I've noticed an interesting behaviour while working on the Debian boot
floppy code. My local CVS work area resides on an NFS mount and is
exported to the machine I use for mroe up-to-date development work.
Under 2.2.x kernels it appears that automagic loop device usage doesn't
work when the regular file being magically looped resides on an NFS share.
The same operation when the client is booted with a recent 2.3.x kernel
(2.3.99-pre[568]) succeeds. Also the same operation performed with the
file on a local mount succeeds on the 2.2.x kernel.
example:
idalton@heathen:~/work-in-progress/boot-floppies$ sudo syslinux -s \
resc1440-s.bin.tmp
ioctl: LOOP_SET_FD: Invalid argument
dmesg reports: loop: device has no block access/not implemented
Here $HOME is mounted NFS.
Machine details:
heathen, the development system, is up-to-date Debian potato prerelease.
Kernel 2.2.15/2.3.99-pre, libc6 2.1.3
the NFS server is running Debian slink (current stable relase and rather
old) kernel 2.2.14 with USB backport, libc6 2.0.7
heathen's kernel is the only thing that changes between working and
not-working.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue May 23 2000 - 21:00:14 EST