Re: [PATCH v2 3/4] target-i386: Allow changing of Hypervisor CPUIDs.

[Don Slutz <Don@CloudSwitch.Com>]

On 09/12/12 13:55, Marcelo Tosatti wrote:
> The problem with integrating this is that it has little or
> no assurance from documentation. The Linux kernel source is a good
> source, then say "accordingly to VMWare guest support code in version xyz"
> in the changelog.
I will work on getting a list of the documentation and sources used to 
generate this.
>
> Also extracting this information in a text file (or comment in the code)
> would be better than just adding code.
I am not sure what information you are talking about here.  Are you 
asking about the known "Hypervisor CPUIDs", or what a lot of Linux 
version look at to determine the Hypervisor they are on, or something else?
>
> On Tue, Sep 11, 2012 at 10:07:46AM -0400, Don Slutz wrote:
>> This is primarily done so that the guest will think it is running
>> under vmware when hypervisor-vendor=vmware is specified as a
>> property of a cpu.
>>
>> Signed-off-by: Don Slutz <Don@CloudSwitch.com>
>> ---
>>   target-i386/cpu.c |  214 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>   target-i386/cpu.h |   21 +++++
>>   target-i386/kvm.c |   33 +++++++--
>>   3 files changed, 262 insertions(+), 6 deletions(-)
>>
>> diff --git a/target-i386/cpu.c b/target-i386/cpu.c
>> index 5f9866a..9f1f390 100644
>> --- a/target-i386/cpu.c
>> +++ b/target-i386/cpu.c
>> @@ -1135,6 +1135,36 @@ static void x86_cpuid_set_model_id(Object *obj, const char *model_id,
>>       }
>>   }
>>   
>> +static void x86_cpuid_set_vmware_extra(Object *obj)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    if ((cpu->env.tsc_khz != 0) &&
>> +        (cpu->env.cpuid_hv_level == CPUID_HV_LEVEL_VMARE_4) &&
>> +        (cpu->env.cpuid_hv_vendor1 == CPUID_HV_VENDOR_VMWARE_1) &&
>> +        (cpu->env.cpuid_hv_vendor2 == CPUID_HV_VENDOR_VMWARE_2) &&
>> +        (cpu->env.cpuid_hv_vendor3 == CPUID_HV_VENDOR_VMWARE_3)) {
>> +        const uint32_t apic_khz = 1000000L;
>> +
>> +        /*
>> +         * From article.gmane.org/gmane.comp.emulators.kvm.devel/22643
>> +         *
>> +         *    Leaf 0x40000010, Timing Information.
>> +         *
>> +         *    VMware has defined the first generic leaf to provide timing
>> +         *    information.  This leaf returns the current TSC frequency and
>> +         *    current Bus frequency in kHz.
>> +         *
>> +         *    # EAX: (Virtual) TSC frequency in kHz.
>> +         *    # EBX: (Virtual) Bus (local apic timer) frequency in kHz.
>> +         *    # ECX, EDX: RESERVED (Per above, reserved fields are set to zero).
>> +         */
>> +        cpu->env.cpuid_hv_extra = 0x40000010;
>> +        cpu->env.cpuid_hv_extra_a = (uint32_t)cpu->env.tsc_khz;
>> +        cpu->env.cpuid_hv_extra_b = apic_khz;
>> +    }
>> +}
> What happens in case you migrate the vmware guest to a host
> with different frequency? How is that transmitted to the
> vmware-guest-running-on-kvm ? Or is migration not supported?
As far as I know, it would be the same as for a non-vmware guest. 
http://lists.nongnu.org/archive/html/qemu-devel/2011-07/msg01656.html is 
related to this.

I did not look to see if this has been done since then.

All this change does is to allow the guest to "read" the tsc-frequency 
instead of trying to calculate it.

I will look into the current state of migration when tsc_freq=X is 
specified.  The machine I have been doing most of the testing on (Intel 
Xeon E3-1260L) when I add tsc_freq=2.0G or tsc_freq=2.4G, the guest does 
not see any difference in accel=kvm.
