Re: [PATCH 2/2] KVM: PPC: booke/bookehv: Add guest debug support

[Scott Wood <scottwood@freescale.com>]

On 08/16/2012 10:12 AM, Bhushan Bharat-R65777 wrote:
> 
> 
>> -----Original Message-----
>> From: Wood Scott-B07421
>> Sent: Tuesday, July 31, 2012 3:31 AM
>> To: Bhushan Bharat-R65777
>> Cc: Wood Scott-B07421; kvm-ppc@vger.kernel.org; kvm@vger.kernel.org;
>> agraf@suse.de
>> Subject: Re: [PATCH 2/2] KVM: PPC: booke/bookehv: Add guest debug support
>>
>> On 07/30/2012 02:37 AM, Bhushan Bharat-R65777 wrote:
>>>
>>>
>>>> -----Original Message-----
>>>> From: Wood Scott-B07421
>>>> Sent: Friday, July 27, 2012 7:00 AM
>>>> To: Bhushan Bharat-R65777
>>>> Cc: kvm-ppc@vger.kernel.org; kvm@vger.kernel.org; agraf@suse.de;
>>>> Bhushan Bharat-
>>>> R65777
>>>> Subject: Re: [PATCH 2/2] KVM: PPC: booke/bookehv: Add guest debug
>>>> support
>>>>
>>>>> +#ifndef CONFIG_PPC_FSL_BOOK3E
>>>>> +	PPC_LD(r7, VCPU_HOST_DBG+KVMPPC_DBG_IAC3, r4)
>>>>> +	PPC_LD(r8, VCPU_HOST_DBG+KVMPPC_DBG_IAC4, r4)
>>>>> +	mtspr	SPRN_IAC3, r7
>>>>> +	mtspr	SPRN_IAC4, r8
>>>>> +#endif
>>>>
>>>> Can you handle this at runtime with a feature section?
>>>
>>> Why you want this to make run time? Removing config_ ?
>>
>> Currently KVM hardcodes the target hardware in a way that is unacceptable in
>> much of the rest of the kernel.  We have a long term goal to stop doing that,
>> and we should avoid making it worse by adding random ifdefs for specific CPUs.
> 
> I do not see any CPU_FTR_* which I can use directly. Should I define a new FTR, something like:
> 
> #define CPU_FTR_DEBUG_E500  LONG_ASM_CONST(0x4000000000000000)
> 
> Use this in: CPU_FTRS_E500_2, CPU_FTRS_E500MC, CPU_FTRS_E5500 etc
> 
> BEGIN_FTR_SECTION
> 	PPC_LD(r7, VCPU_HOST_DBG+KVMPPC_DBG_IAC3, r4)
> 	PPC_LD(r8, VCPU_HOST_DBG+KVMPPC_DBG_IAC4, r4)
> 	mtspr	SPRN_IAC3, r7
> 	mtspr	SPRN_IAC4, r8
> END_FTR_SECTION_IFCLR(CPU_FTR_DEBUG_E500)

It looks like other parts of the kernel use CONFIG_PPC_ADV_DEBUG_IACS
for this, though ideally it would be made runtime in the future.

-Scott