linux-kernel - Re: [PATCH V3] mm/hugetlb: wait for hugetlb folios to be freed

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <950cae5a-bff0-49e6-8fe4-a2447c63d8bc@redhat.com>
Date: Tue, 18 Feb 2025 18:22:12 +0100
From: David Hildenbrand <david@...hat.com>
To: Ge Yang <yangge1116@....com>, akpm@...ux-foundation.org
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org, stable@...r.kernel.org,
 21cnbao@...il.com, baolin.wang@...ux.alibaba.com, muchun.song@...ux.dev,
 osalvador@...e.de, liuzixing@...on.cn
Subject: Re: [PATCH V3] mm/hugetlb: wait for hugetlb folios to be freed

On 18.02.25 13:19, Ge Yang wrote:
> 
> 
> 在 2025/2/18 19:45, David Hildenbrand 写道:
>> On 18.02.25 12:40, yangge1116@....com wrote:
>>> From: Ge Yang <yangge1116@....com>
>>>
>>> Since the introduction of commit c77c0a8ac4c52 ("mm/hugetlb: defer
>>> freeing
>>> of huge pages if in non-task context"), which supports deferring the
>>> freeing of hugetlb pages, the allocation of contiguous memory through
>>> cma_alloc() may fail probabilistically.
>>>
>>> In the CMA allocation process, if it is found that the CMA area is
>>> occupied
>>> by in-use hugetlb folios, these in-use hugetlb folios need to be migrated
>>> to another location. When there are no available hugetlb folios in the
>>> free hugetlb pool during the migration of in-use hugetlb folios, new
>>> folios
>>> are allocated from the buddy system. A temporary state is set on the
>>> newly
>>> allocated folio. Upon completion of the hugetlb folio migration, the
>>> temporary state is transferred from the new folios to the old folios.
>>> Normally, when the old folios with the temporary state are freed, it is
>>> directly released back to the buddy system. However, due to the deferred
>>> freeing of hugetlb pages, the PageBuddy() check fails, ultimately leading
>>> to the failure of cma_alloc().
>>>
>>> Here is a simplified call trace illustrating the process:
>>> cma_alloc()
>>>       ->__alloc_contig_migrate_range() // Migrate in-use hugetlb folios
>>>           ->unmap_and_move_huge_page()
>>>               ->folio_putback_hugetlb() // Free old folios
>>>       ->test_pages_isolated()
>>>           ->__test_page_isolated_in_pageblock()
>>>                ->PageBuddy(page) // Check if the page is in buddy
>>>
>>> To resolve this issue, we have implemented a function named
>>> wait_for_freed_hugetlb_folios(). This function ensures that the hugetlb
>>> folios are properly released back to the buddy system after their
>>> migration
>>> is completed. By invoking wait_for_freed_hugetlb_folios() before calling
>>> PageBuddy(), we ensure that PageBuddy() will succeed.
>>>
>>> Fixes: c77c0a8ac4c52 ("mm/hugetlb: defer freeing of huge pages if in
>>> non-task context")
>>> Signed-off-by: Ge Yang <yangge1116@....com>
>>> Cc: <stable@...r.kernel.org>
>>
>>
>>
>> Acked-by: David Hildenbrand <david@...hat.com>
>>> +void wait_for_freed_hugetlb_folios(void)
>>> +{
>>> +    flush_work(&free_hpage_work);
>>
>> BTW, I was wondering if we could optimize out some calls here by sensing
>> if there is actually work.
>>
> for_each_hstate(h) {
> 	if (hugetlb_vmemmap_optimizable(h)) {
> 		flush_work(&free_hpage_work);
 > 		break;> 	}
> }
> Is this adjustment okay?

I think that's better, except that it would still trigger in scenarios 
where hugetlb is completely unused if 
CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is around.

Can't we check hpage_freelist?

if (llist_empty(&hpage_freelist))
	return;
flush_work(&free_hpage_work);

It should be able to deal with races (we don't care if something is 
getting added concurrently, only if there is something right now).

-- 
Cheers,

David / dhildenb