[lkml]   [2003]   [Jul]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: PROBLEM: Fault in init/do_mounts.c/change_floppy{} ramdisk file system read failure
>      I an trying to build a standalone Linux system using the process described 
> in the Bootdisk-HOWTO. HOWEVER I am using an *uncompressed* root file
> system since I was unable to configure a compressed one that is small
> enough to fit on a single floppy. I built:
> - a boot floppy containing a bzImage Linux kernel and minimal
> infrastructure that gets me to the point of being able to load a file
> system from floppy onto ramdisk. This is the same kernel that I
> use on my desktop every day.
> - an ext2 root file system (4000 blocks) that holds a minimal set of
> files/programs
> - a set of 3 floppy disks that contains an uncompressed copy of the
> ext2 root file system
> When I try to boot the standalone system the following sequence of
> events occurs:
> - I load the boot floppy, and the kernel appears to load and
> initialize correctly.
> - The kernel prompts me to insert a floppy so it can load a file
> system onto ramdisk.
> - The code in init correctly determines that my floppy disk
> contains an uncompressed ext2 file system.
> - in init/do_mounts.c rd_load_image is called with argument "/dev/root"
> - the for loop in this function starts reading from the first
> floppy disk. Cachunk, cachunk.
> - When 1440 1k block have been read, the function change_floppy is
> called to move to the next floppy disk.
> - change_floppy does something BROKEN. See below.
> - The for loop in rd_load_image continues to try and read in more
> uncompressed file system from the floppy, BUT THE READ DOES NOT
> ACCESS the floppy drive. The reads in the for loop appear to
> succeed, but nothing is actually transfered from the floppy
> disk to the ramdisk.
> - The same read failure happens for the third floppy disk.

I saw similar behavior last time I tried to make an uncompressed root
fs that spanned five disks, (around 2.4.21-rc1), but I seem to
remember that it read disk 1, ignored disk 2, read disk 3, etc -
I.E. odd numbered disks were actually accessed.

I worked around the problem by making a _really_ minimal root fs that
fitted on a single disk, booting with that, then extracting everything
else I needed from a tar archive on another disk :-).

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:47    [W:0.046 / U:1.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site