[DNX#2006033142000249] [U-Boot-Users] [PATCH] Support for new board, gth2 [...]

[DENX Support System]

Hello list,

inside the automatic U-Boot patch tracking system a new ticket
[DNX#2006033142000249] was created:

<snip>
> The au1xx0 series cpu:s does not have a built-in
> watchdog or a reset register.
> That means that reset is board dependant for
> au1xx0.
> 
> Looking at cpu.c, it seems that the other mips cpu:s
> are special (or no "true" reset) also:
> 
> #if defined(CONFIG_INCA_IP)
> 	*INCA_IP_WDT_RST_REQ = 0x3f;
> #elif defined(CONFIG_PURPLE) || defined(CONFIG_TB0229)
> 	void (*f)(void) = (void *) 0xbfc00000;
> 
> 	f();
> 
> /Thomas
> 
> Wolfgang Denk wrote:
> >>>It seems like resetting is a pretty board specific thing.  Maybe
> > 
> > 
> > Why do you think so? In almost all cases I've seen it's  a  processor
> > specific thing, not board dependent.
> > 
> > 
> >>>there should be a generic call to board_reset() in cpu.c and lib_mips/board.c
> >>>and have the board directory supply that function.
> >>>
> >>>something like this
> >>>
> >>>#ifdef CONFIG_BOARD_RESET
> >>>    board_reset();
> >>>#else
> >>>    __asm__("la t0, 0xbfc00000; jr t0");
> > 
> > 
> > This is *not* a reset anyway; it should be fixed me thinks.
> > 
> > 
> >>Sounds very good to me. What do you think Wolfgang?
> > 
> > 
> > This has been discussed several times (in length) before. I  have  to
> > admit  that I don;t remember the result of the discussion, but if you
> > cannot find board_reset() for other architectures this probably means
> > that I rejected the idea by then, too...
> > 
> > 
> > Best regards,
> > 
> > Wolfgang Denk
> > 
</snip>

Your U-Boot support team