On Wed, Aug 16, 2017 at 06:11:45AM +0800, kbuild test robot wrote: > tree: https://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git pci/host-xilinx > head: 95b008920fd38e828a37c8a7131b449d33ba03b9 > commit: 13ff772beae556a664c029d5d75eaf7f3c104258 [1/7] PCI: Move enum pci_interrupt_pin to linux/pci.h > config: blackfin-allyesconfig (attached as .config) > compiler: bfin-uclinux-gcc (GCC) 6.2.0 > reproduce: > wget https://raw.githubusercontent.com/01org/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross > chmod +x ~/bin/make.cross > git checkout 13ff772beae556a664c029d5d75eaf7f3c104258 > # save the attached .config to linux build tree > make.cross ARCH=blackfin > > All error/warnings (new ones prefixed by >>): > > In file included from include/linux/pci-epc.h:15:0, > from drivers/pci/endpoint/pci-ep-cfs.c:23: > >> include/linux/pci-epf.h:53:25: error: field 'interrupt_pin' has incomplete type > enum pci_interrupt_pin interrupt_pin; I think the problem with this (and the similar m68k report) is that enum pci_interrupt_pin was originally defined unconditionally in include/linux/pci-epf.h, but after "PCI: Move enum pci_interrupt_pin to linux/pci.h", it was defined in include/linux/pci.h under #ifdef CONFIG_PCI. Neither this blackfin config nor the m68k config defines CONFIG_PCI. CONFIG_PCI enables the host side of the PCI core, e.g., host bridge infrastructure, device enumeration, etc. It is possible to build endpoint support without that, and pci_interrupt_pin is still useful there. I reworked the patch to define enum pci_interrupt_pin outside the #ifdef CONFIG_PCI. Bjorn
