Dear ich,
In message 1312383540.4776.36.camel@debian you wrote:
I know that. But I am not able to port for sh platform to a newer u-boot.
Please note that U-Boot v1.3.1 is about 4 years old and as such no longer supported here.
Your quoting style is strange. Normally the reply _follows_.
Well, if you cannot update, then we cannot help it either. It's yur problem, after all.
[Well, of course there are companies that will happily send you a quotation for such an update, in case you are considering a commercial solution.]
U-Boot is, in it's default configuration, designed to be run on a virgin CPU comingg fresh out of reset, so it naturally has to be installed at the reset vector of your processor. In addition to the start address, many parts of the initialization code expect to find a vorgin, uninitialized system. Such parts must be disabled when you want to change the conditions under which you want to run U-Boot.
I read this several time in the net, but found no solution. I know it is possible, because I saw it in a flash hex-file, but can't reproduce it. I thought changing CFG_MONITOR_BASE / CFG_RESET_ADDRESS would do it.
As mentioned, this is NOT sufficient. See also the FAQ.
Do you say it's not easily possible to boot u-boot from another flash address even with a newer u-boot????
It is possible, and it is not so difficult if you have sufficient experience with U-Boot.
Do I have to make big changes to the uboot-source (start.S,..)?
This depends on your definition of "big". Judging from the questions you are asking, I tend to say: too big for you. No offence meant.
I could put ~256 bytes of assembly code to the reset-vector at 0x0A000000 and jmp in the u-boot; but if you say this is not enough, I will not start to learn the assembly language for my processor.
This has nothing to do with using assembly code.
Best regards,
Wolfgang Denk