[Xen-users] soft lockup detected on CPU#0! (was Determining info Dom0 from within DomU)

[G Perry <gperry@xxxxxxxxxxxxxxxxxxx>]

> But if you can't access Dom0 via the network, then you SHOULD NOT be
> able to access any information owned by Dom0 that isn't specifically
> exported to that particular DomU - if you can, it's a bug and should
> be fixed!

Agreed, although I've been able to see a bit of information about their
Dom0 Hypervisor from a kernel stack trace.  I wanted to get more
information about the Amazon EC2 Dom0 setup prior to posting this, but
they haven't answered the ticket so I'll repost the original thread
here:

BUG: soft lockup detected on CPU#0!
Posted: Jan 29, 2007 4:58 PM            

I am getting the following error message from the EC2 Dom0 Xen
Hypervisor on bootup, using a groupinstall Core bootstrapped FC6 image.

FC6 by default now includes nosegneg friendly libraries so the warning
message about unsupported memory accesses is moot.  I would like to
cross post this to the Xen users list as this appears to be an internal
issue with the Xen Hypervisor, but I can't do that unless I have a bit
more information about the EC2 setup including which revision of Xen
and the operating system you are running for Dom0.  Can someone please
post that information to the group here, as well as the relevant SXP
configuration?

During bootup I am getting the following stack trace from the kernel on
instance ID i-4a5bbf23:

----------

Freeing unused kernel memory: 144k freed
 
  ***************************************************************
  ***************************************************************
  ** WARNING: Currently emulating unsupported memory accesses  **
  **          in /lib/tls glibc libraries. The emulation is    **
  **          slow. To ensure full performance you should      **
  **          install a 'xen-friendly' (nosegneg) version of   **
  **          the library, or disable tls support by executing **
  **          the following as root:                           **
  **          mv /lib/tls /lib/tls.disabled                    **
  ** Offending process: init (pid=1)                           **
  ***************************************************************
  ***************************************************************
 
Pausing... 1<3>BUG: soft lockup detected on CPU#0!

Pid: 1, comm:                 init
EIP: 0061:[<c0107ce2>] CPU: 0
EIP is at delay_tsc+0x12/0x20
 EFLAGS: 00000283    Not tainted  (2.6.16-xenU #1)
EAX: 0024511f EBX: 0024bc33 ECX: d1afe9ee EDX: 00462fea
ESI: 00000001 EDI: c0110950 EBP: ed7c1f18 DS: 007b ES: 007b
CR0: 8005003b CR2: b7efd160 CR3: 003a2000 CR4: 00000660
 [<c0105b83>] show_trace+0x13/0x20
 [<c01033cf>] show_regs+0x1bf/0x1f0
 [<c013ef58>] softlockup_tick+0x88/0xa0
 [<c0128025>] do_timer+0x215/0x440
 [<c010855c>] timer_interrupt+0x1bc/0x650
 [<c013f0af>] handle_IRQ_event+0x3f/0xd0
 [<c013f1ce>] __do_IRQ+0x8e/0xf0
 [<c0106e9d>] do_IRQ+0x1d/0x30
 [<c024ac31>] evtchn_do_upcall+0xa1/0xe0
 [<c01052dc>] hypervisor_callback+0x2c/0x34
 [<c0224b14>] __delay+0x14/0x20
 [<c0224b56>] __const_udelay+0x36/0x40
 [<c0110a7c>] do_fixup_4gb_segment+0x12c/0x150
 [<c01052ab>] error_code+0x2b/0x30
Continuing...

dm_mod: no version for "struct_module" found: kernel tainted.
device-mapper: 4.5.0-ioctl (2005-10-04) initialised: dm-devel@xxxxxxxxxx

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users