Commit d43583b890e7 ("KVM: arm64: Expose PSCI SYSTEM_RESET2 call to the
guest") hooked up the SYSTEM_RESET2 PSCI call for guests but failed to
preserve its arguments for userspace, instead overwriting them with
zeroes via smccc_set_retval(). As Linux only passes zeroes for these
arguments, this appeared to be working for Linux guests. Oh well.
Don't call smccc_set_retval() for a SYSTEM_RESET2 heading to userspace
and instead set X0 (and only X0) explicitly to PSCI_RET_INTERNAL_FAILURE
just in case the vCPU re-enters the guest.
Fixes: d43583b890e7 ("KVM: arm64: Expose PSCI SYSTEM_RESET2 call to the guest")
Reported-by: Andrew Walbran <qwandor@xxxxxxxxxx>
Signed-off-by: Will Deacon <will@xxxxxxxxxx>
---
arch/arm64/kvm/psci.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/arch/arm64/kvm/psci.c b/arch/arm64/kvm/psci.c
index ecb4b74cb12a..b0c920717632 100644
--- a/arch/arm64/kvm/psci.c
+++ b/arch/arm64/kvm/psci.c
@@ -367,14 +367,14 @@ static int kvm_psci_1_x_call(struct kvm_vcpu *vcpu, u32 minor)
if (minor >= 1) {
arg = smccc_get_arg1(vcpu);
- if (arg > PSCI_1_1_RESET_TYPE_SYSTEM_WARM_RESET &&
- arg < PSCI_1_1_RESET_TYPE_VENDOR_START) {
- val = PSCI_RET_INVALID_PARAMS;
- } else {
+ if (arg <= PSCI_1_1_RESET_TYPE_SYSTEM_WARM_RESET ||
+ arg >= PSCI_1_1_RESET_TYPE_VENDOR_START) {
kvm_psci_system_reset2(vcpu);
- val = PSCI_RET_INTERNAL_FAILURE;
- ret = 0;
+ vcpu_set_reg(vcpu, 0, PSCI_RET_INTERNAL_FAILURE);
+ return 0;
}
+
+ val = PSCI_RET_INVALID_PARAMS;
break;
};
fallthrough;
--
2.35.1.616.g0bdcbb4464-goog
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
