On Tue, Dec 08 2020 at 09:33, Andy Lutomirski wrote:
> On Tue, Dec 8, 2020 at 8:26 AM Maxim Levitsky <[email protected]> wrote:
>
> IIRC we introduced mul_u64_u32_shift() for roughly this reason,but we
> don't seem to be using it in the relevant code paths. We should be
> able to use the same basic math with wider intermediates to allow very
> large intervals between updates.
That's 128 bit math and _NO_ we are not going there for various
reasons.
Hint1: There is a world outside of x8664 and TSC.
Hint2: Even on x8664 and TSC it comes with performance overhead
Thanks,
tglx
