I prepared couple of patches to optimize st_table performance.
First patch (st_func.patch) is a trivial translation of some macroses to functions,
but it gives ~4% of improvement on gcc-4.5.2 32bit (sorry for not testing other environments).
Second patch (st_pack_table) provides packing for tables sized upto 12 entries.
It gives countable memory footprint reduction and measurable performance improvement
(~8% compared to trunk, ~4% compared to st_func.patch).
Note: synthetic benchmark could show some performance degradation, so that there
could be some exotic workload, which suffer from patch.
But realworld application (and even make check) runs uniformly faster.
Updated by funny_falcon (Yura Sokolov) about 8 years ago
Another approach to st_table optimization: use lightweight pool allocator for st_table, st_table_entry and bins of smallest size.