Documentation/virt/kvm/api.rst | 11 +++++++++ arch/arm64/include/uapi/asm/kvm.h | 6 +++++ arch/arm64/kvm/hyp/nvhe/psci-relay.c | 2 ++ arch/arm64/kvm/hypercalls.c | 2 ++ arch/arm64/kvm/psci.c | 43 +++++++++++++++++++++++++++++++++++- drivers/firmware/psci/psci.c | 35 +++++++++++++++++++++++++++++ include/kvm/arm_psci.h | 4 +++- include/uapi/linux/psci.h | 20 +++++++++++++++++ kernel/power/hibernate.c | 5 ++++- 9 files changed, 125 insertions(+), 3 deletions(-)
The PSCI v1.3 spec (https://developer.arm.com/documentation/den0022,
currently in Alpha state, hence 'RFC') adds support for a SYSTEM_OFF2
function enabling a HIBERNATE_OFF state which is analogous to ACPI S4.
This will allow hosting environments to determine that a guest is
hibernated rather than just powered off, and ensure that they preserve
the virtual environment appropriately to allow the guest to resume
safely (or bump the hardware_signature in the FACS to trigger a clean
reboot instead).
This updates KVM to support advertising PSCI v1.3, and unconditionally
enables the SYSTEM_OFF2 support when PSCI v1.3 is enabled. For now,
KVM defaults to PSCI v1.2 unless explicitly requested.
For the guest side, add a new SYS_OFF_MODE_POWER_OFF handler with higher
priority than the EFI one, but which *only* triggers when there's a
hibernation in progress. There are other ways to do this (see the commit
message for more details) but this seemed like the simplest.
Version 2 of the patch series splits out the psci.h definitions into a
separate commit (a dependency for both the guest and KVM side), and adds
definitions for the other new functions added in v1.3. It also moves the
pKVM psci-relay support to a separate commit; although in arch/arm64/kvm
that's actually about the *guest* side of SYSTEM_OFF2 (i.e. using it
from the host kernel, relayed through nVHE).
Version 3 dropped the KVM_CAP which allowed userspace to explicitly opt
in to the new feature like with SYSTEM_SUSPEND, and makes it depend only
on PSCI v1.3 being exposed to the guest.
David Woodhouse (5):
firmware/psci: Add definitions for PSCI v1.3 specification (ALPHA)
KVM: arm64: Add support for PSCI v1.2 and v1.3
KVM: arm64: Add PSCI v1.3 SYSTEM_OFF2 function for hibernation
KVM: arm64: nvhe: Pass through PSCI v1.3 SYSTEM_OFF2 call
arm64: Use SYSTEM_OFF2 PSCI call to power off for hibernate
Documentation/virt/kvm/api.rst | 11 +++++++++
arch/arm64/include/uapi/asm/kvm.h | 6 +++++
arch/arm64/kvm/hyp/nvhe/psci-relay.c | 2 ++
arch/arm64/kvm/hypercalls.c | 2 ++
arch/arm64/kvm/psci.c | 43 +++++++++++++++++++++++++++++++++++-
drivers/firmware/psci/psci.c | 35 +++++++++++++++++++++++++++++
include/kvm/arm_psci.h | 4 +++-
include/uapi/linux/psci.h | 20 +++++++++++++++++
kernel/power/hibernate.c | 5 ++++-
9 files changed, 125 insertions(+), 3 deletions(-)
On Tue, Mar 19, 2024 at 12:59:01PM +0000, David Woodhouse wrote: > David Woodhouse (5): > firmware/psci: Add definitions for PSCI v1.3 specification (ALPHA) > KVM: arm64: Add support for PSCI v1.2 and v1.3 > KVM: arm64: Add PSCI v1.3 SYSTEM_OFF2 function for hibernation > KVM: arm64: nvhe: Pass through PSCI v1.3 SYSTEM_OFF2 call > arm64: Use SYSTEM_OFF2 PSCI call to power off for hibernate If we're going down the route of having this PSCI call live in KVM, it really deserves a test. I think you can just pile on the existing psci_test selftest. -- Thanks, Oliver
On Tue, 2024-03-19 at 08:27 -0700, Oliver Upton wrote:
> If we're going down the route of having this PSCI call live in KVM, it
> really deserves a test. I think you can just pile on the existing
> psci_test selftest.
Added to
https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/psci-hibernate
for next time.
