[LTP] [RFC PATCh] lib: redefine the overall timeout logic of test

[Cyril Hrubis]

Hi!
> > The reason for calling tst_set_runtime() in each iteration is that we can
> > directly utilize the real elapsed time as runtime and don't need to do
> > additional measurements.
> >
> 
> Forgot to say, that each iteration has four sub-tcases to accomplish.
> The recommended way (you gave above) is to regard them as one
> big test and reset timeout by heartbeat() in tst_test->tcnt. If this value
> is big enough, that's fine. But I fear that value (come from measurement)
> is still not covered all situations.

I agree that the test is a bit unpredictable, as it synces filesytem and
drops caches twice in each iteration. However the .timeout is a safety
mechanism, so nothing stops us for making it 4x or 8x of the usuall
test execution time on slower hardware.

-- 
Cyril Hrubis
chrubis@suse.cz