[PATCH v6 1/6] arm/arm64: KVM: Introduce armv7 fp/simd vcpu fields and helpers

[Mario Smarduch]

On 1/10/2016 8:32 AM, Christoffer Dall wrote:
> On Tue, Jan 05, 2016 at 11:28:18AM -0800, Mario Smarduch wrote:
>>
>>
>> On 1/5/2016 7:00 AM, Christoffer Dall wrote:
>>> On Sat, Dec 26, 2015 at 01:54:55PM -0800, Mario Smarduch wrote:
>>>> Add helper functions to enable access to fp/smid on guest entry and save host
>>>> fpexc on vcpu put, check if fp/simd registers are dirty and add new vcpu
>>>> fields.
>>>>
>>>> Signed-off-by: Mario Smarduch <m.smarduch at samsung.com>
>>>> ---
>>>>  arch/arm/include/asm/kvm_emulate.h   | 42 ++++++++++++++++++++++++++++++++++++
>>>>  arch/arm/include/asm/kvm_host.h      |  6 ++++++
>>>>  arch/arm64/include/asm/kvm_emulate.h |  8 +++++++
>>>>  3 files changed, 56 insertions(+)
>>>>
>>>> diff --git a/arch/arm/include/asm/kvm_emulate.h b/arch/arm/include/asm/kvm_emulate.h
>>>> index 3095df0..d4d9da1 100644
>>>> --- a/arch/arm/include/asm/kvm_emulate.h
>>>> +++ b/arch/arm/include/asm/kvm_emulate.h
>>>> @@ -24,6 +24,8 @@
>>>>  #include <asm/kvm_mmio.h>
>>>>  #include <asm/kvm_arm.h>
>>>>  #include <asm/cputype.h>
>>>> +#include <asm/vfp.h>
>>>> +#include "../vfp/vfpinstr.h"
>>>
>>> this looks dodgy...
>>>
>>> can you move vfpinstr.h instead?
>> Sure I'll fix it up, it's in couple other places in kernel and kvm
>>  - copied it.
>>>
>>>>  
>>>>  unsigned long *vcpu_reg(struct kvm_vcpu *vcpu, u8 reg_num);
>>>>  unsigned long *vcpu_spsr(struct kvm_vcpu *vcpu);
>>>> @@ -255,4 +257,44 @@ static inline unsigned long vcpu_data_host_to_guest(struct kvm_vcpu *vcpu,
>>>>  	}
>>>>  }
>>>>  
>>>> +#ifdef CONFIG_VFPv3
>>>> +/* Called from vcpu_load - save fpexc and enable guest access to fp/simd unit */
>>>
>>> the comment is misleading, you're not enabling guest access to the
>>> fp/simd unit, you're just setting the enabled bit to ensure guest
>>> accesses trap.
>>
>> That's more accurate.
>>>
>>>> +static inline void vcpu_trap_vfp_enable(struct kvm_vcpu *vcpu)
>>>> +{
>>>> +	u32 fpexc;
>>>> +
>>>> +	/* Save host fpexc, and enable guest access to fp unit */
>>>> +	fpexc = fmrx(FPEXC);
>>>> +	vcpu->arch.host_fpexc = fpexc;
>>>> +	fpexc |= FPEXC_EN;
>>>> +	fmxr(FPEXC, fpexc);
>>>> +
>>>> +	/* Configure HCPTR to trap on tracing and fp/simd access */
>>>> +	vcpu->arch.hcptr = HCPTR_TTA | HCPTR_TCP(10)  | HCPTR_TCP(11);
>>>> +}
>>>> +
>>>> +/* Called from vcpu_put - restore host fpexc */
>>>
>>> I would probably get rid of the "Called from" stuff and just describe
>>> what these functions do locally.  Comments like this are likely to be
>>> out of date soon'ish.
>>
>> Yeah true, will do.
>>>
>>>> +static inline void vcpu_restore_host_fpexc(struct kvm_vcpu *vcpu)
>>>> +{
>>>> +	fmxr(FPEXC, vcpu->arch.host_fpexc);
>>>> +}
>>>> +
>>>> +/* If trap bits are reset then fp/simd registers are dirty */
>>>> +static inline bool vcpu_vfp_isdirty(struct kvm_vcpu *vcpu)
>>>> +{
>>>> +	return !(vcpu->arch.hcptr & (HCPTR_TCP(10) | HCPTR_TCP(11)));
>>>> +}
>>>> +#else
>>>> +static inline void vcpu_trap_vfp_enable(struct kvm_vcpu *vcpu)
>>>> +{
>>>> +	vcpu->arch.hcptr = HCPTR_TTA;
>>>> +}
>>>> +
>>>> +static inline void vcpu_restore_host_fpexc(struct kvm_vcpu *vcpu) {}
>>>> +static inline bool vcpu_vfp_isdirty(struct kvm_vcpu *vcpu)
>>>> +{
>>>> +	return false;
>>>> +}
>>>> +#endif
>>>
>>> this kind of feels like it belongs in its own C-file instead of a header
>>> file, perhaps arch/arm/kvm/vfp.C.
>>>
>>> Marc, what do you think?
>>>
>>
>> That would be starting from vcpu_trap_vfp_enable()? The file is getting
>> little overloaded.
> 
> yes, starting from vcpu_trap_vfp_enable.
> 
> Which file is getting overloaded?
> 
> I'm thinking the assembly file you add in the next patch could be
> inline assembly in a C-file combined with this code, perhaps.

Yes that makes sense, I'll convert it.
> 
>>
>> I'm also thinking that 3rd patch should have one function call for vcpu_put
>> like vcpu_load does instead of exposing arm32/arm64, arm32 only relevant logic.
>> When you have a chance to review that patch please keep that in mind.
>>
> 
> ok, will try ;)
> 
> -Christoffer
>