd); +} + +static void test_subflow(void) +{ + int cgroup_fd, prog_fd, err; + struct mptcp_subflow *skel; + struct nstoken *nstoken; + + cgroup_fd = test__join_cgroup("/mptcp_subflow"); + if (!ASSERT_GE(cgroup_fd, 0, "join_cgroup: mptcp_subflow")) + return; + + skel = mptcp_subflow__open_and_load(); + if (!ASSERT_OK_PTR(skel, "skel_open_load: mptcp_subflow")) + goto close_cgroup; + + err = mptcp_subflow__attach(skel); + if (!ASSERT_OK(err, "skel_attach: mptcp_subflow")) + goto skel_destroy; + + prog_fd = bpf_program__fd(skel->progs.mptcp_subflow); + err = bpf_prog_attach(prog_fd, cgroup_fd, BPF_CGROUP_SOCK_OPS, 0); + if (!ASSERT_OK(err, "prog_attach")) + goto skel_destroy; + + nstoken = create_netns(); + if (!ASSERT_OK_PTR(nstoken, "create_netns: mptcp_subflow")) + goto skel_destroy; + + if (!ASSERT_OK(endpoint_init("subflow"), "endpoint_init")) + goto close_netns; + + run_subflow(skel->data->cc); + +close_netns: + cleanup_netns(nstoken); +skel_destroy: + mptcp_subflow__destroy(skel); +close_cgroup: + close(cgroup_fd); +} + void test_mptcp(void) { if (test__start_subtest("base")) test_base(); if (test__start_subtest("mptcpify")) test_mptcpify(); + if (test__start_subtest("subflow")) + test_subflow(); }