Palacios Public Git Repository

To checkout Palacios execute

  git clone http://v3vee.org/palacios/palacios.web/palacios.git
This will give you the master branch. You probably want the devel branch or one of the release branches. To switch to the devel branch, simply execute
  cd palacios
  git checkout --track -b devel origin/devel
The other branches are similar.


changed the symbiotic calling interface to a synchronous model
[palacios.git] / palacios / src / palacios / svm.c
index 7a1c737..8f87be9 100644 (file)
@@ -46,7 +46,7 @@
 
 
 // This is a global pointer to the host's VMCB
-static addr_t host_vmcbs[CONFIG_MAX_CPUS] = {0};
+static addr_t host_vmcbs[CONFIG_MAX_CPUS] = { [0 ... CONFIG_MAX_CPUS - 1] = 0};
 
 
 
@@ -66,23 +66,15 @@ static vmcb_t * Allocate_VMCB() {
 
 
 
-static void Init_VMCB_BIOS(vmcb_t * vmcb, struct guest_info *vm_info) {
+static void Init_VMCB_BIOS(vmcb_t * vmcb, struct guest_info * vm_info) {
     vmcb_ctrl_t * ctrl_area = GET_VMCB_CTRL_AREA(vmcb);
     vmcb_saved_state_t * guest_state = GET_VMCB_SAVE_STATE_AREA(vmcb);
     uint_t i;
 
 
     //
-    guest_state->rsp = 0x00;
-    guest_state->rip = 0xfff0;
 
 
-    guest_state->cpl = 0;
-
-    guest_state->efer |= EFER_MSR_svm_enable;
-
-
-    guest_state->rflags = 0x00000002; // The reserved bit is always 1
     ctrl_area->svm_instrs.VMRUN = 1;
     ctrl_area->svm_instrs.VMMCALL = 1;
     ctrl_area->svm_instrs.VMLOAD = 1;
@@ -125,51 +117,75 @@ static void Init_VMCB_BIOS(vmcb_t * vmcb, struct guest_info *vm_info) {
     ctrl_area->instrs.PAUSE = 1;
     ctrl_area->instrs.shutdown_evts = 1;
 
+
+    /* DEBUG FOR RETURN CODE */
+    ctrl_area->exit_code = 1;
+
+
+    /* Setup Guest Machine state */
+
+    vm_info->vm_regs.rsp = 0x00;
+    vm_info->rip = 0xfff0;
+
     vm_info->vm_regs.rdx = 0x00000f00;
 
 
-    guest_state->cr0 = 0x60010010; // Set the WP flag so the memory hooks work in real-mode
+    vm_info->cpl = 0;
 
+    vm_info->ctrl_regs.rflags = 0x00000002; // The reserved bit is always 1
+    vm_info->ctrl_regs.cr0 = 0x60010010; // Set the WP flag so the memory hooks work in real-mode
+    vm_info->ctrl_regs.efer |= EFER_MSR_svm_enable;
 
-    guest_state->cs.selector = 0xf000;
-    guest_state->cs.limit = 0xffff;
-    guest_state->cs.base = 0x0000000f0000LL;
-    guest_state->cs.attrib.raw = 0xf3;
 
 
-    /* DEBUG FOR RETURN CODE */
-    ctrl_area->exit_code = 1;
 
 
-    struct vmcb_selector *segregs [] = {&(guest_state->ss), &(guest_state->ds), 
-                                       &(guest_state->es), &(guest_state->fs), 
-                                       &(guest_state->gs), NULL};
+    vm_info->segments.cs.selector = 0xf000;
+    vm_info->segments.cs.limit = 0xffff;
+    vm_info->segments.cs.base = 0x0000000f0000LL;
+
+    // (raw attributes = 0xf3)
+    vm_info->segments.cs.type = 0x3;
+    vm_info->segments.cs.system = 0x1;
+    vm_info->segments.cs.dpl = 0x3;
+    vm_info->segments.cs.present = 1;
+
+
+
+    struct v3_segment * segregs [] = {&(vm_info->segments.ss), &(vm_info->segments.ds), 
+                                     &(vm_info->segments.es), &(vm_info->segments.fs), 
+                                     &(vm_info->segments.gs), NULL};
 
     for ( i = 0; segregs[i] != NULL; i++) {
-       struct vmcb_selector * seg = segregs[i];
+       struct v3_segment * seg = segregs[i];
        
        seg->selector = 0x0000;
        //    seg->base = seg->selector << 4;
        seg->base = 0x00000000;
-       seg->attrib.raw = 0xf3;
        seg->limit = ~0u;
+
+       // (raw attributes = 0xf3)
+       seg->type = 0x3;
+       seg->system = 0x1;
+       seg->dpl = 0x3;
+       seg->present = 1;
     }
 
