mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Yan Zhao <yan.y.zhao@intel.com>
To: iommu@lists.linux.dev, kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: alex.williamson@redhat.com, jgg@nvidia.com, pbonzini@redhat.com,
	seanjc@google.com, joro@8bytes.org, will@kernel.org,
	robin.murphy@arm.com, kevin.tian@intel.com,
	baolu.lu@linux.intel.com, dwmw2@infradead.org,
	yi.l.liu@intel.com, Yan Zhao <yan.y.zhao@intel.com>
Subject: [RFC PATCH 36/42] KVM: x86/mmu: Keep exported TDP root valid
Date: Sat,  2 Dec 2023 17:33:25 +0800	[thread overview]
Message-ID: <20231202093325.15676-1-yan.y.zhao@intel.com> (raw)
In-Reply-To: <20231202091211.13376-1-yan.y.zhao@intel.com>

Keep exported TDP root always valid and zap all leaf entries to replace
the "root role invalid" operation.

Unlike TDP roots accessed by vCPUs only, update of TDP root exported to
external components must be in an atomic way, like
1. allocating new root,
2. updating and notifying new root to external components,
3. making old root invalid,

So, it's more efficient to just zap all leaf entries of the exported TDP.

Though zapping all leaf entries will make "fast zap" not fast enough, as
with commit 0df9dab891ff ("KVM: x86/mmu: Stop zapping invalidated TDP MMU
roots asynchronously"), zap of root is anyway required to be done
synchronously in kvm_mmu_zap_all_fast() before completing memslot removal.

Besides, it's also safe to skip invalidating "exported" root in
kvm_tdp_mmu_invalidate_all_roots() for path kvm_mmu_uninit_tdp_mmu(),
because when the VM is shutting down, as TDP FD will hold reference count
of kvm, kvm_mmu_uninit_tdp_mmu() --> kvm_tdp_mmu_invalidate_all_roots()
will not come until the TDP root is unmarked as "exported" and put. All
children entries are also zapped before the root is put.

Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
---
 arch/x86/kvm/mmu/mmu.c     |  3 +++
 arch/x86/kvm/mmu/tdp_mmu.c | 40 +++++++++++++++++++++++++++++++++-----
 arch/x86/kvm/mmu/tdp_mmu.h |  1 +
 3 files changed, 39 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 3e2475c678c27..37a903fff582a 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -6187,6 +6187,9 @@ static void kvm_mmu_zap_all_fast(struct kvm *kvm)
 
 	kvm_zap_obsolete_pages(kvm);
 
+	if (tdp_mmu_enabled)
+		kvm_tdp_mmu_zap_exported_roots(kvm);
+
 	write_unlock(&kvm->mmu_lock);
 
 	/*
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 47edf54961e89..36a309ad27d47 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -897,12 +897,38 @@ void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm)
 	read_unlock(&kvm->mmu_lock);
 }
 
+void kvm_tdp_mmu_zap_exported_roots(struct kvm *kvm)
+{
+#ifdef CONFIG_HAVE_KVM_EXPORTED_TDP
+	struct kvm_mmu_page *root;
+	bool flush;
+
+	lockdep_assert_held_write(&kvm->mmu_lock);
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(root, &kvm->arch.tdp_mmu_roots, link) {
+		if (!root->exported)
+			continue;
+
+		flush = tdp_mmu_zap_leafs(kvm, root, 0, -1ULL, false, false);
+		if (flush)
+			kvm_flush_remote_tlbs(kvm);
+	}
+
+	rcu_read_unlock();
+#endif
+}
+
 /*
- * Mark each TDP MMU root as invalid to prevent vCPUs from reusing a root that
- * is about to be zapped, e.g. in response to a memslots update.  The actual
- * zapping is done separately so that it happens with mmu_lock with read,
- * whereas invalidating roots must be done with mmu_lock held for write (unless
- * the VM is being destroyed).
+ * Mark each TDP MMU root (except exported root) as invalid to prevent vCPUs from
+ * reusing a root that is about to be zapped, e.g. in response to a memslots
+ * update.
+ * The actual zapping is done separately so that it happens with mmu_lock
+ * with read, whereas invalidating roots must be done with mmu_lock held for write
+ * (unless the VM is being destroyed).
+ * For exported root, zap is done in kvm_tdp_mmu_zap_exported_roots() before
+ * the memslot update completes with mmu_lock held for write.
  *
  * Note, kvm_tdp_mmu_zap_invalidated_roots() is gifted the TDP MMU's reference.
  * See kvm_tdp_mmu_get_vcpu_root_hpa().
@@ -932,6 +958,10 @@ void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm)
 	 * or get/put references to roots.
 	 */
 	list_for_each_entry(root, &kvm->arch.tdp_mmu_roots, link) {
+#ifdef CONFIG_HAVE_KVM_EXPORTED_TDP
+		if (root->exported)
+			continue;
+#endif
 		/*
 		 * Note, invalid roots can outlive a memslot update!  Invalid
 		 * roots must be *zapped* before the memslot update completes,
diff --git a/arch/x86/kvm/mmu/tdp_mmu.h b/arch/x86/kvm/mmu/tdp_mmu.h
index 1d36ed378848b..df42350022a3f 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.h
+++ b/arch/x86/kvm/mmu/tdp_mmu.h
@@ -25,6 +25,7 @@ bool kvm_tdp_mmu_zap_sp(struct kvm *kvm, struct kvm_mmu_page *sp);
 void kvm_tdp_mmu_zap_all(struct kvm *kvm);
 void kvm_tdp_mmu_invalidate_all_roots(struct kvm *kvm);
 void kvm_tdp_mmu_zap_invalidated_roots(struct kvm *kvm);
+void kvm_tdp_mmu_zap_exported_roots(struct kvm *kvm);
 
