On Thursday 08 May 2014 17:56:38 Jingoo Han wrote: > On Wednesday, May 07, 2014 6:26 PM, Arnd Bergmann wrote: > > On Wednesday 07 May 2014 14:52:47 Kishon Vijay Abraham I wrote: > > > On Tuesday 06 May 2014 10:05 PM, Jason Gunthorpe wrote: > > > > On Tue, May 06, 2014 at 07:03:51PM +0530, Kishon Vijay Abraham I wrote: > > > >> +Example: > > > >> +pcie@51000000 { > > > >> + compatible = "ti,dra7xx-pcie"; > > > >> + reg = <0x51002000 0x14c>, <0x51000000 0x2000>; > > > >> + reg-names = "ti_conf", "rc_dbics"; > > > >> + interrupts = <0 232 0x4>, <0 233 0x4>; > > > >> + #address-cells = >; > > > >> + #size-cells = <2>; > > > >> + device_type = "pci"; > > > >> + ti,device_type = >; > > > >> + ranges = <0x00000800 0 0x20001000 0x20001000 0 0x00002000 /* Configuration Space */ > > > > > > > > Configuration space should not show up in the ranges, please don't > > > > copy that mistake from other drivers, put it in reg. > > > > > > But then it needs pcie-designware.c to be modified and it will be breaking > > > other platforms no? > > > > I think the pcie-designware driver should be changed to allow either way. > > Ideally we would deprecate the existing method in a way that for new front-ends > > it doesn't work, but the old front-ends can still deal with it but also work > > if you put it into the reg property. > > (+cc Pratyush Anand, Thierry Reding) > > Hi Arnd, > > Thank you for your comment. > Do you mean the case of Tegra PCIe as below? > > ./arch/arm/boot/dts/tegra20.dts > pcie-controller@80003000 { > ... > reg = <0x80003000 0x00000800 /* PADS registers */ > 0x80003800 0x00000200 /* AFI registers */ > 0x90000000 0x10000000>; /* configuration space */ > ... > ranges = <0x82000000 0 0x80000000 0x80000000 0 0x00001000 /* port 0 registers */ > 0x82000000 0 0x80001000 0x80001000 0 0x00001000 /* port 1 registers */ > 0x81000000 0 0 0x82000000 0 0x00010000 /* downstream I/O */ > 0x82000000 0 0xa0000000 0xa0000000 0 0x08000000 /* non-prefetchable memory */ > 0xc2000000 0 0xa8000000 0xa8000000 0 0x18000000>; /* prefetchable memory */ > ... > > ./drivers/pci/host/pci-tegra.c > /* request configuration space, but remap later, on demand */ > res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cs"); > ... > pcie->cs = devm_request_mem_region(pcie->dev, res->start, > resource_size(res), res->name); Yes, that is how the config space should be handled normally. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html