On 10/7/22 14:25, Venkatesh Yadav Abbarapu wrote:

zynq_gem.c invokes zynq_board_read_rom_ethaddr(), but neglected to declare its declaration, causing the following sparse and compile time warnings:

drivers/net/zynq_gem.c:662:12: warning: no previous prototype for 'zynq_board_read_rom_ethaddr' [-Wmissing-prototypes] 662 | __weak int zynq_board_read_rom_ethaddr(unsigned char *ethaddr)

Include the missing declaration in asm/arch/sys_proto.h.

Signed-off-by: Venkatesh Yadav Abbarapu venkatesh.abbarapu@amd.com

arch/arm/mach-versal/include/mach/sys_proto.h | 1 + 1 file changed, 1 insertion(+)

diff --git a/arch/arm/mach-versal/include/mach/sys_proto.h b/arch/arm/mach-versal/include/mach/sys_proto.h index 8e5712e0c9..9eb91b23d9 100644 --- a/arch/arm/mach-versal/include/mach/sys_proto.h +++ b/arch/arm/mach-versal/include/mach/sys_proto.h @@ -12,6 +12,7 @@ enum {

void tcm_init(u8 mode); void mem_map_fill(void); +int zynq_board_read_rom_ethaddr(unsigned char *ethaddr);

static inline int zynqmp_mmio_write(const u32 address, const u32 mask, const u32 value) {

I think we should remove zynq_board_read_rom_ethaddr() completely as I wrote here. https://lore.kernel.org/u-boot/CAHTX3dJvmQdV2pcBUVP7snWHOK1p0NMs8ANYhO_Sjdnr...

Thanks, Michal