From: Sean Christopherson <seanjc@google.com>
To: Lai Jiangshan <jiangshanlai+lkml@gmail.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Wanpeng Li <wanpengli@tencent.com>,
Jim Mattson <jmattson@google.com>, Joerg Roedel <joro@8bytes.org>,
kvm@vger.kernel.org, LKML <linux-kernel@vger.kernel.org>,
Ben Gardon <bgardon@google.com>,
Junaid Shahid <junaids@google.com>,
Liran Alon <liran.alon@oracle.com>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>,
John Haxby <john.haxby@oracle.com>,
Miaohe Lin <linmiaohe@huawei.com>,
Tom Lendacky <thomas.lendacky@amd.com>
Subject: Re: [PATCH v3 23/37] KVM: nVMX: Add helper to handle TLB flushes on nested VM-Enter/VM-Exit
Date: Thu, 28 Oct 2021 15:22:27 +0000 [thread overview]
Message-ID: <YXrAM9MNqgLTU6+m@google.com> (raw)
In-Reply-To: <CAJhGHyD=S6pVB+OxM7zF0_6LnMUCLqyTfMK4x9GZsdRHZmgN7Q@mail.gmail.com>
-me :-)
On Thu, Oct 28, 2021, Lai Jiangshan wrote:
> On Sat, Mar 21, 2020 at 5:29 AM Sean Christopherson
> <sean.j.christopherson@intel.com> wrote:
>
> > + if (!nested_cpu_has_vpid(vmcs12) || !nested_has_guest_tlb_tag(vcpu)) {
> > + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
> > + } else if (is_vmenter &&
> > + vmcs12->virtual_processor_id != vmx->nested.last_vpid) {
> > + vmx->nested.last_vpid = vmcs12->virtual_processor_id;
> > + vpid_sync_context(nested_get_vpid02(vcpu));
> > + }
> > +}
>
> (I'm sorry to pick this old email to reply to, but the problem has
> nothing to do with this patch nor 5c614b3583e7 and it exists since
> nested vmx is introduced.)
>
> I think kvm_mmu_free_guest_mode_roots() should be called
> if (!enable_ept && vmcs12->virtual_processor_id != vmx->nested.last_vpid)
> just because prev_roots doesn't cache the vpid12.
> (prev_roots caches PCID, which is distinctive)
>
> The problem hardly exists if L1's hypervisor is also kvm, but if L1's
> hypervisor is different or is also kvm with some changes in the way how it
> manages VPID.
Indeed. A more straightforward error case would be if L1 and L2 share CR3, and
vmcs02.VPID is toggled (or used for the first time) on the L1 => L2 VM-Enter.
The fix should simply be:
diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
index eedcebf58004..574823370e7a 100644
--- a/arch/x86/kvm/vmx/nested.c
+++ b/arch/x86/kvm/vmx/nested.c
@@ -1202,17 +1202,15 @@ static void nested_vmx_transition_tlb_flush(struct kvm_vcpu *vcpu,
*
* If a TLB flush isn't required due to any of the above, and vpid12 is
* changing then the new "virtual" VPID (vpid12) will reuse the same
- * "real" VPID (vpid02), and so needs to be flushed. There's no direct
- * mapping between vpid02 and vpid12, vpid02 is per-vCPU and reused for
- * all nested vCPUs. Remember, a flush on VM-Enter does not invalidate
- * guest-physical mappings, so there is no need to sync the nEPT MMU.
+ * "real" VPID (vpid02), and so needs to be flushed. Like the !vpid02
+ * case above, this is a full TLB flush from the guest's perspective.
