On Thu, Jun 27, 2024 at 02:23:38AM +0800, kernel test robot wrote: > tree: https://git.kernel.org/pub/scm/linux/kernel/git/pci/pci.git controller/rockchip > head: 246afbe0f6fca433d8d918b740719170b1b082cc > commit: 9b2ba393b3a659a4695691794dc030b6f7744b23 [10/11] PCI: dw-rockchip: Add endpoint mode support > config: loongarch-randconfig-r081-20240626 (https://download.01.org/0day-ci/archive/20240627/202406270250.k2esVVnL-lkp@xxxxxxxxx/config) > compiler: loongarch64-linux-gcc (GCC) 13.2.0 > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240627/202406270250.k2esVVnL-lkp@xxxxxxxxx/reproduce) > > If you fix the issue in a separate patch/commit (i.e. not just a new version of > the same patch/commit), kindly add following tags > | Reported-by: kernel test robot <lkp@xxxxxxxxx> > | Closes: https://lore.kernel.org/oe-kbuild-all/202406270250.k2esVVnL-lkp@xxxxxxxxx/ > Hello Krzysztof, > All errors (new ones prefixed by >>): > > loongarch64-linux-ld: drivers/pci/controller/dwc/pcie-designware-ep.o: in function `dw_pcie_ep_init_notify': > >> drivers/pci/controller/dwc/pcie-designware-ep.c:26:(.text+0x1e4): undefined reference to `pci_epc_init_notify' Seeing that it is pcie-designware-ep.c that fails to build, I can see the error. I have forgotten to do: diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig index 9c4fb8ba7573..4c38181acffa 100644 --- a/drivers/pci/controller/dwc/Kconfig +++ b/drivers/pci/controller/dwc/Kconfig @@ -329,6 +329,7 @@ config PCIE_ROCKCHIP_DW_EP bool "Rockchip DesignWare PCIe controller (endpoint mode)" depends on ARCH_ROCKCHIP || COMPILE_TEST depends on OF + depends on PCI_ENDPOINT select PCIE_DW_EP select PCIE_ROCKCHIP_DW help (The .config in this build has PCIE_ROCKCHIP_DW_EP selected, but not PCI_ENDPOINT, which should not be allowed...) This is my fault, all the other endpoint drivers have a depends on PCI_ENDPOINT. Will send a patch for this shortly. I'm truly sorry for this mistake. Looking more closely at the pci/controller/rockchip branch, I can also see that we have another potential problem: Commit 7a847796e509 ("PCI: endpoint: Introduce 'epc_deinit' event and notify the EPF drivers"), which is on branch: pci/endpoint has introduced a stub (dummy definition) for pci_epc_init_notify. However, if the test robot builds the pci/controller/rockchip branch with PCIE_ROCKCHIP_DW_HOST selected, and not PCIE_ROCKCHIP_DW_EP selected, we could get a failure that pci_epc_init_notify() does not have a stub... This problem can be solved by cherry-picking 7a847796e509 ("PCI: endpoint: Introduce 'epc_deinit' event and notify the EPF drivers") to the pci/controller/rockchip branch, or by rebasing the pci/controller/rockchip branch on top of the pci/endpoint branch... I'm sorry for this too. Kind regards, Niklas
