Re: [PATCH mptcp-net 3/3] selftests: mptcp: sockopt: fix variable check in error reporting

[Matthieu Baerts <matttbe@kernel.org>]

Hi Geliang,

On 02/09/2025 03:38, Geliang Tang wrote:
> From: Geliang Tang <tanggeliang@kylinos.cn>
> 
> The error message for mptcpi_bytes_acked comparison incorrectly
> references 'ret2' instead of 'ret'. Since the comparison is between
> mptcpi_bytes_acked and 'ret', the error output should consistently
> use 'ret' for both the expected value and difference calculation.
> 
> This patch corrects the variable usage in the error output to ensure
> accurate debugging information when the assertion fails.
> 
> Fixes: 5dcff89e1455 ("selftests: mptcp: explicitly tests aggregate counters")
> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
>  tools/testing/selftests/net/mptcp/mptcp_sockopt.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
> index 8d590629a8e9..a61508270a5c 100644
> --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
> +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c
> @@ -680,8 +680,8 @@ static void process_one_client(int fd, int pipefd)
>  			       s.last_sample.mptcpi_bytes_received - ret);
>  		if (s.last_sample.mptcpi_bytes_acked != ret)
>  			xerror("mptcpi_bytes_acked %" PRIu64 ", expect %" PRIu64,
> -			       s.last_sample.mptcpi_bytes_acked, ret2,
> -			       s.last_sample.mptcpi_bytes_acked - ret2);
> +			       s.last_sample.mptcpi_bytes_acked, ret,
> +			       s.last_sample.mptcpi_bytes_acked - ret);

Good catch! But I also just noticed there is one extra argument: you
have 2 "%", but 3 arguments after the string.

I guess the error should be:

  "mptcpi_bytes_acked %" PRIu64 ", expect %" PRIu64 ", diff %" PRIu64,
  (...)

(Same for the others above.) If yes, can you add this as part of the
same commit?

  selftests: mptcp: sockopt: fix error messages

(with a note about the fact the wrong variable was used)

>  	}
>  
>  	close(fd);

Cheers,
Matt
-- 
Sponsored by the NGI0 Core fund.