"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: VMX: Always clear vmx->fail on emulation_required\n\nRevert a relatively recent change that set vmx->fail if the vCPU is in L2\nand emulation_required is true, as that behavior is completely bogus.\nSetting vmx->fail and synthesizing a VM-Exit is contradictory and wrong:\n\n (a) it's impossible to have both a VM-Fail and VM-Exit\n (b) vmcs.EXIT_REASON is not modified on VM-Fail\n (c) emulation_required refers to guest state and guest state checks are\n always VM-Exits, not VM-Fails.\n\nFor KVM specifically, emulation_required is handled before nested exits\nin __vmx_handle_exit(), thus setting vmx->fail has no immediate effect,\ni.e. KVM calls into handle_invalid_guest_state() and vmx->fail is ignored.\nSetting vmx->fail can ultimately result in a WARN in nested_vmx_vmexit()\nfiring when tearing down the VM as KVM never expects vmx->fail to be set\nwhen L2 is active, KVM always reflects those errors into L1.\n\n ------------[ cut here ]------------\n WARNING: CPU: 0 PID: 21158 at arch/x86/kvm/vmx/nested.c:4548\n nested_vmx_vmexit+0x16bd/0x17e0\n arch/x86/kvm/vmx/nested.c:4547\n Modules linked in:\n CPU: 0 PID: 21158 Comm: syz-executor.1 Not tainted 5.16.0-rc3-syzkaller #0\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011\n RIP: 0010:nested_vmx_vmexit+0x16bd/0x17e0 arch/x86/kvm/vmx/nested.c:4547\n Code: <0f> 0b e9 2e f8 ff ff e8 57 b3 5d 00 0f 0b e9 00 f1 ff ff 89 e9 80\n Call Trace:\n vmx_leave_nested arch/x86/kvm/vmx/nested.c:6220 [inline]\n nested_vmx_free_vcpu+0x83/0xc0 arch/x86/kvm/vmx/nested.c:330\n vmx_free_vcpu+0x11f/0x2a0 arch/x86/kvm/vmx/vmx.c:6799\n kvm_arch_vcpu_destroy+0x6b/0x240 arch/x86/kvm/x86.c:10989\n kvm_vcpu_destroy+0x29/0x90 arch/x86/kvm/../../../virt/kvm/kvm_main.c:441\n kvm_free_vcpus arch/x86/kvm/x86.c:11426 [inline]\n kvm_arch_destroy_vm+0x3ef/0x6b0 arch/x86/kvm/x86.c:11545\n kvm_destroy_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:1189 [inline]\n kvm_put_kvm+0x751/0xe40 arch/x86/kvm/../../../virt/kvm/kvm_main.c:1220\n kvm_vcpu_release+0x53/0x60 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3489\n __fput+0x3fc/0x870 fs/file_table.c:280\n task_work_run+0x146/0x1c0 kernel/task_work.c:164\n exit_task_work include/linux/task_work.h:32 [inline]\n do_exit+0x705/0x24f0 kernel/exit.c:832\n do_group_exit+0x168/0x2d0 kernel/exit.c:929\n get_signal+0x1740/0x2120 kernel/signal.c:2852\n arch_do_signal_or_restart+0x9c/0x730 arch/x86/kernel/signal.c:868\n handle_signal_work kernel/entry/common.c:148 [inline]\n exit_to_user_mode_loop kernel/entry/common.c:172 [inline]\n exit_to_user_mode_prepare+0x191/0x220 kernel/entry/common.c:207\n __syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]\n syscall_exit_to_user_mode+0x2e/0x70 kernel/entry/common.c:300\n do_syscall_64+0x53/0xd0 arch/x86/entry/common.c:86\n entry_SYSCALL_64_after_hwframe+0x44/0xae"
