[PATCH v3 4/6] ARM64: arch_timer: configure and enable event stream

[Sudeep KarkadaNagesha]

On 20/08/13 17:16, Will Deacon wrote:
> On Tue, Aug 20, 2013 at 03:27:33PM +0100, Sudeep KarkadaNagesha wrote:
>> On 20/08/13 14:27, Will Deacon wrote:
>>> On Tue, Aug 13, 2013 at 06:29:42PM +0100, Sudeep KarkadaNagesha wrote:
>>>> diff --git a/arch/arm64/include/asm/hwcap.h b/arch/arm64/include/asm/hwcap.h
>>>> index 6d4482f..effd8f9 100644
>>>> --- a/arch/arm64/include/asm/hwcap.h
>>>> +++ b/arch/arm64/include/asm/hwcap.h
>>>> @@ -30,6 +30,7 @@
>>>>  #define COMPAT_HWCAP_IDIVA	(1 << 17)
>>>>  #define COMPAT_HWCAP_IDIVT	(1 << 18)
>>>>  #define COMPAT_HWCAP_IDIV	(COMPAT_HWCAP_IDIVA|COMPAT_HWCAP_IDIVT)
>>>> +#define COMPAT_HWCAP_EVTSTRM	(1 << 21)
>>>
>>> Looking at the context, we should also add COMPAT_HWCAP_LPAE as a separate
>>> patch (not part of this series).
>>>
>> Ok make sense, I can do that. IIUC even this needs at least under some
>> config option, assuming can't be dynamic ?
> 
> I'm not sure I follow you... if you're asking whether the LPAE HWCAP should
> be dynamic for ARMv8, then no, since the capabilities offered by LPAE are
> included in ARMv8 as far as userspace is concerned.
>
Ok, then it can be in the default list of compat hwcap.

>>>>  #ifndef __ASSEMBLY__
>>>>  /*
>>>> @@ -41,7 +42,8 @@
>>>>  				 COMPAT_HWCAP_FAST_MULT|COMPAT_HWCAP_EDSP|\
>>>>  				 COMPAT_HWCAP_TLS|COMPAT_HWCAP_VFP|\
>>>>  				 COMPAT_HWCAP_VFPv3|COMPAT_HWCAP_VFPv4|\
>>>> -				 COMPAT_HWCAP_NEON|COMPAT_HWCAP_IDIV)
>>>> +				 COMPAT_HWCAP_NEON|COMPAT_HWCAP_IDIV|\
>>>> +				 COMPAT_HWCAP_EVTSTRM)
>>>
>>> So here you're hardcoding COMPAT_HWCAP_EVTSTRM in the COMPAT_ELF_HWCAP, yet
>>> we only enable the native one from the arch-timer driver. The question then is
>>> "Can we rely on the event-stream working for AArch64?". If so, we don't need
>>> the native hwcap at all. If not, then we need to make the compat hwcap
>>> dynamic.
>>
>> How about something like this ? I am not sure if
>> arch/arm64/kernel/setup.c is apt place for compat_elf_hwcap though.
> 
> Looks like the right sort of idea, yes. I can't think of a better place to
> put it. You could adjust it slightly so that COMPAT_ELF_HWCAP stays like it
> is, with an extra '| compat_dyn_elf_hwcap' on the end, but I have no real
> preference.
> 
Yes that's better, will do this way.

Regards,
Sudeep