From: Paolo Bonzini <pbonzini@redhat.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: Jon Kohler <jon@nutanix.com>, Nikunj A Dadhania <nikunj@amd.com>,
Amit Shah <amit.shah@amd.com>,
Sean Christopherson <seanjc@google.com>,
Marcelo Tosatti <mtosatti@redhat.com>
Subject: [PATCH 25/27] KVM: SVM: enable GMET and set it in MMU role
Date: Wed, 8 Apr 2026 11:42:15 -0400 [thread overview]
Message-ID: <20260408154217.458420-26-pbonzini@redhat.com> (raw)
In-Reply-To: <20260408154217.458420-1-pbonzini@redhat.com>
Set the GMET bit in the nested control field. This has effectively
no impact as long as NPT page tables are changed to have U=0.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
arch/x86/kvm/mmu/mmu.c | 6 +++++-
arch/x86/kvm/svm/nested.c | 7 +++++--
arch/x86/kvm/svm/svm.c | 16 ++++++++++++++++
arch/x86/kvm/svm/svm.h | 1 +
4 files changed, 27 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 354f8af5d90a..14c0c2da75bb 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -5840,7 +5840,6 @@ kvm_calc_tdp_mmu_root_page_role(struct kvm_vcpu *vcpu,
{
union kvm_mmu_page_role role = {0};
- role.access = ACC_ALL;
role.cr0_wp = true;
role.cr4_smep = kvm_x86_call(tdp_has_smep)(vcpu->kvm);
role.efer_nx = true;
@@ -5851,6 +5850,11 @@ kvm_calc_tdp_mmu_root_page_role(struct kvm_vcpu *vcpu,
role.direct = true;
role.has_4_byte_gpte = false;
+ /* All TDP pages are supervisor-executable */
+ role.access = ACC_ALL;
+ if (role.cr4_smep && shadow_user_mask)
+ role.access &= ~ACC_USER_MASK;
+
return role;
}
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index a76f2b8cbef9..863040ecec0e 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -829,9 +829,12 @@ static void nested_vmcb02_prepare_control(struct vcpu_svm *svm,
/* Also overwritten later if necessary. */
vmcb02->control.tlb_ctl = TLB_CONTROL_DO_NOTHING;
- /* nested_cr3. */
- if (nested_npt_enabled(svm))
+ /* Use vmcb01 MMU and format if guest does not use nNPT */
+ if (nested_npt_enabled(svm)) {
+ vmcb02->control.nested_ctl &= ~SVM_NESTED_CTL_GMET_ENABLE;
+
nested_svm_init_mmu_context(vcpu);
+ }
vcpu->arch.tsc_offset = kvm_calc_nested_tsc_offset(
vcpu->arch.l1_tsc_offset,
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index e6477affac9a..1705e3cafcb0 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -135,6 +135,9 @@ module_param(pause_filter_count_max, ushort, 0444);
bool npt_enabled = true;
module_param_named(npt, npt_enabled, bool, 0444);
+bool gmet_enabled = true;
+module_param_named(gmet, gmet_enabled, bool, 0444);
+
/* allow nested virtualization in KVM/SVM */
static int nested = true;
module_param(nested, int, 0444);
@@ -1170,6 +1173,10 @@ static void init_vmcb(struct kvm_vcpu *vcpu, bool init_event)
save->g_pat = vcpu->arch.pat;
save->cr3 = 0;
}
+
+ if (gmet_enabled)
+ control->nested_ctl |= SVM_NESTED_CTL_GMET_ENABLE;
+
svm->current_vmcb->asid_generation = 0;
svm->asid = 0;
@@ -4475,6 +4482,11 @@ svm_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
hypercall[2] = 0xd9;
}
+static bool svm_tdp_has_smep(struct kvm *kvm)
+{
+ return gmet_enabled;
+}
+
/*
* The kvm parameter can be NULL (module initialization, or invocation before
* VM creation). Be sure to check the kvm parameter before using it.
