[Findecanor]

RISC-V's vector extension will have at least 128 bits in application processors, so I think you could set VLEN=128 and just use SIMD algorithms.

The P extension is intended more for embedded microcontrollers for which the V extension would be too expensive. It reuses the GPRs at whatever width they are at (32 or 64 bits).

[camel-cdr]

That or you can detect the vector length and specialize for it, just like it's already done on x86 with VLEN 128, 256, and 512 for sse, avx, and avx512.