Re: [PATCH v2] vma remove the unneeded avc bound with non-CoWed folio

[David Hildenbrand <david@redhat.com>]

On 25.08.24 08:42, Lorenzo Stoakes wrote:
> On Sun, Aug 25, 2024 at 01:06:40PM GMT, zhiguojiang wrote:
>>
>>
>> 在 2024/8/25 0:26, Lorenzo Stoakes 写道:
>>> [Some people who received this message don't often get email from lorenzo.stoakes@oracle.com. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
>>>
>>> On Fri, Aug 23, 2024 at 11:02:06PM GMT, Zhiguo Jiang wrote:
>>>> After CoWed by do_wp_page, the vma established a new mapping relationship
>>>> with the CoWed folio instead of the non-CoWed folio. However, regarding
>>>> the situation where vma->anon_vma and the non-CoWed folio's anon_vma are
>>>> not same, the avc binding relationship between them will no longer be
>>>> needed, so it is issue for the avc binding relationship still existing
>>>> between them.
>>>>
>>>> This patch will remove the avc binding relationship between vma and the
>>>> non-CoWed folio's anon_vma, which each has their own independent
>>>> anon_vma. It can also alleviates rmap overhead simultaneously.
>>>>
>>>> Signed-off-by: Zhiguo Jiang <justinjiang@vivo.com>
>>>
>>> NACK (until fixed). This is broken (see below).
>>>
>> Hi Lorenzo Stoakes,
>>
>> Thank you for your comments.
>>> I'm not seeing any numbers to back anything up here as to why we want to
>>> make changes to this incredibly sensitive code?
>> I added a debug trace log (as follows) in wp_page_copy() and observed
>> that a large number of these orphan avc-objects still exist. I believe
>> this will have a certain redundant overhead impact on anonymous folios'
>> rmap avcs, so I want to remove it, which is also the most essential
>> value of this patch.
> 
> Sorry nack to that idea unless you can provide actual _data_ to demonstrate
> an overhead.

Agreed.

-- 
Cheers,

David / dhildenb