Re: [PATCH] string: Use extended timeout for test-mempcpy

On 03/08/26 12:53, Magnus Lindholm wrote:
> The shared memcpy tests use an eight-minute timeout because they can
> take significantly longer than other string tests.

I think it would be better to start split this test, or at least move out
the porting that takes a lot of time.  Having a test that takes that large
amount of time can be be unfeasible to run on small cores and does not help
on regression tests in general.

> 
> test-mempcpy provides its own test setup and includes test-string.h
> before the shared memcpy test functions are included. It therefore
> receives the four-minute default timeout instead of the extended
> timeout used by test-memcpy.
> 
> Define the same eight-minute timeout explicitly for test-mempcpy.
> 
> This prevents the test from being terminated while it is still making
> progress on slower systems.
> 
> Signed-off-by: Magnus Lindholm <[email protected]>
> ---
>  string/test-mempcpy.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/string/test-mempcpy.c b/string/test-mempcpy.c
> index cc7f9d447c..cc2ec23173 100644
> --- a/string/test-mempcpy.c
> +++ b/string/test-mempcpy.c
> @@ -18,6 +18,7 @@
>  
>  #define MEMCPY_RESULT(dst, len) (dst) + (len)
>  #define MIN_PAGE_SIZE 131072
> +#define TIMEOUT (8 * 60)
>  #define TEST_MAIN
>  #define TEST_NAME "mempcpy"
>  #include "test-string.h"