qemu-arm
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-arm] [Qemu-devel] [PATCH v5 07/18] kvm: add kvm_arm_get_max_vm


From: Auger Eric
Subject: Re: [Qemu-arm] [Qemu-devel] [PATCH v5 07/18] kvm: add kvm_arm_get_max_vm_phys_shift
Date: Fri, 1 Feb 2019 11:30:48 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.4.0

Hi Hia,

On 1/29/19 3:25 PM, Jia He wrote:
> Hi Eric
> 
> On 2019/1/23 18:14, Eric Auger wrote:
>> Add the kvm_arm_get_max_vm_phys_shift() helper that returns the
>> log of the maximum IPA size supported by KVM. This capability
>> needs to be known to create the VM with a specific IPA max size
>> (kvm_type passed along KVM_CREATE_VM ioctl.
>>
>> Signed-off-by: Eric Auger <address@hidden>
>>
>> ---
>> v3 -> v4:
>> - s/s/ms in kvm_arm_get_max_vm_phys_shift function comment
>> - check KVM_CAP_ARM_VM_IPA_SIZE extension
>>
>> v1 -> v2:
>> - put this in ARM specific code
>> ---
>>   target/arm/kvm.c     |  8 ++++++++
>>   target/arm/kvm_arm.h | 16 ++++++++++++++++
>>   2 files changed, 24 insertions(+)
>>
>> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
>> index e00ccf9c98..877c588ef5 100644
>> --- a/target/arm/kvm.c
>> +++ b/target/arm/kvm.c
>> @@ -18,6 +18,7 @@
>>   #include "qemu/error-report.h"
>>   #include "sysemu/sysemu.h"
>>   #include "sysemu/kvm.h"
>> +#include "sysemu/kvm_int.h"
>>   #include "kvm_arm.h"
>>   #include "cpu.h"
>>   #include "trace.h"
>> @@ -162,6 +163,13 @@ void kvm_arm_set_cpu_features_from_host(ARMCPU *cpu)
>>       env->features = arm_host_cpu_features.features;
>>   }
>>   +int kvm_arm_get_max_vm_phys_shift(MachineState *ms)
>> +{
>> +    KVMState *s = KVM_STATE(ms->accelerator);
>> +
>> +    return kvm_check_extension(s, KVM_CAP_ARM_VM_IPA_SIZE);
> 
> If KVM_CAP_ARM_VM_IPA_SIZE is not supported in host kernel, would it be
> 
> better if kvm_arm_get_max_vm_phys_shift returns a default value
> 
> (e.g. 40),instead of an errno?
Yes I can change that. At the moment kvm_check_extension() should return
0 if the KVM_CAP_ARM_VM_IPA_SIZE entry is not supported by the kernel.

Thanks

Eric
> 
> ---
> Cheers,
> Jia He
> 
>> +}
>> +
>>   int kvm_arch_init(MachineState *ms, KVMState *s)
>>   {
>>       /* For ARM interrupt delivery is always asynchronous,
>> diff --git a/target/arm/kvm_arm.h b/target/arm/kvm_arm.h
>> index 6393455b1d..5969c41f83 100644
>> --- a/target/arm/kvm_arm.h
>> +++ b/target/arm/kvm_arm.h
>> @@ -207,6 +207,17 @@ bool
>> kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf);
>>    */
>>   void kvm_arm_set_cpu_features_from_host(ARMCPU *cpu);
>>   +/**
>> + * kvm_arm_get_max_vm_phys_shift - Returns log2 of the max IPA size
>> + * supported by KVM
>> + *
>> + * @ms: Machine state handle
>> + *
>> + * Return the max number of IPA bits or a negative value if
>> + * the host kernel does not expose this value.
>> + */
>> +int kvm_arm_get_max_vm_phys_shift(MachineState *ms);
>> +
>>   /**
>>    * kvm_arm_sync_mpstate_to_kvm
>>    * @cpu: ARMCPU
>> @@ -239,6 +250,11 @@ static inline void
>> kvm_arm_set_cpu_features_from_host(ARMCPU *cpu)
>>       cpu->host_cpu_probe_failed = true;
>>   }
>>   +static inline int kvm_arm_get_max_vm_phys_shift(MachineState *ms)
>> +{
>> +    return -ENOENT;
>> +}
>> +
>>   static inline int kvm_arm_vgic_probe(void)
>>   {
>>       return 0;
> 



reply via email to

[Prev in Thread] Current Thread [Next in Thread]