Re: [PATCH] libtraceevent: Fix tests running on big endian arch

[Steven Rostedt <rostedt@goodmis.org>]

On Fri, 29 Mar 2024 15:25:45 +0100
Paul Mars <paul.mars@canonical.com> wrote:

> Tests running on big endian arch (tested on s390x) where failing due to:

Nice, all my big endian machines have died, so I can no longer test on them.

> 
> - a typo in CPUMASK_BYTEPN.
> - a missing call to tep_set_file_bigendian when initializing the test
>   suite.
> 
> Thanks to Mate Kukri <mate.kukri@canonical.com>.

Can you add a signed-off-by please.

And for a change log, do you want:

 Suggested-by: Mate Kukri <mate.kukri@canonical.com>

?

Thanks,

-- Steve



> 
> ---
>  utest/traceevent-utest.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/utest/traceevent-utest.c b/utest/traceevent-utest.c
> index 041843e..b95e478 100644
> --- a/utest/traceevent-utest.c
> +++ b/utest/traceevent-utest.c
> @@ -216,7 +216,7 @@ DECL_CPUMASK_EVENT_DATA(bytep2, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80, 0x00, 0x01);
>  #define CPUMASK_BYTEP2_FMT "cpumask=0,23"
>  
>  DECL_CPUMASK_EVENT_DATA(bytepn, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01);
> -#define CPUMASK_BYTEPN     "ARRAY[80, 00, 00, 00, 00, 00, 80, 01]"
> +#define CPUMASK_BYTEPN     "ARRAY[80, 00, 00, 00, 00, 00, 00, 01]"
>  #define CPUMASK_BYTEPN_FMT "cpumask=0,63"
>  #endif
>  
> @@ -392,6 +392,9 @@ static int test_suite_init(void)
>  	test_tep = tep_alloc();
>  	if (!test_tep)
>  		return 1;
> +#if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
> +	tep_set_file_bigendian(test_tep, TEP_BIG_ENDIAN);
> +#endif
>  	return 0;
>  }
>