From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Liran Alon <liran.alon@oracle.com>
Cc: qemu-devel@nongnu.org, pbonzini@redhat.com, mtosatti@redhat.com,
rth@twiddle.net, ehabkost@redhat.com, kvm@vger.kernel.org,
jmattson@google.com, maran.wilson@oracle.com
Subject: Re: [QEMU PATCH v3 4/9] KVM: i386: Block migration for vCPUs exposed with nested virtualization
Date: Tue, 18 Jun 2019 09:44:20 +0100 [thread overview]
Message-ID: <20190618084420.GA2850@work-vm> (raw)
In-Reply-To: <20190617175658.135869-5-liran.alon@oracle.com>
* Liran Alon (liran.alon@oracle.com) wrote:
> Commit d98f26073beb ("target/i386: kvm: add VMX migration blocker")
> added a migration blocker for vCPU exposed with Intel VMX.
> However, migration should also be blocked for vCPU exposed with
> AMD SVM.
>
> Both cases should be blocked because QEMU should extract additional
> vCPU state from KVM that should be migrated as part of vCPU VMState.
> E.g. Whether vCPU is running in guest-mode or host-mode.
>
> Fixes: d98f26073beb ("target/i386: kvm: add VMX migration blocker")
> Signed-off-by: Liran Alon <liran.alon@oracle.com>
> ---
> target/i386/cpu.c | 6 ------
> target/i386/cpu.h | 24 ++++++++++++++++++++++++
> target/i386/kvm.c | 12 ++++++------
> 3 files changed, 30 insertions(+), 12 deletions(-)
>
> diff --git a/target/i386/cpu.c b/target/i386/cpu.c
> index 536d7d152044..197201087e65 100644
> --- a/target/i386/cpu.c
> +++ b/target/i386/cpu.c
> @@ -5170,12 +5170,6 @@ static int x86_cpu_filter_features(X86CPU *cpu)
> return rv;
> }
>
> -#define IS_INTEL_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 && \
> - (env)->cpuid_vendor2 == CPUID_VENDOR_INTEL_2 && \
> - (env)->cpuid_vendor3 == CPUID_VENDOR_INTEL_3)
> -#define IS_AMD_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_AMD_1 && \
> - (env)->cpuid_vendor2 == CPUID_VENDOR_AMD_2 && \
> - (env)->cpuid_vendor3 == CPUID_VENDOR_AMD_3)
> static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
> {
> CPUState *cs = CPU(dev);
> diff --git a/target/i386/cpu.h b/target/i386/cpu.h
> index fce6660bac00..79d9495ceb0c 100644
> --- a/target/i386/cpu.h
> +++ b/target/i386/cpu.h
> @@ -728,6 +728,13 @@ typedef uint32_t FeatureWordArray[FEATURE_WORDS];
>
> #define CPUID_VENDOR_HYGON "HygonGenuine"
>
> +#define IS_INTEL_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_INTEL_1 && \
> + (env)->cpuid_vendor2 == CPUID_VENDOR_INTEL_2 && \
> + (env)->cpuid_vendor3 == CPUID_VENDOR_INTEL_3)
> +#define IS_AMD_CPU(env) ((env)->cpuid_vendor1 == CPUID_VENDOR_AMD_1 && \
> + (env)->cpuid_vendor2 == CPUID_VENDOR_AMD_2 && \
> + (env)->cpuid_vendor3 == CPUID_VENDOR_AMD_3)
> +
> #define CPUID_MWAIT_IBE (1U << 1) /* Interrupts can exit capability */
> #define CPUID_MWAIT_EMX (1U << 0) /* enumeration supported */
>
> @@ -1866,6 +1873,23 @@ static inline int32_t x86_get_a20_mask(CPUX86State *env)
> }
> }
>
> +static inline bool cpu_has_vmx(CPUX86State *env)
> +{
> + return (IS_INTEL_CPU(env) &&
> + (env->features[FEAT_1_ECX] & CPUID_EXT_VMX));
> +}
> +
> +static inline bool cpu_has_svm(CPUX86State *env)
> +{
> + return (IS_AMD_CPU(env) &&
> + (env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_SVM));
Note that the Hygon Dhyana seems to have SVM set as well, see
target/i386/cpu.c and search for "Dhyana"
(Note the AMD users are going to be a bit more surprised by this
restriction than Intel users, since Nested has been enabled on AMD for a
long time).
