From: Kai Huang <kai.huang@intel.com>
To: isaku.yamahata@intel.com, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: isaku.yamahata@gmail.com, Paolo Bonzini <pbonzini@redhat.com>,
Sean Christopherson <sean.j.christopherson@intel.com>
Subject: Re: [PATCH v7 037/102] KVM: x86/mmu: Track shadow MMIO value/mask on a per-VM basis
Date: Wed, 06 Jul 2022 02:06:27 +1200 [thread overview]
Message-ID: <4c59eddd8f1d5029be8eeac84fbb75131b984568.camel@intel.com> (raw)
In-Reply-To: <242df8a7164b593d3702b9ba94889acd11f43cbb.1656366338.git.isaku.yamahata@intel.com>
On Mon, 2022-06-27 at 14:53 -0700, isaku.yamahata@intel.com wrote:
> From: Sean Christopherson <sean.j.christopherson@intel.com>
>
> 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.
>
> In the case of VMX VM case, the EPT entry for MMIO is non-present PTE
> (present bit cleared) without backing guest physical address (on EPT
> violation, KVM searches backing guest memory and it finds there is no
> backing guest page.) or the value to trigger EPT misconfiguration. Once
> MMIO is triggered on the EPT entry, the EPT entry is updated to trigger EPT
> misconfiguration for the future MMIO on the same GPA. It allows KVM to
> understand the memory access is for MMIO without searching backing guest
> pages.). And then KVM parses guest instruction to figure out
> address/value/width for MMIO.
>
> In the case of the guest TD, the guest memory is protected so that VMM
> can't parse guest instruction to understand the value and access width for
> MMIO. Instead, VMM sets up (Shared) EPT to trigger #VE by clearing
> the VE-suppress bit. When the guest TD issues MMIO, #VE is injected. Guest VE
> handler converts MMIO access into MMIO hypercall to pass
> address/value/width for MMIO to VMM. (or directly paravirtualize MMIO into
> hypercall.) Then VMM can handle the MMIO hypercall without parsing guest
> instructions.
To me only first paragraph is needed. It already describes _why_ we need this
patch and _how_ you are going to implement.
The last two paragraphs only elaborate the _why_ in the first paragraph, but
they does not say this patch will do more. And they have been explained in
previous patches so looks they are not mandatory here.
>
> Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
> ---
> arch/x86/include/asm/kvm_host.h | 4 ++++
> arch/x86/include/asm/vmx.h | 1 +
> arch/x86/kvm/mmu.h | 4 +++-
> arch/x86/kvm/mmu/mmu.c | 20 ++++++++++++----
> arch/x86/kvm/mmu/paging_tmpl.h | 2 +-
> arch/x86/kvm/mmu/spte.c | 41 +++++++++++++++------------------
> arch/x86/kvm/mmu/spte.h | 11 ++++-----
> arch/x86/kvm/mmu/tdp_mmu.c | 6 ++---
> arch/x86/kvm/svm/svm.c | 2 +-
> arch/x86/kvm/vmx/vmx.c | 8 +++++++
> 10 files changed, 59 insertions(+), 40 deletions(-)
>
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index 2c47aab72a1b..39215daa8576 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -1161,6 +1161,10 @@ struct kvm_arch {
> */
> spinlock_t mmu_unsync_pages_lock;
>
> + bool enable_mmio_caching;
> + u64 shadow_mmio_value;
> + u64 shadow_mmio_mask;
> +
> struct list_head assigned_dev_head;
> struct iommu_domain *iommu_domain;
> bool iommu_noncoherent;
> diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h
> index c371ef695fcc..6231ef005a50 100644
> --- a/arch/x86/include/asm/vmx.h
> +++ b/arch/x86/include/asm/vmx.h
> @@ -511,6 +511,7 @@ enum vmcs_field {
> #define VMX_EPT_IPAT_BIT (1ull << 6)
> #define VMX_EPT_ACCESS_BIT (1ull << 8)
> #define VMX_EPT_DIRTY_BIT (1ull << 9)
> +#define VMX_EPT_SUPPRESS_VE_BIT (1ull << 63)
Both the patch title and the changelog say this patch only does per-VM MMIO
value/mask tracking. Why do we need this bit here?
