On 17.02.2019 06:58, Derald D. Woods wrote:

This commit guards against applying the PMECC header in a non-NAND_BOOT scenario. This a modified version of a previous patch found here:

https://patchwork.ozlabs.org/patch/1009885/

Now that SPL_GENERATE_ATMEL_PMECC_HEADER has been converted to Kconfig, this provides an additional guard for building and using the 'atmel_pmecc_params' tool.

Signed-off-by: Derald D. Woods woods.technical@gmail.com

scripts/Makefile.spl | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/scripts/Makefile.spl b/scripts/Makefile.spl index 9d5921606e..8732479f39 100644 --- a/scripts/Makefile.spl +++ b/scripts/Makefile.spl @@ -177,10 +177,12 @@ ifeq ($(CONFIG_SYS_SOC),"at91") MKIMAGEFLAGS_boot.bin = -T atmelimage

ifeq ($(CONFIG_SPL_GENERATE_ATMEL_PMECC_HEADER),y) +ifeq ($(CONFIG_NAND_BOOT),y) MKIMAGEFLAGS_boot.bin += -n $(shell $(obj)/../tools/atmel_pmecc_params)

boot.bin: $(obj)/../tools/atmel_pmecc_params endif +endif

Hello Derald,

This would actually make the functionality "not happening" if the configuration is done badly, but the Kconfig will still be activated, which will contradict it's behavior: enable something, but not working, because inside the Makefile, there is a hack to stop it in some cases... So I would prefer if you can make SPL_GENERATE_ATMEL_PMECC_HEADER depend in Kconfig directly to CONFIG_NAND_BOOT. Does that fit your thoughts as well?

Thanks, Eugen

boot.bin: $(obj)/u-boot-spl.bin FORCE $(call if_changed,mkimage)