kvm_setup_pv_tlb_flush() does not check if kvm_para_available() is true
unlike kvm_guest_init(). Therefore it may proceed even if
kvm_para_available() is 0 (e.g., when booting the VM with `qemu -cpu
host,kvm=off`). In this case, the pr_info message is confusing.
This fixed it.
Signed-off-by: Masanori Misono <m.misono760@xxxxxxxxx>
---
arch/x86/kernel/kvm.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 5c93a65ee1e5..869ac19c52cf 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -69,6 +69,7 @@ early_param("no-steal-acc", parse_no_stealacc);
static DEFINE_PER_CPU_DECRYPTED(struct kvm_vcpu_pv_apf_data, apf_reason) __aligned(64);
static DEFINE_PER_CPU_DECRYPTED(struct kvm_steal_time, steal_time) __aligned(64);
static int has_steal_clock = 0;
+static int has_pv_tlb_flush = 0;
/*
* No need for any "IO delay" on KVM
@@ -641,6 +642,7 @@ static void __init kvm_guest_init(void)
if (kvm_para_has_feature(KVM_FEATURE_PV_TLB_FLUSH) &&
!kvm_para_has_hint(KVM_HINTS_REALTIME) &&
kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
+ has_pv_tlb_flush = 1;
pv_ops.mmu.flush_tlb_others = kvm_flush_tlb_others;
pv_ops.mmu.tlb_remove_table = tlb_remove_table;
}
@@ -748,9 +750,7 @@ static __init int kvm_setup_pv_tlb_flush(void)
{
int cpu;
- if (kvm_para_has_feature(KVM_FEATURE_PV_TLB_FLUSH) &&
- !kvm_para_has_hint(KVM_HINTS_REALTIME) &&
- kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
+ if (has_pv_tlb_flush){
for_each_possible_cpu(cpu) {
zalloc_cpumask_var_node(per_cpu_ptr(&__pv_tlb_mask, cpu),
GFP_KERNEL, cpu_to_node(cpu));
--
2.17.1
