[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[linux-lvm] big endian arm support?



I'm not able to create Volume groups successfully on a Big Endian Xscale
Intel IXP425;  Here is a log of what happens, along with some
potentially helpful debug...

dd if=/dev/zero of=/dev/discs/disc0/disc bs=1M count=1
1+0 records in
1+0 records out

pvcreate /dev/discs/disc0/disc
  Physical volume "/dev/discs/disc0/disc" successfully created

pvdisplay /dev/discs/disc0/disc
  No physical volume label read from /dev/discs/disc0/disc
  Failed to read physical volume "/dev/discs/disc0/disc"

vgcreate vg00 /dev/discs/disc0/disc
  Incorrect metadata area header checksum
  Incorrect metadata area header checksum
  No physical volume label read from /dev/discs/disc0/disc
  /dev/discs/disc0/disc not identified as an existing physical volume
  Unable to add physical volume '/dev/discs/disc0/disc' to volume group
'vg00'.

vgdisplay --version
  LVM version:     2.00.20 (2004-07-03)
  Library version: 1.00.19-ioctl (2004-07-03)
  Driver version:  4.1.0

uname -a
Linux mayfield-blueridge 2.6.7-rc3 #27 Mon Jul 26 19:11:25 MDT 2004
armv5teb unknown

I'm using glibc-2.3.2 and a gcc 3.4.0 FYI.

Also, I read the LVM2 is not focusing on performance and may be
extremely slow compared to LVM1 (factor or 2 was thrown around);  Is
this still true?  I need max speed from my disk subsystem, as I'm
already on a slow platform.  I tried using LVM1, but the compile is
massively broken, and it looks like modern 2.6 kernels don't support
this (No def of LVM_MAJOR, etc, etc);  Perhaps I just need to dig some
more into making LVM1 work for me.

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]