Re: [RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
On 10/11/2011 06:09 PM, Vadim Rozenfeld wrote: > Spinlocks too. TCG does its own round-robin scheduling, so having zero How? You need hypercall page for it. I'm not saying you should implement it, :) in principle TCG could have a hypercall page. Paolo -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
On Tue, 2011-10-11 at 17:52 +0200, Paolo Bonzini wrote: > On 10/11/2011 03:46 PM, Vadim Rozenfeld wrote: > > > These should not be KVM-specific. You should be able to add > > > enlightenments to a TCG VM. At the same time, the KVM leaves could be > > > moved to 0x4100 when enlightenments are active, similar to what Xen > > > does. > > > > > > > IMO, adding Hyper-V features without KVM support has little, if any, > > meaning. Relaxed timing is the only one thing, which can be activated > > without help from hypervisor. > > Spinlocks too. TCG does its own round-robin scheduling, so having zero How? You need hypercall page for it. Best, Vadim. > retries could be the best setting. But even if it is not useful for > TCG, there's nothing KVM-specific, and that's the important point. > > Paolo -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
On 10/11/2011 03:46 PM, Vadim Rozenfeld wrote: > These should not be KVM-specific. You should be able to add > enlightenments to a TCG VM. At the same time, the KVM leaves could be > moved to 0x4100 when enlightenments are active, similar to what Xen > does. > IMO, adding Hyper-V features without KVM support has little, if any, meaning. Relaxed timing is the only one thing, which can be activated without help from hypervisor. Spinlocks too. TCG does its own round-robin scheduling, so having zero retries could be the best setting. But even if it is not useful for TCG, there's nothing KVM-specific, and that's the important point. Paolo -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
On Tue, 2011-10-11 at 09:38 +0200, Paolo Bonzini wrote:
> On 10/09/2011 08:52 PM, Vadim Rozenfeld wrote:
> > ---
> > target-i386/kvm.c | 53
> > -
> > 1 files changed, 52 insertions(+), 1 deletions(-)
> >
> > diff --git a/target-i386/kvm.c b/target-i386/kvm.c
> > index 3840255..74fcc9a 100644
> > --- a/target-i386/kvm.c
> > +++ b/target-i386/kvm.c
> > @@ -29,6 +29,7 @@
> > #include "hw/pc.h"
> > #include "hw/apic.h"
> > #include "ioport.h"
> > +#include "hyperv.h"
> >
> > //#define DEBUG_KVM
> >
> > @@ -379,11 +380,21 @@ int kvm_arch_init_vcpu(CPUState *env)
> > cpuid_i = 0;
> >
> > /* Paravirtualization CPUIDs */
> > -memcpy(signature, "KVMKVMKVM\0\0\0", 12);
>
> These should not be KVM-specific. You should be able to add
> enlightenments to a TCG VM. At the same time, the KVM leaves could be
> moved to 0x4100 when enlightenments are active, similar to what Xen
> does.
>
IMO, adding Hyper-V features without KVM support has little, if any,
meaning. Relaxed timing is the only one thing, which can be activated
without help from hypervisor.
> Paolo
>
> > c =&cpuid_data.entries[cpuid_i++];
> > memset(c, 0, sizeof(*c));
> > c->function = KVM_CPUID_SIGNATURE;
> > +#ifndef CONFIG_HYPERV
> > +memcpy(signature, "KVMKVMKVM\0\0\0", 12);
> > c->eax = 0;
> > +#else
> > +if (!hyperv_enabled()) {
> > +memcpy(signature, "KVMKVMKVM\0\0\0", 12);
> > +c->eax = 0;
> > +} else {
> > +memcpy(signature, "Microsoft Hv", 12);
> > +c->eax = HYPERV_CPUID_MIN;
> > +}
> > +#endif
> > c->ebx = signature[0];
> > c->ecx = signature[1];
> > c->edx = signature[2];
> > @@ -393,6 +404,46 @@ int kvm_arch_init_vcpu(CPUState *env)
> > c->function = KVM_CPUID_FEATURES;
> > c->eax = env->cpuid_kvm_features&
> > kvm_arch_get_supported_cpuid(s, KVM_CPUID_FEATURES, 0, R_EAX);
> > +#ifdef CONFIG_HYPERV
> > +if (hyperv_enabled()) {
> > +memcpy(signature, "Hv#1\0\0\0\0\0\0\0\0", 12);
> > +c->eax = signature[0];
> > +
> > +c =&cpuid_data.entries[cpuid_i++];
> > +memset(c, 0, sizeof(*c));
> > +c->function = HYPERV_CPUID_VERSION;
> > +c->eax = 0x1bbc;
> > +c->ebx = 0x00060001;
> > +
> > +c =&cpuid_data.entries[cpuid_i++];
> > +memset(c, 0, sizeof(*c));
> > +c->function = HYPERV_CPUID_FEATURES;
> > +if (hyperv_relaxed_timing()) {
> > +c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
> > +}
> > +if (hyperv_apic_recommended()) {
> > +c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
> > +c->eax |= HV_X64_MSR_APIC_ACCESS_AVAILABLE;
> > +}
> > +
> > +c =&cpuid_data.entries[cpuid_i++];
> > +memset(c, 0, sizeof(*c));
> > +c->function = HYPERV_CPUID_ENLIGHTMENT_INFO;
> > +if (hyperv_relaxed_timing()) {
> > +c->eax |= HV_X64_RELAXED_TIMING_RECOMMENDED;
> > +}
> > +if (hyperv_apic_recommended()) {
> > +c->eax |= HV_X64_APIC_ACCESS_RECOMMENDED;
> > +}
> > +c->ebx = hyperv_spinlock_retries();
> > +
> > +c =&cpuid_data.entries[cpuid_i++];
> > +memset(c, 0, sizeof(*c));
> > +c->function = HYPERV_CPUID_IMPLEMENT_LIMITS;
> > +c->eax = 0x40;
> > +c->ebx = 0x40;
> > +}
> > +#endif
> >
> > has_msr_async_pf_en = c->eax& (1<< KVM_FEATURE_ASYNC_PF);
> >
>
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
On 10/09/2011 08:52 PM, Vadim Rozenfeld wrote:
---
target-i386/kvm.c | 53 -
1 files changed, 52 insertions(+), 1 deletions(-)
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 3840255..74fcc9a 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -29,6 +29,7 @@
#include "hw/pc.h"
#include "hw/apic.h"
#include "ioport.h"
+#include "hyperv.h"
//#define DEBUG_KVM
@@ -379,11 +380,21 @@ int kvm_arch_init_vcpu(CPUState *env)
cpuid_i = 0;
/* Paravirtualization CPUIDs */
-memcpy(signature, "KVMKVMKVM\0\0\0", 12);
These should not be KVM-specific. You should be able to add
enlightenments to a TCG VM. At the same time, the KVM leaves could be
moved to 0x4100 when enlightenments are active, similar to what Xen
does.