*/
if (!nested_has_guest_tlb_tag(vcpu)) {
kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, vcpu);
} else if (is_vmenter &&
vmcs12->virtual_processor_id != vmx->nested.last_vpid) {
vmx->nested.last_vpid = vmcs12->virtual_processor_id;
- vpid_sync_context(nested_get_vpid02(vcpu));
+ kvm_make_request(KVM_REQ_TLB_FLUSH_GUEST, vcpu);
}
}
next prev parent reply other threads:[~2021-10-28 15:22 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-20 21:27 [PATCH v3 00/37] KVM: x86: TLB flushing fixes and enhancements Sean Christopherson
2020-03-20 21:27 ` [PATCH v3 01/37] KVM: VMX: Flush all EPTP/VPID contexts on remote TLB flush Sean Christopherson
2021-08-03 1:45 ` Lai Jiangshan
2021-08-03 15:39 ` Sean Christopherson
2021-08-04 3:11 ` Lai Jiangshan
2021-08-04 15:33 ` Sean Christopherson
2020-03-20 21:27 ` [PATCH v3 02/37] KVM: nVMX: Validate the EPTP when emulating INVEPT(EXTENT_CONTEXT) Sean Christopherson
2020-03-23 14:51 ` Vitaly Kuznetsov
2020-03-23 15:45 ` Sean Christopherson
2020-03-23 23:46 ` Paolo Bonzini
2020-03-20 21:27 ` [PATCH v3 03/37] KVM: nVMX: Invalidate all EPTP contexts when emulating INVEPT for L1 Sean Christopherson
2020-03-23 15:24 ` Vitaly Kuznetsov
2020-03-23 15:53 ` Sean Christopherson
2020-03-23 16:24 ` Jim Mattson
2020-03-23 16:28 ` Sean Christopherson
2020-03-23 16:36 ` Jim Mattson
2020-03-23 16:44 ` Sean Christopherson
2020-03-23 23:50 ` Paolo Bonzini
2020-03-24 0:12 ` Jim Mattson
2020-03-30 18:38 ` Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 04/37] KVM: nVMX: Invalidate all roots when emulating INVVPID without EPT Sean Christopherson
2020-03-23 15:34 ` Vitaly Kuznetsov
2020-03-23 16:04 ` Sean Christopherson
2020-03-23 16:33 ` Vitaly Kuznetsov
2020-03-23 16:50 ` Sean Christopherson
2020-03-23 16:57 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 05/37] KVM: x86: Export kvm_propagate_fault() (as kvm_inject_emulated_page_fault) Sean Christopherson
2020-03-23 15:47 ` Vitaly Kuznetsov
2020-03-23 16:24 ` Sean Christopherson
2020-03-23 23:56 ` Paolo Bonzini
2020-03-20 21:28 ` [PATCH v3 06/37] KVM: x86: Consolidate logic for injecting page faults to L1 Sean Christopherson
2020-03-24 0:47 ` Paolo Bonzini
2020-03-20 21:28 ` [PATCH v3 07/37] KVM: x86: Sync SPTEs when injecting page/EPT fault into L1 Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 08/37] KVM: VMX: Skip global INVVPID fallback if vpid==0 in vpid_sync_context() Sean Christopherson
2020-03-25 9:33 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 09/37] KVM: VMX: Use vpid_sync_context() directly when possible Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 10/37] KVM: VMX: Move vpid_sync_vcpu_addr() down a few lines Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 11/37] KVM: VMX: Handle INVVPID fallback logic in vpid_sync_vcpu_addr() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 12/37] KVM: VMX: Drop redundant capability checks in low level INVVPID helpers Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 13/37] KVM: nVMX: Use vpid_sync_vcpu_addr() to emulate INVVPID with address Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 14/37] KVM: x86: Move "flush guest's TLB" logic to separate kvm_x86_ops hook Sean Christopherson
2020-03-25 10:23 ` Vitaly Kuznetsov
2020-03-25 15:41 ` Paolo Bonzini
2020-03-25 16:08 ` Vitaly Kuznetsov
2020-03-25 15:48 ` Sean Christopherson
2020-03-25 16:11 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 15/37] KVM: VMX: Clean up vmx_flush_tlb_gva() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 16/37] KVM: x86: Drop @invalidate_gpa param from kvm_x86_ops' tlb_flush() Sean Christopherson
