KVM: x86: simplify handling of PKRU

commit b9dd21e104bcd45e124acfe978a79df71259e59b upstream.

Move it to struct kvm_arch_vcpu, replacing guest_pkru_valid with a
simple comparison against the host value of the register.  The write of
PKRU in addition can be skipped if the guest has not enabled the feature.
Once we do this, we need not test OSPKE in the host anymore, because
guest_CR4.PKE=1 implies host_CR4.PKE=1.

The static PKU test is kept to elide the code on older CPUs.

Suggested-by: Yang Zhang <zy107165@alibaba-inc.com>
Fixes: 1be0e61c1f255faaeab04a390e00c8b9b9042870
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 695605eb1dfbd941b591564fef723bd030c3316f..ed8fdf86acfbcf0aa34e3051698c04ef3ba88e7a 100644 (file)
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -486,6 +486,7 @@ struct kvm_vcpu_arch {
        unsigned long cr4;
        unsigned long cr4_guest_owned_bits;
        unsigned long cr8;
+       u32 pkru;
        u32 hflags;
        u64 efer;
        u64 apic_base;

diff --git a/arch/x86/kvm/kvm_cache_regs.h b/arch/x86/kvm/kvm_cache_regs.h
index 762cdf2595f992fd4ac8bb1e4c2c8914b344db04..e1e89ee4af750dc51f78cfbf8aa71e22d77b4cd1 100644 (file)
--- a/arch/x86/kvm/kvm_cache_regs.h
+++ b/arch/x86/kvm/kvm_cache_regs.h
@@ -84,11 +84,6 @@ static inline u64 kvm_read_edx_eax(struct kvm_vcpu *vcpu)
                | ((u64)(kvm_register_read(vcpu, VCPU_REGS_RDX) & -1u) << 32);
 }
 
-static inline u32 kvm_read_pkru(struct kvm_vcpu *vcpu)
-{
-       return kvm_x86_ops->get_pkru(vcpu);
-}
-
 static inline void enter_guest_mode(struct kvm_vcpu *vcpu)
 {
        vcpu->arch.hflags |= HF_GUEST_MASK;

diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
index 330bf3a811fb07271de382b598be402c871f6496..b0d36a229d2e21f054cec570f194d4a68adad3f2 100644 (file)
--- a/arch/x86/kvm/mmu.h
+++ b/arch/x86/kvm/mmu.h
@@ -182,7 +182,7 @@ static inline u8 permission_fault(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
                * index of the protection domain, so pte_pkey * 2 is
                * is the index of the first bit for the domain.
                */
-               pkru_bits = (kvm_read_pkru(vcpu) >> (pte_pkey * 2)) & 3;
+               pkru_bits = (vcpu->arch.pkru >> (pte_pkey * 2)) & 3;
 
                /* clear present bit, replace PFEC.RSVD with ACC_USER_MASK. */
                offset = (pfec & ~1) +

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index ba9891ac5c568f1798555bfa9dcbc421fff5ae2a..58dbca7f21062f2aac1caaa0efc055439f5cb801 100644 (file)
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1725,11 +1725,6 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
        to_svm(vcpu)->vmcb->save.rflags = rflags;
 }
 
-static u32 svm_get_pkru(struct kvm_vcpu *vcpu)
-{
-       return 0;
-}
-
 static void svm_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg)
 {
        switch (reg) {
@@ -5313,8 +5308,6 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = {
        .get_rflags = svm_get_rflags,
        .set_rflags = svm_set_rflags,
 
-       .get_pkru = svm_get_pkru,
-
        .tlb_flush = svm_flush_tlb,
 
        .run = svm_vcpu_run,

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 270d83da090c24c232af2e9033e33e5513bf3520..2461e1a53f8c60d5bed121e75a06c0c83547d565 100644 (file)
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -636,8 +636,6 @@ struct vcpu_vmx {
 
        u64 current_tsc_ratio;
 
-       bool guest_pkru_valid;
-       u32 guest_pkru;
        u32 host_pkru;
 
        /*
@@ -2368,11 +2366,6 @@ static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
        vmcs_writel(GUEST_RFLAGS, rflags);
 }
 
-static u32 vmx_get_pkru(struct kvm_vcpu *vcpu)
-{
-       return to_vmx(vcpu)->guest_pkru;
-}
-
 static u32 vmx_get_interrupt_shadow(struct kvm_vcpu *vcpu)
 {
        u32 interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO);
@@ -8860,8 +8853,10 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
        if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
                vmx_set_interrupt_shadow(vcpu, 0);
 
-       if (vmx->guest_pkru_valid)
-               __write_pkru(vmx->guest_pkru);
+       if (static_cpu_has(X86_FEATURE_PKU) &&
+           kvm_read_cr4_bits(vcpu, X86_CR4_PKE) &&
+           vcpu->arch.pkru != vmx->host_pkru)
+               __write_pkru(vcpu->arch.pkru);
 
        atomic_switch_perf_msrs(vmx);
        debugctlmsr = get_debugctlmsr();
@@ -9009,13 +9004,11 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
         * back on host, so it is safe to read guest PKRU from current
         * XSAVE.
         */
-       if (boot_cpu_has(X86_FEATURE_OSPKE)) {
-               vmx->guest_pkru = __read_pkru();
-               if (vmx->guest_pkru != vmx->host_pkru) {
-                       vmx->guest_pkru_valid = true;
+       if (static_cpu_has(X86_FEATURE_PKU) &&
+           kvm_read_cr4_bits(vcpu, X86_CR4_PKE)) {
+               vcpu->arch.pkru = __read_pkru();
+               if (vcpu->arch.pkru != vmx->host_pkru)
                        __write_pkru(vmx->host_pkru);
-               } else
-                       vmx->guest_pkru_valid = false;
        }
 
        /*
@@ -11507,8 +11500,6 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
        .get_rflags = vmx_get_rflags,
        .set_rflags = vmx_set_rflags,
 
-       .get_pkru = vmx_get_pkru,
-
        .tlb_flush = vmx_flush_tlb,
 
        .run = vmx_vcpu_run,