Horton

Re: tuple radix sort

John Naylor <johncnaylorls@gmail.com>

View thread
From: John Naylor <johncnaylorls@gmail.com>
To: Chao Li <li.evan.chao@gmail.com>
Cc: Chengpeng Yan <chengpeng_yan@outlook.com>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>, Peter Geoghegan <pg@bowt.ie>
Date: 2025-12-04T05:30:52Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Skip common prefixes during radix sort

  2. Perform radix sort on SortTuples with pass-by-value Datums

On Wed, Dec 3, 2025 at 3:22 PM Chao Li <li.evan.chao@gmail.com> wrote:
> I played with this again today and found an optimization that seems to dramatically improve the performance:
>
> ```
> +static void
> +radix_sort_tuple(SortTuple *begin, size_t n_elems, int level, Tuplesortstate *state)
> +{
> +       RadixPartitionInfo partitions[256] = {0};
> +       uint8_t         remaining_partitions[256] = {0};
> ```
>
> Here partitions and remaining_partitions are just temporary buffers, allocating memory from stack and initialize them seems slow. By passing them as function parameters are much faster. See attached diff for my change.

The lesson here is: you can make it as fast as you like if you
accidentally blow away the state that we needed for this to work
correctly.

--
John Naylor
Amazon Web Services