On Thu, Nov 18, 2021, Juergen Gross wrote: > On 18.11.21 16:05, Sean Christopherson wrote: > > Which is a good segue into pointing out that if a module param is added, it needs > > to be sanity checked against a KVM-defined max. The admin may be trusted to some > > extent, but there is zero reason to let userspace set max_vcspus to 4 billion. > > At that point, it really is just a param vs. capability question. > > I agree. Capping it at e.g. 65536 would probably be a good idea. Any reason to choose 65536 in particular? Why not cap it at the upper limit of NR_CPUS_RANGE_END / MAXSMP, which is currently 8192?
