WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-users

[Xen-users] Problem getting VTPM to work

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Problem getting VTPM to work
From: Martin Hermanowski <xen-user@xxxxxxxxxxxxxx>
Date: Thu, 18 May 2006 10:38:28 +0200
Delivery-date: Thu, 18 May 2006 01:39:09 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/1.5.11+cvs20060403
Hi,

I'm trying to get the VTPM to work with an Atmel TPM. The computer is an
IBM T41p running Debian Etch, and I verified that the TPM is working
well using the libtpm library:

,----
| debian:~# ./libtpm-2.0/utils/tpm_demo
| TPM successfully reset
| TPM version 1.1.0.6
| 16 PCR registers are available
| ...
`----

I installed Xen testing from source (downloaded by `hg clone
http://xenbits.xensource.com/xen-3.0-testing.hg' yesterday), and Xen
itself is running fine.

I compiled my own kernel (make linux-2.6-xen-config
CONFIGMODE=menuconfig, build...) with tpm_atmel and the Xen TPM options
included.

Before starting any DomU, i run `vtpm_managerd', and everything looks
fine:

,----
| INFO[VTPM]: Loaded saved state (dmis = 8).
| INFO[VTSP]: Loading Key into TPM.
| INFO[VTPM]: Re-attaching DMI instance 0 on domain 0 .
| INFO[TCS]: Calling TCS_OpenContext:
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
| INFO[VTPM]: [2]: Waiting for DMI messages.
`----

Once I start my test domain, the following messages are displayed:
,----
| INFO[VTPM]: [1]: RECV[14}: 0x0 0 0 0 1 c1 0 0 0 13 0 0 0 1 0 0 0 0 0 0 0 0 1
| INFO[VTPM]: Re-attaching DMI instance 1 on domain 0 .
| INFO[TCS]: Calling TCS_OpenContext:
| ERROR[VTPM]: Could not exec to launch vtpm
| INFO[VTPM]: [1]: SENT: 0x0 0 0 0 1 c1 0 0 0 a 0 0 0 0
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
| INFO[VTPM]: Launching DMI on PID = 7308
| ERROR[VTPM]: [1]: Error writing to BE. Aborting...
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
| ERROR[VTPM]: [1]: Can't open inbound fh for /dev/vtpm.
`----

(BTW: If I close vtpm_managerd with CTRL+C, vtpm_managerd segfaults and
keys are left loaded in the TPM:
,----
| INFO[VTPM]: VTPM Manager shutting down for signal 2.
| INFO[TCS]: Calling TCS_CloseContext.
| INFO[VTPM]: Child shutting down
| Segmentation fault
`----)

An exec of vtpm fails because there is now /usr/bin/vtpmd. However, if I
modify the Makefile in tools/vtpm_manager/manager to build vtpmd (added
it to the BIN= line), I get the following error:

,----
| INFO[VTPM]: [1]: RECV[14}: 0x0 0 0 0 1 c1 0 0 0 13 0 0 0 1 0 0 0 0 0 0 0 0 1
| INFO[VTPM]: Re-attaching DMI instance 1 on domain 0 .
| INFO[TCS]: Calling TCS_OpenContext:
| INFO[VTPM]: Starting VTPM.
| INFO[TCS]: Constructing new TCS:
| ERROR[TXDATA]: TPM open failedERROR in VTPM_Init_Service at 
vtpm_manager.c:769 code: TPM_IOERROR.
| ERROR[VTPM]: Closing vtpmd due to error during startup.
| INFO[VTPM]: Launching DMI on PID = 7632
| INFO[VTPM]: [1]: SENT: 0x0 0 0 0 1 c1 0 0 0 a 0 0 0 0
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
`----

So it seems this is not the way it should be...

If I replace /usr/bin/vtpmd with an empty shell script, I don't get
errors when starting my test domain, but on the first access (`cat
/sys/bus/platform/devices/tpm_vtpm/pcrs' or tpm_demo again):

,----
| INFO[VTPM]: [1]: RECV[14}: 0x0 0 0 0 1 c1 0 0 0 13 0 0 0 1 0 0 0 0 0 0 0 0 1
| INFO[VTPM]: Re-attaching DMI instance 1 on domain 0 .
| INFO[TCS]: Calling TCS_OpenContext:
| INFO[VTPM]: Launching DMI on PID = 7965
| INFO[VTPM]: [1]: SENT: 0x0 0 0 0 1 c1 0 0 0 a 0 0 0 0
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
(running `cat /sys/bus/platform/devices/tpm_vtpm/pcrs')
| INFO[VTPM]: [1]: RECV[14}: 0x0 0 0 1 0 c1 0 0 0 16 0 0 0 65 0 0 0 5 0 0 0 4 0 
0 1 1
| INFO[VTPM]: [1]: Forwarding command to DMI.
| ERROR[VTPM]: [1]: VTPM ERROR: Can't open outbound fh to dmi.
| INFO[VTPM]: [1]: SENT: 0x0 0 0 1 0 c1 0 0 0 a 0 0 0 1f
| INFO[VTPM]: [1]: Waiting for Guest requests & ctrl messages.
`----

Can you tell me what I am doing wrong?

Thanks in advance,
Martin

-- 
Martin Hermanowski
http://martin.hermanowski.name

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

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-users] Problem getting VTPM to work, Martin Hermanowski <=