[ltt-dev] Problem on combine kernel trace and userspace trace.

[Mathieu Desnoyers]

* Tony Jones (tonyj at suse.de) wrote:
> On Fri, Dec 10, 2010 at 08:19:27AM -0500, Mathieu Desnoyers wrote:
> 
> > Meanwhile, it looks like you forgot to switch your UST from the generic
> > "clock_gettime MONOTONIC" clock source to rdtsc.
> 
> Sorry. I misunderstood the earlier message,
> 
> So, disabling the TRACE_CLOCK_GENERIC define in include/ust/clock.h now gets 
> me the following.  Scaling?

You might want to look at where the cpu frequency value is taken from
/proc/cpuinfo within UST. It it being called or is a 1 cycle = 1 ns
"generic" value used ?

Also, make sure your cpuinfo cpu frequency holds the right value for
your CPU rdtsc frequency.

But as I pointed out, all these problems will go away with the upcoming
vDSO.

Thanks!

Mathieu

> 
> tony
> 
> 
> # lttv -m textDump -t /tmp/traceK | head -3
> Trace set contains 1 traces
> 
> metadata.core_marker_id: 201.791156021 (/tmp/traceK/metadata_0), 0, 0, , , 0, 0x0, MODE_UNKNOWN { channel = "kernel", name = "page_fault_get_user_entry", event_id = 39, int = 4, long = 8, pointer = 8, size_t = 8, alignment = 0 }
> 
> # lttv -m textDump -t /tmp/traceU | head -3
> Trace set contains 1 traces
> 
> metadata.core_marker_id: 433.304628518 (/tmp/traceU/metadata_0), 0, 0, , , 0, 0x0, MODE_UNKNOWN { channel = "ust", name = "potential_exec", event_id = 0, int = 4, long = 8, pointer = 8, size_t = 8, alignment = 0 }
> 
> 
> 

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com