Jeff Davis <pgsql@j-davis.com> writes:
> On Tue, 2009-06-09 at 10:51 -0700, Jeff Davis wrote:
>> I don't know what you mean by "frozen" exactly, but the start point of a
>> synchronized scan is stored in shared memory; otherwise, it wouldn't
>> know where to stop.
> Correction: I didn't actually mean _shared_ memory there. It's just
> backend-local memory.
Well, wherever it's stored, it's a demonstrable fact that we're not
getting the same rows after ExecutorRewind(); and that we do get the
same rows out if we disable synchronize_seqscans in Mark's test case.
I haven't got round to identifying exactly what to change if we decide
to go for a narrow fix instead of storing the query results at a higher
level.
regards, tom lane