From b77d537d00d08fcf0bf641cd3491dd7df0ad1475 Mon Sep 17 00:00:00 2001 From: Manish Jaggi Date: Thu, 30 Mar 2017 18:47:14 -0500 Subject: PCI: Apply Cavium ACS quirk only to CN81xx/CN83xx/CN88xx devices Only apply the Cavium ACS quirk to devices with ID in the range 0xa000-0xa0ff. These are the on-chip PCI devices for CN81xx/CN83xx/CN88xx. Fixes: b404bcfbf035 ("PCI: Add ACS quirk for all Cavium devices") Reported-by: Alex Williamson Signed-off-by: Manish Jaggi Acked-by: David Daney Acked-by: Alex Williamson --- drivers/pci/quirks.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/pci/quirks.c') diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index f754453fe754..178242d0d9df 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4094,6 +4094,9 @@ static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags) acs_flags &= ~(PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT); + if (!((dev->device >= 0xa000) && (dev->device <= 0xa0ff))) + return -ENOTTY; + return acs_flags ? 0 : 1; } -- cgit v1.2.3 From 45a2329367386342d41ea9414c88b023f5a79055 Mon Sep 17 00:00:00 2001 From: Jayachandran C Date: Thu, 13 Apr 2017 20:30:45 +0000 Subject: PCI: Avoid generating invalid ThunderX2 DMA aliases On Cavium ThunderX2 arm64 SoCs (formerly known as Broadcom Vulcan), the PCI topology is slightly unusual. For a multi-node system, it looks like: 00:00.0 PCI bridge to [bus 01-1e] 01:0a.0 PCI-to-PCIe bridge to [bus 02-04] 02:00.0 PCIe Root Port bridge to [bus 03-04] (XLATE_ROOT) 03:00.0 PCIe Endpoint pci_for_each_dma_alias() assumes IOMMU translation is done at the root of the PCI hierarchy. It generates 03:00.0, 01:0a.0, and 00:00.0 as DMA aliases for 03:00.0 because buses 01 and 00 are non-PCIe buses that don't carry the Requester ID. Because the ThunderX2 IOMMU is at 02:00.0, the Requester IDs 01:0a.0 and 00:00.0 are never valid for the endpoint. This quirk stops alias generation at the XLATE_ROOT bridge so we won't generate 01:0a.0 or 00:00.0. The current IOMMU code only maps the last alias (this is a separate bug in itself). Prior to this quirk, we only created IOMMU mappings for the invalid Requester ID 00:00:0, which never matched any DMA transactions. With this quirk, we create IOMMU mappings for a valid Requester ID, which fixes devices with no aliases but leaves devices with aliases still broken. The last alias for the endpoint is also used by the ARM GICv3 MSI-X code. Without this quirk, the GIC Interrupt Translation Tables are setup with the invalid Requester ID, and the MSI-X generated by the device fails to be translated and routed. Link: https://bugzilla.kernel.org/show_bug.cgi?id=195447 Signed-off-by: Jayachandran C Signed-off-by: Bjorn Helgaas Reviewed-by: Robin Murphy Acked-by: David Daney --- drivers/pci/quirks.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'drivers/pci/quirks.c') diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 178242d0d9df..74344db4ed4a 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -3956,6 +3956,20 @@ static void quirk_mic_x200_dma_alias(struct pci_dev *pdev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2260, quirk_mic_x200_dma_alias); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2264, quirk_mic_x200_dma_alias); +/* + * The IOMMU and interrupt controller on Broadcom Vulcan/Cavium ThunderX2 are + * associated not at the root bus, but at a bridge below. This quirk avoids + * generating invalid DMA aliases. + */ +static void quirk_bridge_cavm_thrx2_pcie_root(struct pci_dev *pdev) +{ + pdev->dev_flags |= PCI_DEV_FLAGS_BRIDGE_XLATE_ROOT; +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_BROADCOM, 0x9000, + quirk_bridge_cavm_thrx2_pcie_root); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_BROADCOM, 0x9084, + quirk_bridge_cavm_thrx2_pcie_root); + /* * Intersil/Techwell TW686[4589]-based video capture cards have an empty (zero) * class code. Fix it. -- cgit v1.2.3