From 178a63f8515a461a6625b856d25f4d4154ada960 Mon Sep 17 00:00:00 2001 From: "plai@redhat.com" Date: Mon, 23 Sep 2019 20:40:26 +0200 Subject: [PATCH 10/12] i386: kvm: Disable arch_capabilities if MSR can't be set RH-Author: plai@redhat.com Message-id: <1569271227-28026-10-git-send-email-plai@redhat.com> Patchwork-id: 90858 O-Subject: [RHEL7.8 qemu-kvm PATCH v6 09/10] i386: kvm: Disable arch_capabilities if MSR can't be set Bugzilla: 1709971 RH-Acked-by: Eduardo Habkost RH-Acked-by: Bandan Das RH-Acked-by: Miroslav Rezanina From: Eduardo Habkost KVM has two bugs in the handling of MSR_IA32_ARCH_CAPABILITIES: 1) Linux commit commit 1eaafe91a0df ("kvm: x86: IA32_ARCH_CAPABILITIES is always supported") makes GET_SUPPORTED_CPUID return arch_capabilities even if running on SVM. This makes "-cpu host,migratable=off" incorrectly expose arch_capabilities on CPUID on AMD hosts (where the MSR is not emulated by KVM). 2) KVM_GET_MSR_INDEX_LIST does not return MSR_IA32_ARCH_CAPABILITIES if the MSR is not supported by the host CPU. This makes QEMU not initialize the MSR properly at kvm_put_msrs() on those hosts. Work around both bugs on the QEMU side, by checking if the MSR was returned by KVM_GET_MSR_INDEX_LIST before returning the feature flag on kvm_arch_get_supported_cpuid(). This has the unfortunate side effect of making arch_capabilities unavailable on hosts without hardware support for the MSR until bug #2 is fixed on KVM, but I can't see another way to work around bug #1 without that side effect. Signed-off-by: Eduardo Habkost Message-Id: <20190125220606.4864-2-ehabkost@redhat.com> Signed-off-by: Eduardo Habkost (cherry picked from commit 485b1d256bcb0874bcde0223727c159b6837e6f8) Signed-off-by: Paul Lai Signed-off-by: Miroslav Rezanina Conflicts: target/i386/kvm.c changes to target-i386/kvm.c --- target-i386/kvm.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/target-i386/kvm.c b/target-i386/kvm.c index 12781a8..c79b0ea 100644 --- a/target-i386/kvm.c +++ b/target-i386/kvm.c @@ -235,6 +235,15 @@ uint32_t kvm_arch_get_supported_cpuid(KVMState *s, uint32_t function, if (!kvm_irqchip_in_kernel()) { ret &= ~CPUID_EXT_X2APIC; } + } else if (function == 7 && index == 0 && reg == R_EDX) { + /* + * Linux v4.17-v4.20 incorrectly return ARCH_CAPABILITIES on SVM hosts. + * We can detect the bug by checking if MSR_IA32_ARCH_CAPABILITIES is + * returned by KVM_GET_MSR_INDEX_LIST. + */ + if (!has_msr_arch_capabs) { + ret &= ~CPUID_7_0_EDX_ARCH_CAPABILITIES; + } } else if (function == 0x80000001 && reg == R_EDX) { /* On Intel, kvm returns cpuid according to the Intel spec, * so add missing bits according to the AMD spec: -- 1.8.3.1