-    guest_state->gdtr.limit = 0x0000ffff;
-    guest_state->gdtr.base = 0x0000000000000000LL;
-    guest_state->idtr.limit = 0x0000ffff;
-    guest_state->idtr.base = 0x0000000000000000LL;
+    vm_info->segments.gdtr.limit = 0x0000ffff;
+    vm_info->segments.gdtr.base = 0x0000000000000000LL;
+    vm_info->segments.idtr.limit = 0x0000ffff;
+    vm_info->segments.idtr.base = 0x0000000000000000LL;
 
-    guest_state->ldtr.selector = 0x0000;
-    guest_state->ldtr.limit = 0x0000ffff;
-    guest_state->ldtr.base = 0x0000000000000000LL;
-    guest_state->tr.selector = 0x0000;
-    guest_state->tr.limit = 0x0000ffff;
-    guest_state->tr.base = 0x0000000000000000LL;
+    vm_info->segments.ldtr.selector = 0x0000;
+    vm_info->segments.ldtr.limit = 0x0000ffff;
+    vm_info->segments.ldtr.base = 0x0000000000000000LL;
+    vm_info->segments.tr.selector = 0x0000;
+    vm_info->segments.tr.limit = 0x0000ffff;
+    vm_info->segments.tr.base = 0x0000000000000000LL;
 
 
-    guest_state->dr6 = 0x00000000ffff0ff0LL;
-    guest_state->dr7 = 0x0000000000000400LL;
+    vm_info->dbg_regs.dr6 = 0x00000000ffff0ff0LL;
+    vm_info->dbg_regs.dr7 = 0x0000000000000400LL;
 
 
     v3_init_svm_io_map(vm_info);
@@ -205,7 +221,8 @@ static void Init_VMCB_BIOS(vmcb_t * vmcb, struct guest_info *vm_info) {
        vm_info->shdw_pg_state.guest_cr0 = 0x0000000000000010LL;
        PrintDebug("Created\n");
        
-       guest_state->cr3 = vm_info->direct_map_pt;
+       vm_info->ctrl_regs.cr0 |= 0x80000000;
+       vm_info->ctrl_regs.cr3 = vm_info->direct_map_pt;
 
        ctrl_area->cr_reads.cr0 = 1;
        ctrl_area->cr_writes.cr0 = 1;
@@ -225,7 +242,7 @@ static void Init_VMCB_BIOS(vmcb_t * vmcb, struct guest_info *vm_info) {
 
        guest_state->g_pat = 0x7040600070406ULL;
 
-       guest_state->cr0 |= 0x80000000;
+
 
     } else if (vm_info->shdw_pg_mode == NESTED_PAGING) {
        // Flush the TLB on entries/exits
@@ -266,100 +283,315 @@ static int init_svm_guest(struct guest_info * info, struct v3_vm_config * config
     return 0;
 }
 
-static int start_svm_guest(struct guest_info *info) {
-    //    vmcb_saved_state_t * guest_state = GET_VMCB_SAVE_STATE_AREA((vmcb_t*)(info->vmm_data));
-    //  vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
-    uint_t num_exits = 0;
 
 
+static int update_irq_state_atomic(struct guest_info * info) {
+    vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
 
-    PrintDebug("Launching SVM VM (vmcb=%p)\n", (void *)info->vmm_data);
-    //PrintDebugVMCB((vmcb_t*)(info->vmm_data));
-    
-    info->run_state = VM_RUNNING;
-    rdtscll(info->yield_start_cycle);
+    if ((info->intr_state.irq_pending == 1) && (guest_ctrl->guest_ctrl.V_IRQ == 0)) {
+       
+#ifdef CONFIG_DEBUG_INTERRUPTS
+       PrintDebug("INTAK cycle completed for irq %d\n", info->intr_state.irq_vector);
+#endif
 
+       info->intr_state.irq_started = 1;
+       info->intr_state.irq_pending = 0;
 
-    while (1) {
-       ullong_t tmp_tsc;
-       
-       // Conditionally yield the CPU if the timeslice has expired
-       v3_yield_cond(info);
+       v3_injecting_intr(info, info->intr_state.irq_vector, V3_EXTERNAL_IRQ);
+    }
 
-       /*
-         PrintDebug("SVM Entry to CS=%p  rip=%p...\n", 
-         (void *)(addr_t)info->segments.cs.base, 
-         (void *)(addr_t)info->rip);
-       */
+    if ((info->intr_state.irq_started == 1) && (guest_ctrl->exit_int_info.valid == 0)) {
+#ifdef CONFIG_DEBUG_INTERRUPTS
+       PrintDebug("Interrupt %d taken by guest\n", info->intr_state.irq_vector);
+#endif
+
+       // Interrupt was taken fully vectored
+       info->intr_state.irq_started = 0;
+
+    } else {
+#ifdef CONFIG_DEBUG_INTERRUPTS
+       PrintDebug("EXIT INT INFO is set (vec=%d)\n", guest_ctrl->exit_int_info.vector);
+#endif
+    }
 
-       // disable global interrupts for vm state transition
-       v3_clgi();
+    return 0;
+}
 
 
+static int update_irq_state(struct guest_info * info) {
+    vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
 
-       rdtscll(info->time_state.cached_host_tsc);
-       //    guest_ctrl->TSC_OFFSET = info->time_state.guest_tsc - info->time_state.cached_host_tsc;
+    if (v3_excp_pending(info)) {
+       uint_t excp = v3_get_excp_number(info);
        
-       v3_svm_launch((vmcb_t *)V3_PAddr(info->vmm_data), &(info->vm_regs), (vmcb_t *)host_vmcbs[info->cpu_id]);
+       guest_ctrl->EVENTINJ.type = SVM_INJECTION_EXCEPTION;
        
-       rdtscll(tmp_tsc);
-
+       if (info->excp_state.excp_error_code_valid) {
+           guest_ctrl->EVENTINJ.error_code = info->excp_state.excp_error_code;
+           guest_ctrl->EVENTINJ.ev = 1;
+#ifdef CONFIG_DEBUG_INTERRUPTS
+           PrintDebug("Injecting exception %d with error code %x\n", excp, guest_ctrl->EVENTINJ.error_code);
+#endif
+       }
+       
+       guest_ctrl->EVENTINJ.vector = excp;
        
-       //PrintDebug("SVM Returned\n");
+       guest_ctrl->EVENTINJ.valid = 1;
+
+       PrintDebug("<%d> Injecting Exception %d (CR2=%p) (EIP=%p)\n", 
+                  (int)info->num_exits, 
+                  guest_ctrl->EVENTINJ.vector, 
+                  (void *)(addr_t)info->ctrl_regs.cr2,
+                  (void *)(addr_t)info->rip);
+
 
-       // reenable global interrupts after vm exit
-       v3_stgi();
+#ifdef CONFIG_DEBUG_INTERRUPTS
+       PrintDebug("Injecting Exception %d (EIP=%p)\n", 
+                  guest_ctrl->EVENTINJ.vector, 
+                  (void *)(addr_t)info->rip);
+#endif
+       v3_injecting_excp(info, excp);
 
 
-       // Conditionally yield the CPU if the timeslice has expired
-       v3_yield_cond(info);
+    } else if (info->intr_state.irq_started == 1) {
+#ifdef CONFIG_DEBUG_INTERRUPTS
+       PrintDebug("IRQ pending from previous injection\n");
+#endif
+       guest_ctrl->guest_ctrl.V_IRQ = 1;
+       guest_ctrl->guest_ctrl.V_INTR_VECTOR = info->intr_state.irq_vector;
+       guest_ctrl->guest_ctrl.V_IGN_TPR = 1;
+       guest_ctrl->guest_ctrl.V_INTR_PRIO = 0xf;
 
+    } else {
+       switch (v3_intr_pending(info)) {
+           case V3_EXTERNAL_IRQ: {
+               uint32_t irq = v3_get_intr(info);
+
+               guest_ctrl->guest_ctrl.V_IRQ = 1;
+               guest_ctrl->guest_ctrl.V_INTR_VECTOR = irq;
+               guest_ctrl->guest_ctrl.V_IGN_TPR = 1;
+               guest_ctrl->guest_ctrl.V_INTR_PRIO = 0xf;
+
+#ifdef CONFIG_DEBUG_INTERRUPTS
+               PrintDebug("Injecting Interrupt %d (EIP=%p)\n", 
+                          guest_ctrl->guest_ctrl.V_INTR_VECTOR, 
+                          (void *)(addr_t)info->rip);
+#endif
 
