I'm still trying to understand is twos-complement multiplication. When you're computing the product of two n-bit numbers, the product potentially requires 2n b

I'm still trying to understand is twos-complement multiplication. When you're computing the product of two n-bit numbers, the product potentially requires 2n bits of representation. As I discovered in the RISC-V "M" instructions [0], the n least-significant bits of the result are the same, regardless of whether each operand is signed or unsigned. However, the most-significant n-bits of the result are contingent on the signed-ness of the two operands. Anyone know where I find get a good explanation for that? [0] https://msyksphinz-self.github.io/riscv-isadoc/html/rvm.html https://msyksphinz-self.github.io/riscv-isadoc/html/rvm.html