Message ID | 20240930-imx95_lut-v2-1-3b6467ba539a@nxp.com (mailing list archive) |
---|---|
State | Not Applicable |
Headers | show |
Series | PCI: add enabe(disable)_device() hook for bridge | expand |
Context | Check | Description |
---|---|---|
netdev/tree_selection | success | Not a local patch |
On Mon, Sep 30, 2024 at 03:42:21PM -0400, Frank Li wrote: > Some PCIe bridges require special handling when enabling or disabling > PCIe devices. For example, on the i.MX95 platform, a lookup table must be > configured to inform the hardware how to convert pci_device_id to stream > (bus master) ID, which is used by the IOMMU and MSI controller to identify > bus master device. It's important that this say "PCI *host bridge*" specifically to avoid confusion with PCI-to-PCI bridges. On the PCIe side, it would be better to use "Requester ID" than "pci_device_id" because I think that's the actual key for the lookup table. Possible commit log text, fix my misconceptions as needed: Some PCIe host bridges require special handling when enabling or disabling PCIe Endpoints. For example, the i.MX95 platform has a lookup table to map Requester IDs to StreamIDs, which are used by the SMMU and MSI controller to identify the source of DMA accesses. Without this mapping, DMA accesses may target unintended memory, which would corrupt memory or read the wrong data. Add a host bridge .enable_device() hook the imx6 driver can use to configure the Requester ID to StreamID mapping. The hardware table isn't big enough to map all possible Requester IDs, so this hook may fail if no table space is available. In that case, return failure from pci_enable_device(). It might make more sense to make pci_set_master() decline to enable bus mastering and return failure, but it currently doesn't have a way to return failure. > Enablement will be failure when there is not enough lookup table resource. > Avoid DMA write to wrong position. That is the reason why pci_fixup_enable > can't work since not return value for fixup function. > > Signed-off-by: Frank Li <Frank.Li@nxp.com> > --- > Change from v1 to v2 > - move enable(disable)device ops to pci_host_bridge > --- > drivers/pci/pci.c | 14 ++++++++++++++ > include/linux/pci.h | 2 ++ > 2 files changed, 16 insertions(+) > > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c > index 7d85c04fbba2a..fcdeb12622568 100644 > --- a/drivers/pci/pci.c > +++ b/drivers/pci/pci.c > @@ -2056,6 +2056,7 @@ int __weak pcibios_enable_device(struct pci_dev *dev, int bars) > static int do_pci_enable_device(struct pci_dev *dev, int bars) > { > int err; > + struct pci_host_bridge *host_bridge; > struct pci_dev *bridge; > u16 cmd; > u8 pin; > @@ -2068,6 +2069,13 @@ static int do_pci_enable_device(struct pci_dev *dev, int bars) > if (bridge) > pcie_aspm_powersave_config_link(bridge); > > + host_bridge = pci_find_host_bridge(dev->bus); > + if (host_bridge && host_bridge->enable_device) { > + err = host_bridge->enable_device(host_bridge, dev); > + if (err) > + return err; > + } > + > err = pcibios_enable_device(dev, bars); > if (err < 0) > return err; > @@ -2262,12 +2270,18 @@ void pci_disable_enabled_device(struct pci_dev *dev) > */ > void pci_disable_device(struct pci_dev *dev) > { > + struct pci_host_bridge *host_bridge; > + > dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0, > "disabling already-disabled device"); > > if (atomic_dec_return(&dev->enable_cnt) != 0) > return; > > + host_bridge = pci_find_host_bridge(dev->bus); > + if (host_bridge && host_bridge->disable_device) > + host_bridge->disable_device(host_bridge, dev); > + > do_pci_disable_device(dev); > > dev->is_busmaster = 0; > diff --git a/include/linux/pci.h b/include/linux/pci.h > index 573b4c4c2be61..ac15b02e14ddd 100644 > --- a/include/linux/pci.h > +++ b/include/linux/pci.h > @@ -578,6 +578,8 @@ struct pci_host_bridge { > u8 (*swizzle_irq)(struct pci_dev *, u8 *); /* Platform IRQ swizzler */ > int (*map_irq)(const struct pci_dev *, u8, u8); > void (*release_fn)(struct pci_host_bridge *); > + int (*enable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); > + void (*disable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); > void *release_data; > unsigned int ignore_reset_delay:1; /* For entire hierarchy */ > unsigned int no_ext_tags:1; /* No Extended Tags */ > > -- > 2.34.1 >
