Re: [PATCH v2] mm: remove redundant lru_add_drain() prior to unmapping pages

[Jianfeng Wang <jianfeng.w.wang@oracle.com>]

On 12/14/23 3:00 PM, Matthew Wilcox wrote:
> On Thu, Dec 14, 2023 at 02:27:17PM -0800, Jianfeng Wang wrote:
>> When unmapping VMA pages, pages will be gathered in batch and released by
>> tlb_finish_mmu() if CONFIG_MMU_GATHER_NO_GATHER is not set. The function
>> tlb_finish_mmu() is responsible for calling free_pages_and_swap_cache(),
>> which calls lru_add_drain() to drain cached pages in folio_batch before
>> releasing gathered pages. Thus, it is redundant to call lru_add_drain()
>> before gathering pages, if CONFIG_MMU_GATHER_NO_GATHER is not set.
>>
>> Remove lru_add_drain() prior to gathering and unmapping pages in
>> exit_mmap() and unmap_region() if CONFIG_MMU_GATHER_NO_GATHER is not set.
>>
>> Note that the page unmapping process in oom_killer (e.g., in
>> __oom_reap_task_mm()) also uses tlb_finish_mmu() and does not have
>> redundant lru_add_drain(). So, this commit makes the code more consistent.
> 
> Shouldn't we put this in __tlb_gather_mmu() which already has the
> CONFIG_MMU_GATHER_NO_GATHER ifdefs?  That would presuambly help with, eg
> zap_page_range_single() too.
> 

Thanks. It makes sense to me.
This commit is motivated by a workload that use mmap/unmap heavily.
While the mmu_gather feature is also used by hugetlb, madvise, mprotect,
etc., thus I prefer to have another standalone commit (following this one)
that moves lru_add_drain() to __tlb_gather_mmu() to unify these cases for
not making redundant lru_add_drain() calls when using mmu_gather.

>> Signed-off-by: Jianfeng Wang <jianfeng.w.wang@oracle.com>
>> ---
>>  mm/mmap.c | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/mm/mmap.c b/mm/mmap.c
>> index 1971bfffcc03..da0308eef435 100644
>> --- a/mm/mmap.c
>> +++ b/mm/mmap.c
>> @@ -2330,7 +2330,10 @@ static void unmap_region(struct mm_struct *mm, struct ma_state *mas,
>>  	struct mmu_gather tlb;
>>  	unsigned long mt_start = mas->index;
>>  
>> +	/* Defer lru_add_drain() to tlb_finish_mmu() for the ifndef case. */
>> +#ifdef CONFIG_MMU_GATHER_NO_GATHER
>>  	lru_add_drain();
>> +#endif
>>  	tlb_gather_mmu(&tlb, mm);
>>  	update_hiwater_rss(mm);
>>  	unmap_vmas(&tlb, mas, vma, start, end, tree_end, mm_wr_locked);
>> @@ -3300,7 +3303,10 @@ void exit_mmap(struct mm_struct *mm)
>>  		return;
>>  	}
>>  
>> +	/* Defer lru_add_drain() to tlb_finish_mmu() for the ifndef case. */
>> +#ifdef CONFIG_MMU_GATHER_NO_GATHER
>>  	lru_add_drain();
>> +#endif
>>  	flush_cache_mm(mm);
>>  	tlb_gather_mmu_fullmm(&tlb, mm);
>>  	/* update_hiwater_rss(mm) here? but nobody should be looking */
>> -- 
>> 2.42.1
>>
>>