Re: [PATCH 12/12] swiotlb-xen: this is PV-only on x86

[Stefano Stabellini]

On Mon, 13 Sep 2021, Jan Beulich wrote:
> On 11.09.2021 01:48, Stefano Stabellini wrote:
> > On Wed, 8 Sep 2021, Christoph Hellwig wrote:
> >> On Tue, Sep 07, 2021 at 02:13:21PM +0200, Jan Beulich wrote:
> >>> The code is unreachable for HVM or PVH, and it also makes little sense
> >>> in auto-translated environments. On Arm, with
> >>> xen_{create,destroy}_contiguous_region() both being stubs, I have a hard
> >>> time seeing what good the Xen specific variant does - the generic one
> >>> ought to be fine for all purposes there. Still Arm code explicitly
> >>> references symbols here, so the code will continue to be included there.
> >>
> >> Can the Xen/arm folks look into that?  Getting ARM out of using
> >> swiotlb-xen would be a huge step forward cleaning up some DMA APIs.
> > 
> > On ARM swiotlb-xen is used for a different purpose compared to x86.
> > 
> > Many ARM SoCs still don't have an IOMMU covering all DMA-mastering
> > devices (e.g. Raspberry Pi 4). As a consequence we map Dom0 1:1 (guest
> > physical == physical address).
> > 
> > Now if it was just for Dom0, thanks to the 1:1 mapping, we wouldn't need
> > swiotlb-xen. But when we start using PV drivers to share the network or
> > disk between Dom0 and DomU we are going to get DomU pages mapped in
> > Dom0, we call them "foreign pages".  They are not mapped 1:1. It can
> > happen that one of these foreign pages are used for DMA operations
> > (e.g. related to the NIC). swiotlb-xen is used to detect these
> > situations and translate the guest physical address to physical address
> > of foreign pages appropriately.
> 
> Hmm, you say "translate", which isn't my understanding of swiotlb's
> purpose. As per my understanding swiotlb instead double buffers data
> such that is becomes accessible, or suitably arranges underlying
> machine addresses. The latter part is clearly a PV-only thing, unused
> by Arm as can be seen by there not being any use of XENMEM_exchange.
> So it must be the former part that you're talking about, but that's
> also the purpose of the non-Xen swiotlb code. If only for my own
> education and understanding, could you point me at the difference
> between swiotlb-xen and generic swiotlb which addresses this specific
> aspect of Arm behavior?

If you look at xen_swiotlb_map_page, you'll see the call to
xen_phys_to_dma which eventually calls arch/arm/xen/p2m.c:__pfn_to_mfn.
If everything goes well and we only need to do translation we'll "goto
done". Otherwise, we'll fall back on a swiotlb buffer with
swiotlb_tbl_map_single, the result of which also needs to be translated,
see the second call to xen_phys_to_dma.