Re: [PATCH] xen/nodemask: Remove _unused_nodemask_arg_

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 12/01/2026 10:47 am, Jan Beulich wrote:
> On 12.01.2026 11:40, Andrew Cooper wrote:
>> This only exists to have it's type taken, despite there being a perfectly 
>> good
>> concrete type to use.
>>
>> No functional change.
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> Acked-by: Jan Beulich <jbeulich@xxxxxxxx>

Thanks.

>
>> --- a/xen/include/xen/nodemask.h
>> +++ b/xen/include/xen/nodemask.h
>> @@ -67,8 +67,6 @@ typedef struct { DECLARE_BITMAP(bits, MAX_NUMNODES); } 
>> nodemask_t;
>>  
>>  #define nodemask_bits(src) ((src)->bits)
>>  
>> -extern nodemask_t _unused_nodemask_arg_;
>> -
>>  #define node_set(node, dst) __node_set((node), &(dst))
>>  static inline void __node_set(int node, volatile nodemask_t *dstp)
>>  {
>> @@ -215,7 +213,7 @@ static inline int __last_node(const nodemask_t *srcp, 
>> int nbits)
>>  
>>  #define nodemask_of_node(node)                                              
>> \
>>  ({                                                                  \
>> -    typeof(_unused_nodemask_arg_) m;                                \
>> +    nodemask_t m;                                                   \
>>      if (sizeof(m) == sizeof(unsigned long)) {                       \
>>              m.bits[0] = 1UL<<(node);                                \
>>      } else {                                                        \
> Hard to see why Linux would have introduced that either. (It still has it, 
> btw.)

Yeah, it is a bizarre construct.  I only noticed it when trying to do
some more typeof cleanup.

~Andrew