When using nm to find the address of the kernel_entry symbol, the microMIPS ISA bit is not set for microMIPS kernels. This means that the entry address reported for the kernel is one that is treated as MIPS32 code, which causes problems if it is placed into an image such as a uImage or FIT image from which the bootloader will read the entry address & branch to it as non-microMIPS code. Fix this by instead using the objdump tool to determine the kernel entry address instead. We already pass $(OBJDUMP) around the kernel makefiles, and it reports the entry address with the microMIPS ISA bit intact. Signed-off-by: Paul Burton <paul.burton@xxxxxxxxxx> --- arch/mips/Makefile | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index efd7a9d..9bf611d 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -232,8 +232,9 @@ include arch/mips/Kbuild.platforms ifdef CONFIG_PHYSICAL_START load-y = $(CONFIG_PHYSICAL_START) endif -entry-y = 0x$(shell $(NM) vmlinux 2>/dev/null \ - | grep "\bkernel_entry\b" | cut -f1 -d \ ) +entry-y = $(shell $(OBJDUMP) -f vmlinux 2>/dev/null \ + | grep "^start address " \ + | cut -f3 -d \ ) cflags-y += -I$(srctree)/arch/mips/include/asm/mach-generic drivers-$(CONFIG_PCI) += arch/mips/pci/ -- 2.9.3
