[PULL 40/63] KVM: track whether guest state is encrypted

[Paolo Bonzini]

So far, KVM has allowed KVM_GET/SET_* ioctls to execute even if the
guest state is encrypted, in which case they do nothing.  For the new
API using VM types, instead, the ioctls will fail which is a safer and
more robust approach.

The new API will be the only one available for SEV-SNP and TDX, but it
is also usable for SEV and SEV-ES.  In preparation for that, require
architecture-specific KVM code to communicate the point at which guest
state is protected (which must be after kvm_cpu_synchronize_post_init(),
though that might change in the future in order to suppor migration).
>From that point, skip reading registers so that cpu->vcpu_dirty is
never true: if it ever becomes true, kvm_arch_put_registers() will
fail miserably.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 include/sysemu/kvm.h     |  2 ++
 include/sysemu/kvm_int.h |  1 +
 accel/kvm/kvm-all.c      | 17 ++++++++++++++---
 target/i386/sev.c        |  1 +
 4 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
index 2cba899270c..14b1ddb3be1 100644
--- a/include/sysemu/kvm.h
+++ b/include/sysemu/kvm.h
@@ -539,6 +539,8 @@ bool kvm_dirty_ring_enabled(void);
 
 uint32_t kvm_dirty_ring_size(void);
 
+void kvm_mark_guest_state_protected(void);
+
 /**
  * kvm_hwpoisoned_mem - indicate if there is any hwpoisoned page
  * reported for the VM.
diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h
index 94488d2c1a2..227b61fec3d 100644
--- a/include/sysemu/kvm_int.h
+++ b/include/sysemu/kvm_int.h
@@ -87,6 +87,7 @@ struct KVMState
     bool kernel_irqchip_required;
     OnOffAuto kernel_irqchip_split;
     bool sync_mmu;
+    bool guest_state_protected;
     uint64_t manual_dirty_log_protect;
     /* The man page (and posix) say ioctl numbers are signed int, but
      * they're not.  Linux, glibc and *BSD all treat ioctl numbers as
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index d58916e33ae..d6ebadbf386 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2703,7 +2703,7 @@ bool kvm_cpu_check_are_resettable(void)
 
 static void do_kvm_cpu_synchronize_state(CPUState *cpu, run_on_cpu_data arg)
 {
-    if (!cpu->vcpu_dirty) {
+    if (!cpu->vcpu_dirty && !kvm_state->guest_state_protected) {
         int ret = kvm_arch_get_registers(cpu);
         if (ret) {
             error_report("Failed to get registers: %s", strerror(-ret));
@@ -2717,7 +2717,7 @@ static void do_kvm_cpu_synchronize_state(CPUState *cpu, 
run_on_cpu_data arg)
 
 void kvm_cpu_synchronize_state(CPUState *cpu)
 {
-    if (!cpu->vcpu_dirty) {
+    if (!cpu->vcpu_dirty && !kvm_state->guest_state_protected) {
         run_on_cpu(cpu, do_kvm_cpu_synchronize_state, RUN_ON_CPU_NULL);
     }
 }
@@ -2752,7 +2752,13 @@ static void do_kvm_cpu_synchronize_post_init(CPUState 
*cpu, run_on_cpu_data arg)
 
 void kvm_cpu_synchronize_post_init(CPUState *cpu)
 {
-    run_on_cpu(cpu, do_kvm_cpu_synchronize_post_init, RUN_ON_CPU_NULL);
+    if (!kvm_state->guest_state_protected) {
+        /*
+         * This runs before the machine_init_done notifiers, and is the last
+         * opportunity to synchronize the state of confidential guests.
+         */
+        run_on_cpu(cpu, do_kvm_cpu_synchronize_post_init, RUN_ON_CPU_NULL);
+    }
 }
 
 static void do_kvm_cpu_synchronize_pre_loadvm(CPUState *cpu, run_on_cpu_data 
arg)
@@ -4099,3 +4105,8 @@ void query_stats_schemas_cb(StatsSchemaList **result, 
Error **errp)
         query_stats_schema_vcpu(first_cpu, &stats_args);
     }
 }
+
+void kvm_mark_guest_state_protected(void)
+{
+    kvm_state->guest_state_protected = true;
+}
diff --git a/target/i386/sev.c b/target/i386/sev.c
index b8f79d34d19..c49a8fd55eb 100644
--- a/target/i386/sev.c
+++ b/target/i386/sev.c
@@ -755,6 +755,7 @@ sev_launch_get_measure(Notifier *notifier, void *unused)
         if (ret) {
             exit(1);
         }
+        kvm_mark_guest_state_protected();
     }
 
     /* query the measurement blob length */
-- 
2.44.0