[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <c8184d48-a343-503a-83a3-707e44961f1a@codeaurora.org>
Date: Mon, 21 May 2018 09:25:26 -0400
From: Sinan Kaya <okaya@...eaurora.org>
To: linux-pci@...r.kernel.org, timur@...eaurora.org
Cc: linux-arm-msm@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org,
Bjorn Helgaas <bhelgaas@...gle.com>,
open list <linux-kernel@...r.kernel.org>,
iommu@...ts.linux-foundation.org
Subject: Re: [PATCH] PCI: Do not enable PASID when End-to-End TLP is not
supported
+iommu folks.
On 5/19/2018 12:52 PM, Sinan Kaya wrote:
> A PCIe endpoint carries the process address space identifier (PASID) in
> the TLP prefix as part of the memory read/write transaction. The address
> information in the TLP is relevant only for a given PASID context.
>
> A translation agent takes PASID value and the address information from the
> TLP to look up the physical address in the system.
>
> If a bridge drops the TLP prefix, the translation agent can resolve the
> address to an incorrect location and cause data corruption. Prevent
> this condition by requiring End-to-End TLP prefix to be supported on the
> entire data path between the endpoint and the root port.
>
> Signed-off-by: Sinan Kaya <okaya@...eaurora.org>
> ---
> drivers/pci/ats.c | 16 ++++++++++++++++
> include/uapi/linux/pci_regs.h | 1 +
> 2 files changed, 17 insertions(+)
>
> diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c
> index 89305b5..0bcded5 100644
> --- a/drivers/pci/ats.c
> +++ b/drivers/pci/ats.c
> @@ -265,7 +265,9 @@ EXPORT_SYMBOL_GPL(pci_reset_pri);
> int pci_enable_pasid(struct pci_dev *pdev, int features)
> {
> u16 control, supported;
> + struct pci_dev *bridge;
> int pos;
> + u32 cap;
>
> if (WARN_ON(pdev->pasid_enabled))
> return -EBUSY;
> @@ -274,6 +276,20 @@ int pci_enable_pasid(struct pci_dev *pdev, int features)
> if (!pos)
> return -EINVAL;
>
> + bridge = pci_upstream_bridge(pdev);
> + while (bridge) {
> + if (!pci_find_capability(bridge, PCI_CAP_ID_EXP))
> + return -EINVAL;
> +
> + if (pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap))
> + return -EINVAL;
> +
> + if (!(cap & PCI_EXP_DEVCAP2_E2ETLP))
> + return -EINVAL;
> +
> + bridge = pci_upstream_bridge(bridge);
> + }
> +
> pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported);
> supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV;
>
> diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h
> index 103ba79..d91dea5 100644
> --- a/include/uapi/linux/pci_regs.h
> +++ b/include/uapi/linux/pci_regs.h
> @@ -634,6 +634,7 @@
> #define PCI_EXP_DEVCAP2_OBFF_MASK 0x000c0000 /* OBFF support mechanism */
> #define PCI_EXP_DEVCAP2_OBFF_MSG 0x00040000 /* New message signaling */
> #define PCI_EXP_DEVCAP2_OBFF_WAKE 0x00080000 /* Re-use WAKE# for OBFF */
> +#define PCI_EXP_DEVCAP2_E2ETLP 0x00200000 /* End-to-End TLP Prefix */
> #define PCI_EXP_DEVCTL2 40 /* Device Control 2 */
> #define PCI_EXP_DEVCTL2_COMP_TIMEOUT 0x000f /* Completion Timeout Value */
> #define PCI_EXP_DEVCTL2_COMP_TMOUT_DIS 0x0010 /* Completion Timeout Disable */
>
--
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
Powered by blists - more mailing lists