On 10/23/24 04:54, Sourav Panda wrote:
> alloc_tag_sub() takes bytes as opposed to number of pages as argument.
>
> Currently pgalloc_tag_copy() passes the number of pages. This fix passes
> the correct unit, which is the number of bytes allocated.
>
> Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()")
> Signed-off-by: Sourav Panda <souravpanda@xxxxxxxxxx>
> Acked-by: Suren Baghdasaryan <surenb@xxxxxxxxxx>
> Reviewed-by: Pasha Tatashin <pasha.tatashin@xxxxxxxxxx>
> ---
> Changelog:
> Added Acked-by: Suren Baghdasaryan <surenb@xxxxxxxxxx>
> Added Reviewed-by: Pasha Tatashin <pasha.tatashin@xxxxxxxxxx>
Very minor nit.
Just mentioning that the tags have been picked up should suffice.
> Fixed the nit pointed by Wei in the commit message
>
> v1:
> https://lore.kernel.org/all/20241022205622.133697-1-souravpanda@xxxxxxxxxx/
> ---
> include/linux/mm.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index ecf63d2b0582..2890001c8b97 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -4203,7 +4203,7 @@ static inline void pgalloc_tag_copy(struct folio *new, struct folio *old)
> /* Clear the old ref to the original allocation tag. */
> clear_page_tag_ref(&old->page);
> /* Decrement the counters of the tag on get_new_folio. */
> - alloc_tag_sub(ref, folio_nr_pages(new));
> + alloc_tag_sub(ref, folio_size(new));
>
> __alloc_tag_ref_set(ref, tag);
>
Reviewed-by: Anshuman Khandual <anshuman.khandual@xxxxxxx>