2020-03-25 11:23 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 17/37] KVM: SVM: Wire up ->tlb_flush_guest() directly to svm_flush_tlb() Sean Christopherson
2020-03-25 11:23 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 18/37] KVM: VMX: Move vmx_flush_tlb() to vmx.c Sean Christopherson
2020-03-25 11:25 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 19/37] KVM: nVMX: Move nested_get_vpid02() to vmx/nested.h Sean Christopherson
2020-03-25 11:25 ` Vitaly Kuznetsov
2020-03-20 21:28 ` [PATCH v3 20/37] KVM: VMX: Introduce vmx_flush_tlb_current() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 21/37] KVM: SVM: Document the ASID logic in svm_flush_tlb() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 22/37] KVM: x86: Rename ->tlb_flush() to ->tlb_flush_all() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 23/37] KVM: nVMX: Add helper to handle TLB flushes on nested VM-Enter/VM-Exit Sean Christopherson
2021-10-28 13:11 ` Lai Jiangshan
2021-10-28 15:22 ` Sean Christopherson [this message]
2021-10-29 0:44 ` Lai Jiangshan
2021-10-29 17:10 ` Sean Christopherson
2021-10-30 1:34 ` Lai Jiangshan
2021-11-04 17:47 ` Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 24/37] KVM: x86: Introduce KVM_REQ_TLB_FLUSH_CURRENT to flush current ASID Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 25/37] KVM: x86/mmu: Use KVM_REQ_TLB_FLUSH_CURRENT for MMU specific flushes Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 26/37] KVM: nVMX: Selectively use TLB_FLUSH_CURRENT for nested VM-Enter/VM-Exit Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 27/37] KVM: nVMX: Reload APIC access page on nested VM-Exit only if necessary Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 28/37] KVM: VMX: Retrieve APIC access page HPA only when necessary Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 29/37] KVM: VMX: Don't reload APIC access page if its control is disabled Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 30/37] KVM: x86/mmu: Move fast_cr3_switch() side effects to __kvm_mmu_new_cr3() Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 31/37] KVM: x86/mmu: Add separate override for MMU sync during fast CR3 switch Sean Christopherson
2020-03-24 11:07 ` Paolo Bonzini
2020-03-20 21:28 ` [PATCH v3 32/37] KVM: x86/mmu: Add module param to force TLB flush on root reuse Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 33/37] KVM: nVMX: Skip MMU sync on nested VMX transition when possible Sean Christopherson
2020-03-24 11:19 ` Paolo Bonzini
2020-03-20 21:28 ` [PATCH v3 34/37] KVM: nVMX: Don't flush TLB on nested VMX transition Sean Christopherson
2020-03-24 11:20 ` Paolo Bonzini
2020-03-24 18:10 ` Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 35/37] KVM: nVMX: Free only the affected contexts when emulating INVEPT Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 36/37] KVM: x86: Replace "cr3" with "pgd" in "new cr3/pgd" related code Sean Christopherson
2020-03-20 21:28 ` [PATCH v3 37/37] KVM: VMX: Clean cr3/pgd handling in vmx_load_mmu_pgd() Sean Christopherson
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=YXrAM9MNqgLTU6+m@google.com \
--to=seanjc@google.com \
--cc=bgardon@google.com \
--cc=boris.ostrovsky@oracle.com \
--cc=jiangshanlai+lkml@gmail.com \
--cc=jmattson@google.com \
--cc=john.haxby@oracle.com \
--cc=joro@8bytes.org \
--cc=junaids@google.com \
--cc=kvm@vger.kernel.org \
--cc=linmiaohe@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=liran.alon@oracle.com \
--cc=pbonzini@redhat.com \
--cc=thomas.lendacky@amd.com \
--cc=vkuznets@redhat.com \
--cc=wanpengli@tencent.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).