Paolo
c =&cpuid_data.entries[cpuid_i++];
memset(c, 0, sizeof(*c));
c->function = KVM_CPUID_SIGNATURE;
+#ifndef CONFIG_HYPERV
+memcpy(signature, "KVMKVMKVM\0\0\0", 12);
c->eax = 0;
+#else
+if (!hyperv_enabled()) {
+memcpy(signature, "KVMKVMKVM\0\0\0", 12);
+c->eax = 0;
+} else {
+memcpy(signature, "Microsoft Hv", 12);
+c->eax = HYPERV_CPUID_MIN;
+}
+#endif
c->ebx = signature[0];
c->ecx = signature[1];
c->edx = signature[2];
@@ -393,6 +404,46 @@ int kvm_arch_init_vcpu(CPUState *env)
c->function = KVM_CPUID_FEATURES;
c->eax = env->cpuid_kvm_features&
kvm_arch_get_supported_cpuid(s, KVM_CPUID_FEATURES, 0, R_EAX);
+#ifdef CONFIG_HYPERV
+if (hyperv_enabled()) {
+memcpy(signature, "Hv#1\0\0\0\0\0\0\0\0", 12);
+c->eax = signature[0];
+
+c =&cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_VERSION;
+c->eax = 0x1bbc;
+c->ebx = 0x00060001;
+
+c =&cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_FEATURES;
+if (hyperv_relaxed_timing()) {
+c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
+}
+if (hyperv_apic_recommended()) {
+c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
+c->eax |= HV_X64_MSR_APIC_ACCESS_AVAILABLE;
+}
+
+c =&cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_ENLIGHTMENT_INFO;
+if (hyperv_relaxed_timing()) {
+c->eax |= HV_X64_RELAXED_TIMING_RECOMMENDED;
+}
+if (hyperv_apic_recommended()) {
+c->eax |= HV_X64_APIC_ACCESS_RECOMMENDED;
+}
+c->ebx = hyperv_spinlock_retries();
+
+c =&cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_IMPLEMENT_LIMITS;
+c->eax = 0x40;
+c->ebx = 0x40;
+}
+#endif
has_msr_async_pf_en = c->eax& (1<< KVM_FEATURE_ASYNC_PF);
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
[RFC PATCH 7/7] [hyper-v] init hyper-v cpuid leafs
---
target-i386/kvm.c | 53 -
1 files changed, 52 insertions(+), 1 deletions(-)
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 3840255..74fcc9a 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -29,6 +29,7 @@
#include "hw/pc.h"
#include "hw/apic.h"
#include "ioport.h"
+#include "hyperv.h"
//#define DEBUG_KVM
@@ -379,11 +380,21 @@ int kvm_arch_init_vcpu(CPUState *env)
cpuid_i = 0;
/* Paravirtualization CPUIDs */
-memcpy(signature, "KVMKVMKVM\0\0\0", 12);
c = &cpuid_data.entries[cpuid_i++];
memset(c, 0, sizeof(*c));
c->function = KVM_CPUID_SIGNATURE;
+#ifndef CONFIG_HYPERV
+memcpy(signature, "KVMKVMKVM\0\0\0", 12);
c->eax = 0;
+#else
+if (!hyperv_enabled()) {
+memcpy(signature, "KVMKVMKVM\0\0\0", 12);
+c->eax = 0;
+} else {
+memcpy(signature, "Microsoft Hv", 12);
+c->eax = HYPERV_CPUID_MIN;
+}
+#endif
c->ebx = signature[0];
c->ecx = signature[1];
c->edx = signature[2];
@@ -393,6 +404,46 @@ int kvm_arch_init_vcpu(CPUState *env)
c->function = KVM_CPUID_FEATURES;
c->eax = env->cpuid_kvm_features &
kvm_arch_get_supported_cpuid(s, KVM_CPUID_FEATURES, 0, R_EAX);
+#ifdef CONFIG_HYPERV
+if (hyperv_enabled()) {
+memcpy(signature, "Hv#1\0\0\0\0\0\0\0\0", 12);
+c->eax = signature[0];
+
+c = &cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_VERSION;
+c->eax = 0x1bbc;
+c->ebx = 0x00060001;
+
+c = &cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_FEATURES;
+if (hyperv_relaxed_timing()) {
+c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
+}
+if (hyperv_apic_recommended()) {
+c->eax |= HV_X64_MSR_HYPERCALL_AVAILABLE;
+c->eax |= HV_X64_MSR_APIC_ACCESS_AVAILABLE;
+}
+
+c = &cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_ENLIGHTMENT_INFO;
+if (hyperv_relaxed_timing()) {
+c->eax |= HV_X64_RELAXED_TIMING_RECOMMENDED;
+}
+if (hyperv_apic_recommended()) {
+c->eax |= HV_X64_APIC_ACCESS_RECOMMENDED;
+}
+c->ebx = hyperv_spinlock_retries();
+
+c = &cpuid_data.entries[cpuid_i++];
+memset(c, 0, sizeof(*c));
+c->function = HYPERV_CPUID_IMPLEMENT_LIMITS;
+c->eax = 0x40;
+c->ebx = 0x40;
+}
+#endif
has_msr_async_pf_en = c->eax & (1 << KVM_FEATURE_ASYNC_PF);
--
1.7.4.4
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
