| Message ID | 20210722121242.47838-1-manivannan.sadhasivam@linaro.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | Add Qualcomm PCIe Endpoint driver support | expand |
Hi Manivannan, On 22/07/21 5:42 pm, Manivannan Sadhasivam wrote: > Hello, > > This series adds support for Qualcomm PCIe Endpoint controller found > in platforms like SDX55. The Endpoint controller is based on the designware > core with additional Qualcomm wrappers around the core. > > The driver is added separately unlike other Designware based drivers that > combine RC and EP in a single driver. This is done to avoid complexity and > to maintain this driver autonomously. > > The driver has been validated with an out of tree MHI function driver on > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. Can you also validate it with in-kernel pci-endpoint-test? It would also help if you can test your patches after https://lore.kernel.org/r/20210803050310.27122-1-kishon@ti.com Not expecting any dependencies but just to cross check. Thanks Kishon > > Thanks, > Mani > > Changes in v7: > > * Used existing naming convention for callback functions > * Used active low state for PERST# gpio > > Changes in v6: > > * Removed status property in DT and added reviewed tag from Rob > * Switched to _relaxed variants as suggested by Rob > > Changes in v5: > > * Removed the DBI register settings that are not needed > * Used the standard definitions available in pci_regs.h > * Added defines for all the register fields > * Removed the left over code from previous iteration > > Changes in v4: > > * Removed the active_config settings needed for IPA integration > * Switched to writel for couple of relaxed versions that sneaked in > > Changes in v3: > > * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. > * Noticeable changes are: > - Got rid of _relaxed calls and used readl/writel > - Got rid of separate TCSR memory region and used syscon for getting the > register offsets for Perst registers > - Changed the wake gpio handling logic > - Added remove() callback and removed "suppress_bind_attrs" > - stop_link() callback now just disables PERST IRQ > * Added MMIO region and doorbell interrupt to the binding > * Added logic to write MMIO physicall address to MHI base address as it is > for the function driver to work > > Changes in v2: > > * Addressed the comments from Rob on bindings patch > * Modified the driver as per binding change > * Fixed the warnings reported by Kbuild bot > * Removed the PERST# "enable_irq" call from probe() > > Manivannan Sadhasivam (3): > dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP > controller > PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver > MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding > > .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ > MAINTAINERS | 10 +- > drivers/pci/controller/dwc/Kconfig | 10 + > drivers/pci/controller/dwc/Makefile | 1 + > drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ > 5 files changed, 888 insertions(+), 1 deletion(-) > create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml > create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c >
Hi Kishon, On Tue, Aug 03, 2021 at 10:57:00AM +0530, Kishon Vijay Abraham I wrote: > Hi Manivannan, > > On 22/07/21 5:42 pm, Manivannan Sadhasivam wrote: > > Hello, > > > > This series adds support for Qualcomm PCIe Endpoint controller found > > in platforms like SDX55. The Endpoint controller is based on the designware > > core with additional Qualcomm wrappers around the core. > > > > The driver is added separately unlike other Designware based drivers that > > combine RC and EP in a single driver. This is done to avoid complexity and > > to maintain this driver autonomously. > > > > The driver has been validated with an out of tree MHI function driver on > > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. > > Can you also validate it with in-kernel pci-endpoint-test? > The Qcom EP driver or the DWC IP is designed to work only for MHI bus that has the fixed BAR address and register layout etc... So we can't use the pci-endpoint-test driver to validate the EP driver. But I do have an out of tree MHI EPF driver and validated using that. > It would also help if you can test your patches after > https://lore.kernel.org/r/20210803050310.27122-1-kishon@ti.com > I added below patches (v7) on top and verified that the EP driver works fine. PCI: endpoint: Add virtual function number in pci_epc ops PCI: endpoint: Add support to link a physical function to a virtual function PCI: endpoint: Add support to add virtual function in endpoint core Thanks, Mani > Not expecting any dependencies but just to cross check. > > Thanks > Kishon > > > > > Thanks, > > Mani > > > > Changes in v7: > > > > * Used existing