Re: [PATCH v8 033/103] KVM: x86/mmu: Track shadow MMIO value/mask on a per-VM basis

From: Huang, Kai
Date: Mon Aug 08 2022 - 06:14:57 EST

Next message: Krzysztof Kozlowski: "[PATCH 1/3] MAINTAINERS: Drop Bartlomiej Zolnierkiewicz"
Previous message: Paul Cercueil: "Re: [PATCH 13/28] mfd: sec: Remove #ifdef guards for PM related functions"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Sun, 2022-08-07 at 15:01 -0700, isaku.yamahata@xxxxxxxxx wrote:
> From: Sean Christopherson <sean.j.christopherson@xxxxxxxxx>
>
> TDX will use a different shadow PTE entry value for MMIO from VMX. Add
> members to kvm_arch and track value for MMIO per-VM instead of global
> variables. By using the per-VM EPT entry value for MMIO, the existing VMX
> logic is kept working. To untangle the logic to initialize
> shadow_mmio_access_mask, introduce a separate setter function.
>
> At the same time, disallow MMIO emulation path for protected guest because
> VMM can't parse instructions in protected guest memory.
>
> Signed-off-by: Sean Christopherson <sean.j.christopherson@xxxxxxxxx>
> Signed-off-by: Isaku Yamahata <isaku.yamahata@xxxxxxxxx>
> ---
> arch/x86/include/asm/kvm_host.h | 4 +++
> arch/x86/kvm/mmu.h | 3 ++-
> arch/x86/kvm/mmu/mmu.c | 9 ++++---
> arch/x86/kvm/mmu/spte.c | 45 +++++++++------------------------
> arch/x86/kvm/mmu/spte.h | 10 +++-----
> arch/x86/kvm/mmu/tdp_mmu.c | 13 +++++++---
> arch/x86/kvm/svm/svm.c | 11 +++++---
> arch/x86/kvm/vmx/vmx.c | 26 +++++++++++++++++++
> arch/x86/kvm/vmx/x86_ops.h | 1 +
> 9 files changed, 70 insertions(+), 52 deletions(-)
>
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index 6787d5214fd8..3c4051d4512b 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -1157,6 +1157,10 @@ struct kvm_arch {
> */
> spinlock_t mmu_unsync_pages_lock;
>
> + bool enable_mmio_caching;
> + u64 shadow_mmio_value;
> + u64 shadow_mmio_mask;

Sean suggested we only need pre-VM mmio_value. mmio_caching and mmio_mask can
still be global:

https://lore.kernel.org/all/YumT+6joTz2M1zZP@xxxxxxxxxx/

> +
> struct list_head assigned_dev_head;
> struct iommu_domain *iommu_domain;
> bool iommu_noncoherent;
> diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
> index df9f79ee07d4..dea9f2ed0177 100644
> --- a/arch/x86/kvm/mmu.h
> +++ b/arch/x86/kvm/mmu.h
> @@ -98,7 +98,8 @@ static inline u8 kvm_get_shadow_phys_bits(void)
> return boot_cpu_data.x86_phys_bits;
> }
>
> -void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 mmio_mask, u64 access_mask);
> +void kvm_mmu_set_mmio_spte_mask(struct kvm *kvm, u64 mmio_value, u64 mmio_mask);
> +void kvm_mmu_set_mmio_access_mask(u64 mmio_access_mask);
> void kvm_mmu_set_me_spte_mask(u64 me_value, u64 me_mask);
> void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only);
>
> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> index 216708a433e7..88fc2218fcc3 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -2418,7 +2418,7 @@ static int mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp,
> return kvm_mmu_prepare_zap_page(kvm, child,
> invalid_list);
> }
> - } else if (is_mmio_spte(pte)) {
> + } else if (is_mmio_spte(kvm, pte)) {
> mmu_spte_clear_no_track(spte);
> }
> return 0;
> @@ -3222,7 +3222,8 @@ static int handle_abnormal_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fau
> * and only if L1's MAXPHYADDR is inaccurate with respect to
> * the hardware's).
> */
> - if (unlikely(!enable_mmio_caching) ||
> + if (unlikely(!vcpu->kvm->arch.enable_mmio_caching &&
> + !kvm_gfn_shared_mask(vcpu->kvm)) ||

Sean also suggested we should treat enable_mmio_caching as enabled for TDX:

https://lore.kernel.org/all/YuP66QVxyeT4wd5H@xxxxxxxxxx/
https://lore.kernel.org/all/YuP3zGmpiALuXfW+@xxxxxxxxxx/

So this change isn't necessary.

