Re: [PATCH v7 3/7] selftest/bpf: Fixing powerpc JIT disassembly failure

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The patch introduces a C++ style single-line comment (`//`) which violates the BPF subsystem comment style guidelines.
--

commit 90e32beb297573769f660f2e6d0779da36d4155e
Author: Abhishek Dubey <adubey@linux.ibm.com>

selftest/bpf: Fixing powerpc JIT disassembly failure

This commit ensures that the trampoline stubs JITed at the tail of the
epilogue do not expose the dummy trampoline address to the disassembly flow.
It prevents the disassembler from ingesting this memory address by
truncating the last bytes from JITed buffers before disassembly.

> diff --git a/tools/testing/selftests/bpf/jit_disasm_helpers.c b/tools/testing/selftests/bpf/jit_disasm_helpers.c
> index 364c557c51158..466d62b4ca285 100644
> --- a/tools/testing/selftests/bpf/jit_disasm_helpers.c
> +++ b/tools/testing/selftests/bpf/jit_disasm_helpers.c

[ ... ]

> @@ -216,9 +218,26 @@ int get_jited_program_text(int fd, char *text, size_t text_sz)

[ ... ]

>  	for (pc = 0, i = 0; i < jited_funcs; ++i) {
>  		fprintf(text_out, "func #%d:\n", i);
> -		disasm_one_func(text_out, image + pc, func_lens[i]);
> +		// Disabled JIT have zero func_lens, hence underflow
[Severity: Low]
This isn't a bug, but does this C++ style comment violate the BPF subsystem
conventions which require standard C-style block comments for single-line
comments?

> +		__u32 disasm_len = func_lens[i] > trunc_len ?
> +					func_lens[i] - trunc_len : 0;
> +		disasm_one_func(text_out, image + pc, disasm_len);
>  		fprintf(text_out, "\n");
>  		pc += func_lens[i];
>  	}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260611153826.31187-1-adubey@linux.ibm.com?part=3