diff options
| author | Paolo Bonzini <pbonzini@redhat.com> | 2024-10-11 10:39:58 +0200 |
|---|---|---|
| committer | Paolo Bonzini <pbonzini@redhat.com> | 2024-10-17 19:41:30 +0200 |
| commit | 60de433d4cb17211dbea5c1e6b74d476dec16370 (patch) | |
| tree | 28312f85103f2291a18c414631d61a298c50b654 /accel | |
| parent | 64e0e63ea16aa0122dc0c41a0679da0ae4616208 (diff) | |
accel/kvm: check for KVM_CAP_MULTI_ADDRESS_SPACE on vm
KVM_CAP_MULTI_ADDRESS_SPACE used to be a global capability, but with the
introduction of AMD SEV-SNP confidential VMs, the number of address spaces
can vary by VM type.
Query the extension on the VM level instead of on the KVM level.
Inspired by an analogous patch by Tom Dohrmann.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'accel')
| -rw-r--r-- | accel/kvm/kvm-all.c | 12 |
1 files changed, 6 insertions, 6 deletions
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 99746436bf..d985486a05 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -2618,12 +2618,6 @@ static int kvm_init(MachineState *ms) s->nr_slots_max = KVM_MEMSLOTS_NR_MAX_DEFAULT; } - s->nr_as = kvm_check_extension(s, KVM_CAP_MULTI_ADDRESS_SPACE); - if (s->nr_as <= 1) { - s->nr_as = 1; - } - s->as = g_new0(struct KVMAs, s->nr_as); - type = find_kvm_machine_type(ms); if (type < 0) { ret = -EINVAL; @@ -2637,6 +2631,12 @@ static int kvm_init(MachineState *ms) s->vmfd = ret; + s->nr_as = kvm_vm_check_extension(s, KVM_CAP_MULTI_ADDRESS_SPACE); + if (s->nr_as <= 1) { + s->nr_as = 1; + } + s->as = g_new0(struct KVMAs, s->nr_as); + /* check the vcpu limits */ soft_vcpus_limit = kvm_recommended_vcpus(s); hard_vcpus_limit = kvm_max_vcpus(s); |