> #define VMX_EPT_RWX_MASK (VMX_EPT_READABLE_MASK | \
> VMX_EPT_WRITABLE_MASK | \
> VMX_EPT_EXECUTABLE_MASK)
> diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
> index ccf0ba7a6387..9ba60fd79d33 100644
> --- a/arch/x86/kvm/mmu.h
> +++ b/arch/x86/kvm/mmu.h
> @@ -108,7 +108,9 @@ 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,
> + u64 access_mask);
> +void kvm_mmu_set_default_mmio_spte_mask(u64 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 f239b6cb5d53..496d0d30839b 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -2287,7 +2287,7 @@ 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;
> @@ -3067,8 +3067,13 @@ static int handle_abnormal_pfn(struct kvm_vcpu *vcpu, struct kvm_page_fault *fau
> * by L0 userspace (you can observe gfn > L1.MAXPHYADDR if
> * and only if L1's MAXPHYADDR is inaccurate with respect to
> * the hardware's).
> + *
> + * Excludes the INTEL TD guest. Because TD memory is
> + * protected, the instruction can't be emulated. Instead, use
> + * SPTE value without #VE suppress bit cleared
> + * (kvm->arch.shadow_mmio_value = 0).
> */
Again, I don't think this chunk should be in this patch. It's out-of-scope of
what the patch claims to do.
I see you will make below code change in later patch (couple of patches later):
- if (unlikely(!vcpu->kvm->arch.enable_mmio_caching) ||
+ if (unlikely(!vcpu->kvm->arch.enable_mmio_caching &&
+ !kvm_gfn_shared_mask(vcpu->kvm)) ||
unlikely(fault->gfn > kvm_mmu_max_gfn()))
return RET_PF_EMULATE;
So why not putting the comment and the code change together?
> - if (unlikely(!enable_mmio_caching) ||
> + if (unlikely(!vcpu->kvm->arch.enable_mmio_caching) ||
> unlikely(fault->gfn > kvm_mmu_max_gfn()))
> return RET_PF_EMULATE;
> }
> @@ -3200,7 +3205,8 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
> else
> sptep = fast_pf_get_last_sptep(vcpu, fault->addr, &spte);
>
> - if (!is_shadow_present_pte(spte) || is_mmio_spte(spte))
> + if (!is_shadow_present_pte(spte) ||
> + is_mmio_spte(vcpu->kvm, spte))
> break;
>
> sp = sptep_to_sp(sptep);
> @@ -3907,7 +3913,7 @@ 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);
>
> @@ -4350,7 +4356,7 @@ 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;
> @@ -5864,6 +5870,10 @@ int kvm_mmu_init_vm(struct kvm *kvm)
> node->track_write = kvm_mmu_pte_write;
> node->track_flush_slot = kvm_mmu_invalidate_zap_pages_in_memslot;
> kvm_page_track_register_notifier(kvm, node);
> + kvm_mmu_set_mmio_spte_mask(kvm, shadow_default_mmio_mask,
> + shadow_default_mmio_mask,
> + ACC_WRITE_MASK | ACC_USER_MASK);
> +
This (along with shadow_default_mmio_mask) looks a little bit weird. Please
also see comments below.
> return 0;
> }
>
> diff --git a/arch/x86/kvm/mmu/paging_tmpl.h b/arch/x86/kvm/mmu/paging_tmpl.h
> index ee2fb0c073f3..62ae590d4e5b 100644
> --- a/arch/x86/kvm/mmu/paging_tmpl.h
> +++ b/arch/x86/kvm/mmu/paging_tmpl.h
> @@ -1032,7 +1032,7 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp)
> gfn_t gfn;
>
> if (!is_shadow_present_pte(sp->spt[i]) &&
> - !is_mmio_spte(sp->spt[i]))
> + !is_mmio_spte(vcpu->kvm, sp->spt[i]))
> continue;
>
> pte_gpa = first_pte_gpa + i * sizeof(pt_element_t);
> diff --git a/arch/x86/kvm/mmu/spte.c b/arch/x86/kvm/mmu/spte.c
> index bd441458153f..5194aef60c1f 100644
> --- a/arch/x86/kvm/mmu/spte.c
> +++ b/arch/x86/kvm/mmu/spte.c
> @@ -29,8 +29,7 @@ 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_default_mmio_mask;
This shadow_default_mmio_mask looks a little bit weird. Please also see below.