@@ -5224,6 +5236,7 @@ struct kvm_x86_ops svm_x86_ops __initdata = {
.write_tsc_multiplier = svm_write_tsc_multiplier,
.load_mmu_pgd = svm_load_mmu_pgd,
+ .tdp_has_smep = svm_tdp_has_smep,
.check_intercept = svm_check_intercept,
.handle_exit_irqoff = svm_handle_exit_irqoff,
@@ -5464,6 +5477,9 @@ static __init int svm_hardware_setup(void)
if (!boot_cpu_has(X86_FEATURE_NPT))
npt_enabled = false;
+ if (!npt_enabled || !boot_cpu_has(X86_FEATURE_GMET))
+ gmet_enabled = false;
+
/* Force VM NPT level equal to the host's paging level */
kvm_configure_mmu(npt_enabled, get_npt_level(),
get_npt_level(), PG_LEVEL_1G);
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 6942e6b0eda6..41042379aa48 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -44,6 +44,7 @@ static inline struct page *__sme_pa_to_page(unsigned long pa)
#define IOPM_SIZE PAGE_SIZE * 3
#define MSRPM_SIZE PAGE_SIZE * 2
+extern bool gmet_enabled;
extern bool npt_enabled;
extern int nrips;
extern int vgif;
--
2.52.0
next prev parent reply other threads:[~2026-04-08 15:42 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-08 15:41 [PATCH v3 00/27] KVM: combined patchset for MBEC/GMET support Paolo Bonzini
2026-04-08 15:41 ` [PATCH 01/27] KVM: TDX/VMX: rework EPT_VIOLATION_EXEC_FOR_RING3_LIN into PROT_MASK Paolo Bonzini
2026-04-08 15:41 ` [PATCH 02/27] KVM: x86/mmu: remove SPTE_PERM_MASK Paolo Bonzini
2026-04-08 15:41 ` [PATCH 03/27] KVM: x86/mmu: free up bit 10 of PTEs in preparation for MBEC Paolo Bonzini
2026-04-08 21:29 ` Huang, Kai
2026-04-08 15:41 ` [PATCH 04/27] KVM: x86/mmu: shuffle high bits of SPTEs " Paolo Bonzini
2026-04-17 1:23 ` Jon Kohler
2026-04-08 15:41 ` [PATCH 05/27] KVM: x86/mmu: remove SPTE_EPT_* Paolo Bonzini
2026-04-08 15:41 ` [PATCH 06/27] KVM: x86/mmu: merge make_spte_{non,}executable Paolo Bonzini
2026-04-08 15:41 ` [PATCH 07/27] KVM: x86/mmu: rename and clarify BYTE_MASK Paolo Bonzini
2026-04-08 15:41 ` [PATCH 08/27] KVM: x86/mmu: introduce ACC_READ_MASK Paolo Bonzini
2026-04-08 15:41 ` [PATCH 09/27] KVM: x86/mmu: separate more EPT/non-EPT permission_fault() Paolo Bonzini
2026-04-08 15:42 ` [PATCH 10/27] KVM: x86/mmu: pass PFERR_GUEST_PAGE/FINAL_MASK to kvm_translate_gpa Paolo Bonzini
2026-04-08 15:42 ` [PATCH 11/27] KVM: x86/mmu: pass pte_access for final nGPA->GPA walk Paolo Bonzini
2026-04-08 15:42 ` [PATCH 12/27] KVM: x86: make translate_nested_gpa vendor-specific Paolo Bonzini
2026-04-08 15:42 ` [PATCH 13/27] KVM: x86/mmu: split XS/XU bits for EPT Paolo Bonzini
2026-04-08 15:42 ` [PATCH 14/27] KVM: x86/mmu: move cr4_smep to base role Paolo Bonzini
2026-04-08 15:42 ` [PATCH 15/27] KVM: VMX: enable use of MBEC Paolo Bonzini
2026-04-17 1:19 ` Jon Kohler
2026-04-08 15:42 ` [PATCH 16/27] KVM: nVMX: pass advanced EPT violation vmexit info to guest Paolo Bonzini
2026-04-17 1:18 ` Jon Kohler
2026-04-08 15:42 ` [PATCH 17/27] KVM: nVMX: pass PFERR_USER_MASK to MMU on EPT violations Paolo Bonzini
2026-04-08 15:42 ` [PATCH 18/27] KVM: x86/mmu: add support for MBEC to EPT page table walks Paolo Bonzini
2026-04-08 15:42 ` [PATCH 19/27] KVM: nVMX: advertise MBEC to nested guests Paolo Bonzini
2026-04-08 15:42 ` [PATCH 20/27] KVM: nVMX: allow MBEC with EVMCS Paolo Bonzini
2026-04-08 15:42 ` [PATCH 21/27] KVM: x86/mmu: propagate access mask from root pages down Paolo Bonzini
2026-04-08 15:42 ` [PATCH 22/27] KVM: x86/mmu: introduce cpu_role bit for availability of PFEC.I/D Paolo Bonzini
2026-04-08 15:42 ` [PATCH 23/27] KVM: SVM: add GMET bit definitions Paolo Bonzini
2026-04-08 15:42 ` [PATCH 24/27] KVM: x86/mmu: add support for GMET to NPT page table walks Paolo Bonzini
2026-04-08 15:42 ` Paolo Bonzini [this message]
2026-04-08 15:42 ` [PATCH 26/27] KVM: SVM: work around errata 1218 Paolo Bonzini
2026-04-08 15:42 ` [PATCH 27/27] KVM: nSVM: enable GMET for guests Paolo Bonzini
2026-04-15 7:06 ` [PATCH v3 00/27] KVM: combined patchset for MBEC/GMET support David Riley
2026-04-21 2:07 ` Jon Kohler
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=20260408154217.458420-26-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=amit.shah@amd.com \
--cc=jon@nutanix.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mtosatti@redhat.com \
--cc=nikunj@amd.com \
--cc=seanjc@google.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
all inboxes | Powered by JetHome®