[Xen-bugs] [Bug 106] New: running dmidecode or biosdecode on dom0 crashes Xen but only when another domain exists

[dh286@xxxxxxxxx]

http://bugzilla.xensource.com/cgi-bin/bugzilla/show_bug.cgi?id=106

           Summary: running dmidecode or biosdecode on dom0 crashes Xen but
                    only when another domain exists
           Product: Xen
           Version: unstable
          Platform: x86
        OS/Version: Linux-2.6
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Hypervisor
        AssignedTo: xen-bugs@xxxxxxxxxxxxxxxxxxx
        ReportedBy: dh286@xxxxxxxxx


Running either of dmidecode and biosdecode (and probably other programs that
access the BIOS in a similar way) in a dom0 when it is the only domain running
works in the correct way, but running it when more than one domain exists
produces the following transcript:

>>>
xendemo:~# xm list 
Name              Id  Mem(MB)  CPU VCPU(s)  State  Time(s)  Console
Debian-2           1       47    1      1   -b---     14.0     9601
Domain-0           0      247    0      1   r----     23.2         
xendemo:~# biosdecode
# biosdecode 2.6
(XEN) CPU:    0
(XEN) EIP:    e008:[<ff109f2f>]      
(XEN) EFLAGS: 00010246   CONTEXT: hypervisor
(XEN) eax: 00000001   ebx: ff18ac80   ecx: 00000000   edx: 00010002
(XEN) esi: 00000000   edi: 00000000   ebp: ff188000   esp: ff103e54
(XEN) cr0: 8005003b   cr3: 3cbff000
(XEN) ds: e010   es: e010   fs: 0000   gs: 0000   ss: e010   cs: e008
(XEN) Xen stack trace from esp=ff103e54:
(XEN)    00000000 0ee9deef 0000001f 00000000 00000000 00000004 00000000 000e0025
 
(XEN)    fc401500 ff18a080 ff186080 [ff11f480] ff186080 ff18a080 000000e0 000000
01 
(XEN)    ff189e00 00000004 00000000 0000028f 000e0025 fed4b000 ff199000 [ff1220a
1] 
(XEN)    000e0025 ff18a080 00000001 0032f1c4 fed4b000 ff18a300 ff18a080 [ff12218
a] 
(XEN)    ff18a080 fed4b000 ff199000 ff103fb4 ff18a310 cbc71a3d 00000000 ff18a080
 
(XEN)    ff18a080 b7e97000 00000000 [ff122532] ff18a080 00000000 ff189080 [ff11f
ead] 
(XEN)    ff18a080 ff189080 0000001f 00000286 ff103fb4 ff18a300 cbc71a3c ff18a300
 
(XEN)    00007c71 07c71061 [ff122405] 00007c71 ffffffff 00000000 ff18a080 [ff123
46d] 
(XEN)    00000000 ff18a300 000002df 00000000 ff189080 00007c71 00000000 ff103fb4
 
(XEN)    cbc71a3c 00000000 00000000 ff18a080 cbc71a3c 00ef0000 [ff118e9d] 000000
00 
(XEN)    07952067 080577ea ff189080 ff189080 00007ff0 b7e97000 cd63bd2c [ff13890
3] 
(XEN)    cb45bee0 00000001 00000000 00007ff0 b7e97000 cd63bd2c 0000001a 000e0003
 
(XEN)    c0116f37 00000061 00000246 cb45bee0 00000069 0000007b 0000007b 00000000
 
(XEN)    00000000 00000000 ff189080 
(XEN) Xen call trace from esp=ff103e54:
(XEN)    [<ff11f480>] [<ff1220a1>] [<ff12218a>] [<ff122532>] [<ff11fead>] [<ff12
2405>] 
(XEN)    [<ff12346d>] [<ff118e9d>] [<ff138903>] 

****************************************
Panic on CPU0:
CPU0 FATAL PAGE FAULT
[error_code=0000]
Faulting linear address: 00000004
****************************************

Reboot in five seconds...

<<<
Most of the above is the same each time, but cr3 and the stack trace are
different each time. Also different is the call trace if dmidecode is run
instead of biosdecode:
>>>
(XEN) Xen call trace from esp=ff103e54:
(XEN)    [<ff11f179>] [<ff11f480>] [<ff1220a1>] [<ff12218a>] [<ff122532>]
[<ff11fead>] 
(XEN)    [<ff122405>] [<ff12346d>] [<ff138903>] 
<<<

Tested on felix running unstable from the updated (Debian) demo CD.
Interestingly, the SLES9 install w/ 2.0-testing works fine.

If I get some free moments this week I will look into exactly what
{dmi,bios}decode are doing when they fail, if that will be any help.



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

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