On Tue, May 15, 2018 at 03:00:10AM +0800, kbuild test robot wrote: > tree: https://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git pci/resource > head: 4704207c075cd99a51e58ef75c185652b17fadd8 > commit: 3f3e689a67ba3d0ac2d10fd3b99c1bd0331415be [4/10] PCI: Use devm_kzalloc() in of_pci_get_host_bridge_resources() path > config: x86_64-randconfig-x002-201819 (attached as .config) > compiler: gcc-7 (Debian 7.3.0-16) 7.3.0 > reproduce: > git checkout 3f3e689a67ba3d0ac2d10fd3b99c1bd0331415be > # save the attached .config to linux build tree > make ARCH=x86_64 > > Note: the pci/pci/resource HEAD 4704207c075cd99a51e58ef75c185652b17fadd8 builds fine. > It only hurts bisectibility. > > All errors (new ones prefixed by >>): > > drivers/pci/of.c: In function '__of_pci_get_host_bridge_resources': > >> drivers/pci/of.c:315:4: error: label 'parse_failed' used but not defined > goto parse_failed; > ^~~~ I tried to fold your fixes patch (https://lkml.kernel.org/r/294ff38c-f8df-f89b-3bac-604e79fe6cfe@xxxxxxxxxxx) into the relevant patches, but I didn't do it quite right. I think this whole thing could be avoided if you restructured the patches slightly. What if you: - renamed the of_pci_get_host_bridge_resources() "struct device_node *dev" parameter to "struct device_node *dev_node" without refactoring anything, - added the "struct device *dev" parameter to of_pci_get_host_bridge_resources() and all callers, - renamed of_pci_get_host_bridge_resources() to devm_of_pci_get_host_bridge_resources() and converted it to use devm_kzalloc() internally, and - converted pr_info() to dev_info(). It looks like there aren't any in-tree users that depend on the compatibility wrappers, and I think the patches would end up being easier to read without the refactoring and the wrappers.