> u64 __read_mostly shadow_mmio_access_mask;
> u64 __read_mostly shadow_present_mask;
> u64 __read_mostly shadow_me_value;
> @@ -62,10 +61,11 @@ 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 &&
> + !kvm_gfn_shared_mask(vcpu->kvm));
Chunk shouldn't belong to this patch.
>
> 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;
> @@ -337,7 +337,8 @@ 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,
> + u64 access_mask)
> {
> BUG_ON((u64)(unsigned)access_mask != access_mask);
> WARN_ON(mmio_value & shadow_nonpresent_or_rsvd_lower_gfn_mask);
> @@ -366,11 +367,9 @@ 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;
> + kvm->arch.enable_mmio_caching = !!mmio_value;
> + kvm->arch.shadow_mmio_value = mmio_value;
> + kvm->arch.shadow_mmio_mask = mmio_mask;
> shadow_mmio_access_mask = access_mask;
> }
> EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask);
> @@ -393,24 +392,18 @@ void kvm_mmu_set_ept_masks(bool has_ad_bits, bool has_exec_only)
> shadow_dirty_mask = has_ad_bits ? VMX_EPT_DIRTY_BIT : 0ull;
> shadow_nx_mask = 0ull;
> shadow_x_mask = VMX_EPT_EXECUTABLE_MASK;
> - shadow_present_mask = has_exec_only ? 0ull : VMX_EPT_READABLE_MASK;
> + /* VMX_EPT_SUPPRESS_VE_BIT is needed for W or X violation. */
> + shadow_present_mask =
> + (has_exec_only ? 0ull : VMX_EPT_READABLE_MASK) | VMX_EPT_SUPPRESS_VE_BIT;
Again, this chunk shouldn't be in this patch.
> 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();
>
> @@ -459,9 +452,13 @@ void kvm_mmu_reset_all_pte_masks(void)
> * PTEs and so the reserved PA approach must be disabled.
> */
> if (shadow_phys_bits < 52)
> - mask = BIT_ULL(51) | PT_PRESENT_MASK;
> + shadow_default_mmio_mask = BIT_ULL(51) | PT_PRESENT_MASK;
> else
> - mask = 0;
> + shadow_default_mmio_mask = 0;
> +}
Shadow_default_mmio_mask alone looks a little bit weird with per-VM MMIO
tracking. I think it can be removed by moving this code to vmx_vm_init(), and
call it as VM's MMIO mask/value for non-EPT case. If EPT is enabled, it can
override using new mask/value.
>
> - kvm_mmu_set_mmio_spte_mask(mask, mask, ACC_WRITE_MASK | ACC_USER_MASK);
> +void kvm_mmu_set_default_mmio_spte_mask(u64 mask)
> +{
> + shadow_default_mmio_mask = mask;
> }
> +EXPORT_SYMBOL_GPL(kvm_mmu_set_default_mmio_spte_mask);
> diff --git a/arch/x86/kvm/mmu/spte.h b/arch/x86/kvm/mmu/spte.h
> index 1bfedbe0585f..96312ab4fffb 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
> @@ -160,8 +158,7 @@ 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_default_mmio_mask;
> extern u64 __read_mostly shadow_mmio_access_mask;
> extern u64 __read_mostly shadow_present_mask;
> extern u64 __read_mostly shadow_me_value;
> @@ -233,10 +230,10 @@ static inline bool is_removed_spte(u64 spte)
> */
> 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));
> }
This chunk (checking kvm_gfn_shared_mask(kvm)) should not be in this patch.
>
> 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 2ca03ec3bf52..82f1bfac7ee6 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"
> @@ -1108,7 +1108,7 @@ 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);
> diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
> index 815a07c594f1..0abc43d6a115 100644
> --- a/arch/x86/kvm/svm/svm.c
> +++ b/arch/x86/kvm/svm/svm.c
> @@ -4870,7 +4870,7 @@ static __init void svm_adjust_mmio_mask(void)
> */
> 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);
> + kvm_mmu_set_default_mmio_spte_mask(mask);
SVM doesn't need shadow_default_mmio_mask. Instead, it can define a local
variable in svm.c, and call kvm_mmu_set_mmio_spte_mask(mask, mask,
PT_WRITABLE_MASK | PT_USER_MASK) in svm_vm_init().