-       v3_update_time(info, tmp_tsc - info->time_state.cached_host_tsc);
-       num_exits++;
+               info->intr_state.irq_pending = 1;
+               info->intr_state.irq_vector = irq;
+               
+               break;
+           }
+           case V3_NMI:
+               guest_ctrl->EVENTINJ.type = SVM_INJECTION_NMI;
+               break;
+           case V3_SOFTWARE_INTR:
+               guest_ctrl->EVENTINJ.type = SVM_INJECTION_SOFT_INTR;
+               break;
+           case V3_VIRTUAL_IRQ:
+               guest_ctrl->EVENTINJ.type = SVM_INJECTION_IRQ;
+               break;
+
+           case V3_INVALID_INTR:
+           default:
+               break;
+       }
        
-       if ((num_exits % 5000) == 0) {
-           PrintDebug("SVM Exit number %d\n", num_exits);
+    }
+
+    return 0;
+}
+
+
+/* 
+ * CAUTION and DANGER!!! 
+ * 
+ * The VMCB CANNOT(!!) be accessed outside of the clgi/stgi calls inside this function
+ * When exectuing a symbiotic call the VMCB WILL be overwritten, so any dependencies 
+ * on its contents will cause things to break. The contents at the time of the exit WILL 
+ * change before the exit handler is executed.
+ */
+int v3_svm_enter(struct guest_info * info) {
+    vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
+    vmcb_saved_state_t * guest_state = GET_VMCB_SAVE_STATE_AREA((vmcb_t*)(info->vmm_data)); 
+    ullong_t tmp_tsc;
+    addr_t exit_code = 0, exit_info1 = 0, exit_info2 = 0;
+
+    // Conditionally yield the CPU if the timeslice has expired
+    v3_yield_cond(info);
+
+    // disable global interrupts for vm state transition
+    v3_clgi();
+
+    // Synchronize the guest state to the VMCB
+    guest_state->cr0 = info->ctrl_regs.cr0;
+    guest_state->cr2 = info->ctrl_regs.cr2;
+    guest_state->cr3 = info->ctrl_regs.cr3;
+    guest_state->cr4 = info->ctrl_regs.cr4;
+    guest_state->dr6 = info->dbg_regs.dr6;
+    guest_state->dr7 = info->dbg_regs.dr7;
+    guest_ctrl->guest_ctrl.V_TPR = info->ctrl_regs.cr8 & 0xff;
+    guest_state->rflags = info->ctrl_regs.rflags;
+    guest_state->efer = info->ctrl_regs.efer;
+    
+    guest_state->cpl = info->cpl;
+
+    v3_set_vmcb_segments((vmcb_t*)(info->vmm_data), &(info->segments));
+
+    guest_state->rax = info->vm_regs.rax;
+    guest_state->rip = info->rip;
+    guest_state->rsp = info->vm_regs.rsp;
+    /* ** */
+
+    /*
+      PrintDebug("SVM Entry to CS=%p  rip=%p...\n", 
+      (void *)(addr_t)info->segments.cs.base, 
+      (void *)(addr_t)info->rip);
+    */
+
+#ifdef CONFIG_SYMBIOTIC
+    if (info->sym_state.sym_call_active == 1) {
+       if (guest_ctrl->guest_ctrl.V_IRQ == 1) {
+           V3_Print("!!! Injecting Interrupt during Sym call !!!\n");
        }
+    }
+#endif
+
 