>
>> +static void x86_cpuid_set_hv_level(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +    uint32_t value;
>> +
>> +    visit_type_uint32(v, &value, name, errp);
>> +    if (error_is_set(errp)) {
>> +        return;
>> +    }
>> +
>> +    if ((value != 0) && (value < 0x40000000)) {
>> +        value += 0x40000000;
>> +    }
>> +    cpu->env.cpuid_hv_level = value;
>> +}
>> +
>> +static char *x86_cpuid_get_hv_vendor(Object *obj, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +    CPUX86State *env = &cpu->env;
>> +    char *value;
>> +    int i;
>> +
>> +    value = (char *)g_malloc(CPUID_VENDOR_SZ + 1);
>> +    for (i = 0; i < 4; i++) {
>> +        value[i + 0] = env->cpuid_hv_vendor1 >> (8 * i);
>> +        value[i + 4] = env->cpuid_hv_vendor2 >> (8 * i);
>> +        value[i + 8] = env->cpuid_hv_vendor3 >> (8 * i);
>> +    }
>> +    value[CPUID_VENDOR_SZ] = '\0';
>> +
>> +    /* Convert known names */
>> +    if (!strcmp(value, CPUID_HV_VENDOR_VMWARE)) {
>> +        if (env->cpuid_hv_level == CPUID_HV_LEVEL_VMARE_4) {
>> +            pstrcpy(value, sizeof(value), "vmware4");
>> +        } else if (env->cpuid_hv_level == CPUID_HV_LEVEL_VMARE_3) {
>> +            pstrcpy(value, sizeof(value), "vmware3");
>> +        }
>> +    } else if (!strcmp(value, CPUID_HV_VENDOR_XEN) &&
>> +               env->cpuid_hv_level == CPUID_HV_LEVEL_XEN) {
>> +        pstrcpy(value, sizeof(value), "xen");
>> +    } else if (!strcmp(value, CPUID_HV_VENDOR_KVM) &&
>> +               env->cpuid_hv_level == 0) {
>> +        pstrcpy(value, sizeof(value), "kvm");
>> +    }
>> +    return value;
>> +}
>> +
>> +static void x86_cpuid_set_hv_vendor(Object *obj, const char *value,
>> +                                     Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +    CPUX86State *env = &cpu->env;
>> +    int i;
>> +    char adj_value[CPUID_VENDOR_SZ + 1];
>> +
>> +    memset(adj_value, 0, sizeof(adj_value));
>> +
>> +    /* Convert known names */
>> +    if (!strcmp(value, "vmware") || !strcmp(value, "vmware4")) {
>> +        if (env->cpuid_hv_level == 0) {
>> +            env->cpuid_hv_level = CPUID_HV_LEVEL_VMARE_4;
>> +        }
>> +        pstrcpy(adj_value, sizeof(adj_value), CPUID_HV_VENDOR_VMWARE);
>> +    } else if (!strcmp(value, "vmware3")) {
>> +        if (env->cpuid_hv_level == 0) {
>> +            env->cpuid_hv_level = CPUID_HV_LEVEL_VMARE_3;
>> +        }
>> +        pstrcpy(adj_value, sizeof(adj_value), CPUID_HV_VENDOR_VMWARE);
>> +    } else if (!strcmp(value, "xen")) {
>> +        if (env->cpuid_hv_level == 0) {
>> +            env->cpuid_hv_level = CPUID_HV_LEVEL_XEN;
>> +        }
>> +        pstrcpy(adj_value, sizeof(adj_value), CPUID_HV_VENDOR_XEN);
>> +    } else if (!strcmp(value, "kvm")) {
>> +        pstrcpy(adj_value, sizeof(adj_value), CPUID_HV_VENDOR_KVM);
>> +    } else {
>> +        pstrcpy(adj_value, sizeof(adj_value), value);
>> +    }
>> +
>> +    env->cpuid_hv_vendor1 = 0;
>> +    env->cpuid_hv_vendor2 = 0;
>> +    env->cpuid_hv_vendor3 = 0;
>> +    for (i = 0; i < 4; i++) {
>> +        env->cpuid_hv_vendor1 |= ((uint8_t)adj_value[i + 0]) << (8 * i);
>> +        env->cpuid_hv_vendor2 |= ((uint8_t)adj_value[i + 4]) << (8 * i);
>> +        env->cpuid_hv_vendor3 |= ((uint8_t)adj_value[i + 8]) << (8 * i);
>> +    }
>> +    x86_cpuid_set_vmware_extra(obj);
>> +}
>> +
>> +static void x86_cpuid_get_hv_extra(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    visit_type_uint32(v, &cpu->env.cpuid_hv_extra, name, errp);
>> +}
>> +
>> +static void x86_cpuid_set_hv_extra(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +    uint32_t value;
>> +
>> +    visit_type_uint32(v, &value, name, errp);
