Great read. A question: what is the status of this problem on other architectures such as ARM and RISC-V, would the analysis and solution be the same? e.g. does ARM have invariant TSC?

riscv has mtime. it is somewhat implementation defined, but it should be a single hardware timer shared by all harts. The Zicntr extension defines user space rdtime psuedo instruction to acesss it from userspace.

aarch64 has cntvct_el0 status register that can be read from userspace.