Re: [PATCH] target-i386: Allow changing of Hypervisor CPUIDs.

[Don Slutz <Don@CloudSwitch.Com>]

On 09/05/12 12:48, Marcelo Tosatti wrote:
> On Thu, Aug 30, 2012 at 03:20:35PM -0400, Don Slutz wrote:
>> This is primarily done so that the guest will think it is running
>> under vmware when hypervisor=vmware is specified as a property of a
>> cpu.
>>
>> Also allow this to work in accel=tcg mode.
>>
>> The new cpu properties hyper_level, hyper_extra, hyper_extra_a, and
>> hyper_extra_b can be used to further adjust what the guest sees.
>>
>> Signed-off-by: Don Slutz <Don@CloudSwitch.com>
> For what purpose?
To be able to run bits copied from a ESX(i) server with limited changes 
to the bits.
>
> Is the VMWare interface documented somewhere?
Not that I know of.  All of this change is taken from the Linux source 
(2.6.18-194.32.1.el5 or later) that checks for running on a VMware 
hypervisor.
>
>> ---
>>   target-i386/cpu.c |  178 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>   target-i386/cpu.h |    9 +++
>>   target-i386/kvm.c |   33 ++++++++--
>>   3 files changed, 214 insertions(+), 6 deletions(-)
>>
>> diff --git a/target-i386/cpu.c b/target-i386/cpu.c
>> index f3cac49..a444b95 100644
>> --- a/target-i386/cpu.c
>> +++ b/target-i386/cpu.c
>> @@ -26,6 +26,7 @@
>>   
>>   #include "qemu-option.h"
>>   #include "qemu-config.h"
>> +#include "qemu-timer.h"
>>   
>>   #include "qapi/qapi-visit-core.h"
>>   #include "arch_init.h"
>> @@ -244,6 +245,15 @@ typedef struct x86_def_t {
>>       uint32_t xlevel2;
>>       /* The feature bits on CPUID[EAX=7,ECX=0].EBX */
>>       uint32_t cpuid_7_0_ebx_features;
>> +    /* 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;
>>   } x86_def_t;
>>   
>>   #define I486_FEATURES (CPUID_FP87 | CPUID_VME | CPUID_PSE)
>> @@ -860,6 +870,18 @@ static void x86_cpuid_set_tsc_freq(Object *obj, Visitor *v, void *opaque,
>>       cpu->env.tsc_khz = value / 1000;
>>   }
>>   
>> +static void x86_cpuid_set_hv(x86_def_t *x86_cpu_def, uint32_t level,
>> +                             const char *who)
>> +{
>> +        uint32_t signature[3];
>> +
>> +        memcpy(signature, who, 12);
>> +        x86_cpu_def->cpuid_hv_level = level;
>> +        x86_cpu_def->cpuid_hv_vendor1 = signature[0];
>> +        x86_cpu_def->cpuid_hv_vendor2 = signature[1];
>> +        x86_cpu_def->cpuid_hv_vendor3 = signature[2];
>> +}
>> +
>>   static int cpu_x86_find_by_name(x86_def_t *x86_cpu_def, const char *cpu_model)
>>   {
>>       unsigned int i;
>> @@ -867,6 +889,10 @@ static int cpu_x86_find_by_name(x86_def_t *x86_cpu_def, const char *cpu_model)
>>   
>>       char *s = g_strdup(cpu_model);
>>       char *featurestr, *name = strtok(s, ",");
>> +    bool hyperv_enabled = false;
>> +    bool hv_enabled = false;
>> +    long hyper_level = -1;
>> +    long hyper_extra = -1;
>>       /* Features to be added*/
>>       uint32_t plus_features = 0, plus_ext_features = 0;
>>       uint32_t plus_ext2_features = 0, plus_ext3_features = 0;
>> @@ -993,12 +1019,84 @@ static int cpu_x86_find_by_name(x86_def_t *x86_cpu_def, const char *cpu_model)
>>                   x86_cpu_def->tsc_khz = tsc_freq / 1000;
>>               } else if (!strcmp(featurestr, "hv_spinlocks")) {
>>                   char *err;
>> +
>> +                if (hv_enabled) {
>> +                    fprintf(stderr,
>> +                            "Only one of hypervisor= or hv_* can be used at one time.\n");
>> +                    goto error;
>> +                }
>>                   numvalue = strtoul(val, &err, 0);
>>                   if (!*val || *err) {
>>                       fprintf(stderr, "bad numerical value %s\n", val);
>>                       goto error;
>>                   }
>> +                hyperv_enabled = true;
>>                   hyperv_set_spinlock_retries(numvalue);
>> +            } else if (!strcmp(featurestr, "hyper_level")) {
>> +                char *err;
>> +                long longvalue = strtol(val, &err, 0);
>> +
>> +                if (!*val || *err) {
>> +                    fprintf(stderr, "bad numerical value for hyper_level=%s\n",
>> +                            val);
>> +                    goto error;
>> +                }
>> +                hyper_level = longvalue;
>> +            } else if (!strcmp(featurestr, "hyper_extra")) {
>> +                char *err;