-       if (v3_handle_svm_exit(info) != 0) {
+    rdtscll(info->time_state.cached_host_tsc);
+    //    guest_ctrl->TSC_OFFSET = info->time_state.guest_tsc - info->time_state.cached_host_tsc;
+       
+    v3_svm_launch((vmcb_t *)V3_PAddr(info->vmm_data), &(info->vm_regs), (vmcb_t *)host_vmcbs[info->cpu_id]);
+    
+    rdtscll(tmp_tsc);
+
+    //PrintDebug("SVM Returned\n");
+    
+    info->num_exits++;
+
+    v3_update_time(info, tmp_tsc - info->time_state.cached_host_tsc);
+
+#ifdef CONFIG_SYMBIOTIC
+    if (info->sym_state.sym_call_active == 0) {
+       update_irq_state_atomic(info);
+    }
+#else 
+    update_irq_state_atomic(info);
+#endif
+
+    // Save Guest state from VMCB
+    info->rip = guest_state->rip;
+    info->vm_regs.rsp = guest_state->rsp;
+    info->vm_regs.rax = guest_state->rax;
+
+    info->cpl = guest_state->cpl;
+
+    info->ctrl_regs.cr0 = guest_state->cr0;
+    info->ctrl_regs.cr2 = guest_state->cr2;
+    info->ctrl_regs.cr3 = guest_state->cr3;
+    info->ctrl_regs.cr4 = guest_state->cr4;
+    info->dbg_regs.dr6 = guest_state->dr6;
+    info->dbg_regs.dr7 = guest_state->dr7;
+    info->ctrl_regs.cr8 = guest_ctrl->guest_ctrl.V_TPR;
+    info->ctrl_regs.rflags = guest_state->rflags;
+    info->ctrl_regs.efer = guest_state->efer;
+    
+    v3_get_vmcb_segments((vmcb_t*)(info->vmm_data), &(info->segments));
+    info->cpu_mode = v3_get_vm_cpu_mode(info);
+    info->mem_mode = v3_get_vm_mem_mode(info);
+    /* ** */
+
+
+    // save exit info here
+    exit_code = guest_ctrl->exit_code;
+    exit_info1 = guest_ctrl->exit_info1;
+    exit_info2 = guest_ctrl->exit_info2;
+
+
+    // reenable global interrupts after vm exit
+    v3_stgi();
+
+    // Conditionally yield the CPU if the timeslice has expired
+    v3_yield_cond(info);
+
+
+    if (v3_handle_svm_exit(info, exit_code, exit_info1, exit_info2) != 0) {
+       PrintError("Error in SVM exit handler\n");
+       return -1;
+    }
+
+#ifdef CONFIG_SYMBIOTIC
+    if (info->sym_state.sym_call_active == 0) {
+       update_irq_state(info);
+    }
+#else 
+    update_irq_state(info);
+#endif
+
+    return 0;
+}
+
+
+static int start_svm_guest(struct guest_info *info) {
+    //    vmcb_saved_state_t * guest_state = GET_VMCB_SAVE_STATE_AREA((vmcb_t*)(info->vmm_data));
+    //  vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
+
+
+
+    PrintDebug("Launching SVM VM (vmcb=%p)\n", (void *)info->vmm_data);
+    //PrintDebugVMCB((vmcb_t*)(info->vmm_data));
+    
+    info->run_state = VM_RUNNING;
+    rdtscll(info->yield_start_cycle);
+
+
+    while (1) {
+       if (v3_svm_enter(info) == -1) {
            vmcb_ctrl_t * guest_ctrl = GET_VMCB_CTRL_AREA((vmcb_t*)(info->vmm_data));
            addr_t host_addr;
            addr_t linear_addr = 0;
-
+           
            info->run_state = VM_ERROR;
            
-           PrintDebug("SVM ERROR!!\n"); 
-      
+           V3_Print("SVM ERROR!!\n"); 
+           
            v3_print_guest_state(info);
-
-           PrintDebug("SVM Exit Code: %p\n", (void *)(addr_t)guest_ctrl->exit_code); 
-      
-           PrintDebug("exit_info1 low = 0x%.8x\n", *(uint_t*)&(guest_ctrl->exit_info1));
-           PrintDebug("exit_info1 high = 0x%.8x\n", *(uint_t *)(((uchar_t *)&(guest_ctrl->exit_info1)) + 4));
-      
-           PrintDebug("exit_info2 low = 0x%.8x\n", *(uint_t*)&(guest_ctrl->exit_info2));
-           PrintDebug("exit_info2 high = 0x%.8x\n", *(uint_t *)(((uchar_t *)&(guest_ctrl->exit_info2)) + 4));
-      
+           
+           V3_Print("SVM Exit Code: %p\n", (void *)(addr_t)guest_ctrl->exit_code); 
+           
+           V3_Print("exit_info1 low = 0x%.8x\n", *(uint_t*)&(guest_ctrl->exit_info1));
+           V3_Print("exit_info1 high = 0x%.8x\n", *(uint_t *)(((uchar_t *)&(guest_ctrl->exit_info1)) + 4));
+           
+           V3_Print("exit_info2 low = 0x%.8x\n", *(uint_t*)&(guest_ctrl->exit_info2));
+           V3_Print("exit_info2 high = 0x%.8x\n", *(uint_t *)(((uchar_t *)&(guest_ctrl->exit_info2)) + 4));
+           
            linear_addr = get_addr_linear(info, info->rip, &(info->segments.cs));
-
+           
            if (info->mem_mode == PHYSICAL_MEM) {
                guest_pa_to_host_va(info, linear_addr, &host_addr);
            } else if (info->mem_mode == VIRTUAL_MEM) {
                guest_va_to_host_va(info, linear_addr, &host_addr);
            }
-
-           PrintDebug("Host Address of rip = 0x%p\n", (void *)host_addr);
-
-           PrintDebug("Instr (15 bytes) at %p:\n", (void *)host_addr);
+           
+           V3_Print("Host Address of rip = 0x%p\n", (void *)host_addr);
+           
+           V3_Print("Instr (15 bytes) at %p:\n", (void *)host_addr);
            v3_dump_mem((uint8_t *)host_addr, 15);
-
-
+           
            v3_print_stack(info);
 