Again, I still think splitting this patch into multiple patches would be easier
to review. For instance, support per-VM MMIO value could be one patch which
doesn't have functional change. Allow mmio_value to be 0 when
enable_mmio_caching is true can be a separate patch too. Enabling MMIO caching
for TDX could be another patch, etc.

> unlikely(fault->gfn > kvm_mmu_max_gfn()))
> return RET_PF_EMULATE;
> }
> @@ -4074,7 +4075,7 @@ static int handle_mmio_page_fault(struct kvm_vcpu *vcpu, u64 addr, bool direct)
> if (WARN_ON(reserved))
> return -EINVAL;
>
> - if (is_mmio_spte(spte)) {
> + if (is_mmio_spte(vcpu->kvm, spte)) {
> gfn_t gfn = get_mmio_spte_gfn(spte);
> unsigned int access = get_mmio_spte_access(spte);
>
> @@ -4529,7 +4530,7 @@ static unsigned long get_cr3(struct kvm_vcpu *vcpu)
> static bool sync_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, gfn_t gfn,
> unsigned int access)
> {
> - if (unlikely(is_mmio_spte(*sptep))) {
> + if (unlikely(is_mmio_spte(vcpu->kvm, *sptep))) {
> if (gfn != get_mmio_spte_gfn(*sptep)) {
> mmu_spte_clear_no_track(sptep);
> return true;
> diff --git a/arch/x86/kvm/mmu/spte.c b/arch/x86/kvm/mmu/spte.c
> index 24cba35570ae..3ad16124eeeb 100644
> --- a/arch/x86/kvm/mmu/spte.c
> +++ b/arch/x86/kvm/mmu/spte.c
> @@ -29,8 +29,6 @@ u64 __read_mostly shadow_x_mask; /* mutual exclusive with nx_mask */
> u64 __read_mostly shadow_user_mask;
> u64 __read_mostly shadow_accessed_mask;
> u64 __read_mostly shadow_dirty_mask;
> -u64 __read_mostly shadow_mmio_value;
> -u64 __read_mostly shadow_mmio_mask;
> u64 __read_mostly shadow_mmio_access_mask;
> u64 __read_mostly shadow_present_mask;
> u64 __read_mostly shadow_memtype_mask;
> @@ -60,10 +58,10 @@ u64 make_mmio_spte(struct kvm_vcpu *vcpu, u64 gfn, unsigned int access)
> u64 spte = generation_mmio_spte_mask(gen);
> u64 gpa = gfn << PAGE_SHIFT;
>
> - WARN_ON_ONCE(!shadow_mmio_value);
> + WARN_ON_ONCE(!vcpu->kvm->arch.shadow_mmio_value);
>
> access &= shadow_mmio_access_mask;
> - spte |= shadow_mmio_value | access;
> + spte |= vcpu->kvm->arch.shadow_mmio_value | access;
> spte |= gpa | shadow_nonpresent_or_rsvd_mask;
> spte |= (gpa & shadow_nonpresent_or_rsvd_mask)
> << SHADOW_NONPRESENT_OR_RSVD_MASK_LEN;
> @@ -335,9 +333,8 @@ u64 mark_spte_for_access_track(u64 spte)
> return spte;
> }
>
> -void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 mmio_mask, u64 access_mask)
> +void kvm_mmu_set_mmio_spte_mask(struct kvm *kvm, u64 mmio_value, u64 mmio_mask)
> {
> - BUG_ON((u64)(unsigned)access_mask != access_mask);
> WARN_ON(mmio_value & shadow_nonpresent_or_rsvd_lower_gfn_mask);
>
> if (!enable_mmio_caching)
> @@ -364,12 +361,9 @@ void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 mmio_mask, u64 access_mask)
> WARN_ON(mmio_value && (REMOVED_SPTE & mmio_mask) == mmio_value))
> mmio_value = 0;
>
> - if (!mmio_value)
> - enable_mmio_caching = false;
> -
> - shadow_mmio_value = mmio_value;
> - shadow_mmio_mask = mmio_mask;
> - shadow_mmio_access_mask = access_mask;
> + kvm->arch.enable_mmio_caching = !!mmio_value;
> + kvm->arch.shadow_mmio_value = mmio_value;
> + kvm->arch.shadow_mmio_mask = mmio_mask;
> }
> EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask);
>
> @@ -404,20 +398,12 @@ void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only)
> shadow_acc_track_mask = VMX_EPT_RWX_MASK;
> shadow_host_writable_mask = EPT_SPTE_HOST_WRITABLE;
> shadow_mmu_writable_mask = EPT_SPTE_MMU_WRITABLE;
> -
> - /*
> - * EPT Misconfigurations are generated if the value of bits 2:0
> - * of an EPT paging-structure entry is 110b (write/execute).
> - */
> - kvm_mmu_set_mmio_spte_mask(VMX_EPT_MISCONFIG_WX_VALUE,
> - VMX_EPT_RWX_MASK, 0);
> }
> EXPORT_SYMBOL_GPL(kvm_mmu_set_ept_masks);
>
> void kvm_mmu_reset_all_pte_masks(void)
> {
> u8 low_phys_bits;
> - u64 mask;
>
> shadow_phys_bits = kvm_get_shadow_phys_bits();
>
> @@ -464,18 +450,11 @@ void kvm_mmu_reset_all_pte_masks(void)
>
> shadow_host_writable_mask = DEFAULT_SPTE_HOST_WRITABLE;
> shadow_mmu_writable_mask = DEFAULT_SPTE_MMU_WRITABLE;
> +}
>
> - /*
> - * Set a reserved PA bit in MMIO SPTEs to generate page faults with
> - * PFEC.RSVD=1 on MMIO accesses. 64-bit PTEs (PAE, x86-64, and EPT
> - * paging) support a maximum of 52 bits of PA, i.e. if the CPU supports
> - * 52-bit physical addresses then there are no reserved PA bits in the
> - * PTEs and so the reserved PA approach must be disabled.
> - */
> - if (shadow_phys_bits < 52)
> - mask = BIT_ULL(51) | PT_PRESENT_MASK;
> - else
> - mask = 0;
> -
> - kvm_mmu_set_mmio_spte_mask(mask, mask, ACC_WRITE_MASK | ACC_USER_MASK);