On 2024-09-30 8:42 pm, Frank Li wrote: > Some PCIe bridges require special handling when enabling or disabling > PCIe devices. For example, on the i.MX95 platform, a lookup table must be > configured to inform the hardware how to convert pci_device_id to stream > (bus master) ID, which is used by the IOMMU and MSI controller to identify > bus master device. > > Enablement will be failure when there is not enough lookup table resource. > Avoid DMA write to wrong position. That is the reason why pci_fixup_enable > can't work since not return value for fixup function. > > Signed-off-by: Frank Li <Frank.Li@nxp.com> > --- > Change from v1 to v2 > - move enable(disable)device ops to pci_host_bridge > --- > drivers/pci/pci.c | 14 ++++++++++++++ > include/linux/pci.h | 2 ++ > 2 files changed, 16 insertions(+) > > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c > index 7d85c04fbba2a..fcdeb12622568 100644 > --- a/drivers/pci/pci.c > +++ b/drivers/pci/pci.c > @@ -2056,6 +2056,7 @@ int __weak pcibios_enable_device(struct pci_dev *dev, int bars) > static int do_pci_enable_device(struct pci_dev *dev, int bars) > { > int err; > + struct pci_host_bridge *host_bridge; > struct pci_dev *bridge; > u16 cmd; > u8 pin; > @@ -2068,6 +2069,13 @@ static int do_pci_enable_device(struct pci_dev *dev, int bars) > if (bridge) > pcie_aspm_powersave_config_link(bridge); > > + host_bridge = pci_find_host_bridge(dev->bus); > + if (host_bridge && host_bridge->enable_device) { > + err = host_bridge->enable_device(host_bridge, dev); If the intent is that this call may allocate host bridge resources, what about the existing error returns below? Should those now have a cleanup path to avoid a potential resource leak from here? Thanks, Robin. > + if (err) > + return err; > + } > + > err = pcibios_enable_device(dev, bars); > if (err < 0) > return err; > @@ -2262,12 +2270,18 @@ void pci_disable_enabled_device(struct pci_dev *dev) > */ > void pci_disable_device(struct pci_dev *dev) > { > + struct pci_host_bridge *host_bridge; > + > dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0, > "disabling already-disabled device"); > > if (atomic_dec_return(&dev->enable_cnt) != 0) > return; > > + host_bridge = pci_find_host_bridge(dev->bus); > + if (host_bridge && host_bridge->disable_device) > + host_bridge->disable_device(host_bridge, dev); > + > do_pci_disable_device(dev); > > dev->is_busmaster = 0; > diff --git a/include/linux/pci.h b/include/linux/pci.h > index 573b4c4c2be61..ac15b02e14ddd 100644 > --- a/include/linux/pci.h > +++ b/include/linux/pci.h > @@ -578,6 +578,8 @@ struct pci_host_bridge { > u8 (*swizzle_irq)(struct pci_dev *, u8 *); /* Platform IRQ swizzler */ > int (*map_irq)(const struct pci_dev *, u8, u8); > void (*release_fn)(struct pci_host_bridge *); > + int (*enable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); > + void (*disable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); > void *release_data; > unsigned int ignore_reset_delay:1; /* For entire hierarchy */ > unsigned int no_ext_tags:1; /* No Extended Tags */ >
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 7d85c04fbba2a..fcdeb12622568 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -2056,6 +2056,7 @@ int __weak pcibios_enable_device(struct pci_dev *dev, int bars) static int do_pci_enable_device(struct pci_dev *dev, int bars) { int err; + struct pci_host_bridge *host_bridge; struct pci_dev *bridge; u16 cmd; u8 pin; @@ -2068,6 +2069,13 @@ static int do_pci_enable_device(struct pci_dev *dev, int bars) if (bridge) pcie_aspm_powersave_config_link(bridge); + host_bridge = pci_find_host_bridge(dev->bus); + if (host_bridge && host_bridge->enable_device) { + err = host_bridge->enable_device(host_bridge, dev); + if (err) + return err; + } + err = pcibios_enable_device(dev, bars); if (err < 0) return err; @@ -2262,12 +2270,18 @@ void pci_disable_enabled_device(struct pci_dev *dev) */ void pci_disable_device(struct pci_dev *dev) { + struct pci_host_bridge *host_bridge; + dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0, "disabling already-disabled device"); if (atomic_dec_return(&dev->enable_cnt) != 0) return; + host_bridge = pci_find_host_bridge(dev->bus); + if (host_bridge && host_bridge->disable_device) + host_bridge->disable_device(host_bridge, dev); + do_pci_disable_device(dev); dev->is_busmaster = 0; diff --git a/include/linux/pci.h b/include/linux/pci.h index 573b4c4c2be61..ac15b02e14ddd 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -578,6 +578,8 @@ struct pci_host_bridge { u8 (*swizzle_irq)(struct pci_dev *, u8 *); /* Platform IRQ swizzler */ int (*map_irq)(const struct pci_dev *, u8, u8); void (*release_fn)(struct pci_host_bridge *); + int (*enable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); + void (*disable_device)(struct pci_host_bridge *bridge, struct pci_dev *dev); void *release_data; unsigned int ignore_reset_delay:1; /* For entire hierarchy */ unsigned int no_ext_tags:1; /* No Extended Tags */
Some PCIe bridges require special handling when enabling or disabling PCIe devices. For example, on the i.MX95 platform, a lookup table must be configured to inform the hardware how to convert pci_device_id to stream (bus master) ID, which is used by the IOMMU and MSI controller to identify bus master device. Enablement will be failure when there is not enough lookup table resource. Avoid DMA write to wrong position. That is the reason why pci_fixup_enable can't work since not return value for fixup function. Signed-off-by: Frank Li <Frank.Li@nxp.com> --- Change from v1 to v2 - move enable(disable)device ops to pci_host_bridge --- drivers/pci/pci.c | 14 ++++++++++++++ include/linux/pci.h | 2 ++ 2 files changed, 16 insertions(+)