Xen kernel doesn't boot for me

[Doug Eubanks]

I've tried the xen kernel on Fedora 6, CentOS 5 and now I'm running Fedora 7.

Each time, it seems to hang during or perhaps right after the SCSI hardware is
initialized.  It doesn't panic, although it sometimes prints messages about USB
devices not accepting new addresses (but not always).

I am attaching my SCSI and CPU info, if anyone has a suggestion to fix or troubleshoot
this I would appreciate any information.



[root@DMZServer /]# ls /proc/scsi/
aic7xxx  device_info  gdth  scsi  sg



[root@DMZServer /]# cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 10
model name      : Pentium III (Cascades)
stepping        : 0
cpu MHz         : 701.640
cache size      : 1024 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx
fxsr sse
bogomips        : 1403.76
clflush size    : 32

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 10
model name      : Pentium III (Cascades)
stepping        : 0
cpu MHz         : 701.640
cache size      : 1024 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx
fxsr sse
bogomips        : 1403.02
clflush size    : 32

processor       : 2
vendor_id       : GenuineIntel
cpu family      : 6
model           : 10
model name      : Pentium III (Cascades)
stepping        : 0
cpu MHz         : 701.640
cache size      : 1024 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx
fxsr sse
bogomips        : 1403.02
clflush size    : 32

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 6
model           : 10
model name      : Pentium III (Cascades)
stepping        : 0
cpu MHz         : 701.640
cache size      : 1024 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx
fxsr sse
bogomips        : 1403.02
clflush size    : 32

[Richard W.M. Jones]

Doug Eubanks wrote:
>  I've tried the xen kernel on Fedora 6, CentOS 5 and now I'm
> running 
>  Fedora 7.
>  
>  Each time, it seems to hang during or perhaps right after the SCSI 
>  hardware is initialized.  It doesn't panic, although it sometimes prints 
>  messages about USB devices not accepting new addresses (but not always). 
We really need to see the kernel messages.

Rich.

-- 
Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/
Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod
Street, Windsor, Berkshire, SL4 1TE, United Kingdom.  Registered in
England and Wales under Company Registration No. 03798903

[Doug Eubanks]

So I tried the lastest XEN kernel again on Friday (2.6.20-2925.11.fc7xen).

It gets hung in a loop resetting the AIC7xxx adapter.  It resets, timesouts, resets again
over and over.

I have two adapters that use that driver in my system, the on board (which is disabled)
and an Adaptec 2940 which has my DLT tape drive attached.  Is it possible that the
2.6.20-2925.11.fc7xen kernel has a bug in the AIC7xxxx driver that's been patched in
the newer mainline kernel?

If anyone needs the specific messages, I will be happy to write them down today (I have to
reboot to install another 1G of RAM that's arriving today).

Thanks,
Doug Eubanks


  _____  

From: Richard W.M. Jones [mailto:[e-mail]]
To: Doug Eubanks [mailto:[e-mail]]
Cc: fedora-xen(a)redhat.com
Sent: Fri, 06 Jul 2007 15:11:32 -0400
Subject: Re: [Fedora-xen] Xen kernel doesn't boot for me

Doug Eubanks wrote:
>  I've tried the xen kernel on Fedora 6, CentOS 5 and now I'm
> running 
>  Fedora 7.
>  
>  Each time, it seems to hang during or perhaps right after the SCSI 
>  hardware is initialized.  It doesn't panic, although it sometimes prints 
>  messages about USB devices not accepting new addresses (but not always). 
  
  We really need to see the kernel messages.
  
  Rich.
  
  -- 
  Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/
  Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod
  Street, Windsor, Berkshire, SL4 1TE, United Kingdom.  Registered in
  England and Wales under Company Registration No. 03798903

[Richard W.M. Jones]

Doug Eubanks wrote:
>  So I tried the lastest XEN kernel again on Friday
> (2.6.20-2925.11.fc7xen).
>  
>  It gets hung in a loop resetting the AIC7xxx adapter.  It resets, 
>  timesouts, resets again over and over.
>  
>  I have two adapters that use that driver in my system, the on board 
>  (which is disabled) and an Adaptec 2940 which has my DLT tape drive 
>  attached.  Is it possible that the 2.6.20-2925.11.fc7xen kernel has a 
>  bug in the AIC7xxxx driver that's been patched in the newer mainline kernel? 
The 2.6.20...xen kernel really is based off the 2.6.20 standard (Fedora) 
kernel.  So if the driver works in 2.6.20-2925 and not in the xen 
version, it's not because of a bugfix, but because of something else 
about Xen itself.

>  If anyone needs the specific messages, I will be happy to write them
>
>  down today (I have to reboot to install another 1G of RAM that's 
>  arriving today). 
Very hard to say without seeing the messages.  Assuming that the boot 
isn't getting you into a shell, setting up a serial console is the 
easiest way to capture them.  Alternately you can try taking a 
screenshot with a digital camera.

Rich.

-- 
Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/
Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod
Street, Windsor, Berkshire, SL4 1TE, United Kingdom.  Registered in
England and Wales under Company Registration No. 03798903

[Eduardo Habkost]

On Mon, Jul 09, 2007 at 03:56:39PM +0100, Richard W.M. Jones wrote:
>  Doug Eubanks wrote:
>  >So I tried the lastest XEN kernel again on Friday (2.6.20-2925.11.fc7xen).
>  >
>  >It gets hung in a loop resetting the AIC7xxx adapter.  It resets, 
>  >timesouts, resets again over and over.
>  >
>  >I have two adapters that use that driver in my system, the on board 
>  >(which is disabled) and an Adaptec 2940 which has my DLT tape drive 
>  >attached.  Is it possible that the 2.6.20-2925.11.fc7xen kernel has a 
>  >bug in the AIC7xxxx driver that's been patched in the newer mainline 
>  >kernel?
>  
>  The 2.6.20...xen kernel really is based off the 2.6.20 standard (Fedora) 
>  kernel.  So if the driver works in 2.6.20-2925 and not in the xen 
>  version, it's not because of a bugfix, but because of something else 
>  about Xen itself. 
Actually, it may be because of a missing bugfix, also. It may be because
the F7 non-xen kernel is based on a newer upstream kernel version,
or because I have missed some fix that went to the F7 non-xen kernel.

On FC6, both non-xen and xen kernels come from the same source package,
so this wouldn't happen on FC6. But somewhere between Fedora 6 and 7,
the non-xen and xen kernels were split, because the non-xen kernel was
being rebased faster than we were being able to rebase the xen patches.

Being as close as possible to the non-xen kernel is one of my goals,
but currently there may be some bits missing.

-- 
Eduardo