naming convention for callback functions > > * Used active low state for PERST# gpio > > > > Changes in v6: > > > > * Removed status property in DT and added reviewed tag from Rob > > * Switched to _relaxed variants as suggested by Rob > > > > Changes in v5: > > > > * Removed the DBI register settings that are not needed > > * Used the standard definitions available in pci_regs.h > > * Added defines for all the register fields > > * Removed the left over code from previous iteration > > > > Changes in v4: > > > > * Removed the active_config settings needed for IPA integration > > * Switched to writel for couple of relaxed versions that sneaked in > > > > Changes in v3: > > > > * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. > > * Noticeable changes are: > > - Got rid of _relaxed calls and used readl/writel > > - Got rid of separate TCSR memory region and used syscon for getting the > > register offsets for Perst registers > > - Changed the wake gpio handling logic > > - Added remove() callback and removed "suppress_bind_attrs" > > - stop_link() callback now just disables PERST IRQ > > * Added MMIO region and doorbell interrupt to the binding > > * Added logic to write MMIO physicall address to MHI base address as it is > > for the function driver to work > > > > Changes in v2: > > > > * Addressed the comments from Rob on bindings patch > > * Modified the driver as per binding change > > * Fixed the warnings reported by Kbuild bot > > * Removed the PERST# "enable_irq" call from probe() > > > > Manivannan Sadhasivam (3): > > dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP > > controller > > PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver > > MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding > > > > .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ > > MAINTAINERS | 10 +- > > drivers/pci/controller/dwc/Kconfig | 10 + > > drivers/pci/controller/dwc/Makefile | 1 + > > drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ > > 5 files changed, 888 insertions(+), 1 deletion(-) > > create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml > > create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c > >
On Thu, Jul 22, 2021 at 05:42:39PM +0530, Manivannan Sadhasivam wrote: > Hello, > > This series adds support for Qualcomm PCIe Endpoint controller found > in platforms like SDX55. The Endpoint controller is based on the designware > core with additional Qualcomm wrappers around the core. > > The driver is added separately unlike other Designware based drivers that > combine RC and EP in a single driver. This is done to avoid complexity and > to maintain this driver autonomously. > > The driver has been validated with an out of tree MHI function driver on > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. > Ping on this series! Thanks, Mani > Thanks, > Mani > > Changes in v7: > > * Used existing naming convention for callback functions > * Used active low state for PERST# gpio > > Changes in v6: > > * Removed status property in DT and added reviewed tag from Rob > * Switched to _relaxed variants as suggested by Rob > > Changes in v5: > > * Removed the DBI register settings that are not needed > * Used the standard definitions available in pci_regs.h > * Added defines for all the register fields > * Removed the left over code from previous iteration > > Changes in v4: > > * Removed the active_config settings needed for IPA integration > * Switched to writel for couple of relaxed versions that sneaked in > > Changes in v3: > > * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. > * Noticeable changes are: > - Got rid of _relaxed calls and used readl/writel > - Got rid of separate TCSR memory region and used syscon for getting the > register offsets for Perst registers > - Changed the wake gpio handling logic > - Added remove() callback and removed "suppress_bind_attrs" > - stop_link() callback now just disables PERST IRQ > * Added MMIO region and doorbell interrupt to the binding > * Added logic to write MMIO physicall address to MHI base address as it is > for the function driver to work > > Changes in v2: > > * Addressed the comments from Rob on bindings patch > * Modified the driver as per binding change > * Fixed the warnings reported by Kbuild bot > * Removed the PERST# "enable_irq" call from probe() > > Manivannan Sadhasivam (3): > dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP > controller > PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver > MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding > > .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ > MAINTAINERS | 10 +- > drivers/pci/controller/dwc/Kconfig | 10 + > drivers/pci/controller/dwc/Makefile | 1 + > drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ > 5 files changed, 888 insertions(+), 1 deletion(-) > create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml > create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c > > -- > 2.25.1 >
