Horton

Re: [POC] A better way to expand hash indexes.

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Mithun Cy <mithun.cy@enterprisedb.com>
Cc: Bruce Momjian <bruce@momjian.us>, Amit Kapila <amit.kapila16@gmail.com>, David Steele <david@pgmasters.net>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2017-04-04T11:45:21Z
Lists: pgsql-hackers 
On Tue, Apr 4, 2017 at 6:33 AM, Mithun Cy <mithun.cy@enterprisedb.com> wrote:
> On Tue, Apr 4, 2017 at 9:18 AM, Robert Haas <robertmhaas@gmail.com> wrote:
>> Committed.
>
> Thanks Robert,
>
> And also sorry, one unfortunate thing happened in the last patch while
> fixing one of the review comments a variable disappeared from the
> equation
> @_hash_spareindex.
>
>         splitpoint_phases +=
> -               (((num_bucket - 1) >> (HASH_SPLITPOINT_PHASE_BITS + 1)) &
> +               (((num_bucket - 1) >>
> +                 (splitpoint_group - (HASH_SPLITPOINT_PHASE_BITS + 1))) &
>                  HASH_SPLITPOINT_PHASE_MASK);   /* to 0-based value. */
>
> I wanted most significant 3 bits. And while fixing comments in patch11
> I unknowingly somehow removed splitpoint_group from the equation.
> Extremely sorry for the mistake. Thanks to Ashutosh Sharma for
> pointing the mistake.

Ugh, OK, committed that also.  Please try to be more careful in the future.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Commits

  1. Fix formula in _hash_spareindex.

  2. Expand hash indexes more gradually.

  3. Change hash indexes to store only the hash code rather than the whole indexed