[ltt-dev] Incomplete Traceset statistics on 2.6.27.10?

[compudj@krystal.dyndns.org (Mathieu Desnoyers)]

* Dany R Madden (danymadden at us.ibm.com) wrote:
> Hello,
> 
> I am playing with lttng 0.72 (patch-2.6.27.10-lttng-0.72) on 2.6.27.10 
> kernel. I'm using ltt-control-0.64-14012009 and lttv-0.12.7-20012009. I 
> followed this instruction: 
> http://ltt.polymtl.ca/svn/trunk/lttv/LTTngManual.html#section1. (I did do 
> ltt-armall before starting the trace)
> 
> I am expecting to see more information in the Traceset statistics, but 
> they seem to be blank. What should a working trace data look like? (I 
> can't imagine that this is it.)
> 
> This is the output after doing lttv textDump.
> 

See

lttv -m textDump --help to see how to enable statistics.

In the current lttv versions, the textDump does not show "empty"
statistics. They are only shows when the --stats option is used.

Mathieu

> ...
> kernel.syscall_entry: 1727.121073743 (/tmp/trace1/kernel_0), 0, 0, 
> /bin/dmesg, , 0, 0x0, SYSCALL { ip = 0xffffe430, syscall_id = 4 
> [sys_write+0x0/0x87] }
> fs.open: 1727.121083083 (/tmp/trace1/fs_6), 0, 0, /usr/bin/lttctl, , 0, 
> 0x0, SYSCALL { fd = 4, filename = 
> "/sys/kernel/debug/ltt/control/trace1/enabled" }
> kernel.syscall_exit: 1727.121086156 (/tmp/trace1/kernel_6), 0, 0, 
> /usr/bin/lttctl, , 0, 0x0, SYSCALL { ret = 4 }
> fs.write: 1727.121087465 (/tmp/trace1/fs_0), 0, 0, /bin/dmesg, , 0, 0x0, 
> SYSCALL { fd = 3, count = 64 }
> kernel.syscall_exit: 1727.121090016 (/tmp/trace1/kernel_0), 0, 0, 
> /bin/dmesg, , 0, 0x0, SYSCALL { ret = 64 }
> kernel.syscall_entry: 1727.121092890 (/tmp/trace1/kernel_6), 0, 0, 
> /usr/bin/lttctl, , 0, 0x0, SYSCALL { ip = 0xffffe430, syscall_id = 4 
> [sys_write+0x0/0x87] }
> kernel.syscall_entry: 1727.121094818 (/tmp/trace1/kernel_0), 0, 0, 
> /bin/dmesg, , 0, 0x0, SYSCALL { ip = 0xffffe430, syscall_id = 142 
> [sys_select+0x0/0x148] }
> fs.select: 1727.121098067 (/tmp/trace1/fs_0), 0, 0, /bin/dmesg, , 0, 0x0, 
> SYSCALL { fd = 3, timeout = -1 }
> End trace set
> 
> Traceset statistics:
> 
> statistics summed: 1
> mode_types:
>   MODE_UNKNOWN:
>     submodes:
>       UNKNOWN:
>         event_types:
>         events count: 0
>     events:
>     mode_types:
>       event_types:
>       events count: 0
> event_types:
> events count: 0
>   processes:
>     unknown process:
>       cpu:
>         0:
>           functions:
>             0x0:
>               mode_types:
>                 MODE_UNKNOWN:
>                   submodes:
>                     UNKNOWN:
>                       event_types:
>                       events count: 0
>                   events:
>                   mode_types:
>                     event_types:
>                     events count: 0
>               event_types:
>               events count: 0
>           mode_types:
>             MODE_UNKNOWN:
>               submodes:
>                 UNKNOWN:
>                   event_types:
>                   events count: 0
>               events:
>               mode_types:
>                 event_types:
>                 events count: 0
>           event_types:
>           events count: 0
> 
> ...
> 
> I originally tried lttng on sles 11 rc1 and rc2, but had this result. I am 
> surprised to see the same thing happening with the mainline kernel. So, 
> whatever it is that I am doing wrong, I seems to be repeating it 
> successfully each time.  What am I doing wrong?
> 
> Thanks in advance for the help.
> -------------------------------------------------
> Dany Madden
> IBM Linux Technology Center, MCP Team
> 15400 SW Koll Pkwy, Beaverton, OR 97006
> (503) 578-3086 (TL: 775-3086)
> danymadden at us.ibm.com
> _______________________________________________
> ltt-dev mailing list
> ltt-dev at lists.casi.polymtl.ca
> http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev


-- 
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68