I wrote:
> Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc> writes:
>> samples % symbol name
>> 350318533 98.8618 mergepreread
>> 971822 0.2743 tuplesort_gettuple_common
>> 413674 0.1167 tuplesort_heap_siftup
> I don't have enough memory to really reproduce this, but I've come close
> enough that I believe I see what's happening. It's an artifact of the
> code I added recently to prevent the SortTuple array from growing during
> the merge phase, specifically the "mergeslotsfree" logic.
I've committed a fix for this; if you're still interested, please
retest.
regards, tom lane