 int kvm_tdp_mmu_map(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault);
 
-- 
2.17.1


  parent reply	other threads:[~2023-12-02 10:02 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-02  9:12 [RFC PATCH 00/42] Sharing KVM TDP to IOMMU Yan Zhao
2023-12-02  9:13 ` [RFC PATCH 01/42] KVM: Public header for KVM to export TDP Yan Zhao
2023-12-02  9:15 ` [RFC PATCH 02/42] KVM: x86: Arch header for kvm to export TDP for Intel Yan Zhao
2023-12-02  9:15 ` [RFC PATCH 03/42] KVM: Introduce VM ioctl KVM_CREATE_TDP_FD Yan Zhao
2023-12-02  9:16 ` [RFC PATCH 04/42] KVM: Skeleton of KVM TDP FD object Yan Zhao
2023-12-02  9:16 ` [RFC PATCH 05/42] KVM: Embed "arch" object and call arch init/destroy in TDP FD Yan Zhao
2023-12-02  9:17 ` [RFC PATCH 06/42] KVM: Register/Unregister importers to KVM exported TDP Yan Zhao
2023-12-02  9:18 ` [RFC PATCH 07/42] KVM: Forward page fault requests to arch specific code for " Yan Zhao
2023-12-02  9:18 ` [RFC PATCH 08/42] KVM: Add a helper to notify importers that KVM exported TDP is flushed Yan Zhao
2023-12-02  9:19 ` [RFC PATCH 09/42] iommu: Add IOMMU_DOMAIN_KVM Yan Zhao
2023-12-02  9:20 ` [RFC PATCH 10/42] iommu: Add new iommu op to create domains managed by KVM Yan Zhao
2023-12-04 15:09   ` Jason Gunthorpe
2023-12-02  9:20 ` [RFC PATCH 11/42] iommu: Add new domain op cache_invalidate_kvm Yan Zhao
2023-12-04 15:09   ` Jason Gunthorpe
2023-12-05  6:40     ` Yan Zhao
2023-12-05 14:52       ` Jason Gunthorpe
2023-12-06  1:00         ` Yan Zhao
2023-12-02  9:21 ` [RFC PATCH 12/42] iommufd: Introduce allocation data info and flag for KVM managed HWPT Yan Zhao
2023-12-04 18:29   ` Jason Gunthorpe
2023-12-05  7:08     ` Yan Zhao
2023-12-05 14:53       ` Jason Gunthorpe
2023-12-06  0:58         ` Yan Zhao
2023-12-02  9:21 ` [RFC PATCH 13/42] iommufd: Add a KVM HW pagetable object Yan Zhao
2023-12-02  9:22 ` [RFC PATCH 14/42] iommufd: Enable KVM HW page table object to be proxy between KVM and IOMMU Yan Zhao
2023-12-04 18:34   ` Jason Gunthorpe
2023-12-05  7:09     ` Yan Zhao
2023-12-02  9:22 ` [RFC PATCH 15/42] iommufd: Add iopf handler to KVM hw pagetable Yan Zhao
2023-12-02  9:23 ` [RFC PATCH 16/42] iommufd: Enable device feature IOPF during device attachment to KVM HWPT Yan Zhao
2023-12-04 18:36   ` Jason Gunthorpe
2023-12-05  7:14     ` Yan Zhao
2023-12-05 14:53       ` Jason Gunthorpe
2023-12-06  0:55         ` Yan Zhao
2023-12-02  9:23 ` [RFC PATCH 17/42] iommu/vt-d: Make some macros and helpers to be extern Yan Zhao
2023-12-02  9:24 ` [RFC PATCH 18/42] iommu/vt-d: Support of IOMMU_DOMAIN_KVM domain in Intel IOMMU Yan Zhao
2023-12-02  9:24 ` [RFC PATCH 19/42] iommu/vt-d: Set bit PGSNP in PASIDTE if domain cache coherency is enforced Yan Zhao
2023-12-02  9:25 ` [RFC PATCH 20/42] iommu/vt-d: Support attach devices to IOMMU_DOMAIN_KVM domain Yan Zhao
2023-12-02  9:26 ` [RFC PATCH 21/42] iommu/vt-d: Check reserved bits for " Yan Zhao
2023-12-02  9:26 ` [RFC PATCH 22/42] iommu/vt-d: Support cache invalidate of " Yan Zhao
2023-12-02  9:26 ` [RFC PATCH 23/42] iommu/vt-d: Allow pasid 0 in IOPF Yan Zhao
2023-12-02  9:27 ` [RFC PATCH 24/42] KVM: x86/mmu: Move bit SPTE_MMU_PRESENT from bit 11 to bit 59 Yan Zhao
