Re: [RFC PATCH 14/15] PCI/P2PDMA: Introduce pci_mmap_p2pmem()

[Logan Gunthorpe]

On 2020-11-06 11:09 a.m., Jason Gunthorpe wrote:
>> Ah, hmm, yes. I guess the pages have to be hooked and returned to the
>> genalloc through free_devmap_managed_page(). 
> 
> That sounds about right, but in this case it doesn't need the VMA
> operations.
> 
>> Seems like it might be doable... but it will complicate things for
>> users that don't want to use the genpool (though no such users exist
>> upstream).
> 
> I would like to use this stuff in RDMA pretty much immediately and the
> genpool is harmful for those cases, so please don't make decisions
> that are tying thing to genpool

I certainly can't make decisions for code that isn't currently upstream.
So you will almost certainly have to make changes for the code you want
to add, as is the standard procedure. I can't and should not export APIs
that you might need that have no upstream users, but you are certainly
free to send patches that create them when you add the use case.

Ultimately, if you aren't using the genpool you will have to implement
your own mmap operation that somehow allocates the pages and your own
page_free hook. The latter can be accommodated for by a patch that
splits off pci_p2pdma_add_resource() into a function that doesn't use
the genpool (I've already seen two independent developers create a
similar patch for this but with no upstream user, they couldn't be taken
upstream).

I also don't expect this to be going upstream in the near term so don't
get too excited about using it.

Logan