On 2018-08-28 13:50:43 +1200, Thomas Munro wrote:
> I'm curious about something which may qualify as a stupid question.
>
> What bad thing would happen if we used OIDs directly as hash values in
> internal hash tables (that is, instead of uint32_hash() we'd use
> uint32_identity(), or somehow optimise it away entirely, as you can
> see in some C++ standard libraries for eg std::hash<int>)?
Oids are very much not equally distributed, so in all likelihood you'd
get cases very you currently have a reasonably well averaged out usage
of the hashtable, not be that anymore.
It's also fairly cheap to hash an oid.
> However, as far as I can see OIDs are expected to have an even
> distribution (or at least we don't expect regular sized gaps), so the
> hazard doesn't seem to apply.
Huh? Oids between, say, 1 and FirstNormalObjectId, are vastly more
common than the rest. And even after that, individual tables get large
clusters of sequential values to the global oid counter.
Greetings,
Andres Freund