Messages in this thread | | | From | Ilias Apalodimas <> | Date | Thu, 17 Aug 2023 19:59:37 +0300 | Subject | Re: [PATCH net-next v7 1/6] page_pool: frag API support for 32-bit arch with 64-bit DMA |
| |
Hi Jakub,
On Thu, 17 Aug 2023 at 19:15, Jakub Kicinski <kuba@kernel.org> wrote: > > On Thu, 17 Aug 2023 16:57:16 +0300 Ilias Apalodimas wrote: > > Why should we care about this? Even an architecture that's 32-bit and > > has a 64bit DMA should be allowed to split the pages internally if it > > decides to do so. The trick that drivers usually do is elevate the > > page refcnt and deal with that internally. > > Can we assume the DMA mapping of page pool is page aligned? We should > be, right?
Yes
> That means we're storing 12 bits of 0 at the lower end. > So even with 32b of space we can easily store addresses for 32b+12b => > 16TB of memory. "Ought to be enough" to paraphrase Bill G, and the > problem is only in our heads?
Do you mean moving the pp_frag_count there? I was questioning the need to have PP_FLAG_PAGE_SPLIT_IN_DRIVER overall. With Yunshengs patches such a platform would allocate a page, so why should we prevent it from splitting it internally?
Thanks /Ilias > > Before we go that way - Mina, are the dma-buf "chunks" you're working > with going to be fragment-able? Or rather can driver and/or core take > multiple references on a single buffer?
| |