On Thu, Apr 17 2025 at 17:42, Neeraj Upadhyay wrote:
> On 4/17/2025 4:22 PM, Thomas Gleixner wrote:
>>> +static void savic_update_vector(unsigned int cpu, unsigned int vector, bool set)
>>> +{
>>> + update_vector(cpu, SAVIC_ALLOWED_IRR, vector, set);
>>
>> This indirection is required because otherwise the code is too simple to
>> follow?
>>
> update_vector() is used by send_ipi_dest() in Patch 7. From your comment
> on v3 https://lore.kernel.org/lkml/87y0whv57k.ffs@tglx/ , what I understood
> was that you wanted update_vector() to be defined in the patch where that code
> is added (i.e. this patch) and not at a later patch. Is that not correct
> understanding?
Fair enough. I missed the later usage sites. Again, a short note in the
change log which explains the rationale would avoid this.
Thanks,
tglx