"value":"En el kernel de Linux, se resolvi\u00f3 la siguiente vulnerabilidad: KVM: VMX: borre siempre vmx->fail en emulation_required Revierta un cambio relativamente reciente que establece vmx->fail si la vCPU est\u00e1 en L2 y emulation_required es verdadero, ya que ese comportamiento es completamente falso. Configurar vmx->fail y sintetizar una VM-Exit es contradictorio y incorrecto: (a) es imposible tener VM-Fail y VM-Exit (b) vmcs.EXIT_REASON no se modifica en VM-Fail (c) emulation_required se refiere al estado invitado y las comprobaciones del estado invitado son siempre salidas de VM, no fallas de VM. Para KVM espec\u00edficamente, emulation_required se maneja antes de las salidas anidadas en __vmx_handle_exit(), por lo que configurar vmx->fail no tiene un efecto inmediato, es decir, las llamadas de KVM a handle_invalid_guest_state() y vmx->fail se ignoran. Configurar vmx->fail puede, en \u00faltima instancia, generar una ADVERTENCIA en nested_vmx_vmexit() al desactivar la VM, ya que KVM nunca espera que vmx->fail se configure cuando L2 est\u00e1 activo, KVM siempre refleja esos errores en L1. ------------[ cortar aqu\u00ed ]------------ ADVERTENCIA: CPU: 0 PID: 21158 en arch/x86/kvm/vmx/nested.c:4548 nested_vmx_vmexit+0x16bd/0x17e0 arch/x86/kvm/vmx/nested.c:4547 M\u00f3dulos vinculados en: CPU: 0 PID: 21158 Comm: syz-executor.1 No contaminado 5.16.0-rc3-syzkaller #0 Nombre de hardware: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:nested_vmx_vmexit+0x16bd/0x17e0 arch/x86/kvm/vmx/nested.c:4547 C\u00f3digo: <0f> 0b e9 2e f8 ff ff e8 57 b3 5d 00 0f 0b e9 00 f1 ff ff 89 e9 80 Seguimiento de llamadas: vmx_leave_nested arch/x86/kvm/vmx/nested.c:6220 [en l\u00ednea] nested_vmx_free_vcpu+0x83/0xc0 arch/x86/kvm/vmx/nested.c: 330 vmx_free_vcpu+0x11f/0x2a0 arch/x86/kvm/vmx/vmx.c:6799 kvm_arch_vcpu_destroy+0x6b/0x240 arch/x86/kvm/x86.c:10989 kvm_vcpu_destroy+0x29/0x90 arch/x86/kvm/.. /. ./../virt/kvm/kvm_main.c:441 kvm_free_vcpus arch/x86/kvm/x86.c:11426 [en l\u00ednea] kvm_arch_destroy_vm+0x3ef/0x6b0 arch/x86/kvm/x86.c:11545 kvm_destroy_vm arch/x86/ kvm/../../../virt/kvm/kvm_main.c:1189 [en l\u00ednea] kvm_put_kvm+0x751/0xe40 arch/x86/kvm/../../../virt/kvm/kvm_main.c :1220 kvm_vcpu_release+0x53/0x60 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3489 __fput+0x3fc/0x870 fs/file_table.c:280 task_work_run+0x146/0x1c0 kernel/ task_work.c:164 exit_task_work include/linux/task_work.h:32 [en l\u00ednea] do_exit+0x705/0x24f0 kernel/exit.c:832 do_group_exit+0x168/0x2d0 kernel/exit.c:929 get_signal+0x1740/0x2120 kernel/se\u00f1al .c:2852 arch_do_signal_or_restart+0x9c/0x730 arch/x86/kernel/signal.c:868 handle_signal_work kernel/entry/common.c:148 [en l\u00ednea] exit_to_user_mode_loop kernel/entry/common.c:172 [en l\u00ednea] exit_to_user_mode_prepare+0x191/ 0x220 kernel/entry/common.c:207 __syscall_exit_to_user_mode_work kernel/entry/common.c:289 [en l\u00ednea] syscall_exit_to_user_mode+0x2e/0x70 kernel/entry/common.c:300 do_syscall_64+0x53/0xd0 arch/x86/entry/common. c:86 entrada_SYSCALL_64_after_hwframe+0x44/0xae"
