Camel Coder

Now that there are two RVV 1.0 devboards available, this can probably be revisited. (CanMV k230 with XuanTie C908, and Banana Pi BPI-F3 with SpacemiT X60, I've got them both...

@lemire > Firstly, we have an empty architecture, ppc4, that is currently all falling back on scalar. You can just copy that, so take src/ppc4 and copy it to, say,...

@lemire > So we are assuming that RISC-V runs on Linux. Is that true? It certainly doesn't run on windows today, I think e.g. FreeBSD also has support, but I'd...

@davidlt That's great. I was aware of the XTheadVector patches, good to hear that they are now merged. PS: the mentioned article is done now: https://camel-cdr.github.io/rvv-bench-results/articles/vector-utf.html

So I've run into a bit of a predicament. If I understood it correctly, the current behavior for x86 is to compile all backends using `SIMDUTF_TARGET_REGION`, even if you don't...

This may not be the way you'd like to have the API work, but I wrote down the code in a way I'd write it. Note that I created an...

FYI: none of the code samples above work now, and while memcpy still works, even things like saxpy stopped working on the latest commit: ```c // asm.S .text .balign 8...

Hi @Anzooooo, thanks for the notice. I tested it again just now and realized that `ascii_to_utf16` had a copy and past bug from my benchmark, which I've now adjusted in...

@Anzooooo Ok, this is really weird. The `DefaultConfig` works if I remove any print statements before the RVV code. So for the test cases commenting out the `printf("beg\n");` makes it...

@Anzooooo I haven't ported that yet, only isolated benchmarks, for testing. The problem occurs with the code examples I posted here. When I try executing the `saxpy` code above, then...