Re: Kernel hang with latest snapshot

["bibo,mao" <bibo dot mao at intel dot com>]

I tried stap -e "probe addevent.iosyscall {}" -bMv on my different IA32
machines, system did not crash either. Your configuration seems the same
with mine, only that kernel version is different.

which systemtap version do you have? what is the output of this command
"which stap" and ldd `which stap`.

thanks
bibo,mao

David Boreham wrote:
>  >   :(  I tried to probe using addevent.iosyscall and they seems well
>  > for me. My testing environment:
>  >   IBM T43 thinkpad with Pentium-M 2.0
>
> I've used vmware server running on windowsxp on AMD Athlon, and also
> FC5 native on a 4-processor Xeon 700 machine with 4G ram.
>
>  >   Fedora Core5 updated to the latest
>
> Same here.
>
>  >   gcc 4.1.1 20060525
>
> gcc-4.1.1-1.fc5 here
>
>  >   elfutils 0.124-0.1, rpm packages are from Red Hat ftp server
>
> elfutils-0.124-0.1 here
>
>  >   latest SystemTap
>
> ditto, from cvs
>
> [root@fc5test src]# ls -l /usr/local/bin/stap
> -rwxr-xr-x 1 root root 4178597 Oct 16 09:43 /usr/local/bin/stap
>
>  >   running command: stap -e "probe addevent.iosyscall {}" -bMv
>
> If I run that command I get:
>
> [root@fc5test src]# stap -e "probe addevent.iosyscall {}" -bMv
> Pass 1: parsed user script and 53 library script(s) in
> 440usr/80sys/544real ms.
> Pass 2: analyzed script: 36 probe(s), 52 function(s), 245 global(s) in
> 790usr/460sys/3582real ms.
> Pass 3: translated to C into "/tmp/stapzETvL1/stap_5603.c" in
> 300usr/300sys/781real ms.
> Pass 4: compiled C into "stap_5603.ko" in 31750usr/3730sys/43636real ms.
> Pass 5: starting run.
> <machine is now locked up>