>> +                long longvalue = strtol(val, &err, 0);
>> +
>> +                if (!*val || *err) {
>> +                    fprintf(stderr, "bad numerical value for hyper_extra=%s\n",
>> +                            val);
>> +                    goto error;
>> +                }
>> +                hyper_extra = longvalue;
>> +            } else if (!strcmp(featurestr, "hyper_extra_a")) {
>> +                char *err;
>> +
>> +                numvalue = strtoul(val, &err, 0);
>> +                if (!*val || *err) {
>> +                    fprintf(stderr,
>> +                            "bad numerical value for hyper_extra_a=%s\n",
>> +                            val);
>> +                    goto error;
>> +                }
>> +                x86_cpu_def->cpuid_hv_extra_a = (uint32_t)numvalue;
>> +            } else if (!strcmp(featurestr, "hyper_extra_b")) {
>> +                char *err;
>> +
>> +                numvalue = strtoul(val, &err, 0);
>> +                if (!*val || *err) {
>> +                    fprintf(stderr,
>> +                            "bad numerical value for hyper_extra_b=%s\n",
>> +                            val);
>> +                    goto error;
>> +                }
>> +                x86_cpu_def->cpuid_hv_extra_b = (uint32_t)numvalue;
>> +            } else if (!strcmp(featurestr, "hv") ||
>> +                       !strcmp(featurestr, "hypervisor")) {
>> +                if (hyperv_enabled) {
>> +                    fprintf(stderr,
>> +                            "Only one of hypervisor= or hv_* can be used at one time.\n");
>> +                    goto error;
>> +                }
>> +                hv_enabled = true;
>> +                if (!strcmp(val, "vmware")) {
>> +                    x86_cpuid_set_hv(x86_cpu_def, 0x40000010, "VMwareVMware");
>> +                    minus_kvm_features = ~0;    /* Expected to be zero... */
>> +                } else if (!strcmp(val, "vmware3")) {
>> +                    x86_cpuid_set_hv(x86_cpu_def, 0x40000002, "VMwareVMware");
>> +                    minus_kvm_features = ~0;    /* Expected to be zero... */
>> +                } else if (!strcmp(val, "xen")) {
>> +                    x86_cpuid_set_hv(x86_cpu_def, 0x40000002, "XenVMMXenVMM");
>> +                } else if (!strcmp(val, "kvm")) {
>> +                    x86_cpuid_set_hv(x86_cpu_def, 0, "KVMKVMKVM\0\0\0");
>> +                } else {
>> +                    fprintf(stderr, "unknown hypervisor %s\n",
>> +                            val);
>> +                    goto error;
>> +                }
>>               } else {
>>                   fprintf(stderr, "unrecognized feature %s\n", featurestr);
>>                   goto error;
>> @@ -1008,8 +1106,20 @@ static int cpu_x86_find_by_name(x86_def_t *x86_cpu_def, const char *cpu_model)
>>           } else if (!strcmp(featurestr, "enforce")) {
>>               check_cpuid = enforce_cpuid = 1;
>>           } else if (!strcmp(featurestr, "hv_relaxed")) {
>> +            if (hv_enabled) {
>> +                fprintf(stderr,
>> +                        "Only one of hypervisor= or hv_* can be used at one time.\n");
>> +                goto error;
>> +            }
>> +            hyperv_enabled = true;
>>               hyperv_enable_relaxed_timing(true);
>>           } else if (!strcmp(featurestr, "hv_vapic")) {
>> +            if (hv_enabled) {
>> +                fprintf(stderr,
>> +                        "Only one of hypervisor= or hv_* can be used at one time.\n");
>> +                goto error;
>> +            }
>> +            hyperv_enabled = true;
>>               hyperv_enable_vapic_recommended(true);
>>           } else {
>>               fprintf(stderr, "feature string `%s' not in format (+feature|-feature|feature=xyz)\n", featurestr);
>> @@ -1017,6 +1127,34 @@ static int cpu_x86_find_by_name(x86_def_t *x86_cpu_def, const char *cpu_model)
>>           }
>>           featurestr = strtok(NULL, ",");
>>       }
>> +#ifdef CONFIG_KVM
>> +    if (hyperv_enabled) {
>> +        x86_cpuid_set_hv(x86_cpu_def, HYPERV_CPUID_MIN, "Microsoft Hv");
>> +    }
>> +#endif
>> +    if (hyper_extra >= 0) {
>> +        x86_cpu_def->cpuid_hv_extra = 0x40000000 + hyper_extra;
>> +    } else if (hv_enabled && x86_cpu_def->tsc_khz) {
>> +        /*
>> +         * From http://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).
