On 04/27/2010 10:06 AM, Hauke Mehrtens wrote:
Yes there are some fixups that have to be fixed ;-)
Please have a look at: https://dev.openwrt.org/ticket/3177
Hauke
Jo-Philipp Wich schrieb:
My first best guess is /etc/init.d/nvram, it does some fixups on firstboot.
~ Jow
hmm, thanks for the tips - it really seems, that change of 'sdram_init' in nvram
boot script may be the cause - I guess that 'boardtype' of 0x042f applies to
wl-500gp, so the original sdram_init value was ORed with 0x0009, hm, great:-(
A page describing the memory controller values:
http://wl500g.dyndns.org/sdram.html
sdram_init value of 0x0019 (probably had 0x0011 before, ORed by backfire with
0x0009) is not documented there, having memory controller's ColWidth bitfield
with 110 value now, probably a non sense unsupported by hw...
Is it right assuming that CFE runs initially from flash, without use of ram
because memory controller is not set up yet, then taking values from nvram part
of flash to programm the memory controller before relocating itself/kernel into ram?
I guess the last hope for the router is the pin9 trick - read a lot about it,
checked even datasheets, did not find somebody's explanation, why it should even
work though, so never tried it - shorting an address line of flash chip - what
it might cause, if it looks like CFE is not started as there is nothing on
serial console?
Or is CFE started only partially, and logs are visible on serial console only if
sdram controller is set up properly?
Jan
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
