Re: [PATCH net v5 2/2] bonding: fix u32 overflow in compute_gap()

[Hangbin Liu]

On Fri, Aug 28, 2026 at 12:19:47PM +0300, Nikolay Aleksandrov wrote:
> > > How is that relevant to these patches?
> > > And how does that help at all if today that is done at about 10 second interval?
> > 
> > The 10 seconds is almost certainly completely arbitrary.
> > It is relevant because the 64bit divide is significantly expensive on 32bit.
> > 
> > David
> > 
> 
> Yeah, that is clear. But currently that recalculation is done once every 10
> seconds, such optimizations will be noise. Regardless of that, these changes
> are unrelated to the problem he is fixing with the set.

+1.

Thanks
Hangbin