Re: Xen i386PAE status

[Mikolaj Golub <to.my.trociny%gmail.com@localhost>]

On Sun, 27 Jan 2008 00:41:16 -0500 <marcotte%panix.com@localhost> wrote:

 >> I don't yet figure out why, but on my test the PAE install kernel
 >> crashed when Xen tries to start it on my debian xen host.

 <> I can reproduce that here, but only on some of my hosts.

 <> It works properly on two SuperMicro PDSMI+ systems with 8GB of RAM, but
 <> fails on a SuperMicro X7DVL-E with 12 GB of RAM.

 <> The Dom0 and Xen are the same on both hosts.

 <> Here is what I get on the X7DVL-E:

 <> (XEN) domain_crash_sync called from entry.S (ff17120e)
 <> (XEN) Domain 19 (vcpu#0) crashed on cpu#2:
 <> (XEN) ----[ Xen-3.1.1  x86_32p  debug=n  Not tainted ]----
 <> (XEN) CPU:    2
 <> (XEN) EIP:    0009:[<c01048b3>]
 <> (XEN) EFLAGS: 00010296   CONTEXT: guest
 <> (XEN) eax: 00000000   ebx: 00000000   ecx: 00000130   edx: c0496fc4
 <> (XEN) esi: c0496fc0   edi: c0496fe4   ebp: c054a050   esp: c0549fe8
 <> (XEN) cr0: 8005003b   cr4: 000026f0   cr3: 001b2ca0   cr2: c0549fe8
 <> (XEN) ds: 0011   es: 0011   fs: 0031   gs: 0011   ss: 0011   cs: 0009
 <> (XEN) Guest stack trace from esp=c0549fe8:
 <> (XEN)    00000000 00000000 00000000 00000000 00000000 00000000 00000000 
00000000

I have crash booting current XEN3PAE_DOMU on linux debian etch DOM0 too.

(XEN) domain_crash_sync called from entry.S (ff15ccc9)
(XEN) Domain 5 (vcpu#0) crashed on cpu#0:
(XEN) ----[ Xen-3.0.3-1  x86_32p  debug=n  Not tainted ]----
(XEN) CPU:    0
(XEN) EIP:    e019:[<c03aa4b9>]
(XEN) EFLAGS: 00000292   CONTEXT: guest
(XEN) eax: 00000000   ebx: 00000000   ecx: 00000001   edx: c0458134
(XEN) esi: 00000000   edi: 000011c4   ebp: c044ff80   esp: c044ff54
(XEN) cr0: 8005003b   cr4: 000006f0   cr3: 183d8000   cr2: 00000001
(XEN) ds: e021   es: e021   fs: 0000   gs: 0000   ss: e021   cs: e019
(XEN) Guest stack trace from esp=c044ff54:
(XEN)    00000000 c03aa4b9 0001e019 00010092 c03aa57e 00000000 00000000 00000000
(XEN)    00000000 c04469e8 000011c4 c044ffb0 c030838f 00000000 00000005 00000000
(XEN)    00000000 c044ffbc 00000000 00000000 00000000 c044ffbc 00000100 c044fff0
(XEN)    c030849e c043a53b 00000000 c045a5c0 00000000 00000000 00000000 00000000
(XEN)    00000000 00000000 c0571000 00000000 c044fffc c0450084 00007ff0 c04500a0
(XEN)    c03aa002 c04469e8 c0575000

(gdb) list *0xc03aa4b9
0xc03aa4b9 is in x86_read_psl 
(/arch/netbsd/build/src/sys/arch/xen/x86/xen_intr.c:115).
110
111     u_long
112     x86_read_psl(void)
113     {
114
115             return (curcpu()->ci_vcpu->evtchn_upcall_mask);
116     }
117
118     void
119     x86_write_psl(u_long psl)
(gdb) 

% sudo xm info
host                   : home
release                : 2.6.18-4-xen-vserver-686
version                : #1 SMP Mon Mar 26 22:31:11 UTC 2007
machine                : i686
nr_cpus                : 1
nr_nodes               : 1
sockets_per_node       : 1
cores_per_socket       : 1
threads_per_core       : 1
cpu_mhz                : 2263
hw_caps                : bfebfbff:00000000:00000000:00000180:0000441d
total_memory           : 510
free_memory            : 128
xen_major              : 3
xen_minor              : 0
xen_extra              : .3-1
xen_caps               : xen-3.0-x86_32p
xen_pagesize           : 4096
platform_params        : virt_start=0xf5800000
xen_changeset          : Tue Oct 17 22:09:52 2006 +0100 
cc_compiler            : gcc version 4.1.2 20061028 (prerelease) (Debian 
4.1.1-19)
cc_compile_by          : ultrotter
cc_compile_domain      : debian.org
cc_compile_date        : Thu Nov  2 20:32:28 CET 2006
xend_config_format     : 2

Could it be due to rather old XEN shipped with debian etch? I see we have 3.1
in pkgsrc...

-- 
Mikolaj Golub