[parted-devel] s390 regression on 1.8.1

[Sadrul H Chowdhury]

On 12/5/06, Otavio Salvador <otavio at debian.org> wrote:
>
> "Sadrul H Chowdhury" <imadil at gmail.com> writes:
>
> >    On 12/5/06, Otavio Salvador <otavio at debian.org> wrote:
> >
> >      Hello folks,
> >      Today I managed to get a test of Parted on s390 (using hercules)
> >      and
> >      it fails to reconize the disk label.
> >
> >    I notice the same problem with ext2, ext3 and non-partitioned disk
> >    with the git-repo. 1.7 from the debian unstable/testing repo did
> >    detect the drives/partitions properly.
>
> Please, provide more information and a test case for us if possible.


This is the backtrace:
(gdb) bt
#0  linux_read (dev=0x806d5c0, buffer=0xbffff684, start=0, count=1)
    at arch/linux.c:1459
#1  0x40029cf6 in ped_device_read (dev=0x806d5c0, buffer=0xbffff684,
start=0,
    count=1) at device.c:346
#2  0x4007431f in msdos_probe (dev=0x806d5c0) at dos.c:166
#3  0x4002e371 in ped_disk_probe (dev=0x806d5c0) at disk.c:151
#4  0x4002e420 in ped_disk_new (dev=0x806d5c0) at disk.c:182
#5  0x0804e4f5 in do_print (dev=0xbffffaac) at parted.c:1238
#6  0x0804bb59 in command_run (cmd=0x8069fb0, dev=0xbffffaac) at command.c
:138
#7  0x0805493b in interactive_mode (dev=0xbffffaac, cmd_list=0x80584a0)
    at ui.c:1310
#8  0x08051a3a in main (argc=0, argv=0xbffffb1c) at parted.c:2385
(gdb) p diobuf
$35 = (void *) 0x8070600
(gdb) p read_length
$36 = 512
(gdb) p *arch_specific
$37 = {fd = 5}
(gdb) p *dev
$45 = {next = 0x0, model = 0x806bd18 "VMware, VMware Virtual S",
  path = 0x806bee0 "/dev/sda", type = PED_DEVICE_SCSI, sector_size = 512,
  phys_sector_size = 512, length = 20971520, open_count = 3, read_only = 0,
  external_mode = 0, dirty = 0, boot_dirty = 0, hw_geom = {cylinders = 1305,

    heads = 255, sectors = 63}, bios_geom = {cylinders = 1305, heads = 255,
    sectors = 63}, host = 0, did = 0, arch_specific = 0x806bef0}
(gdb)

arch/linux.c:1459 is:
        status = read (arch_specific->fd, diobuf, read_length);

This read operation returns -1, which results in the following ?ThisThis
read operation returns -1. Tracing through a few more steps, I see the
following exception being thrown:

(gdb) bt 2
#0  ped_exception_throw (ex_type=PED_EXCEPTION_ERROR, ex_opts=112,
    message=0x4008d78a "%s during read on %s") at exception.c:253
#1  0x4003998c in linux_read (dev=0x806d5c0, buffer=0xbffff684, start=0,
    count=1) at arch/linux.c:1469
(More stack frames follow...)
(gdb) p *ex
$44 = {message = 0x8070808 "Invalid argument during read on /dev/sda",
  type = PED_EXCEPTION_ERROR, options = 112}
(gdb)

I can reproduce this every time. So let me know if you need more
information. My kernel version is 2.4.27-3-386, if that's relevant.

Cheers,
Sadrul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.alioth.debian.org/pipermail/parted-devel/attachments/20061206/c2f53f14/attachment.htm