On Saturday 16 July 2011 07:06 PM, Daniel Schwierzeck wrote:
Hi Aneesh,
On 07/16/2011 02:53 PM, Aneesh V wrote:
diff --git a/include/configs/omap4_panda.h b/include/configs/omap4_panda.h index a8dd861..3244fc1 100644 --- a/include/configs/omap4_panda.h +++ b/include/configs/omap4_panda.h @@ -243,4 +243,30 @@ #define CONFIG_SYS_DEFAULT_LPDDR2_TIMINGS #endif
+/* Defines for SPL */ +#define CONFIG_SPL +#define CONFIG_SYS_SPL_TEXT_BASE 0x40304350
Does this work? The SPL framework patches currently use
Hmm.. Good catch. I think it was still working for me because I had this in my linker script:
MEMORY { .sram : ORIGIN = CONFIG_SYS_SPL_TEXT_BASE,\ LENGTH = CONFIG_SYS_SPL_MAX_SIZE }
And then I was using the .sram region for rest of the script.
I think this over-rides the -Ttext on the linker command line.
ifneq ($(CONFIG_SPL_TEXT_BASE),) CPPFLAGS += -DCONFIG_SPL_TEXT_BASE=$(CONFIG_SPL_TEXT_BASE) endif
But CONFIG_SYS_SPL_TEXT_BASE seems to be the right name for this option.
Yes, looks like CONFIG_SYS_SPL_TEXT_BASE is the right name. For u-boot we have CONFIG_SYS_TEXT_BASE
Shall I fix it in the SPL framework patches?
I don't know if Daniel wants to do this himself. As long as you are working on omap and use my linker script, you may still be able to work without any compilation error.
If the framework is going to retain CONFIG_SPL_TEXT_BASE I shall re-submit my patch.
best regards, Aneesh