On 23/11/2017 17:59, David Hildenbrand wrote: > On 21.11.2017 18:22, Mark Kanda wrote: >> The potential performance advantages of a vmcs02 pool have never been >> realized. To simplify the code, eliminate the pool. Instead, a single >> vmcs02 is allocated per VCPU when the VCPU enters VMX operation. > > Did you do any performance measurement to come to the conclusion that we > can remove it? :) This is enough I guess: #define VMCS02_POOL_SIZE 1 so there wasn't really a pool of them... Paolo >> >> Signed-off-by: Jim Mattson <jmattson@xxxxxxxxxx> > > -> why is that in here? > >> Signed-off-by: Mark Kanda <mark.kanda@xxxxxxxxxx> >> Reviewed-by: Ameya More <ameya.more@xxxxxxxxxx> >> --- >> arch/x86/kvm/vmx.c | 146 >
