On Thu, Apr 06, 2023 at 03:33:13PM +0800, lirongqing@xxxxxxxxx wrote: > From: Li RongQing <lirongqing@xxxxxxxxx> > > Check whether vCPU is preempted or not only when HLT is trapped or > there is not realtime hint. In other words, it is unnecessary to check > preemption when vCPU has realtime hint (which means vCPU has dedicated > pCP) and has not PV_UNHALT (which means unintercepted HLT), because > vCPU should not to be marked as preempted in this setup. To what benefit? > Signed-off-by: Li RongQing <lirongqing@xxxxxxxxx> > --- > diff with v1: rewrite changelog and indentation > > arch/x86/kernel/kvm.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c > index 1cceac5..25398d2 100644 > --- a/arch/x86/kernel/kvm.c > +++ b/arch/x86/kernel/kvm.c > @@ -820,8 +820,10 @@ static void __init kvm_guest_init(void) > has_steal_clock = 1; > static_call_update(pv_steal_clock, kvm_steal_clock); > > - pv_ops.lock.vcpu_is_preempted = > - PV_CALLEE_SAVE(__kvm_vcpu_is_preempted); > + if (kvm_para_has_feature(KVM_FEATURE_PV_UNHALT) || > + !kvm_para_has_hint(KVM_HINTS_REALTIME)) This is atrocious coding style, please align on the (. > + pv_ops.lock.vcpu_is_preempted = > + PV_CALLEE_SAVE(__kvm_vcpu_is_preempted); > } > > if (kvm_para_has_feature(KVM_FEATURE_PV_EOI)) > -- > 2.9.4 >