2023-12-02  9:27 ` [RFC PATCH 25/42] KVM: x86/mmu: Abstract "struct kvm_mmu_common" from "struct kvm_mmu" Yan Zhao
2023-12-02  9:28 ` [RFC PATCH 26/42] KVM: x86/mmu: introduce new op get_default_mt_mask to kvm_x86_ops Yan Zhao
2023-12-02  9:28 ` [RFC PATCH 27/42] KVM: x86/mmu: change param "vcpu" to "kvm" in kvm_mmu_hugepage_adjust() Yan Zhao
2023-12-02  9:29 ` [RFC PATCH 28/42] KVM: x86/mmu: change "vcpu" to "kvm" in page_fault_handle_page_track() Yan Zhao
2023-12-02  9:29 ` [RFC PATCH 29/42] KVM: x86/mmu: remove param "vcpu" from kvm_mmu_get_tdp_level() Yan Zhao
2023-12-02  9:30 ` [RFC PATCH 30/42] KVM: x86/mmu: remove param "vcpu" from kvm_calc_tdp_mmu_root_page_role() Yan Zhao
2023-12-02  9:30 ` [RFC PATCH 31/42] KVM: x86/mmu: add extra param "kvm" to kvm_faultin_pfn() Yan Zhao
2023-12-02  9:31 ` [RFC PATCH 32/42] KVM: x86/mmu: add extra param "kvm" to make_mmio_spte() Yan Zhao
2023-12-02  9:31 ` [RFC PATCH 33/42] KVM: x86/mmu: add extra param "kvm" to make_spte() Yan Zhao
2023-12-02  9:32 ` [RFC PATCH 34/42] KVM: x86/mmu: add extra param "kvm" to tdp_mmu_map_handle_target_level() Yan Zhao
2023-12-02  9:32 ` [RFC PATCH 35/42] KVM: x86/mmu: Get/Put TDP root page to be exported Yan Zhao
2023-12-02  9:33 ` Yan Zhao [this message]
2023-12-02  9:33 ` [RFC PATCH 37/42] KVM: x86: Implement KVM exported TDP fault handler on x86 Yan Zhao
2023-12-02  9:35 ` [RFC PATCH 38/42] KVM: x86: "compose" and "get" interface for meta data of exported TDP Yan Zhao
2023-12-02  9:35 ` [RFC PATCH 39/42] KVM: VMX: add config KVM_INTEL_EXPORTED_EPT Yan Zhao
2023-12-02  9:36 ` [RFC PATCH 40/42] KVM: VMX: Compose VMX specific meta data for KVM exported TDP Yan Zhao
2023-12-02  9:36 ` [RFC PATCH 41/42] KVM: VMX: Implement ops .flush_remote_tlbs* in VMX when EPT is on Yan Zhao
2023-12-02  9:37 ` [RFC PATCH 42/42] KVM: VMX: Notify importers of exported TDP to flush TLBs on KVM flushes EPT Yan Zhao
2023-12-04 15:08 ` [RFC PATCH 00/42] Sharing KVM TDP to IOMMU Jason Gunthorpe
2023-12-04 16:38   ` Sean Christopherson
2023-12-05  1:31     ` Yan Zhao
2023-12-05  6:45       ` Tian, Kevin
2023-12-05  1:52   ` Yan Zhao
2023-12-05  6:30   ` Tian, Kevin
2023-12-04 17:00 ` Sean Christopherson
2023-12-04 17:30   ` Jason Gunthorpe
2023-12-04 19:22     ` Sean Christopherson
2023-12-04 19:50       ` Jason Gunthorpe
2023-12-04 20:11         ` Sean Christopherson
2023-12-04 23:49           ` Jason Gunthorpe
2023-12-05  7:17         ` Tian, Kevin
2023-12-05  5:53       ` Yan Zhao
2023-12-05  3:51   ` Yan Zhao

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=20231202093325.15676-1-yan.y.zhao@intel.com \
    --to=yan.y.zhao@intel.com \
    --cc=alex.williamson@redhat.com \
    --cc=baolu.lu@linux.intel.com \
    --cc=dwmw2@infradead.org \
    --cc=iommu@lists.linux.dev \
    --cc=jgg@nvidia.com \
    --cc=joro@8bytes.org \
    --cc=kevin.tian@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=robin.murphy@arm.com \
    --cc=seanjc@google.com \
    --cc=will@kernel.org \
    --cc=yi.l.liu@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

all inboxes | Powered by JetHome®