Paolo Bonzini <[email protected]> wrote: > On 30/03/2015 16:40, Nadav Amit wrote: >> Paolo, >> >> It appears you are right and I have not tested 4 and 5 well enough. I’ll >> repost them (the others were tested presumably well enough). >> >> Two short questions: >> Can I use init.c in the kvm-unit-tests ? >> Why is it disabled? > > Because QEMU support for INIT is incomplete, so the tests would fail > ("Uh, hard reset!"). IIRC sending init to BSP with APIC_DEST_SELF is > also not supported by actual hardware (or at least not supported > "officially") so that test would also have to be changed to not use a > shortcut.
So, I would revive the init unit-test and disable the failing assertions, right? BTW: It appears that there is another bug - DR[0..3] are not reloaded after reset. Regards, Nadav -- >8 -- From: Nadav Amit <[email protected]> Subject: [PATCH] KVM: x86: DR0-DR3 are not clear on reset DR0-DR3 are not cleared as they should during reset and when they are set from userspace. It appears to be caused by c77fb5fe6f03 ("KVM: x86: Allow the guest to run with dirty debug registers"). Force their reload on these situations. Signed-off-by: Nadav Amit <[email protected]> --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index bf5a160..913ae41 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -345,6 +345,7 @@ struct kvm_pmu { enum { KVM_DEBUGREG_BP_ENABLED = 1, KVM_DEBUGREG_WONT_EXIT = 2, + KVM_DEBUGREG_RELOAD = 4, }; struct kvm_vcpu_arch { diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index cc2c759..1f65c3a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -801,6 +801,17 @@ unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_get_cr8); +static void kvm_update_dr0123(struct kvm_vcpu *vcpu) +{ + int i; + + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) { + for (i = 0; i < KVM_NR_DB_REGS; i++) + vcpu->arch.eff_db[i] = vcpu->arch.db[i]; + vcpu->arch.switch_db_regs |= KVM_DEBUGREG_RELOAD; + } +} + static void kvm_update_dr6(struct kvm_vcpu *vcpu) { if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) @@ -3150,6 +3161,7 @@ static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu, return -EINVAL; memcpy(vcpu->arch.db, dbgregs->db, sizeof(vcpu->arch.db)); + kvm_update_dr0123(vcpu); vcpu->arch.dr6 = dbgregs->dr6; kvm_update_dr6(vcpu); vcpu->arch.dr7 = dbgregs->dr7; @@ -6322,6 +6334,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) set_debugreg(vcpu->arch.eff_db[2], 2); set_debugreg(vcpu->arch.eff_db[3], 3); set_debugreg(vcpu->arch.dr6, 6); + vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_RELOAD; } trace_kvm_entry(vcpu->vcpu_id); @@ -7096,6 +7109,7 @@ void kvm_vcpu_reset(struct kvm_vcpu *vcpu) kvm_clear_exception_queue(vcpu); memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db)); + kvm_update_dr0123(vcpu); vcpu->arch.dr6 = DR6_INIT; kvm_update_dr6(vcpu); vcpu->arch.dr7 = DR7_FIXED_1; -- 1.9.1 -- 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