From 8c72a78e6179bc8970edc66a85ab6bee26f581fb Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw@amazon.co.uk>
Date: Tue, 19 Mar 2024 17:07:46 +0000
Subject: [PATCH 4/8] KVM: selftests: Add test for PSCI SYSTEM_OFF2
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
---
.../testing/selftests/kvm/aarch64/psci_test.c | 61 +++++++++++++++++++
1 file changed, 61 insertions(+)
diff --git a/tools/testing/selftests/kvm/aarch64/psci_test.c b/tools/testing/selftests/kvm/aarch64/psci_test.c
index 9b004905d1d3..1c1cf1580d70 100644
--- a/tools/testing/selftests/kvm/aarch64/psci_test.c
+++ b/tools/testing/selftests/kvm/aarch64/psci_test.c
@@ -54,6 +54,15 @@ static uint64_t psci_system_suspend(uint64_t entry_addr, uint64_t context_id)
return res.a0;
}
+static uint64_t psci_system_off2(uint64_t type)
+{
+ struct arm_smccc_res res;
+
+ smccc_hvc(PSCI_1_3_FN64_SYSTEM_OFF2, type, 0, 0, 0, 0, 0, 0, &res);
+
+ return res.a0;
+}
+
static uint64_t psci_features(uint32_t func_id)
{
struct arm_smccc_res res;
@@ -188,11 +197,63 @@ static void host_test_system_suspend(void)
kvm_vm_free(vm);
}
+static void guest_test_system_off2(void)
+{
+ uint64_t ret;
+
+ /* assert that SYSTEM_OFF2 is discoverable */
+ GUEST_ASSERT(psci_features(PSCI_1_3_FN_SYSTEM_OFF2) &
+ (1UL << PSCI_1_3_HIBERNATE_TYPE_OFF));
+ GUEST_ASSERT(psci_features(PSCI_1_3_FN64_SYSTEM_OFF2) &
+ (1UL << PSCI_1_3_HIBERNATE_TYPE_OFF));
+
+ ret = psci_system_off2(PSCI_1_3_HIBERNATE_TYPE_OFF);
+ GUEST_SYNC(ret);
+}
+
+static void host_test_system_off2(void)
+{
+ struct kvm_vcpu *source, *target;
+ uint64_t psci_version = 0;
+ struct kvm_run *run;
+ struct kvm_vm *vm;
+
+ vm = setup_vm(guest_test_system_off2, &source, &target);
+ vcpu_get_reg(target, KVM_REG_ARM_PSCI_VERSION, &psci_version);
+ TEST_ASSERT(psci_version >= PSCI_VERSION(0, 2),
+ "Unexpected PSCI version %lu.%lu",
+ PSCI_VERSION_MAJOR(psci_version),
+ PSCI_VERSION_MINOR(psci_version));
+
+ if (psci_version < PSCI_VERSION(1,3))
+ goto skip;
+
+ vcpu_power_off(target);
+ run = source->run;
+
+ enter_guest(source);
+
+ TEST_ASSERT_KVM_EXIT_REASON(source, KVM_EXIT_SYSTEM_EVENT);
+ TEST_ASSERT(run->system_event.type == KVM_SYSTEM_EVENT_SHUTDOWN,
+ "Unhandled system event: %u (expected: %u)",
+ run->system_event.type, KVM_SYSTEM_EVENT_SHUTDOWN);
+ TEST_ASSERT(run->system_event.ndata >= 1,
+ "Unexpected amount of system event data: %u (expected, >= 1)",
+ run->system_event.ndata);
+ TEST_ASSERT(run->system_event.data[0] & KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2,
+ "PSCI_OFF2 flag not set. Flags %llu (expected %llu)",
+ run->system_event.data[0], KVM_SYSTEM_EVENT_SHUTDOWN_FLAG_PSCI_OFF2);
+
+ skip:
+ kvm_vm_free(vm);
+}
+
int main(void)
{
TEST_REQUIRE(kvm_has_cap(KVM_CAP_ARM_SYSTEM_SUSPEND));
host_test_cpu_on();
host_test_system_suspend();
+ host_test_system_off2();
return 0;
}
--
2.34.1
On Tue, Mar 19, 2024 at 05:14:42PM +0000, David Woodhouse wrote: > On Tue, 2024-03-19 at 08:27 -0700, Oliver Upton wrote: > > If we're going down the route of having this PSCI call live in KVM, it > > really deserves a test. I think you can just pile on the existing > > psci_test selftest. > > Added to > https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/psci-hibernate > for next time. > > From 8c72a78e6179bc8970edc66a85ab6bee26f581fb Mon Sep 17 00:00:00 2001 > From: David Woodhouse <dwmw@amazon.co.uk> > Date: Tue, 19 Mar 2024 17:07:46 +0000 > Subject: [PATCH 4/8] KVM: selftests: Add test for PSCI SYSTEM_OFF2 > > Signed-off-by: David Woodhouse <dwmw@amazon.co.uk> Looks good, thanks! -- Thanks, Oliver
On Tue, 2024-03-19 at 12:41 -0700, Oliver Upton wrote: > On Tue, Mar 19, 2024 at 05:14:42PM +0000, David Woodhouse wrote: > > On Tue, 2024-03-19 at 08:27 -0700, Oliver Upton wrote: > > > If we're going down the route of having this PSCI call live in KVM, it > > > really deserves a test. I think you can just pile on the existing > > > psci_test selftest. > > > > Added to > > https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/psci-hibernate > > for next time. > > > > From 8c72a78e6179bc8970edc66a85ab6bee26f581fb Mon Sep 17 00:00:00 2001 > > From: David Woodhouse <dwmw@amazon.co.uk> > > Date: Tue, 19 Mar 2024 17:07:46 +0000 > > Subject: [PATCH 4/8] KVM: selftests: Add test for PSCI SYSTEM_OFF2 > > > > Signed-off-by: David Woodhouse <dwmw@amazon.co.uk> > > Looks good, thanks! Thanks. Marc, I think I've also addressed your feedback? Is there anything else to do other than wait for the spec to be published? Shall I post a v4 with PSCI v1.3 as default and the self-test? Would you apply that into a branch ready for merging when the spec is ready, or should I just wait and repost it all then?
On Fri, 22 Mar 2024 10:17:58 +0000, David Woodhouse <dwmw2@infradead.org> wrote: > > [1 <text/plain; UTF-8 (quoted-printable)>] > On Tue, 2024-03-19 at 12:41 -0700, Oliver Upton wrote: > > On Tue, Mar 19, 2024 at 05:14:42PM +0000, David Woodhouse wrote: > > > On Tue, 2024-03-19 at 08:27 -0700, Oliver Upton wrote: > > > > If we're going down the route of having this PSCI call live in KVM, it > > > > really deserves a test. I think you can just pile on the existing > > > > psci_test selftest. > > > > > > Added to > > > https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/psci-hibernate > > > for next time. > > > > > > From 8c72a78e6179bc8970edc66a85ab6bee26f581fb Mon Sep 17 00:00:00 2001 > > > From: David Woodhouse <dwmw@amazon.co.uk> > > > Date: Tue, 19 Mar 2024 17:07:46 +0000 > > > Subject: [PATCH 4/8] KVM: selftests: Add test for PSCI SYSTEM_OFF2 > > > > > > Signed-off-by: David Woodhouse <dwmw@amazon.co.uk> > > > > Looks good, thanks! > > Thanks. > > Marc, I think I've also addressed your feedback? Is there anything else > to do other than wait for the spec to be published? Other than the couple of minor nits I mentioned in replies to the individual patches, this looks good to me. > Shall I post a v4 with PSCI v1.3 as default and the self-test? Would > you apply that into a branch ready for merging when the spec is ready, > or should I just wait and repost it all then? I think this can wait for the final spec. I assume that you are directly tracking this anyway, so we don't need to poll for the spec update. Thanks, M. -- Without deviation from the norm, progress is not possible.
On Fri, 2024-03-22 at 16:09 +0000, Marc Zyngier wrote: > > > Marc, I think I've also addressed your feedback? Is there anything else > > to do other than wait for the spec to be published? > > Other than the couple of minor nits I mentioned in replies to the > individual patches, this looks good to me. I believe I've handled all that. And also Sudeep's implicit nudge to use BIT() instead of manually shifting (1<<PSCI_1_3_HIBERNATE_TYPE_OFF). Rebased onto 6.8 and pushed to https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/psci-hibernate-6.8 > > Shall I post a v4 with PSCI v1.3 as default and the self-test? Would > > you apply that into a branch ready for merging when the spec is ready, > > or should I just wait and repost it all then? > > I think this can wait for the final spec. I assume that you are > directly tracking this anyway, so we don't need to poll for the spec > update. Indeed, will post again when the spec is published. Thanks.
© 2016 - 2026 Red Hat, Inc.