From: Sean Christopherson <seanjc@google.com>
To: Yan Zhao <yan.y.zhao@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
Rick Edgecombe <rick.p.edgecombe@intel.com>,
Isaku Yamahata <isaku.yamahata@intel.com>,
Tony Lindgren <tony.lindgren@linux.intel.com>,
Sean Christopherson <sean.j.christopherson@intel.com>,
Kai Huang <kai.huang@intel.com>
Subject: Re: [PATCH 20/30] KVM: TDX: create/destroy VM structure
Date: Fri, 21 Feb 2025 11:43:45 -0800 [thread overview]
Message-ID: <Z7jXcelD0MAk7odP@google.com> (raw)
In-Reply-To: <Z7hdqCqOLJWcR71K@yzhao56-desk.sh.intel.com>
On Fri, Feb 21, 2025, Yan Zhao wrote:
> On Thu, Feb 20, 2025 at 04:55:18PM -0800, Sean Christopherson wrote:
> > TL;DR: Please don't merge this patch to kvm/next or kvm/queue.
...
> > Commit 6fcee03df6a1 ("KVM: x86: avoid loading a vCPU after .vm_destroy was called")
> > papered over an AVIC case, but there are issues, e.g. with the MSR filters[2],
> > and the NULL pointer deref that's blocking the aforementioned fix is a nVMX access
> > to the PIC.
> >
> > I haven't fully tested destroying vCPUs before calling vm_destroy(), but I can't
> > see anything in vmx_vm_destroy() or svm_vm_destroy() that expects to run while
> > vCPUs are still alive. If anything, it's opposite, e.g. freeing VMX's IPIv PID
> > table before vCPUs are destroyed is blatantly unsafe.
> Is it possible to simply move the code like freeing PID table from
> vmx_vm_destroy() to static_call_cond(kvm_x86_vm_free)(kvm) ?
That would fix the potential PID table problem, but not the other issues with
freeing VM state before destroying vCPUs.
> > The good news is, I think it'll lead to a better approach (and naming). KVM already
> > frees MMU state before vCPU state, because while MMUs are largely VM-scoped, all
> > of the common MMU state needs to be freed before any one vCPU is freed.
> >
> > And so my plan is to carved out a kvm_destroy_mmus() helper, which can then call
> > the TDX hook to release/reclaim the HKID, which I assume needs to be done after
> > KVM's general MMU destruction, but before vCPUs are freed.
> >
> > I'll make sure to Cc y'all on the series (typing and testing furiously to try and
> > get it out asap). But to try and avoid posting code that's not usable for TDX,
> > will this work?
> >
> > static void kvm_destroy_mmus(struct kvm *kvm)
> > {
> > struct kvm_vcpu *vcpu;
> > unsigned long i;
> >
> > if (current->mm == kvm->mm) {
> > /*
> > * Free memory regions allocated on behalf of userspace,
> > * unless the memory map has changed due to process exit
> > * or fd copying.
> > */
> > mutex_lock(&kvm->slots_lock);
> > __x86_set_memory_region(kvm, APIC_ACCESS_PAGE_PRIVATE_MEMSLOT,
> > 0, 0);
> > __x86_set_memory_region(kvm, IDENTITY_PAGETABLE_PRIVATE_MEMSLOT,
> > 0, 0);
> > __x86_set_memory_region(kvm, TSS_PRIVATE_MEMSLOT, 0, 0);
> > mutex_unlock(&kvm->slots_lock);
> > }
> >
> > kvm_for_each_vcpu(i, vcpu, kvm) {
> > kvm_clear_async_pf_completion_queue(vcpu);
> > kvm_unload_vcpu_mmu(vcpu);
> > }
> >
> > kvm_x86_call(mmu_destroy)(kvm);
> I suppose this will hook tdx_mmu_release_hkid() ?
Please see my follow-up idea[1] to hook into kvm_arch_pre_destroy_vm(). Turns
out there isn't a hard requirement to unload MMUs prior to destroying vCPUs (at
least, not AFAICT).
