Ok, so I feel as though now I have a little more insight to this problem. Here's a brief rundown. I am using u-boot to boot the latest linux kernel from a large page NAND device (trivial) with the rootfs also on the same NAND device. In my dts the fs is defined as being located at 0x400000 (within the device) and is 0x1c00000 in size. The image for the fs that I burn into the device is 0x240000 in size.
During the boot of linux, after the jffs2 rootfs is mounted it appears that linux attempts to put the magic number 0x1985xxxx into the beginning of each block following the end of the file system through the end of the device, as if to format the NAND flash for future use by the jffs2 fs. Am I correct so far? Here's the error that I'm seeing. Instead of writing this data into the NAND blocks, it's being written into the spare region of the flash, designated for ECC, bad block marker, and so on. This is causing a large problem as now my blocks are being marked as bad. So my second question is, where is the code that performs this "formatting" of the NAND device immediately after mounting of the rootfs? Ron Madrid _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev