On Thu, Sep 10, 2020 at 11:11 AM Yonghong Song <yhs@xxxxxx> wrote:
>
>
>
> On 9/9/20 9:27 AM, Lorenz Bauer wrote:
> > Add a test that exercises a basic sockmap / sockhash iteration. For
> > now we simply count the number of elements seen. Once sockmap update
> > from iterators works we can extend this to perform a full copy.
> >
> > Signed-off-by: Lorenz Bauer <lmb@xxxxxxxxxxxxxx>
> > ---
> > .../selftests/bpf/prog_tests/sockmap_basic.c | 89 +++++++++++++++++++
> > tools/testing/selftests/bpf/progs/bpf_iter.h | 9 ++
> > .../selftests/bpf/progs/bpf_iter_sockmap.c | 43 +++++++++
> > .../selftests/bpf/progs/bpf_iter_sockmap.h | 3 +
> > 4 files changed, 144 insertions(+)
> > create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_sockmap.c
> > create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_sockmap.h
> >
> > diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c
> > index 0b79d78b98db..3215f4d22720 100644
> > --- a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c
> > +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c
> > @@ -6,6 +6,9 @@
> > #include "test_skmsg_load_helpers.skel.h"
> > #include "test_sockmap_update.skel.h"
> > #include "test_sockmap_invalid_update.skel.h"
> > +#include "bpf_iter_sockmap.skel.h"
> > +
> > +#include "progs/bpf_iter_sockmap.h"
> >
> > #define TCP_REPAIR 19 /* TCP sock is under repair right now */
> >
> > @@ -171,6 +174,88 @@ static void test_sockmap_invalid_update(void)
> > test_sockmap_invalid_update__destroy(skel);
> > }
> >
> > +static void test_sockmap_iter(enum bpf_map_type map_type)
> > +{
> > + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts);
> > + int err, len, src_fd, iter_fd, duration;
> > + union bpf_iter_link_info linfo = {0};
> > + __s64 sock_fd[SOCKMAP_MAX_ENTRIES];
> > + __u32 i, num_sockets, max_elems;
> > + struct bpf_iter_sockmap *skel;
> > + struct bpf_link *link;
> > + struct bpf_map *src;
> > + char buf[64];
> > +
> > + skel = bpf_iter_sockmap__open_and_load();
> > + if (CHECK(!skel, "bpf_iter_sockmap__open_and_load", "skeleton open_and_load failed\n"))
> > + return;
> > +
> > + for (i = 0; i < ARRAY_SIZE(sock_fd); i++)
> > + sock_fd[i] = -1;
> > +
> > + /* Make sure we have at least one "empty" entry to test iteration of
> > + * an empty slot.
> > + */
> > + num_sockets = ARRAY_SIZE(sock_fd) - 1;
> > +
> > + if (map_type == BPF_MAP_TYPE_SOCKMAP) {
> > + src = skel->maps.sockmap;
> > + max_elems = bpf_map__max_entries(src);
> > + } else {
> > + src = skel->maps.sockhash;
> > + max_elems = num_sockets;
> > + }
>
> I know you include the shared header progs/bpf_iter_sockmap.h to
> supply SOCKMAP_MAX_ENTRIES in order to define sock_fd array.
>
> I think it is easier to understand if just using bpf_map__max_entries()
> for both sockmap and sockhash to get max_elems and do dynamic allocation
> for sock_fd. WDYT?
imo makes sense as a follow up.
I fixed up "duration = 0" in this test because of build warnings
and pushed the whole set.
Thanks
