[LTP] [PATCH v2] sched: starvation: Autocallibrate the timeout

[Anders Roxell]

On Wed, 12 Jun 2024 at 13:23, Cyril Hrubis <chrubis@suse.cz> wrote:
>
> Instead of hardcoding the values we attempt to measure the CPU speed and
> set the timeout accordingly. Given that the difference in the duration
> of the test when the kernel is buggy is about 30x we do not have to have
> a precise callibration, just very rough estimate if we are running on a
> server or small ARM board would suffice.
>
> So we attempt to measure how long does a bussy loop take and base the
> default timeout on that. On x86_64 CPUs the resulting timeout is about
> double of the value of the actual test runtime and works fine, but we
> need to make sure that the coeficient we divide the result from
> callibrate works for small boards too. So please run the test on as many
> machines as you can and report if we need to make the dividor smaller or
> not.
>
> CC: Steve <xufeifei1@oppo.com>
> Signed-off-by: Cyril Hrubis <chrubis@suse.cz>

I've tested this patch on 1 arm HW and 2 arm64 HW's. see the results below.

arm HW:
===== [1;36mstarvation[0m =====
command: starvation
[   43.278533] /usr/local/bin/kirk[356]: starting test starvation (starvation)
tst_test.c:1734: TINFO: LTP version: 20180118-5676-g4696467fb
tst_test.c:1618: TINFO: Timeout per run is 0h 05m 24s
starvation.c:86: TPASS: sched_setaffinity(0, sizeof(mask), &mask) returned 0
starvation.c:54: TINFO: CPU did 100000000 loops in 89172us
tst_test.c:1626: TINFO: Updating max runtime to 0h 01m 29s
tst_test.c:1618: TINFO: Timeout per run is 0h 06m 53s
starvation.c:132: TPASS: wait_for_pid(child_pid) passed
Summary:
passed   2
failed   0
broken   0
skipped  0
warnings 0
Duration: 1m 2s

arm64 HW1:
===== [1;36mstarvation[0m =====
command: starvation
[   49.213159] /usr/local/bin/kirk[362]: starting test starvation (starvation)
[   49.213159] /usr/local/bin/kirk[362]: starting test starvation (starvation)
tst_test.c:1734: TINFO: LTP version: 20180118-5676-g4696467fb
tst_test.c:1618: TINFO: Timeout per run is 0h 05m 24s
starvation.c:86: TPASS: sched_setaffinity(0, sizeof(mask), &mask) returned 0
starvation.c:54: TINFO: CPU did 100000000 loops in 141688us
tst_test.c:1626: TINFO: Updating max runtime to 0h 02m 21s
tst_test.c:1618: TINFO: Timeout per run is 0h 07m 45s
starvation.c:132: TPASS: wait_for_pid(child_pid) passed
Summary:
passed   2
failed   0
broken   0
skipped  0
warnings 0
Duration: 1m 32s

arm64 HW2:
===== [1;36mstarvation[0m =====
command: starvation
[   25.384826] /usr/local/bin/kirk[775]: starting test starvation (starvation)
tst_test.c:1734: TINFO: LTP version: 20180118-5676-g4696467fb
tst_test.c:1618: TINFO: Timeout per run is 0h 05m 24s
starvation.c:86: TPASS: sched_setaffinity(0, sizeof(mask), &mask) returned 0
starvation.c:54: TINFO: CPU did 100000000 loops in 113584us
tst_test.c:1626: TINFO: Updating max runtime to 0h 01m 53s
tst_test.c:1618: TINFO: Timeout per run is 0h 07m 17s
starvation.c:132: TPASS: wait_for_pid(child_pid) passed
Summary:
passed   2
failed   0
broken   0
skipped  0
warnings 0
Duration: 1m 15s

Tested-by: Anders Roxell <anders.roxell@linaro.org>

Cheers,
Anders