Re: [PATCH next 0/3] lib: Implement mul_u64_u64_div_u64_roundup()

[Nicolas Pitre]

On Fri, 16 May 2025, David Laight wrote:

> On Fri, 16 May 2025 11:47:58 +0200
> Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxx> wrote:
> 
> > Hello David,
> > 
> > On Sat, Apr 05, 2025 at 09:45:27PM +0100, David Laight wrote:
> > > The pwm-stm32.c code wants a 'rounding up' version of mul_u64_u64_div_u64().
> > > This can be done simply by adding 'divisor - 1' to the 128bit product.
> > > Implement mul_u64_add_u64_div_u64(a, b, c, d) = (a * b + c)/d based on the
> > > existing code.
> > > Define mul_u64_u64_div_u64(a, b, d) as mul_u64_add_u64_div_u64(a, b, 0, d) and
> > > mul_u64_u64_div_u64_roundup(a, b, d) as mul_u64_add_u64_div_u64(a, b, d-1, d).
> > > 
> > > Only x86-64 has an optimsed (asm) version of the function.
> > > That is optimised to avoid the 'add c' when c is known to be zero.
> > > In all other cases the extra code will be noise compared to the software
> > > divide code.
> > > 
> > > I've updated the test module to test mul_u64_u64_div_u64_roundup() and
> > > also enhanced it to verify the C division code on x86-64.
> > > 
> > > Note that the code generated by gcc (eg for 32bit x86) just for the multiply
> > > is rather more horrid than one would expect (clang does better).
> > > I dread to think how long the divide loop takes.
> > > And I'm not at all sure the call in kernel/sched/cputime.c isn't in a
> > > relatively common path (rather than just hardware initialisation).
> > > 
> > > David Laight (3):
> > >   lib: Add mul_u64_add_u64_div_u64() and mul_u64_u64_div_u64_roundup()
> > >   lib: Add tests for mul_u64_u64_div_u64_roundup()
> > >   lib: Update the muldiv64 tests to verify the C on x86-64  
> > 
> > I wonder what happend to this series. I'd like to make use of
> > mul_u64_u64_div_u64_roundup() so I'd be interested to get this into the
> > mainline.
> 
> I've a WIP rewrite of the divide code, speeds it up considerably for
> 'not amd-64'.

May I suggest you simply submit the new API now (addressing my latest 
comments if possible) and submit the divide optimization later?


Nicolas