kvm: Fix IRQ injection into full queue

User space may only inject interrupts during kvm_arch_pre_run if
ready_for_interrupt_injection is set in kvm_run. But that field is
updated on exit from KVM_RUN, so we must ensure that we enter the
kernel after potentially queuing an interrupt, otherwise we risk to
loose one - like it happens with the current code against latest
kernel modules (since kvm-86) that started to queue only a single
interrupt.

Fix the problem by reordering kvm_cpu_exec.

Credits go to Gleb Natapov for analyzing the issue in details.

(cherry picked from commit 8c14c173)

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>

kvm-all.c

@@ -451,14 +451,13 @@ int kvm_cpu_exec(CPUState *env)
     dprintf("kvm_cpu_exec()\n");
 
     do {
-        kvm_arch_pre_run(env, run);
-
         if (env->exit_request) {
             dprintf("interrupt exit requested\n");
             ret = 0;
             break;
         }
 
+        kvm_arch_pre_run(env, run);
         ret = kvm_vcpu_ioctl(env, KVM_RUN, 0);
         kvm_arch_post_run(env, run);