[1] https://lore.kernel.org/all/Z7fSIMABm4jp5ADA@google.com
next prev parent reply other threads:[~2025-02-21 19:43 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-20 17:05 [PATCH v3 00/30] TDX initialization + vCPU/VM creation Paolo Bonzini
2025-02-20 17:05 ` [PATCH 01/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX KeyID management Paolo Bonzini
2025-02-20 17:05 ` [PATCH 02/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX TD creation Paolo Bonzini
2025-02-20 17:05 ` [PATCH 03/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX vCPU creation Paolo Bonzini
2025-02-20 17:05 ` [PATCH 04/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX page cache management Paolo Bonzini
2025-02-20 17:05 ` [PATCH 05/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX VM/vCPU field access Paolo Bonzini
2025-02-20 17:05 ` [PATCH 06/30] x86/virt/tdx: Add SEAMCALL wrappers for TDX flush operations Paolo Bonzini
2025-02-20 17:05 ` [PATCH 07/30] x86/virt/tdx: allocate tdx_sys_info in static memory Paolo Bonzini
2025-02-20 21:59 ` Huang, Kai
2025-02-20 23:37 ` Edgecombe, Rick P
2025-02-20 17:05 ` [PATCH 08/30] x86/virt/tdx: Read essential global metadata for KVM Paolo Bonzini
2025-02-20 17:05 ` [PATCH 09/30] x86/virt/tdx: Add tdx_guest_keyid_alloc/free() to alloc and free TDX guest KeyID Paolo Bonzini
2025-02-20 17:05 ` [PATCH 10/30] KVM: Export hardware virtualization enabling/disabling functions Paolo Bonzini
2025-02-20 17:05 ` [PATCH 11/30] KVM: VMX: Refactor VMX module init/exit functions Paolo Bonzini
2025-02-20 21:55 ` Huang, Kai
2025-02-20 17:05 ` [PATCH 12/30] KVM: VMX: Initialize TDX during KVM module load Paolo Bonzini
2025-02-20 23:27 ` Huang, Kai
2025-02-24 18:57 ` Paolo Bonzini
2025-02-24 21:31 ` Huang, Kai
2025-02-20 17:05 ` [PATCH 13/30] KVM: TDX: Get TDX global information Paolo Bonzini
2025-02-21 0:12 ` Huang, Kai
2025-02-20 17:05 ` [PATCH 14/30] KVM: TDX: Add placeholders for TDX VM/vCPU structures Paolo Bonzini
2025-02-20 17:05 ` [PATCH 15/30] KVM: TDX: Define TDX architectural definitions Paolo Bonzini
2025-02-20 17:05 ` [PATCH 16/30] KVM: TDX: Add TDX "architectural" error codes Paolo Bonzini
2025-02-20 17:05 ` [PATCH 17/30] KVM: TDX: Add helper functions to print TDX SEAMCALL error Paolo Bonzini
2025-02-20 17:05 ` [PATCH 18/30] KVM: TDX: Add place holder for TDX VM specific mem_enc_op ioctl Paolo Bonzini
2025-02-25 10:50 ` Huang, Kai
2025-02-20 17:05 ` [PATCH 19/30] KVM: TDX: Get system-wide info about TDX module on initialization Paolo Bonzini
2025-02-20 17:05 ` [PATCH 20/30] KVM: TDX: create/destroy VM structure Paolo Bonzini
2025-02-21 0:55 ` Sean Christopherson
2025-02-21 1:08 ` Sean Christopherson
2025-02-22 0:30 ` Edgecombe, Rick P
2025-02-22 1:38 ` Sean Christopherson
2025-02-24 8:32 ` Yan Zhao
2025-02-21 11:04 ` Yan Zhao
2025-02-21 19:43 ` Sean Christopherson [this message]
2025-02-21 12:25 ` Yan Zhao
2025-02-25 16:24 ` Xiaoyao Li
2025-02-20 17:05 ` [PATCH 21/30] KVM: TDX: Support per-VM KVM_CAP_MAX_VCPUS extension check Paolo Bonzini
2025-02-20 17:05 ` [PATCH 22/30] KVM: x86: expose cpuid_entry2_find for TDX Paolo Bonzini
2025-02-20 17:05 ` [PATCH 23/30] KVM: TDX: initialize VM with TDX specific parameters Paolo Bonzini
2025-02-21 2:31 ` Xiaoyao Li
2025-02-25 17:28 ` Paolo Bonzini
2025-02-20 17:05 ` [PATCH 24/30] KVM: TDX: Make pmu_intel.c ignore guest TD case Paolo Bonzini
2025-02-20 17:05 ` [PATCH 25/30] KVM: TDX: Don't offline the last cpu of one package when there's TDX guest Paolo Bonzini
2025-02-20 17:06 ` [PATCH 26/30] KVM: TDX: create/free TDX vcpu structure Paolo Bonzini
2025-02-20 17:06 ` [PATCH 27/30] KVM: TDX: Do TDX specific vcpu initialization Paolo Bonzini
2025-02-20 17:06 ` [PATCH 28/30] KVM: x86: Introduce KVM_TDX_GET_CPUID Paolo Bonzini
2025-02-20 17:06 ` [PATCH 29/30] KVM: x86/mmu: Taking guest pa into consideration when calculate tdp level Paolo Bonzini
2025-02-20 17:06 ` [PATCH 30/30] KVM: TDX: Register TDX host key IDs to cgroup misc controller Paolo Bonzini
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=Z7jXcelD0MAk7odP@google.com \
--to=seanjc@google.com \
--cc=isaku.yamahata@intel.com \
--cc=kai.huang@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=rick.p.edgecombe@intel.com \
--cc=sean.j.christopherson@intel.com \
--cc=tony.lindgren@linux.intel.com \
--cc=yan.y.zhao@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.