Re: [PATCHv6 06/13] libbpf: Change has_nop_combo to work on top of nop10

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Dropping support for the `nop5` combo in `has_nop_combo()` breaks the BPF USDT selftests.
--

commit dc8c1772a19cbe96786096f839e442eb211041bc
Author: Jiri Olsa <jolsa@kernel.org>

libbpf: Change has_nop_combo to work on top of nop10

This patch modifies the USDT has_nop_combo() function to exclusively check
for an 11-byte nop,nop10 sequence, replacing the previous 6-byte nop,nop5
check. This aligns the detection logic with newer kernel support for faster
uprobe handling using nop10.

> diff --git a/tools/lib/bpf/usdt.c b/tools/lib/bpf/usdt.c
> index 57fb82bb81b58..d2ecd3daab961 100644
> --- a/tools/lib/bpf/usdt.c
> +++ b/tools/lib/bpf/usdt.c

[ ... ]

> @@ -604,14 +604,14 @@ static int parse_usdt_spec(struct usdt_spec *spec, const struct usdt_note *note,
>  #if defined(__x86_64__)
>  static bool has_nop_combo(int fd, long off)
>  {
> -	unsigned char nop_combo[6] = {
> -		0x90, 0x0f, 0x1f, 0x44, 0x00, 0x00 /* nop,nop5 */
> +	unsigned char nop_combo[11] = {
> +		0x90, 0x66, 0x2e, 0x0f, 0x1f, 0x84, 0x00, 0x00, 0x00, 0x00, 0x00,
>  	};
> -	unsigned char buf[6];
> +	unsigned char buf[11];
>  
> -	if (pread(fd, buf, 6, off) != 6)
> +	if (pread(fd, buf, 11, off) != 11)
>  		return false;
> -	return memcmp(buf, nop_combo, 6) == 0;
> +	return memcmp(buf, nop_combo, 11) == 0;
>  }

[Severity: Medium]
Will dropping support for the nop5 combo here break the BPF USDT selftests?

The selftest in tools/testing/selftests/bpf/prog_tests/usdt.c still generates
and expects the nop,nop5 combination for usdt_2. Because this function now
returns false for nop5, collect_usdt_targets() skips the usdt_abs_ip++
optimization.

This causes the uprobe to be placed directly at addr_2, overwriting the nop
with an int3 (0xcc) breakpoint, which makes the test assertion fail:

tools/testing/selftests/bpf/prog_tests/usdt.c:subtest_optimized_attach() {
    ...
    /* usdt_2 USDT probe has nop,nop5 instructions combo */
    addr_2 = find_instr(usdt_2, nop1_nop5_combo, 6);
    if (!ASSERT_OK_PTR(addr_2, "usdt_2_find_nop1_nop5_combo"))
        return;
    ...
    /* nop stays on addr_2 address */
    ASSERT_EQ(*addr_2, 0x90, "nop");
}

Does the test suite also need to be updated to generate and expect nop,nop10?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260703114917.238144-7-jolsa@kernel.org?part=1