Comment 21 for bug 1833322

Hi Paride

> Back in the day I asked upstream their take on irqbalance usefulness with
> newer kernels, here is their reply:
> https://github.com/Irqbalance/irqbalance/issues/151

Thanks for this and the other extra pointers.
The Debian bug was referenced before, AFAIC it is mostly around
a) the kernel got smarter in many cases (true)
b) bad in virtual environments (we already removed it from those)

And in that discussion the upstream comments (it is good to see that
they are still convinced of their code) revolved around:
c) There should be no conflict with running irqbalance (with the new kernel)
d) The kernel policy is driver centric (irqbalance has a full picture)

Both - as I read them - are more arguments to keep it than to remove.
But as all other, not with enough data to make it a clear yes/no.

As I said much earlier in this case, I feel this is system and workload
dependent and hence there will never be a clear generic yes/no.
The best we can achieve is finding sets (like images used in virtual
environments - or as suggested desktop systems) and drop it being the
default there.