Re: DBT-3 with SF=20 got failed

On 09/11/2015 07:16 PM, Robert Haas wrote:
> On Fri, Sep 11, 2015 at 1:12 PM, Tomas Vondra
> <tomas.vondra@2ndquadrant.com> wrote:
>> I'm arguing for fixing the existing bug, and then addressing the case of
>> over-estimation separately, with proper analysis.
>
> Well, this is part of how we're looking it differently.  I think the
> bug is "we're passing a value to palloc that is too large, so
> sometimes it fails" and the way to fix that is to properly limit the
> value.  You are clearly defining the bug a bit differently.

Yes, I see it differently.

I don't quite understand why limiting the value is more "proper" than 
using a function that can handle the actual value.

The proposed bugfix addresses the issue in the most straightforward way, 
without introducing additional considerations about possible 
over-estimations (which the current code completely ignores, so this is 
a new thing). I think bugfixes should not introduce such changes to 
behavior (albeit internal), especially not without any numbers.

regards

-- 
Tomas Vondra                   http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services