Dave
> +}
> +
> +static inline bool cpu_has_nested_virt(CPUX86State *env)
> +{
> + return (cpu_has_vmx(env) || cpu_has_svm(env));
> +}
> +
> /* fpu_helper.c */
> void update_fp_status(CPUX86State *env);
> void update_mxcsr_status(CPUX86State *env);
> diff --git a/target/i386/kvm.c b/target/i386/kvm.c
> index c8fd53055d37..f43e2d69859e 100644
> --- a/target/i386/kvm.c
> +++ b/target/i386/kvm.c
> @@ -906,7 +906,7 @@ static int hyperv_init_vcpu(X86CPU *cpu)
> }
>
> static Error *invtsc_mig_blocker;
> -static Error *vmx_mig_blocker;
> +static Error *nested_virt_mig_blocker;
>
> #define KVM_MAX_CPUID_ENTRIES 100
>
> @@ -1270,13 +1270,13 @@ int kvm_arch_init_vcpu(CPUState *cs)
> !!(c->ecx & CPUID_EXT_SMX);
> }
>
> - if ((env->features[FEAT_1_ECX] & CPUID_EXT_VMX) && !vmx_mig_blocker) {
> - error_setg(&vmx_mig_blocker,
> - "Nested VMX virtualization does not support live migration yet");
> - r = migrate_add_blocker(vmx_mig_blocker, &local_err);
> + if (cpu_has_nested_virt(env) && !nested_virt_mig_blocker) {
> + error_setg(&nested_virt_mig_blocker,
> + "Nested virtualization does not support live migration yet");
> + r = migrate_add_blocker(nested_virt_mig_blocker, &local_err);
> if (local_err) {
> error_report_err(local_err);
> - error_free(vmx_mig_blocker);
> + error_free(nested_virt_mig_blocker);
> return r;
> }
> }
> --
> 2.20.1
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
next prev parent reply other threads:[~2019-06-18 8:44 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-17 17:56 [QEMU PATCH v3 0/9]: KVM: i386: Add support for save and restore of nested state Liran Alon
2019-06-17 17:56 ` [QEMU PATCH v3 1/9] KVM: Introduce kvm_arch_destroy_vcpu() Liran Alon
2019-06-18 22:15 ` [Qemu-devel] " Maran Wilson
2019-06-17 17:56 ` [QEMU PATCH v3 2/9] KVM: i386: Use symbolic constant for #DB/#BP exception constants Liran Alon
2019-06-17 17:56 ` [QEMU PATCH v3 3/9] KVM: i386: Re-inject #DB to guest with updated DR6 Liran Alon
2019-06-17 17:56 ` [QEMU PATCH v3 4/9] KVM: i386: Block migration for vCPUs exposed with nested virtualization Liran Alon
2019-06-18 8:44 ` Dr. David Alan Gilbert [this message]
2019-06-18 22:16 ` [Qemu-devel] " Maran Wilson
2019-06-17 17:56 ` [QEMU PATCH v3 5/9] linux-headers: i386: Modify struct kvm_nested_state to have explicit fields for data Liran Alon
2019-06-18 22:16 ` [Qemu-devel] " Maran Wilson
2019-06-17 17:56 ` [QEMU PATCH v3 6/9] vmstate: Add support for kernel integer types Liran Alon
2019-06-18 8:55 ` Dr. David Alan Gilbert
2019-06-18 15:36 ` Liran Alon
2019-06-18 15:42 ` Dr. David Alan Gilbert
2019-06-18 16:44 ` Paolo Bonzini
2019-06-17 17:56 ` [QEMU PATCH v3 7/9] KVM: i386: Add support for save and restore nested state Liran Alon
2019-06-18 9:03 ` Dr. David Alan Gilbert
2019-06-18 15:40 ` Liran Alon
2019-06-18 15:48 ` Dr. David Alan Gilbert
2019-06-18 15:50 ` Liran Alon
2019-06-18 16:16 ` Paolo Bonzini
2019-06-18 22:16 ` [Qemu-devel] " Maran Wilson
2019-06-17 17:56 ` [QEMU PATCH v3 8/9] KVM: i386: Add support for KVM_CAP_EXCEPTION_PAYLOAD Liran Alon
2019-06-18 9:07 ` Dr. David Alan Gilbert
2019-06-18 15:45 ` Liran Alon
2019-06-17 17:56 ` [QEMU PATCH v3 9/9] KVM: i386: Remove VMX migration blocker Liran Alon
2019-06-18 22:17 ` [Qemu-devel] " Maran Wilson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190618084420.GA2850@work-vm \
--to=dgilbert@redhat.com \
--cc=ehabkost@redhat.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=liran.alon@oracle.com \
--cc=maran.wilson@oracle.com \
--cc=mtosatti@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).