Not as yet. I can get things setup to the point of creating the page table entries and about to jump into the kernel start and at which point it hangs.

I built for the i.MX27ADS and used a modified loader (putting in the "right" stuff) from the iMX31Litekit information in the forum elsewhere and I'm slowly debugging the setup.

Thilo

I managed to get it going a little bit. I'm using the 2.6.25-rc7-ptxmxc7 kernel from http://www.pengutronix.de...x-i.MX/index_en.html
and made some changes, see https://support.ultrawaves.com/imx27/imx27-uartfix.patch

The board hangs when reading PBC_VERSION_REG, which is on CS4, so I guess that's an external chip on the ADS board that's not there on the litekit.

I also keep the UART1 clock enabled (in PCCR1), and disable all baud rate programming.

My output is at https://support.ultrawaves.com/imx27/imx27-output.txt
Kernel config is at https://support.ultrawaves.com/imx27/imx27.config

Now I need to really figure out what's going on with those clocks...

Eric

I'm trying to run a Linux kernel on ZOOM iMX27 LITEKIT too.
I compile the kernel for iMX27ADS board and make some adjustments.
This adjustments are due to the existence of a external IO controller in ADS board.
To fix the clock issue, I changed the function "board_get_ckih_rate" in file
"arch/arm/mach-mx27/board-mx27ads.c" to only return 25000000 (25Hz).
The ADS board use a register to decide what clock to use, the BOARD VERSION REGISTER.

The things seems to be ok here.

My problem is the root file system. I'm trying to mount the root fs on NAND, but I'm not having success.
The problem seems to be that the Yaffs marks wrongly one block as bad.
I try to copy some file to NAND before erasing it. And the write attempt have marked the block as bad.
Anyone knows how I can rescan NAND for badblocks?

Thanks.

Hi, Yuri
I attempt to find board-mx27ads.c in Freescale ltib install path, but didn't get it.
Could you tell me which BSP you use, LTIB or Pengutronix?
Thank you!