Received-SPF: pass (google.com: domain of linux-kernel+bounces-203958-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Subject: Re: [PATCH] mm/page_alloc: skip THP-sized PCP list when allocating
 non-CMA THP-sized page
To: Baolin Wang <baolin.wang@linux.alibaba.com>, akpm@linux-foundation.org
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, liuzixing@hygon.cn,
 Johannes Weiner <hannes@cmpxchg.org>, Vlastimil Babka <vbabka@suse.cz>,
 Zi Yan <ziy@nvidia.com>
References: <1717492460-19457-1-git-send-email-yangge1116@126.com>
 <c180d2a0-1e34-41f0-bae8-1205d04a5f6b@linux.alibaba.com>
 <82d31425-86d7-16fa-d09b-fcb203de0986@126.com>
 <7087d0af-93d8-4d49-94f4-dc846a4e2b98@linux.alibaba.com>
From: yangge1116 <yangge1116@126.com>
Message-ID: <1250569c-c428-5a0c-f1b7-fe5013eace3f@126.com>
Date: Thu, 6 Jun 2024 17:10:58 +0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.10.0
Precedence: bulk
MIME-Version: 1.0
In-Reply-To: <7087d0af-93d8-4d49-94f4-dc846a4e2b98@linux.alibaba.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit


在 2024/6/6 上午11:06, Baolin Wang 写道:
> 
> 
> On 2024/6/4 20:36, yangge1116 wrote:
>>
>>
>> 在 2024/6/4 下午8:01, Baolin Wang 写道:
>>> Cc Johannes, Zi and Vlastimil.
>>>
>>> On 2024/6/4 17:14, yangge1116@126.com wrote:
>>>> From: yangge <yangge1116@126.com>
>>>>
>>>> Since commit 5d0a661d808f ("mm/page_alloc: use only one PCP list for
>>>> THP-sized allocations") no longer differentiates the migration type
>>>> of pages in THP-sized PCP list, it's possible to get a CMA page from
>>>> the list, in some cases, it's not acceptable, for example, allocating
>>>> a non-CMA page with PF_MEMALLOC_PIN flag returns a CMA page.
>>>>
>>>> The patch forbids allocating non-CMA THP-sized page from THP-sized
>>>> PCP list to avoid the issue above.
>>>>
>>>> Fixes: 5d0a661d808f ("mm/page_alloc: use only one PCP list for 
>>>> THP-sized allocations")
>>>> Signed-off-by: yangge <yangge1116@126.com>
>>>> ---
>>>>   mm/page_alloc.c | 10 ++++++++++
>>>>   1 file changed, 10 insertions(+)
>>>>
>>>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>>>> index 2e22ce5..0bdf471 100644
>>>> --- a/mm/page_alloc.c
>>>> +++ b/mm/page_alloc.c
>>>> @@ -2987,10 +2987,20 @@ struct page *rmqueue(struct zone 
>>>> *preferred_zone,
>>>>       WARN_ON_ONCE((gfp_flags & __GFP_NOFAIL) && (order > 1));
>>>>       if (likely(pcp_allowed_order(order))) {
>>>> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
>>>> +        if (!IS_ENABLED(CONFIG_CMA) || alloc_flags & ALLOC_CMA ||
>>>> +                        order != HPAGE_PMD_ORDER) {
>>>
>>> Seems you will also miss the non-CMA THP from the PCP, so I wonder if 
>>> we can add a migratetype comparison in __rmqueue_pcplist(), and if 
>>> it's not suitable, then fallback to buddy?
>>
>> Yes, we may miss some non-CMA THPs in the PCP. But, if add a 
>> migratetype comparison in __rmqueue_pcplist(), we may need to compare 
>> many times because of pcp batch.
> 
> I mean we can only compare once, focusing on CMA pages.

pcp_list may contains CMA and no-CMA pages, why only compare once, just 
increase one chance of using the pcp_list?


> 
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 3734fe7e67c0..960a3b5744d8 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -2973,6 +2973,11 @@ struct page *__rmqueue_pcplist(struct zone *zone, 
> unsigned int order,
>                  }
> 
>                  page = list_first_entry(list, struct page, pcp_list);
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +               if (order == HPAGE_PMD_ORDER && 
> !is_migrate_movable(migratetype) &&
> +                   is_migrate_cma(get_pageblock_migratetype(page)))
> +                       return NULL;
> +#endif
>                  list_del(&page->pcp_list);
>                  pcp->count -= 1 << order;
>          } while (check_new_pages(page, order));