Sorry perhaps Sean will not like this code change. In Sean's MMIO caching bug
fixing series I suggested to remove setting MMIO mask/value from
kvm_mmu_reset_all_pte_masks() but enforce vendor specific code to always call it
depending on what features vendor is going to use. But Sean didn't like it:

https://lore.kernel.org/all/YuP3zGmpiALuXfW+@xxxxxxxxxx/

Will leave this to Sean.

> +void kvm_mmu_set_mmio_access_mask(u64 mmio_access_mask)
> +{
> + BUG_ON((u64)(unsigned)mmio_access_mask != mmio_access_mask);
> + shadow_mmio_access_mask = mmio_access_mask;
> }
> +EXPORT_SYMBOL(kvm_mmu_set_mmio_access_mask);
> diff --git a/arch/x86/kvm/mmu/spte.h b/arch/x86/kvm/mmu/spte.h
> index 30f456e59e58..824ab5490d5c 100644
> --- a/arch/x86/kvm/mmu/spte.h
> +++ b/arch/x86/kvm/mmu/spte.h
> @@ -5,8 +5,6 @@
>
> #include "mmu_internal.h"
>
> -extern bool __read_mostly enable_mmio_caching;
> -
> /*
> * A MMU present SPTE is backed by actual memory and may or may not be present
> * in hardware. E.g. MMIO SPTEs are not considered present. Use bit 11, as it
> @@ -156,8 +154,6 @@ extern u64 __read_mostly shadow_x_mask; /* mutual exclusive with nx_mask */
> extern u64 __read_mostly shadow_user_mask;
> extern u64 __read_mostly shadow_accessed_mask;
> extern u64 __read_mostly shadow_dirty_mask;
> -extern u64 __read_mostly shadow_mmio_value;
> -extern u64 __read_mostly shadow_mmio_mask;
> extern u64 __read_mostly shadow_mmio_access_mask;
> extern u64 __read_mostly shadow_present_mask;
> extern u64 __read_mostly shadow_memtype_mask;
> @@ -231,10 +227,10 @@ static inline int spte_index(u64 *sptep)
> */
> extern u64 __read_mostly shadow_nonpresent_or_rsvd_lower_gfn_mask;
>
> -static inline bool is_mmio_spte(u64 spte)
> +static inline bool is_mmio_spte(struct kvm *kvm, u64 spte)
> {
> - return (spte & shadow_mmio_mask) == shadow_mmio_value &&
> - likely(enable_mmio_caching);
> + return (spte & kvm->arch.shadow_mmio_mask) == kvm->arch.shadow_mmio_value &&
> + likely(kvm->arch.enable_mmio_caching || kvm_gfn_shared_mask(kvm));
> }
>
> static inline bool is_shadow_present_pte(u64 pte)
> diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
> index af510dd31ebc..8bc3a8d1803e 100644
> --- a/arch/x86/kvm/mmu/tdp_mmu.c
> +++ b/arch/x86/kvm/mmu/tdp_mmu.c
> @@ -569,8 +569,8 @@ static void __handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn,
> * impact the guest since both the former and current SPTEs
> * are nonpresent.
> */
> - if (WARN_ON(!is_mmio_spte(old_spte) &&
> - !is_mmio_spte(new_spte) &&
> + if (WARN_ON(!is_mmio_spte(kvm, old_spte) &&
