On 24.10.2023 15:40, Nicola Vetrini wrote:
> On 24/10/2023 10:12, Jan Beulich wrote:
>> On 24.10.2023 09:58, Nicola Vetrini wrote:
>>> On 24/10/2023 09:32, Jan Beulich wrote:
>>>> On 23.10.2023 11:56, Nicola Vetrini wrote:
>>>>> --- a/xen/arch/x86/include/asm/asm_defns.h
>>>>> +++ b/xen/arch/x86/include/asm/asm_defns.h
>>>>> @@ -31,6 +31,7 @@ asm ( "\t.equ CONFIG_INDIRECT_THUNK, "
>>>>> * gets set up by the containing function.
>>>>> */
>>>>> #ifdef CONFIG_FRAME_POINTER
>>>>> +/* SAF-1-safe */
>>>>> register unsigned long current_stack_pointer asm("rsp");
>>>>> # define ASM_CALL_CONSTRAINT , "+r" (current_stack_pointer)
>>>>> #else
>>>>
>>>> SAF-1-safe is about symbols "used only by asm modules". This doesn't
>>>> apply
>>>> to the declaration here.
>>>>
>>>
>>> The wording could change to "asm code" if that is deemed clearer.
>>
>> Question is what would be meant by "asm code"; "asm modules" is quite
>> clear.
>>
>
> Well, I don't know. It's up to the community to decide that. It can be
> an ad-hoc
> justification, but I don't see much value in doing so. What do you
> propose to get this patch
> approved (at least on your account)?.
Drop this change and have Eclair recognize that what we're talking
about here is just a declaration, not a definition.
Jan
