Re: pgsql: Fix theoretical bug in tuplesort

[Tom Lane]

... btw, wouldn't it be advisable to set stup->tuple to NULL
after dropping the referenced storage?

                        regards, tom lane