[LTP] [PATCH v2 0/3] Refine overall timeout and runtime management in LTP

[Li Wang]

On Thu, Jan 9, 2025 at 12:56 PM Li Wang <liwang@redhat.com> wrote:

> This patch series introduces a restructured approach to managing
> timeouts and runtime limits in the LTP framework. The changes aim to
> simplify timeout handling, improve clarity, and provide better
> separation between the phases of a test (setup, execution, and cleanup).
>
> The main updates include refining the description and usage of .timeout,
> deprecating TST_UNLIMITED_RUNTIME, and introducing the tst_set_timeout()
> function. Additionally, test cases that previously relied on .max_runtime
> have been updated to adopt the revised logic, either by moving to .timeout
> directly or transitioning to .runtime depending on their behavior.
>
> The series ends with an update to the set_mempolicy01 test, demonstrating
> the use of tst_set_timeout() to determine an upper bound for its execution.
>
> CI job: https://github.com/wangli5665/ltp/actions/runs/12683826906/
>
> Major changes V1 to V2:
>
>   * Refined the description of .timeout usage and its relationship with
> .runtime.
>
>   * Moved .max_runtime to .timeout directly for tests where runtime is not
> explicitly used.
>
>   * Transitioned .max_runtime to .runtime for tests that call
> tst_remaining_runtime()
>     or use the fuzzy_sync library.
>
>   * Added the tst_set_timeout() function to allow test cases to set the
> overall
>     timeout dynamically, particularly useful for tests that aim to
> determine
>     upper execution bounds.
>
>   * Updated set_mempolicy01 to make use of tst_set_timeout().
>
>
> Li Wang (3):
>   lib: redefine the overall timeout logic of test
>   lib: add tst_set_timeout for reset entire timeout
>   set_mempolicy01: make use of tst_set_timeout
>

Note:
Seems I forgot to update the cover-letter title and add the 4/4 patch,
but that is not a big deal, it already covers the major changes there.


-- 
Regards,
Li Wang