On Mon, 12 Dec 2005, Neil Conway wrote:
>
> Whether we should go to the trouble of second-guessing glibc is a
> separate question, though: it would be good to see some performance
> figures for real-world queries.
>
For qsort, due to its simple usage, I think simulation test should be
enough. But we have to consider many situations like cardinality, data
distribution etc. Maybe not easy to find real world queries providing so
many variations.
> BTW, Luke Lonergan recently posted some performance results for a fairly
> efficient public domain implementation of qsort to the bizgres list:
>
> http://lists.pgfoundry.org/pipermail/bizgres-general/2005-December/000294.html
>
Ooops, more interesting than the thread itself ;-)
Regards,
Qingqing