> }
>
> 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 1d87885245cc..e2415ac55317 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -7289,6 +7289,14 @@ 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, 0);
> +
As commented above, I think we can remove shadow_default_mmio_mask by moving the
logic in kvm_mmu_reset_all_pte_mask() here.
Or use SVM similar way, use a local variable 'mask' in vmx.c, calculate the
'mask' during hardware_setup(), and use it here for non-EPT case.
> if (boot_cpu_has(X86_BUG_L1TF) && enable_ept) {
> switch (l1tf_mitigation) {
> case L1TF_MITIGATION_OFF:
next prev parent reply other threads:[~2022-07-05 14:14 UTC|newest]
Thread overview: 219+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-27 21:52 [PATCH v7 000/102] KVM TDX basic feature support isaku.yamahata
2022-06-27 21:52 ` [PATCH v7 001/102] KVM: x86: Move check_processor_compatibility from init ops to runtime ops isaku.yamahata
2022-06-27 21:52 ` [PATCH v7 002/102] Partially revert "KVM: Pass kvm_init()'s opaque param to additional arch funcs" isaku.yamahata
2022-07-13 1:55 ` Kai Huang
2022-07-26 23:57 ` Isaku Yamahata
2022-06-27 21:52 ` [PATCH v7 003/102] KVM: Refactor CPU compatibility check on module initialiization isaku.yamahata
2022-07-12 1:15 ` Kai Huang
2022-07-13 3:16 ` Kai Huang
2022-07-13 3:11 ` Kai Huang
2022-07-27 22:04 ` Isaku Yamahata
2022-06-27 21:52 ` [PATCH v7 004/102] KVM: VMX: Move out vmx_x86_ops to 'main.c' to wrap VMX and TDX isaku.yamahata
2022-06-27 21:52 ` [PATCH v7 005/102] x86/virt/vmx/tdx: export platform_tdx_enabled() isaku.yamahata
2022-06-27 21:52 ` [PATCH v7 006/102] KVM: TDX: Detect CPU feature on kernel module initialization isaku.yamahata
2022-06-28 3:43 ` Kai Huang
2022-07-11 23:48 ` Isaku Yamahata
2022-07-12 0:45 ` Kai Huang
2022-06-27 21:52 ` [PATCH v7 007/102] KVM: Enable hardware before doing arch VM initialization isaku.yamahata
2022-06-28 2:59 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 008/102] KVM: x86: Refactor KVM VMX module init/exit functions isaku.yamahata
2022-06-28 3:53 ` Kai Huang
2022-07-12 0:38 ` Isaku Yamahata
2022-07-12 1:30 ` Kai Huang
2022-07-27 0:44 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 009/102] KVM: TDX: Add placeholders for TDX VM/vcpu structure isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 010/102] x86/virt/tdx: Add a helper function to return system wide info about TDX module isaku.yamahata
2022-07-07 2:46 ` Yuan Yao
2022-07-12 0:39 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 011/102] KVM: TDX: Initialize TDX module when loading kvm_intel.ko isaku.yamahata
2022-06-28 4:31 ` Kai Huang
2022-07-12 0:46 ` Isaku Yamahata
2022-07-12 1:13 ` Kai Huang
2022-07-27 0:39 ` Isaku Yamahata
2022-07-27 4:38 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 012/102] KVM: x86: Introduce vm_type to differentiate default VMs from confidential VMs isaku.yamahata
2022-06-28 2:52 ` Kai Huang
2022-07-04 6:44 ` Kai Huang
2022-07-12 1:01 ` Isaku Yamahata
2022-07-12 1:24 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 013/102] KVM: TDX: Make TDX VM type supported isaku.yamahata
2022-07-07 2:55 ` Yuan Yao
2022-07-12 1:06 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 014/102] [MARKER] The start of TDX KVM patch series: TDX architectural definitions isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 015/102] KVM: TDX: Define " isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 016/102] KVM: TDX: Add TDX "architectural" error codes isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 017/102] KVM: TDX: Add C wrapper functions for SEAMCALLs to the TDX module isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 018/102] KVM: TDX: Add helper functions to print TDX SEAMCALL error isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 019/102] [MARKER] The start of TDX KVM patch series: TD VM creation/destruction isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 020/102] KVM: TDX: Stub in tdx.h with structs, accessors, and VMCS helpers isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 021/102] x86/cpu: Add helper functions to allocate/free TDX private host key id isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 022/102] KVM: TDX: create/destroy VM structure isaku.yamahata
