On 2011-07-19 16:12, Juan Quintela wrote: > Marcus Meissner <[email protected]> wrote: >> Hi, >> >> Reported to us by Stephan Mueller of atsec. >> >> Several .pad struct members are not set to something, so they >> will leak stack content back to user space. >> >> Ciao, Marcus >> >> Reported-by: Stephan Mueller <[email protected]> >> Signed-off-by: Marcus Meissner <[email protected]> >> --- >> arch/x86/kvm/x86.c | 1 + >> 1 files changed, 1 insertions(+), 0 deletions(-) >> >> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c >> index 77c9d86..621ffb6 100644 >> --- a/arch/x86/kvm/x86.c >> +++ b/arch/x86/kvm/x86.c >> @@ -3002,6 +3002,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp, >> case KVM_GET_VCPU_EVENTS: { >> struct kvm_vcpu_events events; >> >> + memset(&events, 0, sizeof(events)); >> kvm_vcpu_ioctl_x86_get_vcpu_events(vcpu, &events); >> >> r = -EFAULT; > > Looking at arch/x86/include/asm/kvm.h & arch/x86/kvm/x86.c I can't see > what pad fields are not initialized. My reading is that everything is > initialized in kvm_vcpu_ioctl_x86_get_vcpu_events(). What field are you > refering to?
Good question. Information leaks were once addressed by 97e69aa62f, and kvm_vcpu_events was not changed since then. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