Hi, On Thu, Jul 22, 2021 at 05:42:39PM +0530, Manivannan Sadhasivam wrote: > Hello, > > This series adds support for Qualcomm PCIe Endpoint controller found > in platforms like SDX55. The Endpoint controller is based on the designware > core with additional Qualcomm wrappers around the core. > > The driver is added separately unlike other Designware based drivers that > combine RC and EP in a single driver. This is done to avoid complexity and > to maintain this driver autonomously. > > The driver has been validated with an out of tree MHI function driver on > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. > Any chance this driver could make v5.15? Thanks, Mani > Thanks, > Mani > > Changes in v7: > > * Used existing naming convention for callback functions > * Used active low state for PERST# gpio > > Changes in v6: > > * Removed status property in DT and added reviewed tag from Rob > * Switched to _relaxed variants as suggested by Rob > > Changes in v5: > > * Removed the DBI register settings that are not needed > * Used the standard definitions available in pci_regs.h > * Added defines for all the register fields > * Removed the left over code from previous iteration > > Changes in v4: > > * Removed the active_config settings needed for IPA integration > * Switched to writel for couple of relaxed versions that sneaked in > > Changes in v3: > > * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. > * Noticeable changes are: > - Got rid of _relaxed calls and used readl/writel > - Got rid of separate TCSR memory region and used syscon for getting the > register offsets for Perst registers > - Changed the wake gpio handling logic > - Added remove() callback and removed "suppress_bind_attrs" > - stop_link() callback now just disables PERST IRQ > * Added MMIO region and doorbell interrupt to the binding > * Added logic to write MMIO physicall address to MHI base address as it is > for the function driver to work > > Changes in v2: > > * Addressed the comments from Rob on bindings patch > * Modified the driver as per binding change > * Fixed the warnings reported by Kbuild bot > * Removed the PERST# "enable_irq" call from probe() > > Manivannan Sadhasivam (3): > dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP > controller > PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver > MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding > > .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ > MAINTAINERS | 10 +- > drivers/pci/controller/dwc/Kconfig | 10 + > drivers/pci/controller/dwc/Makefile | 1 + > drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ > 5 files changed, 888 insertions(+), 1 deletion(-) > create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml > create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c > > -- > 2.25.1 >
Hi, On Thu, Jul 22, 2021 at 05:42:39PM +0530, Manivannan Sadhasivam wrote: > Hello, > > This series adds support for Qualcomm PCIe Endpoint controller found > in platforms like SDX55. The Endpoint controller is based on the designware > core with additional Qualcomm wrappers around the core. > > The driver is added separately unlike other Designware based drivers that > combine RC and EP in a single driver. This is done to avoid complexity and > to maintain this driver autonomously. > > The driver has been validated with an out of tree MHI function driver on > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. > Ping again! Do I need to resend this series on top of v5.15-rc1? I thought this one could go in for v5.15 but... Thanks, Mani > Thanks, > Mani > > Changes in v7: > > * Used existing naming convention for callback functions > * Used active low state for PERST# gpio > > Changes in v6: > > * Removed status property in DT and added reviewed tag from Rob > * Switched to _relaxed variants as suggested by Rob > > Changes in v5: > > * Removed the DBI register settings that are not needed > * Used the standard definitions available in pci_regs.h > * Added defines for all the register fields > * Removed the left over code from previous iteration > > Changes in v4: > > * Removed the active_config settings needed for IPA integration > * Switched to writel for couple of relaxed versions that sneaked in > > Changes in v3: > > * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. > * Noticeable changes are: > - Got rid of _relaxed calls and used readl/writel > - Got rid of separate TCSR memory region and used syscon for getting the > register offsets for Perst registers > - Changed the wake gpio handling logic > - Added remove() callback and removed "suppress_bind_attrs" > - stop_link() callback now just disables PERST IRQ > * Added MMIO region and doorbell interrupt to the binding > * Added logic to write MMIO physicall address to MHI base address as it is > for the function driver to work > > Changes in v2: > > * Addressed the comments from Rob on bindings patch > * Modified the driver as per binding change > * Fixed the warnings reported by Kbuild bot > * Removed the PERST# "enable_irq" call from probe() > > Manivannan Sadhasivam (3): > dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP > controller > PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver > MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding > > .