Re: [PATCH bpf-next v2 4/5] selftests/bpf: Add trampoline and raw_tp preemption recursion test

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Medium] selftests/bpf: raw #pragma unroll in fill_stack() breaks
  bpf-gcc build
--

commit 85924bc5107b6604b6bb8852a70f1660aa030b1c
Author: Mike Bazov <mikeba@sentinelone.com>

selftests/bpf: Add trampoline and raw_tp preemption recursion test

This commit adds BPF selftests verifying preemption, private-stack
isolation, and recursion rejection across trampoline and raw tracepoint
program types.

> diff --git a/tools/testing/selftests/bpf/progs/trampoline_preempt.c b/tools/testing/selftests/bpf/progs/trampoline_preempt.c
> new file mode 100644
> index 0000000000000..78ed74627a79d
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/progs/trampoline_preempt.c
> @@ -0,0 +1,265 @@

[ ... ]

> +static __always_inline void fill_stack(__u64 *buf, __u32 tid)
> +{
> +	#pragma unroll
> +	for (int i = 0; i < 8; i++)
> +		buf[i] = ((__u64)tid << 32) | (0x5a5a0000U + i);
> +}
> +
> +static __always_inline void check_stack(const __u64 *buf, __u32 tid)
> +{
> +	#pragma unroll
> +	for (int i = 0; i < 8; i++) {
> +		if (buf[i] != (((__u64)tid << 32) | (0x5a5a0000U + i)))
> +			stack_corrupt = 1;
> +	}
> +}

[Severity: Medium]
Does raw #pragma unroll in fill_stack() and check_stack() break the
test_progs-bpf_gcc build with -Werror=unknown-pragmas?  Including
bpf_compiler.h and using __pragma_loop_unroll (or
__pragma_loop_unroll_full), as introduced in 52dbd67dff5d ("bpf: Abstract
loop unrolling pragmas in BPF selftests"), avoids the bpf-gcc build
failure.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261111-bpf-preempt-v2-r2-v2-0-9d8f6b952ad2@sentinelone.com?part=4