> On Feb 20, 2020, at 8:20 PM, Andrii Nakryiko <andriin@xxxxxx> wrote:
>
> On 2/20/20 6:06 PM, Song Liu wrote:
>> On Thu, Feb 20, 2020 at 3:07 PM Andrii Nakryiko <andriin@xxxxxx> wrote:
>>>
>>> Libbpf's Travis CI tests caught this issue. Ensure bpf_link and bpf_object
>>> clean up is performed correctly.
>>>
>>> Fixes: d633d57902a5 ("selftest/bpf: Add test for allowed trampolines count")
>>> Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
>>> Signed-off-by: Andrii Nakryiko <andriin@xxxxxx>
>>> ---
>>> .../bpf/prog_tests/trampoline_count.c | 25 +++++++++++++------
>>> 1 file changed, 18 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/tools/testing/selftests/bpf/prog_tests/trampoline_count.c b/tools/testing/selftests/bpf/prog_tests/trampoline_count.c
>>> index 1f6ccdaed1ac..781c8d11604b 100644
>>> --- a/tools/testing/selftests/bpf/prog_tests/trampoline_count.c
>>> +++ b/tools/testing/selftests/bpf/prog_tests/trampoline_count.c
>>> @@ -55,31 +55,40 @@ void test_trampoline_count(void)
>>> /* attach 'allowed' 40 trampoline programs */
>>> for (i = 0; i < MAX_TRAMP_PROGS; i++) {
>>> obj = bpf_object__open_file(object, NULL);
>>> - if (CHECK(IS_ERR(obj), "obj_open_file", "err %ld\n", PTR_ERR(obj)))
>>> + if (CHECK(IS_ERR(obj), "obj_open_file", "err %ld\n", PTR_ERR(obj))) {
>>> + obj = NULL;
>> I think we don't need obj and link in cleanup? Did I miss anything?
>
> We do set obj below (line 87) after this loop, so need to clean it up as well. As for link, yeah, technically link doesn't have to be set to NULL, but I kind of did it for completeness without thinking too much.
I meant "obj = NULL;" before "goto cleanup;", as we don't use obj in the
cleanup path.
Anyway, this is not a real issue.
Thanks,
Song
