Re: [PATCH v7] arm/kvm: Enable support for KVM_ARM_VCPU_PMU_V3_FILTER

[Eric Auger]

Hi Daniel,

On 3/19/24 16:22, Daniel P. Berrangé wrote:
> On Wed, Feb 21, 2024 at 01:34:31AM -0500, Shaoqin Huang wrote:
>> The KVM_ARM_VCPU_PMU_V3_FILTER provides the ability to let the VMM decide
>> which PMU events are provided to the guest. Add a new option
>> `kvm-pmu-filter` as -cpu sub-option to set the PMU Event Filtering.
>> Without the filter, all PMU events are exposed from host to guest by
>> default. The usage of the new sub-option can be found from the updated
>> document (docs/system/arm/cpu-features.rst).
>>
>> Here is an example which shows how to use the PMU Event Filtering, when
>> we launch a guest by use kvm, add such command line:
>>
>>   # qemu-system-aarch64 \
>>         -accel kvm \
>>         -cpu host,kvm-pmu-filter="D:0x11-0x11"
> 
> snip
> 
>> @@ -517,6 +533,12 @@ void kvm_arm_add_vcpu_properties(ARMCPU *cpu)
>>                               kvm_steal_time_set);
>>      object_property_set_description(obj, "kvm-steal-time",
>>                                      "Set off to disable KVM steal time.");
>> +
>> +    object_property_add_str(obj, "kvm-pmu-filter", kvm_pmu_filter_get,
>> +                            kvm_pmu_filter_set);
>> +    object_property_set_description(obj, "kvm-pmu-filter",
>> +                                    "PMU Event Filtering description for "
>> +                                    "guest PMU. (default: NULL, disabled)");
>>  }
> 
> Passing a string property, but....[1]
> 
>>  
>>  bool kvm_arm_pmu_supported(void)
>> @@ -1706,6 +1728,62 @@ static bool kvm_arm_set_device_attr(ARMCPU *cpu, 
>> struct kvm_device_attr *attr,
>>      return true;
>>  }
>>  
>> +static void kvm_arm_pmu_filter_init(ARMCPU *cpu)
>> +{
>> +    static bool pmu_filter_init;
>> +    struct kvm_pmu_event_filter filter;
>> +    struct kvm_device_attr attr = {
>> +        .group      = KVM_ARM_VCPU_PMU_V3_CTRL,
>> +        .attr       = KVM_ARM_VCPU_PMU_V3_FILTER,
>> +        .addr       = (uint64_t)&filter,
>> +    };
>> +    int i;
>> +    g_auto(GStrv) event_filters;
>> +
>> +    if (!cpu->kvm_pmu_filter) {
>> +        return;
>> +    }
>> +    if (kvm_vcpu_ioctl(CPU(cpu), KVM_HAS_DEVICE_ATTR, &attr)) {
>> +        warn_report("The KVM doesn't support the PMU Event Filter!");
> 
> If the user requested a filter and it can't be supported, QEMU
> must exit with an error, not ignore the user's request.
> 
>> +        return;
>> +    }
>> +
>> +    /*
>> +     * The filter only needs to be initialized through one vcpu ioctl and it
>> +     * will affect all other vcpu in the vm.
>> +     */
>> +    if (pmu_filter_init) {
>> +        return;
>> +    } else {
>> +        pmu_filter_init = true;
>> +    }
>> +
>> +    event_filters = g_strsplit(cpu->kvm_pmu_filter, ";", -1);
>> +    for (i = 0; event_filters[i]; i++) {
>> +        unsigned short start = 0, end = 0;
>> +        char act;
>> +
>> +        if (sscanf(event_filters[i], "%c:%hx-%hx", &act, &start, &end) != 
>> 3) {
>> +            warn_report("Skipping invalid PMU filter %s", event_filters[i]);
>> +            continue;
> 
> Warning on user syntax errors is undesirable - it should be a fatal
> error of the user gets this wrong.
> 
>> +        }
>> +
>> +        if ((act != 'A' && act != 'D') || start > end) {
>> +            warn_report("Skipping invalid PMU filter %s", event_filters[i]);
>> +            continue;
> 
> Likewise should be fatal.
> 
>> +        }
>> +
>> +        filter.base_event = start;
>> +        filter.nevents = end - start + 1;
>> +        filter.action = (act == 'A') ? KVM_PMU_EVENT_ALLOW :
>> +                                       KVM_PMU_EVENT_DENY;
>> +
>> +        if (!kvm_arm_set_device_attr(cpu, &attr, "PMU_V3_FILTER")) {
>> +            break;
>> +        }
>> +    }
>> +}
> 
> ..[1] then implementing a custom parser is rather a QEMU design anti-pattern,
> especially when the proposed syntax is incapable of being mapped into the
> normal QAPI syntax for a list of structs should we want to fully convert
> -cpu to QAPI parsing later. I wonder if can we model this property with
> QAPI now ?
I guess you mean creating a new property like those in
hw/core/qdev-properties-system.c for instance  and populating an array
of those at CPU object level?

Note there is v8 but most of your comments still apply
20240312074849.71475-1-shahuang@redhat.com/">https://lore.kernel.org/all/20240312074849.71475-1-shahuang@redhat.com/

Thanks

Eric
> 
> With regards,
> Daniel