No, this is totally normal: the kernel tries to read from the second

harddisk, and fails.  No problem.  The reason that 0.10 works, but 0.11

doesn't, is that I changed 0.11 so that it does a hd-reset at bootup,

mostly to test the code.  Ok it wasn't so great an idea, but it showed

up the bug in the reset-code.
If you can get linux up and running on some machine (0.10 works, but has

bugs, so 0.11 is better), you can make a new 0.11 that works somewhat

just by changing the line
static int reset = 1;
to
static int reset = 0;
at the head of hd.c, and recompiling the kernel.  This works, but

ignores the deeper problem of not being able to reset the disk.  I still

think you'd better wait for 0.12: it's ready, but I want to test

everything out for a bit yet.  I was still adding things to it

yesterday, now I'm just playing around, and trying to see if 0.12 fails.
		Linus
PS. Ignore my warning about bugs in bash-1.10: bash 1.11 came out

yesterday (the day before?), and was very easy to port to linux. I

haven't found any bugs yet.