> + !is_mmio_spte(kvm, new_spte) &&
> !is_removed_spte(new_spte)))
> pr_err("Unexpected SPTE change! Nonpresent SPTEs\n"
> "should not be replaced with another,\n"
> @@ -1094,7 +1094,7 @@ static int tdp_mmu_map_handle_target_level(struct kvm_vcpu *vcpu,
> }
>
> /* If a MMIO SPTE is installed, the MMIO will need to be emulated. */
> - if (unlikely(is_mmio_spte(new_spte))) {
> + if (unlikely(is_mmio_spte(vcpu->kvm, new_spte))) {
> vcpu->stat.pf_mmio_spte_created++;
> trace_mark_mmio_spte(rcu_dereference(iter->sptep), iter->gfn,
> new_spte);
> @@ -1863,6 +1863,13 @@ int kvm_tdp_mmu_get_walk(struct kvm_vcpu *vcpu, u64 addr, u64 *sptes,
>
> *root_level = vcpu->arch.mmu->root_role.level;
>
> + /*
> + * mmio page fault isn't supported for protected guest because
> + * instructions in protected guest memory can't be parsed by VMM.
> + */
> + if (WARN_ON(kvm_gfn_shared_mask(vcpu->kvm)))
> + return leaf;
> +

Why not putting to handle_mmio_page_fault() directly?

> tdp_mmu_for_each_pte(iter, mmu, gfn, gfn + 1) {
> leaf = iter.level;
> sptes[leaf] = iter.old_spte;
> diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
> index 8aa3c95e8b6e..07829be93c93 100644
> --- a/arch/x86/kvm/svm/svm.c
> +++ b/arch/x86/kvm/svm/svm.c
> @@ -229,6 +229,7 @@ module_param(dump_invalid_vmcb, bool, 0644);
> bool intercept_smi = true;
> module_param(intercept_smi, bool, 0444);
>
> +static u64 __read_mostly svm_shadow_mmio_mask;
>
> static bool svm_gp_erratum_intercept = true;
>
> @@ -4729,6 +4730,9 @@ static bool svm_is_vm_type_supported(unsigned long type)
>
> static int svm_vm_init(struct kvm *kvm)
> {
> + kvm_mmu_set_mmio_spte_mask(kvm, svm_shadow_mmio_mask,
> + svm_shadow_mmio_mask);
> +
> if (!pause_filter_count || !pause_filter_thresh)
> kvm->arch.pause_in_guest = true;
>
> @@ -4878,7 +4882,7 @@ static struct kvm_x86_ops svm_x86_ops __initdata = {
> static __init void svm_adjust_mmio_mask(void)
> {
> unsigned int enc_bit, mask_bit;
> - u64 msr, mask;
> + u64 msr;
>
> /* If there is no memory encryption support, use existing mask */
> if (cpuid_eax(0x80000000) < 0x8000001f)
> @@ -4905,9 +4909,8 @@ static __init void svm_adjust_mmio_mask(void)
> *
> * If the mask bit location is 52 (or above), then clear the mask.
> */
> - mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0;
> -
> - kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK);
> + svm_shadow_mmio_mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0;
> + kvm_mmu_set_mmio_access_mask(PT_WRITABLE_MASK | PT_USER_MASK);

svm_shadow_mmio_mask is left as '0' if AMD memory encryption isn't supported:

Also see here:

https://lore.kernel.org/all/YuP3zGmpiALuXfW+@xxxxxxxxxx/

> }
>
> static __init void svm_set_cpu_caps(void)
> diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> index 0bce352f81b8..ec2bd4df0684 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -141,6 +141,8 @@ module_param_named(preemption_timer, enable_preemption_timer, bool, S_IRUGO);
> extern bool __read_mostly allow_smaller_maxphyaddr;
> module_param(allow_smaller_maxphyaddr, bool, S_IRUGO);
>
> +u64 __ro_after_init vmx_shadow_mmio_mask;
> +
> #define KVM_VM_CR0_ALWAYS_OFF (X86_CR0_NW | X86_CR0_CD)
> #define KVM_VM_CR0_ALWAYS_ON_UNRESTRICTED_GUEST X86_CR0_NE
> #define KVM_VM_CR0_ALWAYS_ON \
> @@ -7359,6 +7361,17 @@ int vmx_vm_init(struct kvm *kvm)
> if (!ple_gap)
> kvm->arch.pause_in_guest = true;
>
> + /*
> + * EPT Misconfigurations can be generated if the value of bits 2:0
> + * of an EPT paging-structure entry is 110b (write/execute).
> + */
> + if (enable_ept)
> + kvm_mmu_set_mmio_spte_mask(kvm, VMX_EPT_MISCONFIG_WX_VALUE,
> + VMX_EPT_RWX_MASK);
> + else
> + kvm_mmu_set_mmio_spte_mask(kvm, vmx_shadow_mmio_mask,
> + vmx_shadow_mmio_mask);
> +
> if (boot_cpu_has(X86_BUG_L1TF) && enable_ept) {
> switch (l1tf_mitigation) {
> case L1TF_MITIGATION_OFF:
> @@ -8357,6 +8370,19 @@ int __init vmx_init(void)
> if (!enable_ept)
> allow_smaller_maxphyaddr = true;
>
> + /*
> + * Set a reserved PA bit in MMIO SPTEs to generate page faults with
> + * PFEC.RSVD=1 on MMIO accesses. 64-bit PTEs (PAE, x86-64, and EPT
> + * paging) support a maximum of 52 bits of PA, i.e. if the CPU supports
> + * 52-bit physical addresses then there are no reserved PA bits in the
> + * PTEs and so the reserved PA approach must be disabled.
> + */
> + if (kvm_get_shadow_phys_bits() < 52)
> + vmx_shadow_mmio_mask = BIT_ULL(51) | PT_PRESENT_MASK;
> + else
> + vmx_shadow_mmio_mask = 0;
> + kvm_mmu_set_mmio_access_mask(0);
> +
> return 0;
> }
>
> diff --git a/arch/x86/kvm/vmx/x86_ops.h b/arch/x86/kvm/vmx/x86_ops.h
> index b4ffa1590d41..62f1d1cdd44b 100644
> --- a/arch/x86/kvm/vmx/x86_ops.h
> +++ b/arch/x86/kvm/vmx/x86_ops.h
> @@ -13,6 +13,7 @@ void hv_vp_assist_page_exit(void);
> void __init vmx_init_early(void);
> int __init vmx_init(void);
> void vmx_exit(void);
> +extern u64 __ro_after_init vmx_shadow_mmio_mask;
>
> __init int vmx_cpu_has_kvm_support(void);
> __init int vmx_disabled_by_bios(void);

Next message: Krzysztof Kozlowski: "[PATCH 1/3] MAINTAINERS: Drop Bartlomiej Zolnierkiewicz"
Previous message: Paul Cercueil: "Re: [PATCH 13/28] mfd: sec: Remove #ifdef guards for PM related functions"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]