Hi,
On 2018-04-11 11:57:20 +1200, Thomas Munro wrote:
> Rebased, but I don't actually like this patch any more. Over in
> another thread[1] I proposed that we should just make exit(1) the
> default behaviour built into latch.c for those cases that don't want
> to do something special (eg SyncRepWaitForLSN()), so we don't finish
> up duplicating the ugly exit(1) code everywhere (or worse, forgetting
> to). Tom Lane seemed to think that was an idea worth pursuing.
>
> I think what we need for PG12 is a patch that does that, and also
> introduces a reused WaitEventSet object in several of these places.
> Then eg SyncRepWaitForLSN() won't be interacting with contended kernel
> objects every time (assuming an implementation like epoll or
> eventually kqueue, completion ports etc is available).
>
> Then if pgarch_ArchiverCopyLoop() and HandleStartupProcInterrupts()
> (ie loops without waiting) adopt a prctl(PR_SET_PDEATHSIG)-based
> approach where available as suggested by Andres[2] or fall back to
> polling a reusable WaitEventSet (timeout = 0), then there'd be no more
> calls to PostmasterIsAlive() outside latch.c.
I'm still unconvinced by this. There's good reasons why code might be
busy-looping without checking the latch, and we shouldn't force code to
add more latch checks if unnecessary. Resetting the latch frequently can
actually increase the amount of context switches considerably.
- Andres