RE: [PATCH v5 2/2] PCI: xilinx-cpm: Add support for Versal CPM5 Root Port

From: Gogada, Bharat Kumar
Date: Fri Jun 17 2022 - 22:36:39 EST


>
> On Thu, Jun 16, 2022 at 06:14:29PM +0530, Bharat Kumar Gogada wrote:
> > Xilinx Versal Premium series has CPM5 block which supports Root Port
> > functioning at Gen5 speed.
> >
> > Xilinx Versal CPM5 has few changes with existing CPM block.
> > - CPM5 has dedicated register space for control and status registers.
> > - CPM5 legacy interrupt handling needs additional register bit
> > to enable and handle legacy interrupts.
> >
> > Signed-off-by: Bharat Kumar Gogada <bharat.kumar.gogada@xxxxxxxxxx>
> > ---
> > drivers/pci/controller/pcie-xilinx-cpm.c | 62
> > ++++++++++++++++++++++--
> > 1 file changed, 59 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/pci/controller/pcie-xilinx-cpm.c
> > b/drivers/pci/controller/pcie-xilinx-cpm.c
> > index c7cd44ed4dfc..0bcd11d27eeb 100644
> > --- a/drivers/pci/controller/pcie-xilinx-cpm.c
> > +++ b/drivers/pci/controller/pcie-xilinx-cpm.c
> > @@ -35,6 +35,10 @@
> > #define XILINX_CPM_PCIE_MISC_IR_ENABLE 0x00000348
> > #define XILINX_CPM_PCIE_MISC_IR_LOCAL BIT(1)
> >
> > +#define XILINX_CPM_PCIE_IR_STATUS 0x000002A0
> > +#define XILINX_CPM_PCIE_IR_ENABLE 0x000002A8
> > +#define XILINX_CPM_PCIE_IR_LOCAL BIT(0)
> > +
> > /* Interrupt registers definitions */
> > #define XILINX_CPM_PCIE_INTR_LINK_DOWN 0
> > #define XILINX_CPM_PCIE_INTR_HOT_RESET 3
> > @@ -98,6 +102,16 @@
> > /* Phy Status/Control Register definitions */
> > #define XILINX_CPM_PCIE_REG_PSCR_LNKUP BIT(11)
> >
> > +/**
> > + * struct xilinx_cpm_variant - CPM variant information
> > + * @cpm_version: CPM5 has few changes compared to CPM block.
> > + * CPM5 has dedicated register space for control and status registers.
> > + *
>
> Superfluous blank line.
>
> > + */
> > +struct xilinx_cpm_variant {
> > + bool cpm_version;
>
> This is not really a bool, unless you want to preclude the possibility of ever
> having a CPM6 or other future variants.
>
> > +};
> > +
> > /**
> > * struct xilinx_cpm_pcie - PCIe port information
> > * @dev: Device pointer
> > @@ -109,6 +123,7 @@
> > * @intx_irq: legacy interrupt number
> > * @irq: Error interrupt number
> > * @lock: lock protecting shared register access
> > + * @is_cpm5: value to check cpm version
> > */
> > struct xilinx_cpm_pcie {
> > struct device *dev;
> > @@ -120,6 +135,7 @@ struct xilinx_cpm_pcie {
> > int intx_irq;
> > int irq;
> > raw_spinlock_t lock;
> > + bool is_cpm5;
> > };
> >
> > static u32 pcie_read(struct xilinx_cpm_pcie *port, u32 reg) @@ -285,6
> > +301,14 @@ static void xilinx_cpm_pcie_event_flow(struct irq_desc *desc)
> > generic_handle_domain_irq(port->cpm_domain, i);
> > pcie_write(port, val, XILINX_CPM_PCIE_REG_IDR);
> >
> > + if (port->is_cpm5) {
> > + val = readl_relaxed(port->cpm_base +
> XILINX_CPM_PCIE_IR_STATUS);
> > + if (val)
> > + writel_relaxed(val,
> > + port->cpm_base +
> > + XILINX_CPM_PCIE_IR_STATUS);
> > + }
> > +
> > /*
> > * XILINX_CPM_PCIE_MISC_IR_STATUS register is mapped to
> > * CPM SLCR block.
> > @@ -484,6 +508,12 @@ static void xilinx_cpm_pcie_init_port(struct
> xilinx_cpm_pcie *port)
> > */
> > writel(XILINX_CPM_PCIE_MISC_IR_LOCAL,
> > port->cpm_base + XILINX_CPM_PCIE_MISC_IR_ENABLE);
> > +
> > + if (port->is_cpm5) {
> > + writel(XILINX_CPM_PCIE_IR_LOCAL,
> > + port->cpm_base + XILINX_CPM_PCIE_IR_ENABLE);
> > + }
> > +
> > /* Enable the Bridge enable bit */
> > pcie_write(port, pcie_read(port, XILINX_CPM_PCIE_REG_RPSC) |
> > XILINX_CPM_PCIE_REG_RPSC_BEN, @@ -503,6 +533,10 @@
> > static int xilinx_cpm_pcie_parse_dt(struct xilinx_cpm_pcie *port,
> > struct device *dev = port->dev;
> > struct platform_device *pdev = to_platform_device(dev);
> > struct resource *res;
> > + const struct xilinx_cpm_variant *variant =
> > + of_device_get_match_data(dev);
> > +
> > + port->is_cpm5 = variant->cpm_version;
>
> It's a little clunky to use booleans for a potentially multi-valued version and
> to copy them around. Something like this might leave room for future
> variants that need more knobs:

