Horton

Re: A qsort template

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: John Naylor <john.naylor@enterprisedb.com>
Cc: Thomas Munro <thomas.munro@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2021-07-30T07:10:49Z
Lists: pgsql-hackers 
On Fri, Jul 30, 2021 at 3:34 AM John Naylor
<john.naylor@enterprisedb.com> wrote:
> I'm also attaching your tuplesort patch so others can see what exactly I'm comparing.

If you're going to specialize the sort routine for unsigned integer
style abbreviated keys then you might as well cover all relevant
opclasses/types. Almost all abbreviated key schemes produce
conditioned datums that are designed to use simple 3-way unsigned int
comparator. It's not just text. (Actually, the only abbreviated key
scheme that doesn't do it that way is numeric.)

Offhand I know that UUID, macaddr, and inet all have abbreviated keys
that can use your new ssup_datum_binary_cmp() comparator instead of
their own duplicated comparator (which will make them use the
corresponding specialized sort routine inside tuplesort.c).

-- 
Peter Geoghegan

Commits

  1. Remove debug messages from tuplesort_sort_memtuples()

  2. Fix performance regression in tuplesort specializations

  3. Fix tuplesort optimization for CLUSTER-on-expression.

  4. Initial pgindent and pgperltidy run for v14.

  5. Specialize checkpointer sort functions.

  6. Use sort_template.h for qsort() and qsort_arg().

  7. Use sort_template.h for qsort_tuple() and qsort_ssup().

  8. Add sort_template.h for making sort functions.

  9. Use abbreviated keys for faster sorting of text datums.