Date: Sat, 16 Nov 2019 02:36:42 +0100
From: Tomas Vondra <tomas.vondra@2ndquadrant.com>
To: tuomas.leikola@gmail.com, pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #16111: Unexpected query compilation error =?utf-8?B?4oCc?=
 =?utf-8?Q?negative_bitmapset_member_not_allowed=E2=80=9D?=
Message-ID: <20191116013642.t45tcb5r2av5nhn5@development>
References: <16111-687799584c3a7e73@postgresql.org>
 <20191113134114.rxcli5c6zjmryvf6@development>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20191113134114.rxcli5c6zjmryvf6@development>
User-Agent: NeoMutt/20180716
Precedence: bulk

On Wed, Nov 13, 2019 at 02:41:14PM +0100, Tomas Vondra wrote:
>On Wed, Nov 13, 2019 at 10:42:32AM +0000, PG Bug reporting form wrote:
>>The following bug has been logged on the website:
>>
>>Bug reference:      16111
>>Logged by:          Tuomas Leikola
>>Email address:      tuomas.leikola@gmail.com
>>PostgreSQL version: 12.0
>>Operating system:   windows, linux
>>Description:
>>
>>I’ve been trying to cook up some kind of online reorganizer, which requires
>>looking at rows’ physical locations and rearranging those that are
>>suboptimal. In this endeavor, I stumbled on what appears to be a compiler
>>bug introduced with multivariate statistics.
>>
>
>Yeah, this is a bug in estimate_multivariate_ndistinct, which fails to
>ignore system attributes :-(
>
>The attached patch fixes it, but unfortunatly we've missed this round of
>minor releases :-( I'll get it pushed in a couple of days.
>
>Thanks for the report and the reproducer, very helpful!
>

I've pushed the fix, after a minor improvement (it wasn't quite right,
because the bitmap was updated again a bit later). Backpatched to 10+.


regards

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