Re: [PATCH bpf 2/2] selftests/bpf: Cover RHASH batch cursor loss

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] selftests/bpf: standalone rhash_batch.c file instead of
  extending rhash.c
--

commit bf9586f48a54e730e2740cce0745ca08b2ddac63
Author: Tianyi Chen <hi@tychen.cc>

selftests/bpf: Cover RHASH batch cursor loss

This commit adds BPF selftests covering normal completion, missing input
cursors, and concurrent deletions for BPF_MAP_TYPE_RHASH batch lookup and
lookup-and-delete operations.

> diff --git a/tools/testing/selftests/bpf/prog_tests/rhash_batch.c b/tools/testing/selftests/bpf/prog_tests/rhash_batch.c
> new file mode 100644
> index 0000000000000..bba9062afd123
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/prog_tests/rhash_batch.c

[ ... ]

> +void test_rhash_batch(void)
> +{
> +	if (test__start_subtest("lookup"))
> +		test_batch(false, false);
> +	if (test__start_subtest("lookup_delete"))
> +		test_batch(true, false);
> +	if (test__start_subtest("lost_lookup_cursor"))
> +		test_batch(false, true);
> +	if (test__start_subtest("lost_delete_cursor"))
> +		test_batch(true, true);
> +	if (test__start_subtest("concurrent_delete"))
> +		test_concurrent_delete();
> +}

[Severity: Low]
This isn't a bug, but could these subtests be added to test_rhash() in
tools/testing/selftests/bpf/prog_tests/rhash.c instead of introducing a
separate file?  rhash.c already houses userspace map syscall and iteration
subtests for BPF_MAP_TYPE_RHASH, and rhash_batch.c does not load a separate
BPF program skeleton.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1794056839.git.hi@tychen.cc?part=2