2022-07-07 6:16 ` Yuan Yao
2022-07-12 6:21 ` Isaku Yamahata
2022-08-02 19:46 ` Sean Christopherson
2022-08-11 18:29 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 023/102] KVM: TDX: x86: Add ioctl to get TDX systemwide parameters isaku.yamahata
2022-07-07 6:48 ` Yuan Yao
2022-06-27 21:53 ` [PATCH v7 024/102] KVM: TDX: Add place holder for TDX VM specific mem_enc_op ioctl isaku.yamahata
2022-07-07 7:12 ` Yuan Yao
2022-06-27 21:53 ` [PATCH v7 025/102] KVM: TDX: initialize VM with TDX specific parameters isaku.yamahata
2022-06-28 8:30 ` Xiaoyao Li
2022-07-12 7:11 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 026/102] KVM: TDX: Make pmu_intel.c ignore guest TD case isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 027/102] [MARKER] The start of TDX KVM patch series: TD vcpu creation/destruction isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 028/102] KVM: TDX: allocate/free TDX vcpu structure isaku.yamahata
2022-08-02 19:56 ` Sean Christopherson
2022-06-27 21:53 ` [PATCH v7 029/102] " isaku.yamahata
2022-06-28 11:34 ` Kai Huang
2022-07-12 7:55 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 030/102] KVM: TDX: Do TDX specific vcpu initialization isaku.yamahata
2022-07-08 2:14 ` Yuan Yao
2022-07-12 20:35 ` Isaku Yamahata
2022-07-13 0:22 ` Xiaoyao Li
2022-06-27 21:53 ` [PATCH v7 031/102] [MARKER] The start of TDX KVM patch series: KVM MMU GPA shared bits isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 032/102] KVM: x86/mmu: introduce config for PRIVATE KVM MMU isaku.yamahata
2022-07-08 1:53 ` Kai Huang
2022-07-13 1:25 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 033/102] KVM: x86/mmu: Add address conversion functions for TDX shared bits isaku.yamahata
2022-07-08 2:15 ` Kai Huang
2022-07-13 4:52 ` Isaku Yamahata
2022-07-13 10:41 ` Kai Huang
2022-07-14 0:14 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 034/102] [MARKER] The start of TDX KVM patch series: KVM TDP refactoring for TDX isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 035/102] KVM: x86/mmu: Explicitly check for MMIO spte in fast page fault isaku.yamahata
2022-06-30 11:37 ` Kai Huang
2022-07-13 8:35 ` Isaku Yamahata
2022-07-13 10:29 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 036/102] KVM: x86/mmu: Allow non-zero value for non-present SPTE isaku.yamahata
2022-06-30 11:03 ` Kai Huang
2022-07-14 18:05 ` Isaku Yamahata
2022-07-08 5:18 ` Yuan Yao
2022-07-08 15:30 ` Sean Christopherson
2022-07-11 7:05 ` Yuan Yao
2022-07-11 14:47 ` Sean Christopherson
2022-07-14 18:41 ` Isaku Yamahata
2022-07-20 2:44 ` Kai Huang
2022-07-20 3:12 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 037/102] KVM: x86/mmu: Track shadow MMIO value/mask on a per-VM basis isaku.yamahata
2022-06-30 11:45 ` Kai Huang
2022-07-05 14:06 ` Kai Huang [this message]
2022-07-19 8:47 ` Isaku Yamahata
2022-07-20 3:45 ` Kai Huang
2022-07-27 23:20 ` Isaku Yamahata
2022-07-28 0:48 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 038/102] KVM: x86/mmu: Disallow fast page fault on private GPA isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 039/102] KVM: x86/mmu: Allow per-VM override of the TDP max page level isaku.yamahata
2022-06-30 12:27 ` Kai Huang
2022-07-19 10:26 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 040/102] KVM: x86/mmu: Zap only leaf SPTEs for deleted/moved memslot for private mmu isaku.yamahata
2022-07-01 10:41 ` Kai Huang
2022-07-19 11:06 ` Isaku Yamahata
2022-07-19 23:17 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 041/102] KVM: VMX: Introduce test mode related to EPT violation VE isaku.yamahata
2022-07-08 2:23 ` Kai Huang
2022-07-19 14:49 ` Isaku Yamahata
2022-07-20 5:13 ` Kai Huang
2022-07-27 23:39 ` Isaku Yamahata
2022-07-28 0:54 ` Kai Huang
2022-07-28 20:11 ` Sean Christopherson
2022-08-09 0:48 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 042/102] [MARKER] The start of TDX KVM patch series: KVM TDP MMU hooks isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 043/102] KVM: x86/mmu: Focibly use TDP MMU for TDX isaku.yamahata
2022-07-11 5:48 ` Yuan Yao
2022-07-11 14:56 ` Sean Christopherson
