Re: [PATCH bpf-next v5 09/10] selftests/bpf: Check callback map value lock identity

["Kumar Kartikeya Dwivedi" <memxor@gmail.com>]

On Thu Sep 17, 2026 at 2:50 AM CEST, Alexei Starovoitov wrote:
> On Wed, Sep 16, 2026 at 11:20 PM Kumar Kartikeya Dwivedi <memxor@gmail.com> wrote:
>> diff --git a/tools/testing/selftests/bpf/prog_tests/cb_refs.c b/tools/testing/selftests/bpf/prog_tests/cb_refs.c
>> index 78566b817fd7..490e15e7126d 100644
>> --- a/tools/testing/selftests/bpf/prog_tests/cb_refs.c
>> +++ b/tools/testing/selftests/bpf/prog_tests/cb_refs.c
>> @@ -13,7 +13,7 @@ struct {
>>  } cb_refs_tests[] = {
>>  	{ "underflow_prog", "release kfunc bpf_kfunc_call_test_release expects referenced PTR_TO_BTF_ID passed to R1" },
>>  	{ "leak_prog", "Possibly NULL pointer passed to helper R2" },
>> -	{ "nested_cb", "Unreleased reference id=4 alloc_insn=2" }, /* alloc_insn=2{4,5} */
>> +	{ "nested_cb", "Unreleased reference id=5 alloc_insn=2" }, /* alloc_insn=2{4,5} */
>>  	{ "non_cb_transfer_ref", "Unreleased reference id=4 alloc_insn=1" }, /* alloc_insn=1{1,2} */
>>  };
>
> \robot voice
>
> This is the hunk that doesn't apply. The two context lines above are
> not what bpf-next has, and on b
> pf-next leak_prog still expects
> "Unreleased reference id=4 alloc_insn=3" for an acquire done inside a
> for_each callback, so it will need the same +1 after patch 8 hands out
> an id to R3 at callback push time.
>
> Chasing verifier-internal id numbers in expected strings is fragile.
> While rebasing, either drop the id from these patterns (match on
> "Unreleased reference" and alloc_insn) or at least fold the expectation
> update into patch 8, so test_progs is green at every step of the
> series.
>
> pw-bot: cr

It got mistagged by me, should have been targeting bpf tree. I did a CI run in
https://github.com/kernel-patches/bpf/pull/13872, which is green for bpf tree,
so if there are no other objections it can be applied there. Otherwise, I will
respin tomorrow.