Re: [RFC PATCH v1 1/2] mm: swap: Remove CLUSTER_FLAG_HUGE from swap_cluster_info:flags

From: Ryan Roberts
Date: Wed Oct 11 2023 - 06:16:58 EST


On 11/10/2023 09:17, Kefeng Wang wrote:
>
>
> On 2023/10/10 22:21, Ryan Roberts wrote:
>> As preparation for supporting small-sized THP in the swap-out path,
>> without first needing to split to order-0, Remove the CLUSTER_FLAG_HUGE,
>> which, when present, always implies PMD-sized THP, which is the same as
>> the cluster size.
>>
>> The only use of the flag was to determine whether a swap entry refers to
>> a single page or a PMD-sized THP in swap_page_trans_huge_swapped().
>> Instead of relying on the flag, we now pass in nr_pages, which
>> originates from the folio's number of pages. This allows the logic to
>> work for folios of any order.
>>
>> The one snag is that one of the swap_page_trans_huge_swapped() call
>> sites does not have the folio. But it was only being called there to
>> avoid bothering to call __try_to_reclaim_swap() in some cases.
>> __try_to_reclaim_swap() gets the folio and (via some other functions)
>> calls swap_page_trans_huge_swapped(). So I've removed the problematic
>> call site and believe the new logic should be equivalent.
>>
>> Removing CLUSTER_FLAG_HUGE also means we can remove split_swap_cluster()
>> which used to be called during folio splitting, since
>> split_swap_cluster()'s only job was to remove the flag.
>>
>> Signed-off-by: Ryan Roberts <ryan.roberts@xxxxxxx>
>> ---
>>   include/linux/swap.h | 10 ----------
>>   mm/huge_memory.c     |  3 ---
>>   mm/swapfile.c        | 47 ++++++++------------------------------------
>>   3 files changed, 8 insertions(+), 52 deletions(-)
>>
>> diff --git a/include/linux/swap.h b/include/linux/swap.h
>> index 19f30a29e1f1..a073366a227c 100644
>> --- a/include/linux/swap.h
>> +++ b/include/linux/swap.h
>> @@ -259,7 +259,6 @@ struct swap_cluster_info {
>>   };
>>   #define CLUSTER_FLAG_FREE 1 /* This cluster is free */
>>   #define CLUSTER_FLAG_NEXT_NULL 2 /* This cluster has no next cluster */
>> -#define CLUSTER_FLAG_HUGE 4 /* This cluster is backing a transparent huge
>> page */
>>
>>   /*
>>    * We assign a cluster to each CPU, so each CPU can allocate swap entry from
>> @@ -595,15 +594,6 @@ static inline int add_swap_extent(struct swap_info_struct
>> *sis,
>>   }
>>   #endif /* CONFIG_SWAP */
>>
>> -#ifdef CONFIG_THP_SWAP
>> -extern int split_swap_cluster(swp_entry_t entry);
>> -#else
>> -static inline int split_swap_cluster(swp_entry_t entry)
>> -{
>> -    return 0;
>> -}
>> -#endif
>> -
>>   #ifdef CONFIG_MEMCG
>>   static inline int mem_cgroup_swappiness(struct mem_cgroup *memcg)
>>   {
>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>> index c9cbcbf6697e..46b3fb943207 100644
>> --- a/mm/huge_memory.c
>> +++ b/mm/huge_memory.c
>> @@ -2597,9 +2597,6 @@ static void __split_huge_page(struct page *page, struct
>> list_head *list,
>>           shmem_uncharge(head->mapping->host, nr_dropped);
>>       remap_page(folio, nr);
>>
>> -    if (folio_test_swapcache(folio))
>> -        split_swap_cluster(folio->swap);
>> -
>>       for (i = 0; i < nr; i++) {
>>           struct page *subpage = head + i;
>>           if (subpage == page)
>> diff --git a/mm/swapfile.c b/mm/swapfile.c
>> index e52f486834eb..c668838fa660 100644
>> --- a/mm/swapfile.c
>> +++ b/mm/swapfile.c
>> @@ -342,18 +342,6 @@ static inline void cluster_set_null(struct
>> swap_cluster_info *info)
>>       info->data = 0;
>>   }
>>
>> -static inline bool cluster_is_huge(struct swap_cluster_info *info)
>> -{
>> -    if (IS_ENABLED(CONFIG_THP_SWAP))
>> -        return info->flags & CLUSTER_FLAG_HUGE;
>> -    return false;
>> -}
>> -
>> -static inline void cluster_clear_huge(struct swap_cluster_info *info)
>> -{
>> -    info->flags &= ~CLUSTER_FLAG_HUGE;
>> -}
>> -
>>   static inline struct swap_cluster_info *lock_cluster(struct swap_info_struct
>> *si,
>>                                unsigned long offset)
>>   {
>> @@ -1021,7 +1009,7 @@ static int swap_alloc_cluster(struct swap_info_struct
>> *si, swp_entry_t *slot)
>>       offset = idx * SWAPFILE_CLUSTER;
>>       ci = lock_cluster(si, offset);
>>       alloc_cluster(si, idx);
>> -    cluster_set_count_flag(ci, SWAPFILE_CLUSTER, CLUSTER_FLAG_HUGE);
>> +    cluster_set_count_flag(ci, SWAPFILE_CLUSTER, 0);
>
> Maybe just use cluster_set_count() and kill cluster_set_count_flag().

Yes, good point - I'll do this in the next version. Thanks!