"Florian G. Pflug" <fgp(at)phlo(dot)org> writes:
> Tom Lane wrote:
>> This is wrong --- it will break early detection of aborted
>> subtransactions.
> I don't see how - TransactionIdIsInProgress doesn't even touch
> latestCompletedXid.
Nah, I take that back --- I was worried that XidInMVCCSnapshot might
say it was still running when it wasn't, but we never apply
XidInMVCCSnapshot to XIDs that we haven't already determined to be
committed, so it doesn't matter.
It's still a bad idea though. Let me go respond to your -hackers
post (which arrived here after the other one).
regards, tom lane