Re: [PATCH] mm: move idle swap cache pages to the tail of LRU after COW

From: "Huang, Ying" <ying.huang@intel.com>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	Matthew Wilcox <willy@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Xu <peterx@redhat.com>, Hugh Dickins <hughd@google.com>,
	Mel Gorman <mgorman@suse.de>, Rik van Riel <riel@surriel.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Michal Hocko <mhocko@kernel.org>,
	Dave Hansen <dave.hansen@intel.com>,
	Tim Chen <tim.c.chen@intel.com>
Subject: Re: [PATCH] mm: move idle swap cache pages to the tail of LRU after COW
Date: Thu, 20 May 2021 09:22:45 +0800	[thread overview]
Message-ID: <87r1i28ahm.fsf@yhuang6-desk1.ccr.corp.intel.com> (raw)
In-Reply-To: <YKUlfeAiq/vv+dHl@cmpxchg.org> (Johannes Weiner's message of "Wed, 19 May 2021 10:49:33 -0400")

Johannes Weiner <hannes@cmpxchg.org> writes:

> On Wed, May 19, 2021 at 09:33:13AM +0800, Huang Ying wrote:
>> diff --git a/mm/memory.c b/mm/memory.c
>> index b83f734c4e1d..2b6847f4c03e 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -3012,6 +3012,11 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
>>  				munlock_vma_page(old_page);
>>  			unlock_page(old_page);
>>  		}
>> +		if (page_copied && PageSwapCache(old_page) &&
>> +		    !page_mapped(old_page) && trylock_page(old_page)) {
>> +			try_to_free_idle_swapcache(old_page);
>> +			unlock_page(old_page);
>
> If there are no more swap or pte references, can we just attempt to
> free the page right away, like we do during regular unmap?
>
> 		if (page_copied)
> 			free_swap_cache(old_page);
> 		put_page(old_page);

A previous version of the patch does roughly this.

https://lore.kernel.org/lkml/20210113024241.179113-1-ying.huang@intel.com/

But Linus has concerns with the overhead introduced in the hot COW path.

Another possibility is to move the idle swap cache page to the tail of
the file LRU list.  But the question is how to identify the page.

Best Regards,
Huang, Ying