Re: [Qemu-devel] [Qemu-ppc] [PATCH 3/5] target-ppc: Synchronize more SPRs to KVM using ONE_REG interface

[Alexander Graf]

On 23.01.2013, at 23:41, David Gibson wrote:

> On Wed, Jan 23, 2013 at 11:52:54AM -0600, Scott Wood wrote:
>> On 01/22/2013 11:04:59 PM, David Gibson wrote:
>>> diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
>>> index 2f4f068..e84b993 100644
>>> --- a/target-ppc/kvm.c
>>> +++ b/target-ppc/kvm.c
>>> @@ -60,7 +60,7 @@ static int cap_booke_sregs;
>>> static int cap_ppc_smt;
>>> static int cap_ppc_rma;
>>> static int cap_spapr_tce;
>>> -static int cap_hior;
>>> +static int cap_one_reg;
>>> 
>>> /* XXX We have a race condition where we actually have a level
>>> triggered
>>> *     interrupt, but the infrastructure can't expose that yet,
>>> so the guest
>>> @@ -89,7 +89,11 @@ int kvm_arch_init(KVMState *s)
>>>    cap_ppc_smt = kvm_check_extension(s, KVM_CAP_PPC_SMT);
>>>    cap_ppc_rma = kvm_check_extension(s, KVM_CAP_PPC_RMA);
>>>    cap_spapr_tce = kvm_check_extension(s, KVM_CAP_SPAPR_TCE);
>>> -    cap_hior = kvm_check_extension(s, KVM_CAP_PPC_HIOR);
>>> +    /* This capability is misnamed - it was introduced with the
>>> +     * KVM_SET_ONE_REG ioctl(), which at the time only supported the
>>> +     * HIOR.  We don't want a different capability for every register
>>> +     * the interface can support though. */
>>> +    cap_one_reg = kvm_check_extension(s, KVM_CAP_PPC_HIOR);
>> 
>> So what happens when we want to use onereg for booke, which doesn't
>> have KVM_CAP_PPC_HIOR?  Into what variable would we put a check for
>> KVM_CAP_ONE_REG?
> 
> Drat, good point.  There is no KVM_CAP_ONE_REG

Of course there is a KVM_CAP_ONE_REG :).

Alex