From: Will Deacon <will@kernel.org>
To: kvmarm@lists.linux.dev
Cc: linux-arm-kernel@lists.infradead.org,
Will Deacon <will@kernel.org>, Marc Zyngier <maz@kernel.org>,
Oliver Upton <oupton@kernel.org>, Joey Gouly <joey.gouly@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Zenghui Yu <yuzenghui@huawei.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Quentin Perret <qperret@google.com>,
Fuad Tabba <tabba@google.com>,
Vincent Donnefort <vdonnefort@google.com>,
Mostafa Saleh <smostafa@google.com>
Subject: [PATCH v2 11/35] KVM: arm64: Split teardown hypercall into two phases
Date: Mon, 19 Jan 2026 12:46:04 +0000 [thread overview]
Message-ID: <20260119124629.2563-12-will@kernel.org> (raw)
In-Reply-To: <20260119124629.2563-1-will@kernel.org>
In preparation for reclaiming protected guest VM pages from the host
during teardown, split the current 'pkvm_teardown_vm' hypercall into
separate 'start' and 'finalise' calls.
The 'pkvm_start_teardown_vm' hypercall puts the VM into a new 'is_dying'
state, which is a point of no return past which no vCPU of the pVM is
allowed to run any more. Once in this new state,
'pkvm_finalize_teardown_vm' can be used to reclaim meta-data and
page-table pages from the VM. A subsequent patch will add support for
reclaiming the individual guest memory pages.
Co-developed-by: Quentin Perret <qperret@google.com>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Will Deacon <will@kernel.org>
---
arch/arm64/include/asm/kvm_asm.h | 3 ++-
arch/arm64/include/asm/kvm_host.h | 7 +++++
arch/arm64/kvm/hyp/include/nvhe/pkvm.h | 4 ++-
arch/arm64/kvm/hyp/nvhe/hyp-main.c | 14 +++++++---
arch/arm64/kvm/hyp/nvhe/pkvm.c | 36 ++++++++++++++++++++++----
arch/arm64/kvm/pkvm.c | 7 ++++-
6 files changed, 60 insertions(+), 11 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index 2076005e9253..2b5ceaed0d7e 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h
@@ -89,7 +89,8 @@ enum __kvm_host_smccc_func {
__KVM_HOST_SMCCC_FUNC___pkvm_unreserve_vm,
__KVM_HOST_SMCCC_FUNC___pkvm_init_vm,
__KVM_HOST_SMCCC_FUNC___pkvm_init_vcpu,
- __KVM_HOST_SMCCC_FUNC___pkvm_teardown_vm,
+ __KVM_HOST_SMCCC_FUNC___pkvm_start_teardown_vm,
+ __KVM_HOST_SMCCC_FUNC___pkvm_finalize_teardown_vm,
__KVM_HOST_SMCCC_FUNC___pkvm_vcpu_load,
__KVM_HOST_SMCCC_FUNC___pkvm_vcpu_put,
__KVM_HOST_SMCCC_FUNC___pkvm_tlb_flush_vmid,
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index ac7f970c7883..3191d10a2622 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -255,6 +255,13 @@ struct kvm_protected_vm {
struct kvm_hyp_memcache stage2_teardown_mc;
bool is_protected;
bool is_created;
+
+ /*
+ * True when the guest is being torn down. When in this state, the
+ * guest's vCPUs can't be loaded anymore, but its pages can be
+ * reclaimed by the host.
