On Mon, Sep 12, 2016 at 8:57 AM, Will Deacon <will.deacon@xxxxxxx> wrote: > > On Mon, Sep 12, 2016 at 06:33:43PM +0300, Michael S. Tsirkin wrote: > > On Mon, Sep 12, 2016 at 01:10:41PM +0100, Will Deacon wrote: > > > Legacy virtio defines the virtqueue base using a 32-bit PFN field, with > > > a read-only register indicating a fixed page size of 4k. > > > > > > This can cause problems for DMA allocators that allocate top down from > > > the DMA mask, which is set to 64 bits. In this case, the addresses are > > > silently truncated to 44-bit, leading to IOMMU faults, failure to read > > > from the queue or data corruption. > > > > > > This patch restricts the DMA mask for legacy PCI virtio devices to > > > 44 bits, which matches the specification. > > > > > > Cc: Andy Lutomirski <luto@xxxxxxxxxx> > > > Cc: Michael S. Tsirkin <mst@xxxxxxxxxx> > > > Signed-off-by: Will Deacon <will.deacon@xxxxxxx> > > > > Hmm - IIUC it's actually only the case for the virtio rings > > themselves. The buffer addresses put in the rings are full 64 bit ones. > > I think that's right, yes. Yes. > > > > It so happens that virtio doesn't use coherent allocs except > > for the rings. > > So I'm inclined to say the coherent mask should be set to 44, > > with a comment explaning that this is for the rings. > > I can certainly add that in v2, along with your suggestion to use > 32 + VIRTIO_PCI_QUEUE_ADDR_SHIFT instead of the 44. > > > In case we start using coherent allocations in virtio, > > it might be cleaner to relax the mask after allocating > > the rings, but I'm not sure that's allowed by the DMA API. > > thoughts? > > Hmm, that *might* work, but I could certainly imagine some DMA > implementations going wrong if they assume the mask is fixed. There's probably a hack to make the buffer addresses be 64-bit capable while letting the queue addresses be 44-bit, but is it worth the trouble? Maybe assume that guests with more than 16 TB of RAM are rare today, and catch the Virtio 1.x spec to fix the 32-bit PFN limit? Alternately, you could keep the DMA mask at 64 bits but try to allocate the rings early, and abort if they can't get a < 16 TB address. I think I prefer the 44 bit mask approach, though. > > > Will > _______________________________________________ > Virtualization mailing list > Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx > https://lists.linuxfoundation.org/mailman/listinfo/virtualization _______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization
