Re: [PATCH net-next 04/12] net: homa: create homa_pool.h and homa_pool.c

On Wed, Oct 30, 2024 at 9:03 AM Andrew Lunn [off-list ref] wrote:

On Wed, Oct 30, 2024 at 08:46:33AM -0700, John Ousterhout wrote:

quoted

On Wed, Oct 30, 2024 at 5:54 AM Andrew Lunn [off-list ref] wrote:

quoted

quoted

I think this is a different problem from what page pools solve. Rather
than the application providing a buffer each time it calls recvmsg, it
provides a large region of memory in its virtual address space in
advance;

Ah, O.K. Yes, page pool is for kernel memory. However, is the virtual
address space mapped to pages and pinned? Or do you allocate pages
into that VM range as you need them? And then free them once the
application says it has completed? If you are allocating and freeing
pages, the page pool might be useful for these allocations.

Homa doesn't allocate or free pages for this: the application mmap's a
region and passes the virtual address range to Homa. Homa doesn't need
to pin the pages. This memory is used in a fashion similar to how a
buffer passed to recvmsg would be used, except that Homa maintains
access to the region for the lifetime of the associated socket. When
the application finishes processing an incoming message, it notifies
Homa so that Homa can reuse the message's buffer space for future
messages; there's no page allocation or freeing in this process.

I clearly don't know enough about memory management! I would of
expected the kernel to do lazy allocation of pages to VM addresses as
needed. Maybe it is, and when you actually access one of these missing
pages, you get a page fault and the MM code is kicking in to put an
actual page there? This could all be hidden inside the copy_to_user()
call.

Yes, this is all correct.  MM code gets called during copy_to_user to
allocate pages as needed (I should have been more clear: Homa doesn't
allocate or free pages directly). Homa tries to be clever about using
the buffer region to minimize the number of physical pages that
actually need to be allocated (it tries to allocate at the beginning
of the region, only using higher addresses when the lower addresses
are in use).

quoted

quoted

Taking a step back here, the kernel already has a number of allocators
and ideally we don't want to add yet another one unless it is really
required. So it would be good to get some reviews from the MM people.

I'm happy to do that if you still think it's necessary; how do I do that?

Reach out to Andrew Morton [off-list ref], the main
Memory Management Maintainer. Ask who a good person would be to review
this code.

I have started this process in a separate email.

-John-