I wrote:
> ... so my guess is that something messed up in transmitting or combining a
> parallel worker's execution time.
Hmph, no, a look at explain.c shows that the "Execution Time" is just
based on the difference of INSTR_TIME_SET_CURRENT measurements taken
within the current process. It's difficult to conclude anything except
that the clock went backwards. Which is weird, because according to [1]
that system does have clock_gettime(CLOCK_MONOTONIC), which'd be our
preferred choice of INSTR_TIME time base; and such clocks are not
supposed to go backwards ever.
Tis puzzling, but it's hard to avoid the suspicion of a kernel bug.
regards, tom lane
[1] https://man.openbsd.org/OpenBSD-6.8/clock_gettime