>> +         */
>> +        x86_cpu_def->cpuid_hv_extra = 0x40000010;
>> +        x86_cpu_def->cpuid_hv_extra_a = (uint32_t)x86_cpu_def->tsc_khz;
>> +        x86_cpu_def->cpuid_hv_extra_b = (uint32_t)(get_ticks_per_sec() / 1000);
>> +    }
>> +    if (hyper_level >= 0) {
>> +        x86_cpu_def->cpuid_hv_level = 0x40000000 + hyper_level;
>> +    }
>>       x86_cpu_def->features |= plus_features;
>>       x86_cpu_def->ext_features |= plus_ext_features;
>>       x86_cpu_def->ext2_features |= plus_ext2_features;
>> @@ -1192,6 +1330,13 @@ int cpu_x86_register(X86CPU *cpu, const char *cpu_model)
>>       env->cpuid_ext4_features = def->ext4_features;
>>       env->cpuid_7_0_ebx = def->cpuid_7_0_ebx_features;
>>       env->cpuid_xlevel2 = def->xlevel2;
>> +    env->cpuid_hv_level = def->cpuid_hv_level;
>> +    env->cpuid_hv_vendor1 = def->cpuid_hv_vendor1;
>> +    env->cpuid_hv_vendor2 = def->cpuid_hv_vendor2;
>> +    env->cpuid_hv_vendor3 = def->cpuid_hv_vendor3;
>> +    env->cpuid_hv_extra = def->cpuid_hv_extra;
>> +    env->cpuid_hv_extra_a = def->cpuid_hv_extra_a;
>> +    env->cpuid_hv_extra_b = def->cpuid_hv_extra_b;
>>       object_property_set_int(OBJECT(cpu), (int64_t)def->tsc_khz * 1000,
>>                               "tsc-frequency", &error);
>>       if (!kvm_enabled()) {
>> @@ -1390,6 +1535,16 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
>>                   index =  env->cpuid_xlevel;
>>               }
>>           }
>> +    } else if (index & 0x40000000) {
>> +        if (env->cpuid_hv_level > 0) {
>> +            /* Handle Paravirtualization CPUIDs */
>> +            if (index > env->cpuid_hv_level) {
>> +                index = env->cpuid_hv_level;
>> +            }
>> +        } else {
>> +            if (index > env->cpuid_level)
>> +                index = env->cpuid_level;
>> +        }
>>       } else {
>>           if (index > env->cpuid_level)
>>               index = env->cpuid_level;
>> @@ -1528,6 +1683,29 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
>>               *edx = 0;
>>           }
>>           break;
>> +    case 0x40000000:
>> +        *eax = env->cpuid_hv_level;
>> +        *ebx = env->cpuid_hv_vendor1;
>> +        *ecx = env->cpuid_hv_vendor2;
>> +        *edx = env->cpuid_hv_vendor3;
>> +        break;
>> +    case 0x40000001:
>> +        *eax = env->cpuid_kvm_features;
>> +        *ebx = 0;
>> +        *ecx = 0;
>> +        *edx = 0;
>> +        break;
>> +    case 0x40000002 ... 0x400000FF:
>> +        if (index == env->cpuid_hv_extra) {
>> +            *eax = env->cpuid_hv_extra_a;
>> +            *ebx = env->cpuid_hv_extra_b;
>> +        } else {
>> +            *eax = 0;
>> +            *ebx = 0;
>> +        }
>> +        *ecx = 0;
>> +        *edx = 0;
>> +        break;
>>       case 0x80000000:
>>           *eax = env->cpuid_xlevel;
>>           *ebx = env->cpuid_vendor1;
>> diff --git a/target-i386/cpu.h b/target-i386/cpu.h
>> index 0677502..dc2039a 100644
>> --- a/target-i386/cpu.h
>> +++ b/target-i386/cpu.h
>> @@ -746,6 +746,15 @@ typedef struct CPUX86State {
>>       uint32_t cpuid_ext4_features;
>>       /* Flags from CPUID[EAX=7,ECX=0].EBX */
>>       uint32_t cpuid_7_0_ebx;
>> +    /* Paravirtualization 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 ffc294e..d01a5f8 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);
>> -- 
>> 1.7.1
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe kvm" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
v2 is on it way now that QEMU 1.3 is open.
     -Don