Re: [PATCH RFC 3/8] memory-provider: dmabuf devmem memory provider

[Yunsheng Lin]

On 2023/11/16 1:44, Mina Almasry wrote:
> On Wed, Nov 15, 2023 at 1:21 AM Yunsheng Lin <linyunsheng@xxxxxxxxxx> wrote:
>>
>> On 2023/11/14 21:16, Jason Gunthorpe wrote:
>>> On Tue, Nov 14, 2023 at 04:21:26AM -0800, Mina Almasry wrote:
>>>
>>>> Actually because you put the 'strtuct page for devmem' in
>>>> skb->bv_frag, the net stack will grab the 'struct page' for devmem
>>>> using skb_frag_page() then call things like page_address(), kmap,
>>>> get_page, put_page, etc, etc, etc.
>>>
>>> Yikes, please no. If net has its own struct page look alike it has to
>>> stay entirely inside net. A non-mm owned struct page should not be
>>> passed into mm calls. It is just way too hacky to be seriously
>>> considered :(
>>
>> Yes, that is something this patchset is trying to do, defining its own
>> struct page look alike for page pool to support devmem.
>>
>> struct page for devmem will not be called into the mm subsystem, so most
>> of the mm calls is avoided by calling into the devmem memory provider'
>> ops instead of calling mm calls.
>>
>> As far as I see for now, only page_ref_count(), page_is_pfmemalloc() and
>> PageTail() is called for devmem page, which should be easy to ensure that
>> those call for devmem page is consistent with the struct page owned by mm.
> 
> I'm not sure this is true. These 3 calls are just the calls you're
> aware of. In your proposal you're casting mirror pages into page* and
> releasing them into the net stack. You need to scrub the entire net
> stack for mm calls, i.e. all driver code and all skb_frag_page() call
> sites. Of the top of my head, the driver is probably calling
> page_address() and illegal_highdma() is calling PageHighMem(). TCP
> zerocopy receive is calling vm_insert_pages().

For net stack part, I believe your patch below is handling to aovid those
mm calls? I don't include it in this patchset as I thought it is obvious
that whatever the proposal is, we always need those checking.
Maybe we should have included it to avoid this kind of confusion.
https://lore.kernel.org/all/20231106024413.2801438-10-almasrymina@xxxxxxxxxx/

For driver part, I was thinking if the driver supports devmem, it should check
that if it can call page_address() related call on a specific 'stuct page', or
maybe we should introduce a new helper to make it obvious?