2022-07-19 15:04 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 044/102] KVM: x86/mmu: Add a private pointer to struct kvm_mmu_page isaku.yamahata
2022-07-01 11:12 ` Kai Huang
2022-07-19 15:35 ` Isaku Yamahata
2022-07-11 6:28 ` Yuan Yao
2022-07-28 19:41 ` David Matlack
2022-08-09 23:52 ` Isaku Yamahata
2022-07-28 20:13 ` David Matlack
2022-08-09 23:50 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 045/102] KVM: x86/tdp_mmu: refactor kvm_tdp_mmu_map() isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 046/102] KVM: x86/tdp_mmu: Support TDX private mapping for TDP MMU isaku.yamahata
2022-07-08 3:44 ` Kai Huang
2022-07-26 23:39 ` Isaku Yamahata
2022-07-11 8:28 ` Yuan Yao
2022-07-26 23:41 ` Isaku Yamahata
2022-07-12 2:36 ` Yuan Yao
2022-07-26 23:42 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 047/102] [MARKER] The start of TDX KVM patch series: TDX EPT violation isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 048/102] KVM: x86/mmu: Disallow dirty logging for x86 TDX isaku.yamahata
2022-07-08 2:30 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 049/102] KVM: x86/tdp_mmu: Ignore unsupported mmu operation on private GFNs isaku.yamahata
2022-07-12 2:58 ` Yuan Yao
2022-07-19 18:03 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 050/102] KVM: VMX: Split out guts of EPT violation to common/exposed function isaku.yamahata
2022-07-08 10:25 ` Kai Huang
2022-06-27 21:53 ` [PATCH v7 051/102] KVM: VMX: Move setting of EPT MMU masks to common VT-x code isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 052/102] KVM: TDX: Add load_mmu_pgd method for TDX isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 053/102] KVM: TDX: don't request KVM_REQ_APIC_PAGE_RELOAD isaku.yamahata
2022-07-12 3:47 ` Yuan Yao
2022-07-12 6:14 ` Chao Gao
2022-07-19 18:12 ` Isaku Yamahata
2022-06-27 21:53 ` [PATCH v7 054/102] KVM: TDX: TDP MMU TDX support isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 055/102] [MARKER] The start of TDX KVM patch series: KVM TDP MMU MapGPA isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 056/102] KVM: x86/mmu: steal software usable git to record if GFN is for shared or not isaku.yamahata
2022-07-18 8:37 ` Yuan Yao
2022-06-27 21:53 ` [PATCH v7 057/102] KVM: x86/tdp_mmu: implement MapGPA hypercall for TDX isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 058/102] KVM: x86/mmu: Introduce kvm_mmu_map_tdp_page() for use by TDX isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 059/102] [MARKER] The start of TDX KVM patch series: TD finalization isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 060/102] KVM: TDX: Create initial guest memory isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 061/102] KVM: TDX: Finalize VM initialization isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 062/102] [MARKER] The start of TDX KVM patch series: TD vcpu enter/exit isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 063/102] KVM: TDX: Add helper assembly function to TDX vcpu isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 064/102] KVM: TDX: Implement TDX vcpu enter/exit path isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 065/102] KVM: TDX: vcpu_run: save/restore host state(host kernel gs) isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 066/102] KVM: TDX: restore host xsave state when exit from the guest TD isaku.yamahata
2022-06-27 21:53 ` [PATCH v7 067/102] KVM: x86: Allow to update cached values in kvm_user_return_msrs w/o wrmsr isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 068/102] KVM: TDX: restore user ret MSRs isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 069/102] [MARKER] The start of TDX KVM patch series: TD vcpu exits/interrupts/hypercalls isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 070/102] KVM: TDX: complete interrupts after tdexit isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 071/102] KVM: TDX: restore debug store when TD exit isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 072/102] KVM: TDX: handle vcpu migration over logical processor isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 073/102] KVM: x86: Add a switch_db_regs flag to handle