-
            break;
        }
+       
+       if ((info->num_exits % 5000) == 0) {
+           V3_Print("SVM Exit number %d\n", (uint32_t)info->num_exits);
+       }
+
+
+       
+
     }
     return 0;
 }
@@ -379,7 +611,7 @@ int v3_is_svm_capable() {
     PrintDebug("CPUID_EXT_FEATURE_IDS_ecx=0x%x\n", ecx);
 
     if ((ecx & CPUID_EXT_FEATURE_IDS_ecx_svm_avail) == 0) {
-      PrintDebug("SVM Not Available\n");
+      V3_Print("SVM Not Available\n");
       return 0;
     }  else {
        v3_get_msr(SVM_VM_CR_MSR, &vm_cr_high, &vm_cr_low);
@@ -387,21 +619,21 @@ int v3_is_svm_capable() {
        PrintDebug("SVM_VM_CR_MSR = 0x%x 0x%x\n", vm_cr_high, vm_cr_low);
        
        if ((vm_cr_low & SVM_VM_CR_MSR_svmdis) == 1) {
-           PrintDebug("SVM is available but is disabled.\n");
+           V3_Print("SVM is available but is disabled.\n");
            
            v3_cpuid(CPUID_SVM_REV_AND_FEATURE_IDS, &eax, &ebx, &ecx, &edx);
            
            PrintDebug("CPUID_SVM_REV_AND_FEATURE_IDS_edx=0x%x\n", edx);
            
            if ((edx & CPUID_SVM_REV_AND_FEATURE_IDS_edx_svml) == 0) {
-               PrintDebug("SVM BIOS Disabled, not unlockable\n");
+               V3_Print("SVM BIOS Disabled, not unlockable\n");
            } else {
-               PrintDebug("SVM is locked with a key\n");
+               V3_Print("SVM is locked with a key\n");
            }
            return 0;
 
        } else {
-           PrintDebug("SVM is available and  enabled.\n");
+           V3_Print("SVM is available and  enabled.\n");
 
            v3_cpuid(CPUID_SVM_REV_AND_FEATURE_IDS, &eax, &ebx, &ecx, &edx);
            PrintDebug("CPUID_SVM_REV_AND_FEATURE_IDS_eax=0x%x\n", eax);
@@ -409,13 +641,6 @@ int v3_is_svm_capable() {
            PrintDebug("CPUID_SVM_REV_AND_FEATURE_IDS_ecx=0x%x\n", ecx);
            PrintDebug("CPUID_SVM_REV_AND_FEATURE_IDS_edx=0x%x\n", edx);
 
-
-           if ((edx & CPUID_SVM_REV_AND_FEATURE_IDS_edx_np) == 0) {
-               PrintDebug("SVM Nested Paging not supported\n");
-           } else {
-               PrintDebug("SVM Nested Paging supported\n");
-           }
-
            return 1;
        }
     }
@@ -429,10 +654,10 @@ static int has_svm_nested_paging() {
     //PrintDebug("CPUID_EXT_FEATURE_IDS_edx=0x%x\n", edx);
 
     if ((edx & CPUID_SVM_REV_AND_FEATURE_IDS_edx_np) == 0) {
-       PrintDebug("SVM Nested Paging not supported\n");
+       V3_Print("SVM Nested Paging not supported\n");
        return 0;
     } else {
-       PrintDebug("SVM Nested Paging supported\n");
+       V3_Print("SVM Nested Paging supported\n");
        return 1;
     }
 }
@@ -447,7 +672,7 @@ void v3_init_svm_cpu(int cpu_id) {
     msr.e_reg.low |= EFER_MSR_svm_enable;
     v3_set_msr(EFER_MSR, 0, msr.e_reg.low);
 
-    PrintDebug("SVM Enabled\n");
+    V3_Print("SVM Enabled\n");
 
     // Setup the host state save area
     host_vmcbs[cpu_id] = (addr_t)V3_AllocPages(4);