Re: [RFC PATCH net-next] page_pool: Track DMA-mapped pages and unmap them when destroying the pool

[Shuah <shuah@kernel.org>]

On 3/12/25 06:05, Yunsheng Lin wrote:
> On 2025/3/11 23:11, Matthew Wilcox wrote:
>> On Tue, Mar 11, 2025 at 08:25:25PM +0800, Yunsheng Lin wrote:
>>>> struct page {
>>>> 	unsigned long flags;
>>>> 	unsigned long memdesc;
>>>
>>> It seems there may be memory behind the above 'memdesc' with different size
>>> and layout for different subsystem?
>>
>> Yes.
>>
>>> I am not sure if I understand the case of the same page might be handle in
>>> two subsystems concurrently or a page is allocated in one subsystem and
>>> then passed to be handled in other subsystem, for examlpe:
>>> page_pool owned page is mmap'ed into user space through tcp zero copy,
>>> see tcp_zerocopy_vm_insert_batch(), it seems the same page is handled in
>>> both networking/page_pool and vm subsystem?
>>
>> It's not that arbitrary.  I mean, you could read all the documentation
>> I've written about this concept, listen to the talks I've given.

You can't point to talk given on the concept - people don't have to go
find your talks to understand the concept. You are expected to answer
the question and explain it to us here in this thread.

But
>> sure, you're a special fucking snowflake and deserve your own unique
>> explanation.

Yunsheng Lin, This message is a rude personal attack. This isn't the
way to treat your peers in the community. Apology is warranted.

> 
> If you don't like responding to the above question/comment, I would rather
> you strip out them like the other question/comment or just ignore it:(
> 
> I am not sure how to interpret the comment, but I am sure it is not a kind
> one, so CC 'Code of Conduct Committee' in case there is more coming.
> 

Thank you Mathew for letting us know about this.

thanks,
-- Shuah ((on behalf of the CoC committee)