Knut Anders Hatlen <[email protected]> writes: > I think the memory footprint and the available memory are also used to > determine the cost of a hash scan (if it doesn't fit in main memory, it > will be considered more expensive).
That seems a good condicate culprit here, especially when running the entire test suite. I guess people run with different heap sizes.. and/or we have some leakage. Not sure if anyone has tested to leaks recently? Dag --
