From: Radim Krčmář Date: Tue, 23 Oct 2018 14:31:38 +0000 (+0200) Subject: Revert "kvm: x86: optimize dr6 restore" X-Git-Tag: v4.20-rc1~113^2~1 X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=f9dcf08e2017cad12d771cdd862c8c314b885615;p=users%2Fhch%2Fconfigfs.git Revert "kvm: x86: optimize dr6 restore" This reverts commit 0e0a53c551317654e2d7885fdfd23299fee99b6b. As Christian Ehrhardt noted: The most common case is that vcpu->arch.dr6 and the host's %dr6 value are not related at all because ->switch_db_regs is zero. To do this all correctly, we must handle the case where the guest leaves an arbitrary unused value in vcpu->arch.dr6 before disabling breakpoints again. However, this means that vcpu->arch.dr6 is not suitable to detect the need for a %dr6 clear. Signed-off-by: Radim Krčmář --- diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index bdcb5babfb68..66d66d77caee 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3271,16 +3271,11 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) kvm_x86_ops->vcpu_put(vcpu); vcpu->arch.last_host_tsc = rdtsc(); /* - * Here dr6 is either zero or, if the guest has run and userspace - * has not set any breakpoints or watchpoints, it can be set to - * the guest dr6 (stored in vcpu->arch.dr6). do_debug expects dr6 - * to be cleared after it runs, so clear the host register. However, - * MOV to DR can be expensive when running nested, omit it if - * vcpu->arch.dr6 is already zero: in that case, the host dr6 cannot - * currently be nonzero. + * If userspace has set any breakpoints or watchpoints, dr6 is restored + * on every vmexit, but if not, we might have a stale dr6 from the + * guest. do_debug expects dr6 to be cleared after it runs, do the same. */ - if (vcpu->arch.dr6) - set_debugreg(0, 6); + set_debugreg(0, 6); } static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu,