quadsort VS mountain-sort

https://github.com/scandum/quadsort/blob/f171a0b26cf6bd6f6dc...

As you can see, quadsort 1.1.4.1 used 2 instead of 4 writes in the bi-directional parity merges. This was in June 2021, and would have compiled as branchless with clang, but as branched with gcc.

When I added a compile time check to use ternary operations for clang I was not adapting your work. I was well aware that clang compiled ternary operations as branchless, but I wasn't aware that rust did as well. I added the compile time check to use ternary operations for a fair performance comparison against glidesort.

https://raw.githubusercontent.com/scandum/fluxsort/main/imag...

As for ipnsort's small sort, it is very similar to quadsort's small sort, which uses stable sorting networks, instead of unstable sorting networks. From my perspective it's not exactly novel. I didn't go for unstable sorting networks in crumsort to increase code reuse, and to not reduce adaptivity.

In the code it looks like the seed to the benchmark can be provided as the 4th command line argument: https://github.com/scandum/quadsort/blob/master/src/bench.c#...

If you look at sorting for example, it's been proven that you can't do a comparison-based sort faster than O(n logn). You may then think that we've already found the fastest possible sorting algorithms since Quicksort and Mergesort are already O(n logn). However, new sorting algorithms keep being invented, for example Quadsort. They're all still O(n logn), but they do offer a considerable speed improvement over more traditional algorithms

It depends on the size of the structs. For struct pointers you're likely better off sorting keys and pointers simultaneously. It doesn't matter much until you get to large sizes (millions), but sorting indices and then selecting with them is random access. If the original ordering is messy, selecting can be slower than the sorting step. For structs a few words long, the unit you're moving is a larger portion of a cache line, and I'd expect the data movement to drown out any SIMD advantage. A radix sort might be all right because it moves less, but I'd probably go with sorting indices as the first thing to try unless I knew the arrays were huge. For very large structs there's an interesting effort called mountain sort[0], "probably the best sorting algorithm if you need to sort actual mountains by height". Given that it minimizes number of moves it's ignoring cache entirely. I haven't benchmarked so I can't say much about how practical it is.

[0] https://github.com/Morwenn/mountain-sort