| Message ID | 1401345500-20188-5-git-send-email-kishon@ti.com (mailing list archive) |
|---|---|
| State | New, archived |
| Delegated to: | Bjorn Helgaas |
| Headers | show |
Hi Arnd, On Thursday 29 May 2014 12:08 PM, Kishon Vijay Abraham I wrote: > In DRA7, the cpu sees 32bit address, but the pcie controller can see only 28bit > address. So whenever the cpu issues a read/write request, the 4 most > significant bits are used by L3 to determine the target controller. > For example, the cpu reserves 0x2000_0000 - 0x2FFF_FFFF for PCIe controller but > the PCIe controller will see only (0x000_0000 - 0xFFF_FFF). So for programming > the outbound translation window the *base* should be programmed as 0x000_0000. > Whenever we try to write to say 0x2000_0000, it will be translated to whatever > we have programmed in the translation window with base as 0x000_0000. > > This is needed when the dt node is modelled something like below > axi { > compatible = "simple-bus"; > #size-cells = <1>; > #address-cells = <1>; > ranges = <0x0 0x20000000 0x10000000 // 28-bit bus > 0x51000000 0x51000000 0x3000>; > pcie@51000000 { > reg = <0x1000 0x2000>, <0x51002000 0x14c>, <0x51000000 0x2000>; > reg-names = "config", "ti_conf", "rc_dbics"; > #address-cells = <3>; > #size-cells = <2>; > ranges = <0x81000000 0 0 0x03000 0 0x00010000 > 0x82000000 0 0x20013000 0x13000 0 0xffed000>; > }; > }; > > Here the CPU address for configuration space is 0x20013000 and the controller > address for configuration space is 0x13000. The controller address should be > used while programming the ATU (in order for translation to happen properly in > DRA7xx). I've fixed this up with what you suggested in the v1 of the series. Do you think this is fine? Thanks Kishon > > Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> > Cc: Bjorn Helgaas <bhelgaas@google.com> > Cc: Mohit Kumar <mohit.kumar@st.com> > Cc: Jingoo Han <jg1.han@samsung.com> > Cc: Marek Vasut <marex@denx.de> > Cc: Arnd Bergmann <arnd@arndb.de> > Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com> > --- > drivers/pci/host/pcie-designware.c | 49 ++++++++++++++++++++++++++++-------- > drivers/pci/host/pcie-designware.h | 4 +++ > 2 files changed, 42 insertions(+), 11 deletions(-) > > diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c > index 603b386..9dfd2d4 100644 > --- a/drivers/pci/host/pcie-designware.c > +++ b/drivers/pci/host/pcie-designware.c > @@ -397,8 +397,15 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > struct of_pci_range range; > struct of_pci_range_parser parser; > struct resource *cfg_res; > - u32 val; > - int i; > + u32 val, na, ns; > + const __be32 *addrp; > + int i, index; > + > + /* Find the address cell size and the number of cells in order to get > + * the untranslated address. > + */ > + of_property_read_u32(np, "#address-cells", &na); > + ns = of_n_size_cells(np); > > cfg_res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "config"); > if (cfg_res) { > @@ -406,6 +413,12 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > pp->config.cfg1_size = resource_size(cfg_res)/2; > pp->cfg0_base = cfg_res->start; > pp->cfg1_base = cfg_res->start + pp->config.cfg0_size; > + > + /* Find the untranslated configuration space address */ > + index = of_property_match_string(np, "reg-names", "config"); > + addrp = of_get_address(np, index, false, false); > + pp->cfg0_mod_addr = of_read_number(addrp, ns); > + pp->cfg1_mod_addr = pp->cfg0_mod_addr + pp->config.cfg0_size; > } else { > dev_err(pp->dev, "missing *config* reg space\n"); > } > @@ -431,12 +444,20 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > pp->config.io_size = resource_size(&pp->io); > pp->config.io_bus_addr = range.pci_addr; > pp->io_base = range.cpu_addr; > + > + /* Find the untranslated IO space address */ > + pp->io_mod_addr = of_read_number(parser.range - > + parser.np + na, ns); > } > if (restype == IORESOURCE_MEM) { > of_pci_range_to_resource(&range, np, &pp->mem); > pp->mem.name = "MEM"; > pp->config.mem_size = resource_size(&pp->mem); > pp->config.mem_bus_addr = range.pci_addr; > + > + /* Find the untranslated MEM space address */ > + pp->mem_mod_addr = of_read_number(parser.range - > + parser.np + na, ns); > } > if (restype == 0) { > of_pci_range_to_resource(&range, np, &pp->cfg); > @@ -444,6 +465,12 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > pp->config.cfg1_size = resource_size(&pp->cfg)/2; > pp->cfg0_base = pp->cfg.start; > pp->cfg1_base = pp->cfg.start + pp->config.cfg0_size; > + > + /* Find the untranslated configuration space address */ > + pp->cfg0_mod_addr = of_read_number(parser.range - > + parser.np + na, ns); > + pp->cfg1_mod_addr = pp->cfg0_mod_addr + > + pp->config.cfg0_size; > } > } > > @@ -518,9 +545,9 @@ static void dw_pcie_prog_viewport_cfg0(struct pcie_port *pp, u32 busdev) > /* Program viewport 0 : OUTBOUND : CFG0 */ > dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0, > PCIE_ATU_VIEWPORT); > - dw_pcie_writel_rc(pp, pp->cfg0_base, PCIE_ATU_LOWER_BASE); > - dw_pcie_writel_rc(pp, (pp->cfg0_base >> 32), PCIE_ATU_UPPER_BASE); > - dw_pcie_writel_rc(pp, pp->cfg0_base + pp->config.cfg0_size - 1, > + dw_pcie_writel_rc(pp, pp->cfg0_mod_addr, PCIE_ATU_LOWER_BASE); > + dw_pcie_writel_rc(pp, (pp->cfg0_mod_addr >> 32), PCIE_ATU_UPPER_BASE); > + dw_pcie_writel_rc(pp, pp->cfg0_mod_addr + pp->config.cfg0_size - 1, > PCIE_ATU_LIMIT); > dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET); > dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET); > @@ -534,9 +561,9 @@ static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev) > dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1, > PCIE_ATU_VIEWPORT); > dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1); > - dw_pcie_writel_rc(pp, pp->cfg1_base, PCIE_ATU_LOWER_BASE); > - dw_pcie_writel_rc(pp, (pp->cfg1_base >> 32), PCIE_ATU_UPPER_BASE); > - dw_pcie_writel_rc(pp, pp->cfg1_base + pp->config.cfg1_size - 1, > + dw_pcie_writel_rc(pp, pp->cfg1_mod_addr, PCIE_ATU_LOWER_BASE); > + dw_pcie_writel_rc(pp, (pp->cfg1_mod_addr >> 32), PCIE_ATU_UPPER_BASE); > + dw_pcie_writel_rc(pp, pp->cfg1_mod_addr + pp->config.cfg1_size - 1, > PCIE_ATU_LIMIT); > dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET); > dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET); > @@ -549,9 +576,9 @@ static void dw_pcie_prog_viewport_mem_outbound(struct pcie_port *pp) > dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0, > PCIE_ATU_VIEWPORT); > dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1); > - dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE); > - dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE); > - dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1, > + dw_pcie_writel_rc(pp, pp->mem_mod_addr, PCIE_ATU_LOWER_BASE); > + dw_pcie_writel_rc(pp, (pp->mem_mod_addr >> 32), PCIE_ATU_UPPER_BASE); > + dw_pcie_writel_rc(pp, pp->mem_mod_addr + pp->config.mem_size - 1, > PCIE_ATU_LIMIT); > dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET); > dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr), > diff --git a/drivers/pci/host/pcie-designware.h b/drivers/pci/host/pcie-designware.h > index 3063b35..d1f7d5e 100644 > --- a/drivers/pci/host/pcie-designware.h > +++ b/drivers/pci/host/pcie-designware.h > @@ -36,11 +36,15 @@ struct pcie_port { > u8 root_bus_nr; > void __iomem *dbi_base; > u64 cfg0_base; > + u64 cfg0_mod_addr; > void __iomem *va_cfg0_base; > u64 cfg1_base; > + u64 cfg1_mod_addr; > void __iomem *va_cfg1_base; > u64 io_base; > + u64 io_mod_addr; > u64 mem_base; > + u64 mem_mod_addr; > spinlock_t conf_lock; > struct resource cfg; > struct resource io; > -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wednesday 18 June 2014 14:38:43 Kishon Vijay Abraham I wrote: > On Thursday 29 May 2014 12:08 PM, Kishon Vijay Abraham I wrote: > > > > Here the CPU address for configuration space is 0x20013000 and the controller > > address for configuration space is 0x13000. The controller address should be > > used while programming the ATU (in order for translation to happen properly in > > DRA7xx). > > I've fixed this up with what you suggested in the v1 of the series. Do you > think this is fine? Hi Kishon, Yes, I think this will work and do the right thing in all cases. I'm not completely sure if the change should be generalized into the range parser or better kept in the pcie-designware driver as you do it here: > > @@ -406,6 +413,12 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > > pp->config.cfg1_size = resource_size(cfg_res)/2; > > pp->cfg0_base = cfg_res->start; > > pp->cfg1_base = cfg_res->start + pp->config.cfg0_size; > > + > > + /* Find the untranslated configuration space address */ > > + index = of_property_match_string(np, "reg-names", "config"); > > + addrp = of_get_address(np, index, false, false); > > + pp->cfg0_mod_addr = of_read_number(addrp, ns); > > + pp->cfg1_mod_addr = pp->cfg0_mod_addr + pp->config.cfg0_size; I think I've recently seen some objections to hand-parsing standard properties in driver code. > > } else { > > dev_err(pp->dev, "missing *config* reg space\n"); > > } > > @@ -431,12 +444,20 @@ int __init dw_pcie_host_init(struct pcie_port *pp) > > pp->config.io_size = resource_size(&pp->io); > > pp->config.io_bus_addr = range.pci_addr; > > pp->io_base = range.cpu_addr; > > + > > + /* Find the untranslated IO space address */ > > + pp->io_mod_addr = of_read_number(parser.range - > > + parser.np + na, ns); > > } So the of_read_number() call could be moved into of_pci_range_to_resource(), with another field added to struct of_pci_range to carry the address on the parent bus. Any other thoughts on this? Maybe Rob Herring or Andrew Murray have a strong opinion on this. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, May 29, 2014 at 1:38 AM, Kishon Vijay Abraham I <kishon@ti.com> wrote: > In DRA7, the cpu sees 32bit address, but the pcie controller can see only 28bit > address. So whenever the cpu issues a read/write request, the 4 most > significant bits are used by L3 to determine the target controller. > For example, the cpu reserves 0x2000_0000 - 0x2FFF_FFFF for PCIe controller but > the PCIe controller will see only (0x000_0000 - 0xFFF_FFF). So for programming > the outbound translation window the *base* should be programmed as 0x000_0000. > Whenever we try to write to say 0x2000_0000, it will be translated to whatever > we have programmed in the translation window with base as 0x000_0000. > > This is needed when the dt node is modelled something like below > axi { > compatible = "simple-bus"; > #size-cells = <1>; > #address-cells = <1>; > ranges = <0x0 0x20000000 0x10000000 // 28-bit bus > 0x51000000 0x51000000 0x3000>; > pcie@51000000 { > reg = <0x1000 0x2000>, <0x51002000 0x14c>, <0x51000000 0x2000>; > reg-names = "config", "ti_conf", "rc_dbics"; > #address-cells = <3>; > #size-cells = <2>; > ranges = <0x81000000 0 0 0x03000 0 0x00010000 > 0x82000000 0 0x20013000 0x13000 0 0xffed000>; > }; > }; > > Here the CPU address for configuration space is 0x20013000 and the controller > address for configuration space is 0x13000. The controller address should be > used while programming the ATU (in order for translation to happen properly in > DRA7xx). This talks about config space, but the ranges field is PCI memory space. Also, does this actually work because I though Linux expects memory BARs to be 1MB aligned. Getting the controller offset should work whether you specify the address as 0x13000 with translation or the absolute address 0x20013000. In other words, the driver should know how many bits to mask off to get the offset. Rob -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Friday 20 June 2014 12:45:46 Rob Herring wrote: > On Thu, May 29, 2014 at 1:38 AM, Kishon Vijay Abraham I <kishon@ti.com> wrote: > > In DRA7, the cpu sees 32bit address, but the pcie controller can see only 28bit > > address. So whenever the cpu issues a read/write request, the 4 most > > significant bits are used by L3 to determine the target controller. > > For example, the cpu reserves 0x2000_0000 - 0x2FFF_FFFF for PCIe controller but > > the PCIe controller will see only (0x000_0000 - 0xFFF_FFF). So for programming > > the outbound translation window the *base* should be programmed as 0x000_0000. > > Whenever we try to write to say 0x2000_0000, it will be translated to whatever > > we have programmed in the translation window with base as 0x000_0000. > > > > This is needed when the dt node is modelled something like below > > axi { > > compatible = "simple-bus"; > > #size-cells = <1>; > > #address-cells = <1>; > > ranges = <0x0 0x20000000 0x10000000 // 28-bit bus > > 0x51000000 0x51000000 0x3000>; > > pcie@51000000 { > > reg = <0x1000 0x2000>, <0x51002000 0x14c>, <0x51000000 0x2000>; > > reg-names = "config", "ti_conf", "rc_dbics"; > > #address-cells = >; > > #size-cells = <2>; > > ranges = <0x81000000 0 0 0x03000 0 0x00010000 > > 0x82000000 0 0x20013000 0x13000 0 0xffed000>; > > }; > > }; > > > > Here the CPU address for configuration space is 0x20013000 and the controller > > address for configuration space is 0x13000. The controller address should be > > used while programming the ATU (in order for translation to happen properly in > > DRA7xx). > > This talks about config space, but the ranges field is PCI memory > space. Also, does this actually work because I though Linux expects > memory BARs to be 1MB aligned. Good point about the alignment. Actually both the config space and memory space are set up through the intermediate ranges of the parent bus. > Getting the controller offset should work whether you specify the > address as 0x13000 with translation or the absolute address > 0x20013000. In other words, the driver should know how many bits to > mask off to get the offset. That's what the first version of the patch did, and I didn't like that because the masking is not actually a property of the controller, but it's based on how the controller is connected to the parent bus, in this case by using a narrower connection. We can describe the connection just fine using standard DT properties and I think we should. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c index 603b386..9dfd2d4 100644 --- a/drivers/pci/host/pcie-designware.c +++ b/drivers/pci/host/pcie-designware.c @@ -397,8 +397,15 @@ int __init dw_pcie_host_init(struct pcie_port *pp) struct of_pci_range range; struct of_pci_range_parser parser; struct resource *cfg_res; - u32 val; - int i; + u32 val, na, ns; + const __be32 *addrp; + int i, index; + + /* Find the address cell size and the number of cells in order to get + * the untranslated address. + */ + of_property_read_u32(np, "#address-cells", &na); + ns = of_n_size_cells(np); cfg_res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "config"); if (cfg_res) { @@ -406,6 +413,12 @@ int __init dw_pcie_host_init(struct pcie_port *pp) pp->config.cfg1_size = resource_size(cfg_res)/2; pp->cfg0_base = cfg_res->start; pp->cfg1_base = cfg_res->start + pp->config.cfg0_size; + + /* Find the untranslated configuration space address */ + index = of_property_match_string(np, "reg-names", "config"); + addrp = of_get_address(np, index, false, false); + pp->cfg0_mod_addr = of_read_number(addrp, ns); + pp->cfg1_mod_addr = pp->cfg0_mod_addr + pp->config.cfg0_size; } else { dev_err(pp->dev, "missing *config* reg space\n"); } @@ -431,12 +444,20 @@ int __init dw_pcie_host_init(struct pcie_port *pp) pp->config.io_size = resource_size(&pp->io); pp->config.io_bus_addr = range.pci_addr; pp->io_base = range.cpu_addr; + + /* Find the untranslated IO space address */ + pp->io_mod_addr = of_read_number(parser.range - + parser.np + na, ns); } if (restype == IORESOURCE_MEM) { of_pci_range_to_resource(&range, np, &pp->mem); pp->mem.name = "MEM"; pp->config.mem_size = resource_size(&pp->mem); pp->config.mem_bus_addr = range.pci_addr; + + /* Find the untranslated MEM space address */ + pp->mem_mod_addr = of_read_number(parser.range - + parser.np + na, ns); } if (restype == 0) { of_pci_range_to_resource(&range, np, &pp->cfg); @@ -444,6 +465,12 @@ int __init dw_pcie_host_init(struct pcie_port *pp) pp->config.cfg1_size = resource_size(&pp->cfg)/2; pp->cfg0_base = pp->cfg.start; pp->cfg1_base = pp->cfg.start + pp->config.cfg0_size; + + /* Find the untranslated configuration space address */ + pp->cfg0_mod_addr = of_read_number(parser.range - + parser.np + na, ns); + pp->cfg1_mod_addr = pp->cfg0_mod_addr + + pp->config.cfg0_size; } } @@ -518,9 +545,9 @@ static void dw_pcie_prog_viewport_cfg0(struct pcie_port *pp, u32 busdev) /* Program viewport 0 : OUTBOUND : CFG0 */ dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0, PCIE_ATU_VIEWPORT); - dw_pcie_writel_rc(pp, pp->cfg0_base, PCIE_ATU_LOWER_BASE); - dw_pcie_writel_rc(pp, (pp->cfg0_base >> 32), PCIE_ATU_UPPER_BASE); - dw_pcie_writel_rc(pp, pp->cfg0_base + pp->config.cfg0_size - 1, + dw_pcie_writel_rc(pp, pp->cfg0_mod_addr, PCIE_ATU_LOWER_BASE); + dw_pcie_writel_rc(pp, (pp->cfg0_mod_addr >> 32), PCIE_ATU_UPPER_BASE); + dw_pcie_writel_rc(pp, pp->cfg0_mod_addr + pp->config.cfg0_size - 1, PCIE_ATU_LIMIT); dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET); dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET); @@ -534,9 +561,9 @@ static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev) dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1, PCIE_ATU_VIEWPORT); dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1); - dw_pcie_writel_rc(pp, pp->cfg1_base, PCIE_ATU_LOWER_BASE); - dw_pcie_writel_rc(pp, (pp->cfg1_base >> 32), PCIE_ATU_UPPER_BASE); - dw_pcie_writel_rc(pp, pp->cfg1_base + pp->config.cfg1_size - 1, + dw_pcie_writel_rc(pp, pp->cfg1_mod_addr, PCIE_ATU_LOWER_BASE); + dw_pcie_writel_rc(pp, (pp->cfg1_mod_addr >> 32), PCIE_ATU_UPPER_BASE); + dw_pcie_writel_rc(pp, pp->cfg1_mod_addr + pp->config.cfg1_size - 1, PCIE_ATU_LIMIT); dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET); dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET); @@ -549,9 +576,9 @@ static void dw_pcie_prog_viewport_mem_outbound(struct pcie_port *pp) dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0, PCIE_ATU_VIEWPORT); dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1); - dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE); - dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE); - dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1, + dw_pcie_writel_rc(pp, pp->mem_mod_addr, PCIE_ATU_LOWER_BASE); + dw_pcie_writel_rc(pp, (pp->mem_mod_addr >> 32), PCIE_ATU_UPPER_BASE); + dw_pcie_writel_rc(pp, pp->mem_mod_addr + pp->config.mem_size - 1, PCIE_ATU_LIMIT); dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET); dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr), diff --git a/drivers/pci/host/pcie-designware.h b/drivers/pci/host/pcie-designware.h index 3063b35..d1f7d5e 100644 --- a/drivers/pci/host/pcie-designware.h +++ b/drivers/pci/host/pcie-designware.h @@ -36,11 +36,15 @@ struct pcie_port { u8 root_bus_nr; void __iomem *dbi_base; u64 cfg0_base; + u64 cfg0_mod_addr; void __iomem *va_cfg0_base; u64 cfg1_base; + u64 cfg1_mod_addr; void __iomem *va_cfg1_base; u64 io_base; + u64 io_mod_addr; u64 mem_base; + u64 mem_mod_addr; spinlock_t conf_lock; struct resource cfg; struct resource io;
In DRA7, the cpu sees 32bit address, but the pcie controller can see only 28bit address. So whenever the cpu issues a read/write request, the 4 most significant bits are used by L3 to determine the target controller. For example, the cpu reserves 0x2000_0000 - 0x2FFF_FFFF for PCIe controller but the PCIe controller will see only (0x000_0000 - 0xFFF_FFF). So for programming the outbound translation window the *base* should be programmed as 0x000_0000. Whenever we try to write to say 0x2000_0000, it will be translated to whatever we have programmed in the translation window with base as 0x000_0000. This is needed when the dt node is modelled something like below axi { compatible = "simple-bus"; #size-cells = <1>; #address-cells = <1>; ranges = <0x0 0x20000000 0x10000000 // 28-bit bus 0x51000000 0x51000000 0x3000>; pcie@51000000 { reg = <0x1000 0x2000>, <0x51002000 0x14c>, <0x51000000 0x2000>; reg-names = "config", "ti_conf", "rc_dbics"; #address-cells = <3>; #size-cells = <2>; ranges = <0x81000000 0 0 0x03000 0 0x00010000 0x82000000 0 0x20013000 0x13000 0 0xffed000>; }; }; Here the CPU address for configuration space is 0x20013000 and the controller address for configuration space is 0x13000. The controller address should be used while programming the ATU (in order for translation to happen properly in DRA7xx). Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: Mohit Kumar <mohit.kumar@st.com> Cc: Jingoo Han <jg1.han@samsung.com> Cc: Marek Vasut <marex@denx.de> Cc: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com> --- drivers/pci/host/pcie-designware.c | 49 ++++++++++++++++++++++++++++-------- drivers/pci/host/pcie-designware.h | 4 +++ 2 files changed, 42 insertions(+), 11 deletions(-)