On Tuesday, March 11, 2014 at 03:05:11 AM, Masahiro Yamada wrote:
Signed-off-by: Masahiro Yamada yamada.m@jp.panasonic.com Cc: Marek Vasut marex@denx.de Cc: Fabio Estevam fabio.estevam@freescale.com Cc: Otavio Salvador otavio@ossystems.com.br Cc: Lauri Hintsala lauri.hintsala@bluegiga.com
You can drop Lauri from CC , I think this address is dead.
Hi MXS board maitainers,
I think this commit is probably correct. (I mean, it still generates correct "u-boot.sb" image)
But I could not corfirm it by binary comparison.
Every build generates a different u-boot.sb even if no source file is changed.
Note sure about that:
marex@bfu:/tmp/u-boot$ make -j9 u-boot.sb CHK include/config/uboot.release CHK include/generated/timestamp_autogenerated.h UPD include/generated/timestamp_autogenerated.h UPD include/config/uboot.release CHK include/generated/version_autogenerated.h UPD include/generated/version_autogenerated.h HOSTCC tools/dumpimage.o HOSTCC tools/image-host.o HOSTCC tools/mkenvimage.o HOSTCC tools/mkimage.o HOSTLD tools/envcrc HOSTLD tools/mkenvimage HOSTLD tools/dumpimage HOSTLD tools/mkimage AS arch/arm/cpu/arm926ejs/start.o CC arch/arm/lib/board.o CC drivers/video/cfb_console.o LD arch/arm/lib/built-in.o CC lib/display_options.o CC common/main.o CC common/cmd_version.o AS spl/arch/arm/cpu/arm926ejs/mxs/start.o LD lib/built-in.o CC spl/lib/display_options.o LD spl/lib/built-in.o LD spl/u-boot-spl OBJCOPY spl/u-boot-spl.bin LD common/built-in.o LD drivers/video/built-in.o LD drivers/built-in.o LD u-boot OBJCOPY u-boot.bin MXSIMAGE u-boot.sb marex@bfu:/tmp/u-boot$ sha1sum u-boot.bin a31b6fb66fab4a0f05634987585140ffe463452e u-boot.bin marex@bfu:/tmp/u-boot$ make -j9 u-boot.sb CHK include/config/uboot.release CHK include/generated/timestamp_autogenerated.h UPD include/generated/timestamp_autogenerated.h CHK include/generated/version_autogenerated.h HOSTCC tools/dumpimage.o HOSTCC tools/image-host.o HOSTCC tools/mkenvimage.o HOSTCC tools/mkimage.o HOSTLD tools/envcrc HOSTLD tools/mkenvimage HOSTLD tools/dumpimage HOSTLD tools/mkimage AS arch/arm/cpu/arm926ejs/start.o CC arch/arm/lib/board.o CC drivers/video/cfb_console.o LD arch/arm/lib/built-in.o CC lib/display_options.o CC common/main.o CC common/cmd_version.o AS spl/arch/arm/cpu/arm926ejs/mxs/start.o CC spl/lib/display_options.o LD lib/built-in.o LD spl/lib/built-in.o LD spl/u-boot-spl OBJCOPY spl/u-boot-spl.bin LD common/built-in.o LD drivers/video/built-in.o LD drivers/built-in.o LD u-boot OBJCOPY u-boot.bin MXSIMAGE u-boot.sb marex@bfu:/tmp/u-boot$ sha1sum u-boot.bin b5d5feab4a6974fe00272d3a3408300d8c8a473d u-boot.bin
Given that u-boot.bin and spl/u-boot-spl.bin are packed into u-boot.sb , this is why you always see a different u-boot.sb .
I wish someone could do run test this patch.
The patch is good, I have just a minor rant below, otherwise add my:
Acked-by: Marek Vasut marex@denx.de
btw. you can check if the image was not corrupted by issuing:
$ mkimage -l u-boot.sb
;-)
[...]
diff --git a/arch/arm/cpu/arm926ejs/mxs/Makefile b/arch/arm/cpu/arm926ejs/mxs/Makefile index 152546e..1eee661 100644 --- a/arch/arm/cpu/arm926ejs/mxs/Makefile +++ b/arch/arm/cpu/arm926ejs/mxs/Makefile @@ -14,11 +14,12 @@ obj-y += spl_boot.o spl_lradc_init.o spl_mem_init.o spl_power_init.o endif
# Specify the target for use in elftosb call -MKIMAGE_TARGET-$(CONFIG_MX23) = mx23 -MKIMAGE_TARGET-$(CONFIG_MX28) = mx28 +MKIMAGE_TARGET-$(CONFIG_MX23) = mxsimage.mx23.cfg +MKIMAGE_TARGET-$(CONFIG_MX28) = mxsimage.mx28.cfg
-$(OBJTREE)/mxsimage.cfg: $(SRCTREE)/$(CPUDIR)/$(SOC)/mxsimage.$(MKIMAGE_TARGET-y).cfg - sed "s@OBJTREE@$(OBJTREE)@g" $^ > $@ +quiet_cmd_mkimage_mxs = UIMAGE $@
Why don't we use 'MKIMAGE' here to be consistent ? Or 'MXSIMAGE' if we want to differentiate ?
+cmd_mkimage_mxs = $(objtree)/tools/mkimage -n $< -T mxsimage $@ \
- $(if $(KBUILD_VERBOSE:1=), >/dev/null)
Invokation seems OK to me. Getting rid of the 'sed' call is really to my liking here :)
Thank you!