TDX's auto-switched behavior isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 074/102] KVM: TDX: Add support for find pending IRQ in a protected local APIC isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 075/102] KVM: x86: Assume timer IRQ was injected if APIC state is proteced isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 076/102] KVM: TDX: remove use of struct vcpu_vmx from posted_interrupt.c isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 077/102] KVM: TDX: Implement interrupt injection isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 078/102] KVM: TDX: Implements vcpu request_immediate_exit isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 079/102] KVM: TDX: Implement methods to inject NMI isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 080/102] KVM: VMX: Modify NMI and INTR handlers to take intr_info as function argument isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 081/102] KVM: VMX: Move NMI/exception handler to common helper isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 082/102] KVM: x86: Split core of hypercall emulation to helper function isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 083/102] KVM: TDX: Add a place holder to handle TDX VM exit isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 084/102] KVM: TDX: handle EXIT_REASON_OTHER_SMI isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 085/102] KVM: TDX: handle ept violation/misconfig exit isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 086/102] KVM: TDX: handle EXCEPTION_NMI and EXTERNAL_INTERRUPT isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 087/102] KVM: TDX: Add a place holder for handler of TDX hypercalls (TDG.VP.VMCALL) isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 088/102] KVM: TDX: handle KVM hypercall with TDG.VP.VMCALL isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 089/102] KVM: TDX: Handle TDX PV CPUID hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 090/102] KVM: TDX: Handle TDX PV HLT hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 091/102] KVM: TDX: Handle TDX PV port io hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 092/102] KVM: TDX: Handle TDX PV MMIO hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 093/102] KVM: TDX: Implement callbacks for MSR operations for TDX isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 094/102] KVM: TDX: Handle TDX PV rdmsr/wrmsr hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 095/102] KVM: TDX: Handle TDX PV report fatal error hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 096/102] KVM: TDX: Handle TDX PV map_gpa hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 097/102] KVM: TDX: Handle TDG.VP.VMCALL<GetTdVmCallInfo> hypercall isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 098/102] KVM: TDX: Silently discard SMI request isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 099/102] KVM: TDX: Silently ignore INIT/SIPI isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 100/102] KVM: TDX: Add methods to ignore accesses to CPU state isaku.yamahata
2022-06-27 21:54 ` [PATCH v7 101/102] Documentation/virtual/kvm: Document on Trust Domain Extensions(TDX) isaku.yamahata
2022-07-08 1:34 ` Kai Huang
2022-06-27 21:54 ` [PATCH v7 102/102] KVM: x86: design documentation on TDX support of x86 KVM TDP MMU isaku.yamahata
2022-07-11 15:17 ` [PATCH v7 000/102] KVM TDX basic feature support Isaku Yamahata
2022-07-12 5:07 ` Chao Gao
2022-07-12 10:54 ` Chao Peng
2022-07-12 17:22 ` Isaku Yamahata
2022-07-13 7:37 ` Chao Peng
2022-07-12 10:49 ` Chao Peng
2022-07-12 17:35 ` Isaku Yamahata
2022-07-14 1:03 ` Sean Christopherson
2022-07-14 4:09 ` Xiaoyao Li
2022-07-20 14:59 ` Chao Peng
2022-07-25 13:46 ` Nikunj A. Dadhania
2022-07-26 14:32 ` Chao Peng
2022-07-27 9:26 ` Nikunj A. Dadhania
2022-08-03 10:48 ` Chao Peng
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4c59eddd8f1d5029be8eeac84fbb75131b984568.camel@intel.com \
--to=kai.huang@intel.com \
--cc=isaku.yamahata@gmail.com \
--cc=isaku.yamahata@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=sean.j.christopherson@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).