Hi,
On 2017-03-07 02:57:30 +1300, Thomas Munro wrote:
> I'm not sure why nodeHashjoin.c is doing raw batchfile read/write
> operations anyway; why not use tuplestore.c for that (as
> tuplestore.c's comments incorrectly say is the case)?
Another reason presumably is that using tuplestores would make it harder
to control the amount of memory used - we do *not* want an extra set of
work_mem used here, right?
- Andres