From: Chuck R. <shi...@ro...> - 2008-08-20 21:16:23
|
On Aug 20, 2008, at 11:19 AM, Harry J Mason wrote: > On Wed, 20 Aug 2008, Chuck Randall wrote: > >> On Aug 20, 2008, at 3:31 AM, Harry J Mason wrote: >> >>> On Tue, 19 Aug 2008, Chuck Randall wrote: >>> >>>> I just reflashed my connex-400xm with a newly built gumstix-basic- >>>> image. Now I get a "bad magic number" when I try to boot it. What >>>> could have gone wrong? The log of the reflash procedure is show >>>> below. >>> >>> [snip] >>> >>>> GUM> fatload mmc 1 a20000000 rootfs >>>> GUM> cp.b a20000000 40000 ${filesize} >>> >>> Too many zeros here. The correct address is a2000000 (6 zeros). >> >> Yes, I just noticed that, too. When I use a2000000 I get a bunch of >> errors. See the log listing below. > >> jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at >> 0x00000000: 0x0f0f id >> jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at >> 0x00000004: 0x0404 id > > It looks like there's incorrect data instead of a valid jffs2, and the > kernel is trying and failing to mount it. > > Did you reflash the filesystem with the correct address, before you > reflashed the kernel? > Did you protect on 1:0-1 ; erase all first? > Did you flash the correct file from oe/tmp/deploy/glibc/images ? Well, I thought I did! ;-) I'll re-do all of the steps again and see what happens. Thx, Chuck |