+ */
+ bool is_dying;
};
struct kvm_mpidr_data {
diff --git a/arch/arm64/kvm/hyp/include/nvhe/pkvm.h b/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
index 184ad7a39950..04c7ca703014 100644
--- a/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
+++ b/arch/arm64/kvm/hyp/include/nvhe/pkvm.h
@@ -73,7 +73,9 @@ int __pkvm_init_vm(struct kvm *host_kvm, unsigned long vm_hva,
unsigned long pgd_hva);
int __pkvm_init_vcpu(pkvm_handle_t handle, struct kvm_vcpu *host_vcpu,
unsigned long vcpu_hva);
-int __pkvm_teardown_vm(pkvm_handle_t handle);
+
+int __pkvm_start_teardown_vm(pkvm_handle_t handle);
+int __pkvm_finalize_teardown_vm(pkvm_handle_t handle);
struct pkvm_hyp_vcpu *pkvm_load_hyp_vcpu(pkvm_handle_t handle,
unsigned int vcpu_idx);
diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
index eb5cfe32b2c9..ebcbdd1cb232 100644
--- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c
+++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
@@ -550,11 +550,18 @@ static void handle___pkvm_init_vcpu(struct kvm_cpu_context *host_ctxt)
cpu_reg(host_ctxt, 1) = __pkvm_init_vcpu(handle, host_vcpu, vcpu_hva);
}
-static void handle___pkvm_teardown_vm(struct kvm_cpu_context *host_ctxt)
+static void handle___pkvm_start_teardown_vm(struct kvm_cpu_context *host_ctxt)
{
DECLARE_REG(pkvm_handle_t, handle, host_ctxt, 1);
- cpu_reg(host_ctxt, 1) = __pkvm_teardown_vm(handle);
+ cpu_reg(host_ctxt, 1) = __pkvm_start_teardown_vm(handle);
+}
+
+static void handle___pkvm_finalize_teardown_vm(struct kvm_cpu_context *host_ctxt)
+{
+ DECLARE_REG(pkvm_handle_t, handle, host_ctxt, 1);
+
+ cpu_reg(host_ctxt, 1) = __pkvm_finalize_teardown_vm(handle);
}
typedef void (*hcall_t)(struct kvm_cpu_context *);
@@ -595,7 +602,8 @@ static const hcall_t host_hcall[] = {
HANDLE_FUNC(__pkvm_unreserve_vm),
HANDLE_FUNC(__pkvm_init_vm),
HANDLE_FUNC(__pkvm_init_vcpu),
- HANDLE_FUNC(__pkvm_teardown_vm),
+ HANDLE_FUNC(__pkvm_start_teardown_vm),
+ HANDLE_FUNC(__pkvm_finalize_teardown_vm),
HANDLE_FUNC(__pkvm_vcpu_load),
HANDLE_FUNC(__pkvm_vcpu_put),
HANDLE_FUNC(__pkvm_tlb_flush_vmid),
diff --git a/arch/arm64/kvm/hyp/nvhe/pkvm.c b/arch/arm64/kvm/hyp/nvhe/pkvm.c
index 8911338961c5..7f8191f96fc3 100644
--- a/arch/arm64/kvm/hyp/nvhe/pkvm.c
+++ b/arch/arm64/kvm/hyp/nvhe/pkvm.c
@@ -256,7 +256,10 @@ struct pkvm_hyp_vcpu *pkvm_load_hyp_vcpu(pkvm_handle_t handle,
hyp_spin_lock(&vm_table_lock);
hyp_vm = get_vm_by_handle(handle);
- if (!hyp_vm || hyp_vm->kvm.created_vcpus <= vcpu_idx)
+ if (!hyp_vm || hyp_vm->kvm.arch.pkvm.is_dying)
+ goto unlock;
+
+ if (hyp_vm->kvm.created_vcpus <= vcpu_idx)
goto unlock;
hyp_vcpu = hyp_vm->vcpus[vcpu_idx];
@@ -829,7 +832,32 @@ teardown_donated_memory(struct kvm_hyp_memcache *mc, void *addr, size_t size)
unmap_donated_memory_noclear(addr, size);
}
-int __pkvm_teardown_vm(pkvm_handle_t handle)
+int __pkvm_start_teardown_vm(pkvm_handle_t handle)
+{
+ struct pkvm_hyp_vm *hyp_vm;
+ int ret = 0;
+
+ hyp_spin_lock(&vm_table_lock);
+ hyp_vm = get_vm_by_handle(handle);
+ if (!hyp_vm) {
+ ret = -ENOENT;
+ goto unlock;
+ } else if (WARN_ON(hyp_page_count(hyp_vm))) {
+ ret = -EBUSY;
+ goto unlock;
+ } else if (hyp_vm->kvm.arch.pkvm.is_dying) {
+ ret = -EINVAL;
+ goto unlock;
+ }
+
+ hyp_vm->kvm.arch.pkvm.is_dying = true;
+unlock:
+ hyp_spin_unlock(&vm_table_lock);
+
+ return ret;
+}
+
+int __pkvm_finalize_teardown_vm(pkvm_handle_t handle)
{
struct kvm_hyp_memcache *mc, *stage2_mc;
struct pkvm_hyp_vm *hyp_vm;
@@ -843,9 +871,7 @@ int __pkvm_teardown_vm(pkvm_handle_t handle)
if (!hyp_vm) {
err = -ENOENT;
goto err_unlock;
- }
-
- if (WARN_ON(hyp_page_count(hyp_vm))) {
+ } else if (!hyp_vm->kvm.arch.pkvm.is_dying) {
err = -EBUSY;
goto err_unlock;
}
diff --git a/arch/arm64/kvm/pkvm.c b/arch/arm64/kvm/pkvm.c
index 20d50abb3b94..a39dacd1d617 100644
--- a/arch/arm64/kvm/pkvm.c
+++ b/arch/arm64/kvm/pkvm.c
@@ -88,7 +88,7 @@ void __init kvm_hyp_reserve(void)
static void __pkvm_destroy_hyp_vm(struct kvm *kvm)
{
if (pkvm_hyp_vm_is_created(kvm)) {
- WARN_ON(kvm_call_hyp_nvhe(__pkvm_teardown_vm,
+ WARN_ON(kvm_call_hyp_nvhe(__pkvm_finalize_teardown_vm,
kvm->arch.pkvm.handle));
} else if (kvm->arch.pkvm.handle) {
/*
@@ -350,6 +350,11 @@ void pkvm_pgtable_stage2_destroy_range(struct kvm_pgtable *pgt,
if (!handle)
return;
+ if (pkvm_hyp_vm_is_created(kvm) && !kvm->arch.pkvm.is_dying) {
+ WARN_ON(kvm_call_hyp_nvhe(__pkvm_start_teardown_vm, handle));
+ kvm->arch.pkvm.is_dying = true;
+ }
+
__pkvm_pgtable_stage2_unshare(pgt, addr, addr + size);
}
--
2.52.0.457.g6b5491de43-goog
next prev parent reply other threads:[~2026-01-19 12:47 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-19 12:45 [PATCH v2 00/35] KVM: arm64: Add support for protected guest memory with pKVM Will Deacon
2026-01-19 12:45 ` [PATCH v2 01/35] KVM: arm64: Invert KVM_PGTABLE_WALK_HANDLE_FAULT to fix pKVM walkers Will Deacon
2026-01-19 12:45 ` [PATCH v2 02/35] KVM: arm64: Don't leak stage-2 page-table if VM fails to init under pKVM Will Deacon
2026-01-19 12:45 ` [PATCH v2 03/35] KVM: arm64: Move handle check into pkvm_pgtable_stage2_destroy_range() Will Deacon
2026-01-19 12:45 ` [PATCH v2 04/35] KVM: arm64: Rename __pkvm_pgtable_stage2_unmap() Will Deacon
2026-01-19 12:45 ` [PATCH v2 05/35] KVM: arm64: Don't advertise unsupported features for protected guests Will Deacon
2026-01-19 12:45 ` [PATCH v2 06/35] KVM: arm64: Expose self-hosted debug regs as RAZ/WI " Will Deacon
2026-01-19 12:46 ` [PATCH v2 07/35] KVM: arm64: Remove is_protected_kvm_enabled() checks from hypercalls Will Deacon
2026-02-10 14:53 ` Alexandru Elisei
2026-03-03 15:45 ` Will Deacon
2026-03-06 11:33 ` Alexandru Elisei
2026-01-19 12:46 ` [PATCH v2 08/35] KVM: arm64: Ignore MMU notifier callbacks for protected VMs Will Deacon
2026-01-19 12:46 ` [PATCH v2 09/35] KVM: arm64: Prevent unsupported memslot operations on " Will Deacon
2026-01-19 12:46 ` [PATCH v2 10/35] KVM: arm64: Ignore -EAGAIN when mapping in pages for the pKVM host Will Deacon
2026-01-19 12:46 ` Will Deacon [this message]
2026-01-19 12:46 ` [PATCH v2 12/35] KVM: arm64: Introduce __pkvm_host_donate_guest() Will Deacon
2026-01-19 12:46 ` [PATCH v2 13/35] KVM: arm64: Hook up donation hypercall to pkvm_pgtable_stage2_map() Will Deacon
2026-01-19 12:46 ` [PATCH v2 14/35] KVM: arm64: Handle aborts from protected VMs Will Deacon
2026-02-12 10:37 ` Alexandru Elisei
2026-03-04 14:06 ` Will Deacon
2026-03-06 11:34 ` Alexandru Elisei
2026-03-11 10:24 ` Fuad Tabba
2026-01-19 12:46 ` [PATCH v2 15/35] KVM: arm64: Introduce __pkvm_reclaim_dying_guest_page() Will Deacon
2026-01-19 12:46 ` [PATCH v2 16/35] KVM: arm64: Hook up reclaim hypercall to pkvm_pgtable_stage2_destroy() Will Deacon
2026-01-19 12:46 ` [PATCH v2 17/35] KVM: arm64: Refactor enter_exception64() Will Deacon
2026-01-19 12:46 ` [PATCH v2 18/35] KVM: arm64: Inject SIGSEGV on illegal accesses Will Deacon
2026-01-19 12:46 ` [PATCH v2 19/35] KVM: arm64: Avoid pointless annotation when mapping host-owned pages Will Deacon
2026-01-19 12:46 ` [PATCH v2 20/35] KVM: arm64: Generalise kvm_pgtable_stage2_set_owner() Will Deacon
2026-01-19 12:46 ` [PATCH v2 21/35] KVM: arm64: Introduce host_stage2_set_owner_metadata_locked() Will Deacon
2026-01-19 12:46 ` [PATCH v2 22/35] KVM: arm64: Change 'pkvm_handle_t' to u16 Will Deacon
2026-01-28 10:28 ` Fuad Tabba
2026-01-19 12:46 ` [PATCH v2 23/35] KVM: arm64: Annotate guest donations with handle and gfn in host stage-2 Will Deacon
2026-01-28 10:29 ` Fuad Tabba
2026-01-19 12:46 ` [PATCH v2 24/35] KVM: arm64: Introduce hypercall to force reclaim of a protected page Will Deacon
2026-02-12 17:18 ` Alexandru Elisei
2026-03-04 14:08 ` Will Deacon
2026-01-19 12:46 ` [PATCH v2 25/35] KVM: arm64: Reclaim faulting page from pKVM in spurious fault handler Will Deacon
2026-02-12 17:22 ` Alexandru Elisei
2026-03-04 14:06 ` Will Deacon
2026-01-19 12:46 ` [PATCH v2 26/35] KVM: arm64: Return -EFAULT from VCPU_RUN on access to a poisoned pte Will Deacon
2026-01-19 12:46 ` [PATCH v2 27/35] KVM: arm64: Add hvc handler at EL2 for hypercalls from protected VMs Will Deacon
2026-01-19 12:46 ` [PATCH v2 28/35] KVM: arm64: Implement the MEM_SHARE hypercall for " Will Deacon
2026-01-19 12:46 ` [PATCH v2 29/35] KVM: arm64: Implement the MEM_UNSHARE " Will Deacon
2026-01-19 12:46 ` [PATCH v2 30/35] KVM: arm64: Allow userspace to create protected VMs when pKVM is enabled Will Deacon
2026-01-19 12:46 ` [PATCH v2 31/35] KVM: arm64: Add some initial documentation for pKVM Will Deacon
2026-01-19 12:46 ` [PATCH v2 32/35] KVM: arm64: Extend pKVM page ownership selftests to cover guest donation Will Deacon
2026-01-19 12:46 ` [PATCH v2 33/35] KVM: arm64: Register 'selftest_vm' in the VM table Will Deacon
2026-01-19 12:46 ` [PATCH v2 34/35] KVM: arm64: Extend pKVM page ownership selftests to cover forced reclaim Will Deacon
2026-01-19 12:46 ` [PATCH v2 35/35] KVM: arm64: Extend pKVM page ownership selftests to cover guest hvcs Will Deacon
2026-02-10 18:58 ` [PATCH v2 00/35] KVM: arm64: Add support for protected guest memory with pKVM Trilok Soni
2026-02-10 19:03 ` Fuad Tabba
2026-02-16 10:58 ` Venkata Rao Kakani
2026-02-16 11:00 ` Fuad Tabba
2026-02-17 10:43 ` Venkata Rao Kakani
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=20260119124629.2563-12-will@kernel.org \
--to=will@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=joey.gouly@arm.com \
--cc=kvmarm@lists.linux.dev \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=maz@kernel.org \
--cc=oupton@kernel.org \
--cc=qperret@google.com \
--cc=smostafa@google.com \
--cc=suzuki.poulose@arm.com \
--cc=tabba@google.com \
--cc=vdonnefort@google.com \
--cc=yuzenghui@huawei.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