From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.linutronix.de (146.0.238.70:993) by crypto-ml.lab.linutronix.de with IMAP4-SSL for ; 12 Jul 2018 14:39:34 -0000 Received: from localhost ([127.0.0.1] helo=nanos.tec.linutronix.de) by Galois.linutronix.de with esmtp (Exim 4.80) (envelope-from ) id 1fdcft-0006vG-20 for speck@linutronix.de; Thu, 12 Jul 2018 16:34:29 +0200 Message-ID: <20180712142957.114708196@linutronix.de> Date: Thu, 12 Jul 2018 16:19:04 +0200 From: Thomas Gleixner References: <20180712141902.576562442@linutronix.de> Subject: [patch V10 02/10] Control knobs and Documentation 2 Mime-Version: 1 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit To: speck@linutronix.de List-ID: Subject: [patch V10 02/10] x86/kvm: Drop L1TF MSR list approach From: Thomas Gleixner The VMX module parameter to control the L1D flush should become writeable. The MSR list is set up at VM init per guest VCPU, but the run time switching is based on a static key which is global. Toggling the MSR list at run time might be feasible, but for now drop this optimization and use the regular MSR write to make run-time switching possible. The default mitigation is the conditional flush anyway, so for extra paranoid setups this will add some small overhead, but the extra code executed is in the noise compared to the flush itself. Aside of that the EPT disabled case is not handled correctly at the moment and the MSR list magic is in the way for fixing that as well. If it's really providing a significant advantage, then this needs to be revisited after the code is correct and the control is writable. Signed-off-by: Thomas Gleixner --- arch/x86/kvm/vmx.c | 43 +++++++------------------------------------ 1 file changed, 7 insertions(+), 36 deletions(-) --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -6231,16 +6231,6 @@ static void ept_set_mmio_spte_mask(void) VMX_EPT_MISCONFIG_WX_VALUE); } -static bool vmx_l1d_use_msr_save_list(void) -{ - if (!enable_ept || !boot_cpu_has_bug(X86_BUG_L1TF) || - static_cpu_has(X86_FEATURE_HYPERVISOR) || - !static_cpu_has(X86_FEATURE_FLUSH_L1D)) - return false; - - return vmentry_l1d_flush == VMENTER_L1D_FLUSH_ALWAYS; -} - #define VMX_XSS_EXIT_BITMAP 0 /* * Sets up the vmcs for emulated real mode. @@ -6362,12 +6352,6 @@ static void vmx_vcpu_setup(struct vcpu_v vmcs_write64(PML_ADDRESS, page_to_phys(vmx->pml_pg)); vmcs_write16(GUEST_PML_INDEX, PML_ENTITY_NUM - 1); } - /* - * If flushing the L1D cache on every VMENTER is enforced and the - * MSR is available, use the MSR save list. - */ - if (vmx_l1d_use_msr_save_list()) - add_atomic_switch_msr(vmx, MSR_IA32_FLUSH_CMD, L1D_FLUSH, 0, true); } static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) @@ -9617,26 +9601,14 @@ static void vmx_l1d_flush(struct kvm_vcp bool always; /* - * This code is only executed when: - * - the flush mode is 'cond' - * - the flush mode is 'always' and the flush MSR is not - * available - * - * If the CPU has the flush MSR then clear the flush bit because - * 'always' mode is handled via the MSR save list. - * - * If the MSR is not avaibable then act depending on the mitigation - * mode: If 'flush always', keep the flush bit set, otherwise clear - * it. + * This code is only executed when the the flush mode is 'cond' or + * 'always' * - * The flush bit gets set again either from vcpu_run() or from one - * of the unsafe VMEXIT handlers. + * If 'flush always', keep the flush bit set, otherwise clear + * it. The flush bit gets set again either from vcpu_run() or from + * one of the unsafe VMEXIT handlers. */ - if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) - always = false; - else - always = vmentry_l1d_flush == VMENTER_L1D_FLUSH_ALWAYS; - + always = vmentry_l1d_flush == VMENTER_L1D_FLUSH_ALWAYS; vcpu->arch.l1tf_flush_l1d = always; vcpu->stat.l1d_flush++; @@ -13234,8 +13206,7 @@ static int __init vmx_setup_l1d_flush(vo l1tf_vmx_mitigation = vmentry_l1d_flush; - if (vmentry_l1d_flush == VMENTER_L1D_FLUSH_NEVER || - vmx_l1d_use_msr_save_list()) + if (vmentry_l1d_flush == VMENTER_L1D_FLUSH_NEVER) return 0; if (!boot_cpu_has(X86_FEATURE_FLUSH_L1D)) {