diff options
| author | Wanpeng Li <wanpengli@tencent.com> | 2018-02-13 09:05:41 +0800 |
|---|---|---|
| committer | Radim Krčmář <rkrcmar@redhat.com> | 2018-03-06 18:40:45 +0100 |
| commit | b2798ba0b8769b42f00899b44a538b5fcecb480d (patch) | |
| tree | d62c2b5aadf73a3f60523e8cd65444c7605c3808 | |
| parent | a4429e53c9b3082b05e51224c3d58dbdd39306c5 (diff) | |
| download | linux-0-day-b2798ba0b8769b42f00899b44a538b5fcecb480d.tar.gz linux-0-day-b2798ba0b8769b42f00899b44a538b5fcecb480d.tar.xz | |
KVM: X86: Choose qspinlock when dedicated physical CPUs are available
Waiman Long mentioned that:
> Generally speaking, unfair lock performs well for VMs with a small
> number of vCPUs. Native qspinlock may perform better than pvqspinlock
> if there is vCPU pinning and there is no vCPU over-commitment.
This patch uses the KVM_HINTS_DEDICATED performance hint, which is
provided by the hypervisor admin, to choose the qspinlock algorithm
when a dedicated physical CPU is available.
PV_DEDICATED = 1, PV_UNHALT = anything: default is qspinlock
PV_DEDICATED = 0, PV_UNHALT = 1: default is Hybrid PV queued/unfair lock
PV_DEDICATED = 0, PV_UNHALT = 0: default is tas
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
| -rw-r--r-- | arch/x86/kernel/kvm.c | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 8c9d98c46f840..f803e89e0b474 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -735,6 +735,11 @@ void __init kvm_spinlock_init(void) if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) return; + if (kvm_para_has_hint(KVM_HINTS_DEDICATED)) { + static_branch_disable(&virt_spin_lock_key); + return; + } + __pv_init_lock_hash(); pv_lock_ops.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath; pv_lock_ops.queued_spin_unlock = PV_CALLEE_SAVE(__pv_queued_spin_unlock); |