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ > MAINTAINERS | 10 +- > drivers/pci/controller/dwc/Kconfig | 10 + > drivers/pci/controller/dwc/Makefile | 1 + > drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ > 5 files changed, 888 insertions(+), 1 deletion(-) > create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml > create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c > > -- > 2.25.1 >
On Tue, Sep 14, 2021 at 01:39:11PM +0530, Manivannan Sadhasivam wrote: > Hi, > > On Thu, Jul 22, 2021 at 05:42:39PM +0530, Manivannan Sadhasivam wrote: > > Hello, > > > > This series adds support for Qualcomm PCIe Endpoint controller found > > in platforms like SDX55. The Endpoint controller is based on the designware > > core with additional Qualcomm wrappers around the core. > > > > The driver is added separately unlike other Designware based drivers that > > combine RC and EP in a single driver. This is done to avoid complexity and > > to maintain this driver autonomously. > > > > The driver has been validated with an out of tree MHI function driver on > > SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. > > Ping again! Do I need to resend this series on top of v5.15-rc1? I > thought this one could go in for v5.15 but... It's still "New" in patchwork: https://patchwork.kernel.org/project/linux-pci/list/?series=519709 so if it still applies cleanly on top of v5.15-rc1, you needn't post it again. If it requires any tweaks to apply on v5.15-rc1, please rebase it and post a v8. Bjorn
Hello, This series adds support for Qualcomm PCIe Endpoint controller found in platforms like SDX55. The Endpoint controller is based on the designware core with additional Qualcomm wrappers around the core. The driver is added separately unlike other Designware based drivers that combine RC and EP in a single driver. This is done to avoid complexity and to maintain this driver autonomously. The driver has been validated with an out of tree MHI function driver on SDX55 based Telit FN980 EVB connected to x86 host machine over PCIe. Thanks, Mani Changes in v7: * Used existing naming convention for callback functions * Used active low state for PERST# gpio Changes in v6: * Removed status property in DT and added reviewed tag from Rob * Switched to _relaxed variants as suggested by Rob Changes in v5: * Removed the DBI register settings that are not needed * Used the standard definitions available in pci_regs.h * Added defines for all the register fields * Removed the left over code from previous iteration Changes in v4: * Removed the active_config settings needed for IPA integration * Switched to writel for couple of relaxed versions that sneaked in Changes in v3: * Lot of minor cleanups to the driver patch based on review from Bjorn and Stan. * Noticeable changes are: - Got rid of _relaxed calls and used readl/writel - Got rid of separate TCSR memory region and used syscon for getting the register offsets for Perst registers - Changed the wake gpio handling logic - Added remove() callback and removed "suppress_bind_attrs" - stop_link() callback now just disables PERST IRQ * Added MMIO region and doorbell interrupt to the binding * Added logic to write MMIO physicall address to MHI base address as it is for the function driver to work Changes in v2: * Addressed the comments from Rob on bindings patch * Modified the driver as per binding change * Fixed the warnings reported by Kbuild bot * Removed the PERST# "enable_irq" call from probe() Manivannan Sadhasivam (3): dt-bindings: pci: Add devicetree binding for Qualcomm PCIe EP controller PCI: qcom-ep: Add Qualcomm PCIe Endpoint controller driver MAINTAINERS: Add entry for Qualcomm PCIe Endpoint driver and binding .../devicetree/bindings/pci/qcom,pcie-ep.yaml | 158 ++++ MAINTAINERS | 10 +- drivers/pci/controller/dwc/Kconfig | 10 + drivers/pci/controller/dwc/Makefile | 1 + drivers/pci/controller/dwc/pcie-qcom-ep.c | 710 ++++++++++++++++++ 5 files changed, 888 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml create mode 100644 drivers/pci/controller/dwc/pcie-qcom-ep.c