Hi Rob, On Wed, Mar 22, 2017 at 2:47 PM, Rob Herring <robh@xxxxxxxxxx> wrote: > On Wed, Mar 22, 2017 at 3:58 AM, Geert Uytterhoeven > <geert@xxxxxxxxxxxxxx> wrote: >> On Wed, Mar 22, 2017 at 3:03 AM, Rob Herring <robh@xxxxxxxxxx> wrote: >>> dtc recently added PCI bus checks. Fix these warnings. >> >> It's always a good idea to put the warnings in the commit message: > > Well, the warnings are a bit noisy in that they get repeated, can be > misleading, change as I fix problems (device nodes are not checked if > the bridge node fails), and most importantly I'm lazy having fixed > these treewide. Sub-arch maintainers are welcome to make better > patches. :) > >> arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_bridge): Node >> /pci@ee090000/usb@0,1 node name is not "pci" or "pcie" [...] >> The above indeed go away with your patch, but I don't know why ;-) > > Most of these are because 'device_type = "pci";' should only be on PCI > bridges, not PCI devices. Thanks, retrying... arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_reg): Node /pci@ee090000/usb@0,1 PCI unit address format error, expected "1,0" arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_reg): Node /pci@ee090000/usb@0,2 PCI unit address format error, expected "2,0" arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_reg): Node /pci@ee0d0000/usb@0,1 PCI unit address format error, expected "1,0" arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_reg): Node /pci@ee0d0000/usb@0,2 PCI unit address format error, expected "2,0" arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_bus_num): Node /pci@ee0d0000/usb@0,1 PCI bus number 0 out of range, expected (2 - 2) arch/arm/boot/dts/r8a7790-lager.dtb: Warning (pci_device_bus_num): Node /pci@ee0d0000/usb@0,2 PCI bus number 0 out of range, expected (2 - 2) OK, without 'device_type = "pci"' it makes much more sense... >> Both Documentation/devicetree/bindings/pci/pci.txt and ePAPR refer to >> http://www.firmware.org/1275/bindings/pci/pci2_1.pdf >> http://www.firmware.org/1275/practice/imap/imap0_9d.pdf >> which no longer exist. > > Yeah, they are still out there. Just need to google "PCI supplement 1275". :) > > I have the full archive saved off. I need to get them added to devicetree.org. Having working links in the DT bindings would be great. BTW, seems like I still have an offline copy of v2.0 of the PCI spec in PostScript format, from my CHRP days[*], when hard disks where smaller than CD-Rs ;-) [*] I never managed to memorize the PCI DT bindings, though. Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
