On 2023/2/28 10:52, Waiman Long wrote:
> On 2/27/23 21:40, Haifeng Xu wrote:
>>
>> On 2023/2/27 22:56, Waiman Long wrote:
>>> On 2/27/23 03:07, Haifeng Xu wrote:
>>>> Commit 002f290627c2 ("cpuset: use static key better and convert to new API")
>>>> has used __cpuset_node_allowed instead of cpuset_node_allowed to check
>>>> whether we can allocate on a memory node. Now this function isn't used by
>>>> anyone, so we can remove it safely.
>>>>
>>>> Signed-off-by: Haifeng Xu <haifeng.xu@xxxxxxxxxx>
>>>> ---
>>>> include/linux/cpuset.h | 12 ------------
>>>> 1 file changed, 12 deletions(-)
>>>>
>>>> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
>>>> index d58e0476ee8e..7fad5afe3bba 100644
>>>> --- a/include/linux/cpuset.h
>>>> +++ b/include/linux/cpuset.h
>>>> @@ -82,13 +82,6 @@ int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask);
>>>> extern bool __cpuset_node_allowed(int node, gfp_t gfp_mask);
>>>> -static inline bool cpuset_node_allowed(int node, gfp_t gfp_mask)
>>>> -{
>>>> - if (cpusets_enabled())
>>>> - return __cpuset_node_allowed(node, gfp_mask);
>>>> - return true;
>>>> -}
>>>> -
>>>> static inline bool __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
>>>> {
>>>> return __cpuset_node_allowed(zone_to_nid(z), gfp_mask);
>>>> @@ -223,11 +216,6 @@ static inline int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask)
>>>> return 1;
>>>> }
>>>> -static inline bool cpuset_node_allowed(int node, gfp_t gfp_mask)
>>>> -{
>>>> - return true;
>>>> -}
>>>> -
>>>> static inline bool __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
>>>> {
>>>> return true;
>>> The kernel convention is to add a "__" prefix to a function name if there is higher level helper without the "__" prefix that uses it. Since cpuset_node_allowed() is no longer used. We should just rename __cpuset_node_allowed() to cpuset_node_allowed() and get rid of the unused helper. A bit more code changes are needed for this, though.
>>>
>>> Cheers,
>>> Longman
>>>
>> Maybe we can still use cpuset_node_allowed in __cpuset_zone_allowed? If so, less code need to be changed.
>>
> mm/page_alloc.c has 2 references to __cpuset_node_allowed() in their comments. These references should also be updated as well. That is why I said a bit more code needs to be touched.
>
> Cheers,
> Longman
>
Ok. I mean maybe we could keep cpuset_node_allowed() and use it in __cpuset_zone_allowed(). So we don't need rename __cpuset_node_allowed() or update releated comments.
Thanks.
