This bridge sometimes shows up as a root complex device and sometimes
as a discrete PCIe-to-PCI bridge. Testing indicates that in the
latter case, we need to enable the PCIe bridge DMA alias quirk.
Signed-off-by: Alex Williamson <[email protected]>
Reported-by: Milos Kaurin <[email protected]>
Tested-by: Milos Kaurin <[email protected]>
---
drivers/pci/quirks.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 78a7df6..460c354 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -3405,6 +3405,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
/* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
+/* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
+DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
/*
* AMD has indicated that the devices below do not support peer-to-peer
On Mon, Jun 23, 2014 at 04:36:25PM -0600, Alex Williamson wrote:
> This bridge sometimes shows up as a root complex device and sometimes
> as a discrete PCIe-to-PCI bridge. Testing indicates that in the
> latter case, we need to enable the PCIe bridge DMA alias quirk.
>
> Signed-off-by: Alex Williamson <[email protected]>
> Reported-by: Milos Kaurin <[email protected]>
> Tested-by: Milos Kaurin <[email protected]>
Applied to for-linus for v3.16, thanks!
> ---
>
> drivers/pci/quirks.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 78a7df6..460c354 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -3405,6 +3405,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
> DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
> /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
> DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> +/* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
> +DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
>
> /*
> * AMD has indicated that the devices below do not support peer-to-peer
>
[+cc Joerg]
On Sat, Jul 05, 2014 at 10:31:43AM -0600, Bjorn Helgaas wrote:
> On Mon, Jun 23, 2014 at 04:36:25PM -0600, Alex Williamson wrote:
> > This bridge sometimes shows up as a root complex device and sometimes
> > as a discrete PCIe-to-PCI bridge. Testing indicates that in the
> > latter case, we need to enable the PCIe bridge DMA alias quirk.
> >
> > Signed-off-by: Alex Williamson <[email protected]>
> > Reported-by: Milos Kaurin <[email protected]>
> > Tested-by: Milos Kaurin <[email protected]>
>
> Applied to for-linus for v3.16, thanks!
Actually, this quirk doesn't make any difference until the iommu changes
are in, so there's no reason to have this in v3.16, is there?
I saw that Joerg applied those iommu changes to his core branch, which I
assume will be merged for v3.17. So I'll move this to a pci/iommu branch,
to be merged during the v3.17 merge window.
> > ---
> >
> > drivers/pci/quirks.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> > index 78a7df6..460c354 100644
> > --- a/drivers/pci/quirks.c
> > +++ b/drivers/pci/quirks.c
> > @@ -3405,6 +3405,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
> > DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
> > /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
> > DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> > +/* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
> > +DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
> >
> > /*
> > * AMD has indicated that the devices below do not support peer-to-peer
> >
On Sat, Jul 05, 2014 at 03:26:29PM -0600, Bjorn Helgaas wrote:
> I saw that Joerg applied those iommu changes to his core branch, which I
> assume will be merged for v3.17. So I'll move this to a pci/iommu branch,
> to be merged during the v3.17 merge window.
Yes, these changes are queued for v3.17.
Joerg