Request a Demo
Home   >   mailing lists

Re: Two division by 0 errors in optimizer/plan/planner.c and optimizer/path/costsize.c - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Two division by 0 errors in optimizer/plan/planner.c and optimizer/path/costsize.c
Date
Msg-id 11392.1459117324@sss.pgh.pa.us
Whole thread Raw
In response to Re: Two division by 0 errors in optimizer/plan/planner.c and optimizer/path/costsize.c  (Piotr Stefaniak <postgres@piotr-stefaniak.me>)
List pgsql-hackers
Tree view 
Piotr Stefaniak <postgres@piotr-stefaniak.me> writes:
> On the exact same note, something like this (again reduced from what 
> sqlsmith produced):
> leads to vardata.rel->tuples being zero here:
>     if (vardata.rel)
>         ndistinct *= vardata.rel->rows / vardata.rel->tuples;

Ugh.  That's a bit worse because it'll be 0/0, ie you get a NaN.
Thanks for the report.
        regards, tom lane

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Two division by 0 errors in optimizer/plan/planner.c and optimizer/path/costsize.c
Next
From: Andres Freund
Date:
Subject: Re: silent data loss with ext4 / all current versions