Julia position in the Debian Benchmark Game can be improved, and categorization of some Julia there is unfair

[Palli]

@foobar_lv2

Smallish allocations can be done with a bump allocator, which is very fast.

See my experiment with Immix GC:

Improved allocation design, with 4-byte pointers, and sometimes 5-byte in effect

@Oscar_Smith, @GeorgeGkountouras Immix GC (./julia) is confidently 33% faster than Julia 1.11 (with its default GC), for three threads, which is the optimal number for at least than program with (at least) Immix, on this worst-case outlier from Debian Benchmark Game, though only 11% faster than 1.10. Julia 1.11 (default GC) is 17% slower than that Julia 1.10 (default GC), i.e. Immix (1.12.0-DEV.1745) is 33% faster than 1.11. Since the benchmark game uses four threads (all the cores), and I beli…

Such 33% faster (37% faster based on mean) would make Julia climb up to 4.7436 sec. (4.46 sec. if their benchmark uses mean, calculated from 37%) on that benchmark (and improve Julia’s overall rank likely), e.g. climb past C#, and Erlang which is actually faster, and approach Java at 2.62 sec. still 81% slower than it, and 4.9x slower than fastest (C++). We can and should tune it to be able to exploit threads better (now 3 threads, not 4, is optimal).