Peter Geoghegan <peter@2ndquadrant.com> writes:
> On 14 June 2012 19:28, Robert Haas <robertmhaas@gmail.com> wrote:
>> I thought that doubling repeatedly would be overly aggressive in terms
>> of memory usage.
> I fail to understand how this sortsupport buffer fundamentally differs
> from a generic dynamic array abstraction built to contain chars. That
> being the case, I see no reason not to just do what everyone else does
> when expanding dynamic arrays, and no reason why we shouldn't make
> essentially the same time-space trade-off here as others do elsewhere.
I agree with Peter on this one; not only is double-each-time the most
widespread plan, but it is what we do in just about every other place
in Postgres that needs a dynamically expansible buffer. If you do it
randomly differently here, readers of the code will be constantly
stopping to wonder why it's different here and if that's a bug or not.
(And from a performance standpoint, I'm not entirely convinced it's not
a bug, anyway. Worst-case behavior could be pretty bad.)
regards, tom lane