Re: [PATCH v2 1/2] mm/filemap: count only the faulting address as a mmap hit

[Fujunjie <fujunjie1@qq.com>]

On Tue, Apr 28, 2026 at 02:40, Matthew Wilcox wrote:
> On Mon, Apr 27, 2026 at 04:57:50PM +0000, fujunjie wrote:
>> -		/*
>> -		 * If there are too many folios that are recently evicted
>> -		 * in a file, they will probably continue to be evicted.
>> -		 * In such situation, read-ahead is only a waste of IO.
>> -		 * Don't decrease mmap_miss in this scenario to make sure
>> -		 * we can stop read-ahead.
>> -		 */
> 
> I'm sad to lose this comment.  Why not move it to ...
> 
>> +		ret |= map_ret;
> 
> here?
> 
>> +		if ((map_ret & VM_FAULT_NOPAGE) &&
>> +		    !folio_test_workingset(folio)) {
>> +			unsigned short mmap_miss;
>> +
>> +			mmap_miss = READ_ONCE(file->f_ra.mmap_miss);
>> +			if (mmap_miss)
>> +				WRITE_ONCE(file->f_ra.mmap_miss,
>> +					   mmap_miss - 1);
>> +		}

Yes, that makes sense. I'll move the comment there in the next version.

Best regards,
fujunjie