Re: [PATCH] xen: fix for_each_cpu when NR_CPUS=1

[Jan Beulich <jbeulich@xxxxxxxx>]

On 11.03.2021 10:40, Dario Faggioli wrote:
> When running an hypervisor build with NR_CPUS=1 for_each_cpu does not
> take into account whether the bit of the CPU is set or not in the
> provided mask.
> 
> This means that whatever we have in the bodies of these loops is always
> done once, even if the mask was empty and it should never be done. This
> is clearly a bug and was in fact causing an assert to trigger in credit2
> code.
> 
> Removing the special casing of NR_CPUS == 1 makes things work again.
> 
> Signed-off-by: Dario Faggioli <dfaggioli@xxxxxxxx>

Doesn't this want a Reported-by: Roger?

Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>

And FTR I don't really mind the other NR_CPUS == 1 piece of logic to
remain there.

> I'm not really sure whether this should be 4.15 material.
> 
> It's definitely a bug, IMO. The risk is also pretty low, considering
> that no one should really run Xen in this configuration (NR_CPUS=1, I
> mean). Which is also the reason why it's probably not really important
> that we fix it at this point of the release cycle.

I agree; I'll put it in the 4.16 bucket.

Jan