On 13.11.2012, at 03:47, David Gibson wrote:
> From: Alexey Kardashevskiy <a...@ozlabs.ru>
>
> In future (with VFIO) we will have multiple PCI host bridges on
> pseries. Each one needs a unique LIOBN (IOMMU id). At the moment we
> derive these from the pci domain number, but the whole notion of
> domain numbers on the qemu side is bogus and in any case they're not
> actually uniquely allocated at this point.
>
> This patch, therefore uses a simple sequence counter to generate
> unique LIOBNs for PCI host bridges.
>
> Signed-off-by: Alexey Kardashevskiy <a...@ozlabs.ru>
> Signed-off-by: David Gibson <da...@gibson.dropbear.id.au>
I don't really like the idea of having a global variable just because our
domain ID generation seems to not work as expected. Michael, any comments here?
Alex
> ---
> hw/spapr_pci.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/hw/spapr_pci.c b/hw/spapr_pci.c
> index 3c5b855..f6544d7 100644
> --- a/hw/spapr_pci.c
> +++ b/hw/spapr_pci.c
> @@ -521,6 +521,7 @@ static int spapr_phb_init(SysBusDevice *s)
> char *namebuf;
> int i;
> PCIBus *bus;
> + static int phbnum;
>
> sphb->dtbusname = g_strdup_printf("pci@%" PRIx64, sphb->buid);
> namebuf = alloca(strlen(sphb->dtbusname) + 32);
> @@ -572,7 +573,7 @@ static int spapr_phb_init(SysBusDevice *s)
> PCI_DEVFN(0, 0), PCI_NUM_PINS);
> phb->bus = bus;
>
> - sphb->dma_liobn = SPAPR_PCI_BASE_LIOBN | (pci_find_domain(bus) << 16);
> + sphb->dma_liobn = SPAPR_PCI_BASE_LIOBN | (++phbnum << 16);
> sphb->dma_window_start = 0;
> sphb->dma_window_size = 0x40000000;
> sphb->dma = spapr_tce_new_dma_context(sphb->dma_liobn,
> sphb->dma_window_size);
> --
> 1.7.10.4
>
