On Thu, 1 Aug 2024 at 11:44, Alexandre Oliva wrote:
>
> When we get to test_pr91486_wait_until(), we're about 10s past the
> float_steady_clock epoch. This is enough for the 1s delta for the
> timeout to come out slightly lower when the futex-less wait_until
> converts the deadline from float_ste
On Aug 7, 2024, Alexandre Oliva wrote:
> On Aug 1, 2024, Alexandre Oliva wrote:
>> Each iteration calls float_steady_clock::now() [...] an extra iteration
>> will reach 5 and cause the test to fail.
>> (Do we really want to use floats, that even with this tweak have
>> borderline precision fo
On Aug 1, 2024, Alexandre Oliva wrote:
> Each iteration calls float_steady_clock::now() [...] an extra iteration
> will reach 5 and cause the test to fail.
> (Do we really want to use floats, that even with this tweak have
> borderline precision for sub-µs vs 1s deltas? Do we want to make sure
When we get to test_pr91486_wait_until(), we're about 10s past the
float_steady_clock epoch. This is enough for the 1s delta for the
timeout to come out slightly lower when the futex-less wait_until
converts the deadline from float_steady_clock to __clock_t. So we may
wake up a little too early,
On Jul 29, 2024, Alexandre Oliva wrote:
> - auto status = f1.wait_for(wait_time);
> + auto status __attribute__ (__unused__) = f1.wait_for(wait_time);
Sorry, it looks like I posted the patch before refreshing it. Make it:
+ auto status __attribute__ ((__unused__)) = f1.wait_for(wait_time)
When we get to test_pr91486_wait_until(), we're about 10s past the
float_steady_clock epoch. This is enough for the 1s delta for the
timeout to come out slightly lower when the futex-less wait_until
converts the deadline from float_steady_clock to __clock_t. So we may
wake up a little too early
