public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google•com>
To: Paolo Bonzini <pbonzini@redhat•com>,
	Marc Zyngier <maz@kernel•org>,
	 Oliver Upton <oliver.upton@linux•dev>,
	Sean Christopherson <seanjc@google•com>
Cc: kvm@vger•kernel.org, linux-arm-kernel@lists•infradead.org,
	 kvmarm@lists•linux.dev, linux-kernel@vger•kernel.org,
	 Ira Weiny <ira.weiny@intel•com>, Gavin Shan <gshan@redhat•com>,
	Shivank Garg <shivankg@amd•com>,
	 Vlastimil Babka <vbabka@suse•cz>,
	Xiaoyao Li <xiaoyao.li@intel•com>,
	 David Hildenbrand <david@redhat•com>,
	Fuad Tabba <tabba@google•com>,
	 Ackerley Tng <ackerleytng@google•com>,
	Tao Chan <chentao@kylinos•cn>,
	 James Houghton <jthoughton@google•com>
Subject: [PATCH v17 18/24] KVM: arm64: Handle guest_memfd-backed guest page faults
Date: Tue, 29 Jul 2025 15:54:49 -0700	[thread overview]
Message-ID: <20250729225455.670324-19-seanjc@google.com> (raw)
In-Reply-To: <20250729225455.670324-1-seanjc@google.com>

From: Fuad Tabba <tabba@google•com>

Add arm64 architecture support for handling guest page faults on memory
slots backed by guest_memfd.

This change introduces a new function, gmem_abort(), which encapsulates
the fault handling logic specific to guest_memfd-backed memory. The
kvm_handle_guest_abort() entry point is updated to dispatch to
gmem_abort() when a fault occurs on a guest_memfd-backed memory slot (as
determined by kvm_slot_has_gmem()).

Until guest_memfd gains support for huge pages, the fault granule for
these memory regions is restricted to PAGE_SIZE.

Reviewed-by: Gavin Shan <gshan@redhat•com>
Reviewed-by: James Houghton <jthoughton@google•com>
Reviewed-by: Marc Zyngier <maz@kernel•org>
Signed-off-by: Fuad Tabba <tabba@google•com>
Signed-off-by: Sean Christopherson <seanjc@google•com>
---
 arch/arm64/kvm/mmu.c | 86 ++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 83 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
index b3eacb400fab..8c82df80a835 100644
--- a/arch/arm64/kvm/mmu.c
+++ b/arch/arm64/kvm/mmu.c
@@ -1512,6 +1512,82 @@ static void adjust_nested_fault_perms(struct kvm_s2_trans *nested,
 	*prot |= kvm_encode_nested_level(nested);
 }
 
+#define KVM_PGTABLE_WALK_MEMABORT_FLAGS (KVM_PGTABLE_WALK_HANDLE_FAULT | KVM_PGTABLE_WALK_SHARED)
+
+static int gmem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
+		      struct kvm_s2_trans *nested,
+		      struct kvm_memory_slot *memslot, bool is_perm)
+{
+	bool write_fault, exec_fault, writable;
+	enum kvm_pgtable_walk_flags flags = KVM_PGTABLE_WALK_MEMABORT_FLAGS;
+	enum kvm_pgtable_prot prot = KVM_PGTABLE_PROT_R;
+	struct kvm_pgtable *pgt = vcpu->arch.hw_mmu->pgt;
+	unsigned long mmu_seq;
+	struct page *page;
+	struct kvm *kvm = vcpu->kvm;
+	void *memcache;
+	kvm_pfn_t pfn;
+	gfn_t gfn;
+	int ret;
+
+	ret = prepare_mmu_memcache(vcpu, true, &memcache);
+	if (ret)
+		return ret;
+
+	if (nested)
+		gfn = kvm_s2_trans_output(nested) >> PAGE_SHIFT;
+	else
+		gfn = fault_ipa >> PAGE_SHIFT;
+
+	write_fault = kvm_is_write_fault(vcpu);
+	exec_fault = kvm_vcpu_trap_is_exec_fault(vcpu);
+
+	VM_WARN_ON_ONCE(write_fault && exec_fault);
+
+	mmu_seq = kvm->mmu_invalidate_seq;
+	/* Pairs with the smp_wmb() in kvm_mmu_invalidate_end(). */
+	smp_rmb();
+
+	ret = kvm_gmem_get_pfn(kvm, memslot, gfn, &pfn, &page, NULL);
+	if (ret) {
+		kvm_prepare_memory_fault_exit(vcpu, fault_ipa, PAGE_SIZE,
+					      write_fault, exec_fault, false);
+		return ret;
+	}
+
+	writable = !(memslot->flags & KVM_MEM_READONLY);
+
+	if (nested)
+		adjust_nested_fault_perms(nested, &prot, &writable);
+
+	if (writable)
+		prot |= KVM_PGTABLE_PROT_W;
+
+	if (exec_fault ||
+	    (cpus_have_final_cap(ARM64_HAS_CACHE_DIC) &&
+	     (!nested || kvm_s2_trans_executable(nested))))
+		prot |= KVM_PGTABLE_PROT_X;
+
+	kvm_fault_lock(kvm);
+	if (mmu_invalidate_retry(kvm, mmu_seq)) {
+		ret = -EAGAIN;
+		goto out_unlock;
+	}
+
+	ret = KVM_PGT_FN(kvm_pgtable_stage2_map)(pgt, fault_ipa, PAGE_SIZE,
+						 __pfn_to_phys(pfn), prot,
+						 memcache, flags);
+
+out_unlock:
+	kvm_release_faultin_page(kvm, page, !!ret, writable);
+	kvm_fault_unlock(kvm);
+
+	if (writable && !ret)
+		mark_page_dirty_in_slot(kvm, memslot, gfn);
+
+	return ret != -EAGAIN ? ret : 0;
+}
+
 static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 			  struct kvm_s2_trans *nested,
 			  struct kvm_memory_slot *memslot, unsigned long hva,
@@ -1536,7 +1612,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	enum kvm_pgtable_prot prot = KVM_PGTABLE_PROT_R;
 	struct kvm_pgtable *pgt;
 	struct page *page;
-	enum kvm_pgtable_walk_flags flags = KVM_PGTABLE_WALK_HANDLE_FAULT | KVM_PGTABLE_WALK_SHARED;
+	enum kvm_pgtable_walk_flags flags = KVM_PGTABLE_WALK_MEMABORT_FLAGS;
 
 	if (fault_is_perm)
 		fault_granule = kvm_vcpu_trap_get_perm_fault_granule(vcpu);
@@ -1961,8 +2037,12 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu)
 	VM_WARN_ON_ONCE(kvm_vcpu_trap_is_permission_fault(vcpu) &&
 			!write_fault && !kvm_vcpu_trap_is_exec_fault(vcpu));
 
-	ret = user_mem_abort(vcpu, fault_ipa, nested, memslot, hva,
-			     esr_fsc_is_permission_fault(esr));
+	if (kvm_slot_has_gmem(memslot))
+		ret = gmem_abort(vcpu, fault_ipa, nested, memslot,
+				 esr_fsc_is_permission_fault(esr));
+	else
+		ret = user_mem_abort(vcpu, fault_ipa, nested, memslot, hva,
+				     esr_fsc_is_permission_fault(esr));
 	if (ret == 0)
 		ret = 1;
 out:
-- 
2.50.1.552.g942d659e1b-goog



  parent reply	other threads:[~2025-07-29 23:45 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-29 22:54 [PATCH v17 00/24] KVM: Enable mmap() for guest_memfd Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 01/24] KVM: Rename CONFIG_KVM_PRIVATE_MEM to CONFIG_KVM_GUEST_MEMFD Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 02/24] KVM: x86: Have all vendor neutral sub-configs depend on KVM_X86, not just KVM Sean Christopherson
2025-07-31  8:08   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 03/24] KVM: x86: Select KVM_GENERIC_PRIVATE_MEM directly from KVM_SW_PROTECTED_VM Sean Christopherson
2025-07-31  8:08   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 04/24] KVM: x86: Select TDX's KVM_GENERIC_xxx dependencies iff CONFIG_KVM_INTEL_TDX=y Sean Christopherson
2025-07-31  8:07   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 05/24] KVM: Rename CONFIG_KVM_GENERIC_PRIVATE_MEM to CONFIG_HAVE_KVM_ARCH_GMEM_POPULATE Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 06/24] KVM: Rename kvm_slot_can_be_private() to kvm_slot_has_gmem() Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 07/24] KVM: Fix comments that refer to slots_lock Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 08/24] KVM: Fix comment that refers to kvm uapi header path Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 09/24] KVM: x86: Enable KVM_GUEST_MEMFD for all 64-bit builds Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 10/24] KVM: guest_memfd: Add plumbing to host to map guest_memfd pages Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 11/24] KVM: guest_memfd: Track guest_memfd mmap support in memslot Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 12/24] KVM: x86/mmu: Rename .private_max_mapping_level() to .gmem_max_mapping_level() Sean Christopherson
2025-07-31  8:15   ` Fuad Tabba
2025-07-31  8:29     ` David Hildenbrand
2025-07-31  8:33       ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 13/24] KVM: x86/mmu: Hoist guest_memfd max level/order helpers "up" in mmu.c Sean Christopherson
2025-07-31  7:59   ` David Hildenbrand
2025-07-31  8:06   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 14/24] KVM: x86/mmu: Enforce guest_memfd's max order when recovering hugepages Sean Christopherson
2025-07-30  7:33   ` Xiaoyao Li
2025-07-31  8:06     ` David Hildenbrand
2025-07-31  8:10   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 15/24] KVM: x86/mmu: Extend guest_memfd's max mapping level to shared mappings Sean Christopherson
2025-07-30  7:36   ` Xiaoyao Li
2025-07-31  8:01   ` David Hildenbrand
2025-07-31  8:05   ` Fuad Tabba
2025-07-29 22:54 ` [PATCH v17 16/24] KVM: x86/mmu: Handle guest page faults for guest_memfd with shared memory Sean Christopherson
2025-07-30  7:37   ` Xiaoyao Li
2025-07-29 22:54 ` [PATCH v17 17/24] KVM: arm64: Refactor user_mem_abort() Sean Christopherson
2025-07-29 22:54 ` Sean Christopherson [this message]
2025-07-29 22:54 ` [PATCH v17 19/24] KVM: arm64: nv: Handle VNCR_EL2-triggered faults backed by guest_memfd Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 20/24] KVM: arm64: Enable support for guest_memfd backed memory Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 21/24] KVM: Allow and advertise support for host mmap() on guest_memfd files Sean Christopherson
2025-07-29 22:54 ` [PATCH v17 22/24] KVM: selftests: Do not use hardcoded page sizes in guest_memfd test Sean Christopherson
2025-07-30 11:04   ` Xiaoyao Li
2025-07-29 22:54 ` [PATCH v17 23/24] KVM: selftests: guest_memfd mmap() test when mmap is supported Sean Christopherson
2025-07-30 11:39   ` Xiaoyao Li
2025-07-30 12:57     ` Sean Christopherson
2025-07-31  7:49       ` Xiaoyao Li
2025-08-07  8:12   ` Shivank Garg
2025-07-29 22:54 ` [PATCH v17 24/24] KVM: selftests: Add guest_memfd testcase to fault-in on !mmap()'d memory Sean Christopherson
2025-07-30  8:20   ` Xiaoyao Li
2025-07-30 15:51   ` Fuad Tabba
2026-03-30  6:21   ` Zenghui Yu
2026-04-17 16:47     ` Sean Christopherson
2026-05-12  7:28       ` Zenghui Yu
2026-05-12 15:53         ` Sean Christopherson
2025-07-30 21:34 ` [PATCH v17 00/24] KVM: Enable mmap() for guest_memfd Ackerley Tng
2025-07-30 22:44   ` Ackerley Tng
2025-08-27  8:43 ` Paolo Bonzini
2025-08-27 12:57   ` Sean Christopherson
2025-08-27 13:08   ` Marc Zyngier
2025-08-27 13:11     ` Paolo Bonzini
2025-08-27 13:14       ` Marc Zyngier

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=20250729225455.670324-19-seanjc@google.com \
    --to=seanjc@google$(echo .)com \
    --cc=ackerleytng@google$(echo .)com \
    --cc=chentao@kylinos$(echo .)cn \
    --cc=david@redhat$(echo .)com \
    --cc=gshan@redhat$(echo .)com \
    --cc=ira.weiny@intel$(echo .)com \
    --cc=jthoughton@google$(echo .)com \
    --cc=kvm@vger$(echo .)kernel.org \
    --cc=kvmarm@lists$(echo .)linux.dev \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    --cc=linux-kernel@vger$(echo .)kernel.org \
    --cc=maz@kernel$(echo .)org \
    --cc=oliver.upton@linux$(echo .)dev \
    --cc=pbonzini@redhat$(echo .)com \
    --cc=shivankg@amd$(echo .)com \
    --cc=tabba@google$(echo .)com \
    --cc=vbabka@suse$(echo .)cz \
    --cc=xiaoyao.li@intel$(echo .)com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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