Agreed will fix this in next patch.

Regards,
Bharat

> enum xilinx_cpm_version {
> CPM,
> CPM5,
> };
>
> struct xilinx_cpm_variant {
> enum xilinx_cpm_version version;
> };
>
> struct xilinx_cpm_pcie {
> ...
> const struct xilinx_cpm_variant *variant;
>
> static const struct xilinx_cpm_variant cpm5 = {
> .version = CPM5,
> };
>
> xilinx_cpm_pcie_event_flow()
> {
> ...
> if (port->variant->version == CPM5)
>
> xilinx_cpm_pcie_probe()
> {
> ...
> port->variant = of_device_get_match_data(dev);
>
>
> > port->cpm_base = devm_platform_ioremap_resource_byname(pdev,
> >
> > "cpm_slcr"); @@ -518,7 +552,14 @@ static int
> xilinx_cpm_pcie_parse_dt(struct xilinx_cpm_pcie *port,
> > if (IS_ERR(port->cfg))
> > return PTR_ERR(port->cfg);
> >
> > - port->reg_base = port->cfg->win;
> > + if (port->is_cpm5) {
> > + port->reg_base =
> devm_platform_ioremap_resource_byname(pdev,
> > + "cpm_csr");
> > + if (IS_ERR(port->reg_base))
> > + return PTR_ERR(port->reg_base);
> > + } else {
> > + port->reg_base = port->cfg->win;
> > + }
> >
> > return 0;
> > }
> > @@ -591,9 +632,24 @@ static int xilinx_cpm_pcie_probe(struct
> platform_device *pdev)
> > return err;
> > }
> >
> > +static const struct xilinx_cpm_variant cpm_host = {
> > + .cpm_version = false,
> > +};
> > +
> > +static const struct xilinx_cpm_variant cpm5_host = {
> > + .cpm_version = true,
> > +};
> > +
> > static const struct of_device_id xilinx_cpm_pcie_of_match[] = {
> > - { .compatible = "xlnx,versal-cpm-host-1.00", },
> > - {}
> > + {
> > + .compatible = "xlnx,versal-cpm-host-1.00",
> > + .data = &cpm_host,
> > + },
> > + {
> > + .compatible = "xlnx,versal-cpm5-host",
> > + .data = &cpm5_host,
> > + },
> > + {},
> > };
> >
> > static struct platform_driver xilinx_cpm_pcie_driver = {
> > --
> > 2.17.1
> >