On 02/03/20 10:28, Nadav Amit wrote: > > Yet, this might be a slippery slope, making Linux optimized to run on KVM > (and maybe Xen). In other words, I wonder whether a similar change was > acceptable for a paravirtual feature that is only supported by a proprietary > hypervisor, such as Hyper-V or VMware. Your concern is understandable. I think in this case Andy's patch makes sense since there is a hooking mechanism that is used in exactly one case and there are no performance issues in removing it. In fact if more hypervisors implemented a #PF extension I would keep the static branch, but use it to choose between a pvop and the default case; this would be a much more localized change, so arguably better than trap_init. Paolo