>> +    if (error_is_set(errp)) {
>> +        return;
>> +    }
>> +
>> +    if ((value != 0) && (value < 0x40000000)) {
>> +        value += 0x40000000;
>> +    }
>> +    cpu->env.cpuid_hv_extra = value;
>> +}
>> +
>> +static void x86_cpuid_get_hv_extra_a(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    visit_type_uint32(v, &cpu->env.cpuid_hv_extra_a, name, errp);
>> +}
>> +
>> +static void x86_cpuid_set_hv_extra_a(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    visit_type_uint32(v, &cpu->env.cpuid_hv_extra_a, name, errp);
>> +}
>> +
>> +static void x86_cpuid_get_hv_extra_b(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    visit_type_uint32(v, &cpu->env.cpuid_hv_extra_b, name, errp);
>> +}
>> +
>> +static void x86_cpuid_set_hv_extra_b(Object *obj, Visitor *v, void *opaque,
>> +                                const char *name, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    visit_type_uint32(v, &cpu->env.cpuid_hv_extra_b, name, errp);
>>   }
>>   
>>   #if !defined(CONFIG_USER_ONLY)
>> +static void x86_set_hyperv(Object *obj, Error **errp)
>> +{
>> +    X86CPU *cpu = X86_CPU(obj);
>> +
>> +    cpu->env.cpuid_hv_level = HYPERV_CPUID_MIN;
>> +    x86_cpuid_set_hv_vendor(obj, "Microsoft Hv", errp);
>> +}
>> +
>>   static void x86_get_hv_spinlocks(Object *obj, Visitor *v, void *opaque,
>>                                    const char *name, Error **errp)
>>   {
>> @@ -1189,6 +1385,7 @@ static void x86_set_hv_spinlocks(Object *obj, Visitor *v, void *opaque,
>>           return;
>>       }
>>       hyperv_set_spinlock_retries(value);
>> +    x86_set_hyperv(obj, errp);
>>   }
>>   
>>   static void x86_get_hv_relaxed(Object *obj, Visitor *v, void *opaque,
>> @@ -1209,6 +1406,7 @@ static void x86_set_hv_relaxed(Object *obj, Visitor *v, void *opaque,
>>           return;
>>       }
>>       hyperv_enable_relaxed_timing(value);
>> +    x86_set_hyperv(obj, errp);
>>   }
> This seems unrelated to the rest of the patch?
The new call to x86_set_hyperv() is how the 1st change to 
target-i386/kvm.c still does the same thing as before.
>
>>   static void x86_get_hv_vapic(Object *obj, Visitor *v, void *opaque,
>> @@ -1229,6 +1427,7 @@ static void x86_set_hv_vapic(Object *obj, Visitor *v, void *opaque,
>>           return;
>>       }
>>       hyperv_enable_vapic_recommended(value);
>> +    x86_set_hyperv(obj, errp);
>>   }
>>   #endif
>>   
>> @@ -2061,6 +2260,21 @@ static void x86_cpu_initfn(Object *obj)
>>       object_property_add(obj, "enforce", "bool",
>>                           x86_cpuid_get_enforce,
>>                           x86_cpuid_set_enforce, NULL, NULL, NULL);
>> +    object_property_add(obj, "hypervisor-level", "int",
>> +                        x86_cpuid_get_hv_level,
>> +                        x86_cpuid_set_hv_level, NULL, NULL, NULL);
>> +    object_property_add_str(obj, "hypervisor-vendor",
>> +                            x86_cpuid_get_hv_vendor,
>> +                            x86_cpuid_set_hv_vendor, NULL);
>> +    object_property_add(obj, "hypervisor-extra", "int",
>> +                        x86_cpuid_get_hv_extra,
>> +                        x86_cpuid_set_hv_extra, NULL, NULL, NULL);
>> +    object_property_add(obj, "hypervisor-extra-a", "int",
>> +                        x86_cpuid_get_hv_extra_a,
>> +                        x86_cpuid_set_hv_extra_a, NULL, NULL, NULL);
>> +    object_property_add(obj, "hypervisor-extra-b", "int",
>> +                        x86_cpuid_get_hv_extra_b,
>> +                        x86_cpuid_set_hv_extra_b, NULL, NULL, NULL);
>>   #if !defined(CONFIG_USER_ONLY)
>>       object_property_add(obj, "hv_spinlocks", "int",
