Re: [XEN PATCH v2 1/5] x86/acpi: remove acpi_pic_sci_set_trigger()

[Jan Beulich <jbeulich@xxxxxxxx>]

On 07.12.2023 11:09, Federico Serafini wrote:
> On 07/12/23 10:19, Jan Beulich wrote:
>> On 07.12.2023 09:47, Federico Serafini wrote:
>>> Remove apci_pic_set_trigger() declaration: there is no definition and there 
>>> are
>>> no calls to such function in the XEN project.
>>>
>>> Signed-off-by: Federico Serafini <federico.serafini@xxxxxxxxxxx>
>>
>> Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
>>
>> A reference to the offending commit would have been nice, the more that iirc
>> I had already gone and fished that out for you.
> 
> Is it correct to use Fixes: <id> ("subj") even if there are other
> useless entities left?

It was specifically because of being uncertain in this case that I didn't
mention a possible Fixes: tag. There's no breakage from stray declarations,
so it doesn't really feel to me like a "fix".

> In particular, this is what I think should be removed:
> 
> Functions:
> __acpi_acquire_global_lock()
> __acpi_release_global_lock()
> acpi_save_state{mem,disk}()
> acpi_restore_state_mem()
> acpi_enter_state()
> {save,restore}_rest_processor_state()
> 
> Variables:
> acpi_wakeup_address
> 
> Macros:
> ACPI_ACQUIRE_GLOBAL_LOCK()
> 
> If you agree with me, I can propose a new patch which includes such 
> removals and refers to the offending commit with a Fixes.

I haven't checked the entities above, but yes, I agree other stray
declarations would want removing as well. In fact I was assuming that
removal of just one item meant no other stray ones exist (right here).

Jan