riscv-gnu-toolchain VS kendryte-gcc

> no absurdely and grotesquely massive and complex compilers anywhere

Absence of evidence is not evidence of absence, and anyway there's not even an absence: https://github.com/riscv-collab/riscv-gnu-toolchain https://llvm.org/docs/RISCVUsage.html

> feature creeps on computer language syntax nowhere to be found

At least one of us is very confused, and in case it's me, how do language details matter to RISC-V?

I then looked at the .cargo/config.toml provided by the guide and saw that it wasn't actually statically compiling the code. After a bit of tinkering and building my own toolchain from here, I ended up with this config.toml file:

b) collabriscv - essentially gcc 12.2 + binutils master/2.40 as per https://github.com/riscv-collab/riscv-gnu-toolchain

Or would it be better to take what is in https://github.com/riscv-collab/riscv-gnu-toolchain which is gcc 12.2 and start from there?

I'm not able to build https://github.com/riscv-collab/riscv-gnu-toolchain.git like this:

Kendryte had their own chain: https://github.com/kendryte/kendryte-gnu-toolchain and it contained some Kendrye-specific magic in atomics, douible/floating point, in the CSRs, and a few other comparatively esoteric areas. Even if you dig through https://github.com/kendryte/kendryte-gcc/commits/5b254d7c602ec112161daa0981e1b9ace1052c47 it can be really hard to tell what was a cherrypick from upstream and what was new. (Those S390 changes seem unlikely to add much value...)

I thus have no idea why Kendryte has its own (seemingly abandoned) branch of gcc ; it's not for intrinsics, like I thought. (I don't care enough to pull it and diff it.)