>>                           x86_get_hv_spinlocks,
>> diff --git a/target-i386/cpu.h b/target-i386/cpu.h
>> index 5265c5a..a2d3588 100644
>> --- a/target-i386/cpu.h
>> +++ b/target-i386/cpu.h
>> @@ -488,6 +488,18 @@
>>   
>>   #define CPUID_VENDOR_VIA   "CentaurHauls"
>>   
>> +#define CPUID_HV_VENDOR_VMWARE_1 0x61774d56 /* "VMwa" */
>> +#define CPUID_HV_VENDOR_VMWARE_2 0x4d566572 /* "reVM" */
>> +#define CPUID_HV_VENDOR_VMWARE_3 0x65726177 /* "ware" */
>> +#define CPUID_HV_VENDOR_VMWARE "VMwareVMware"
>> +#define CPUID_HV_LEVEL_VMARE_3 0x40000002
>> +#define CPUID_HV_LEVEL_VMARE_4 0x40000010
>> +
>> +#define CPUID_HV_VENDOR_XEN "XenVMMXenVMM"
>> +#define CPUID_HV_LEVEL_XEN  0x40000002
>> +
>> +#define CPUID_HV_VENDOR_KVM "KVMKVMKVM"
>> +
>>   #define CPUID_MWAIT_IBE     (1 << 1) /* Interrupts can exit capability */
>>   #define CPUID_MWAIT_EMX     (1 << 0) /* enumeration supported */
>>   
>> @@ -782,6 +794,15 @@ typedef struct CPUX86State {
>>       uint32_t cpuid_ext4_features;
>>       /* Flags from CPUID[EAX=7,ECX=0].EBX */
>>       uint32_t cpuid_7_0_ebx;
>> +    /* Hypervisor CPUIDs */
>> +    uint32_t cpuid_hv_level;
>> +    uint32_t cpuid_hv_vendor1;
>> +    uint32_t cpuid_hv_vendor2;
>> +    uint32_t cpuid_hv_vendor3;
>> +    /* VMware extra data */
>> +    uint32_t cpuid_hv_extra;
>> +    uint32_t cpuid_hv_extra_a;
>> +    uint32_t cpuid_hv_extra_b;
>>   
>>       /* MTRRs */
>>       uint64_t mtrr_fixed[11];
>> diff --git a/target-i386/kvm.c b/target-i386/kvm.c
>> index 895d848..17c72bc 100644
>> --- a/target-i386/kvm.c
>> +++ b/target-i386/kvm.c
>> @@ -389,16 +389,18 @@ int kvm_arch_init_vcpu(CPUX86State *env)
>>       c = &cpuid_data.entries[cpuid_i++];
>>       memset(c, 0, sizeof(*c));
>>       c->function = KVM_CPUID_SIGNATURE;
>> -    if (!hyperv_enabled()) {
>> +    if (env->cpuid_hv_level == 0) {
>>           memcpy(signature, "KVMKVMKVM\0\0\0", 12);
>>           c->eax = 0;
>> +        c->ebx = signature[0];
>> +        c->ecx = signature[1];
>> +        c->edx = signature[2];
>>       } else {
>> -        memcpy(signature, "Microsoft Hv", 12);
>> -        c->eax = HYPERV_CPUID_MIN;
>> +        c->eax = env->cpuid_hv_level;
>> +        c->ebx = env->cpuid_hv_vendor1;
>> +        c->ecx = env->cpuid_hv_vendor2;
>> +        c->edx = env->cpuid_hv_vendor3;
>>       }
>> -    c->ebx = signature[0];
>> -    c->ecx = signature[1];
>> -    c->edx = signature[2];
>>   
>>       c = &cpuid_data.entries[cpuid_i++];
>>       memset(c, 0, sizeof(*c));
>> @@ -452,6 +454,25 @@ int kvm_arch_init_vcpu(CPUX86State *env)
>>           c->ebx = signature[0];
>>           c->ecx = signature[1];
>>           c->edx = signature[2];
>> +    } else if (env->cpuid_hv_level > 0) {
>> +        for (i = KVM_CPUID_FEATURES + 1; i <= env->cpuid_hv_level; i++) {
>> +            c = &cpuid_data.entries[cpuid_i++];
>> +            memset(c, 0, sizeof(*c));
>> +            c->function = i;
>> +            if (i == env->cpuid_hv_extra) {
>> +                c->eax = env->cpuid_hv_extra_a;
>> +                c->ebx = env->cpuid_hv_extra_b;
>> +            }
>> +        }
>> +
>> +        c = &cpuid_data.entries[cpuid_i++];
>> +        memset(c, 0, sizeof(*c));
>> +        c->function = KVM_CPUID_SIGNATURE_NEXT;
>> +        memcpy(signature, "KVMKVMKVM\0\0\0", 12);
>> +        c->eax = 0;
>> +        c->ebx = signature[0];
>> +        c->ecx = signature[1];
>> +        c->edx = signature[2];
>>       }
>>   
>>       has_msr_async_pf_en = c->eax & (1 << KVM_FEATURE_ASYNC_PF);
>
> Its one big patch, better split in logically correlated patches
> (with better changelog). This would help reviewers.
Will re-work into a set of smaller changes.
    -Don