Messages in this thread |  | | | Date | Mon, 4 Jun 2007 16:17:18 +0200 | | From | Ingo Molnar <> | | Subject | Re: [bug] very high non-preempt latency in context_struct_compute_av() |
| |
* James Morris <jmorris@namei.org> wrote:
> > the latency is caused by a _very_ long loop in the SELinux code: > > > > sshd-4828 0.N.. 465894us : avtab_search_node (context_struct_compute_av) > > What do the 0DNs fields mean and what did you use to create this > trace?
i used the latency tracer from -rt. Here's the meaning of the fields:
_------=> CPU# / _-----=> irqs-off | / _----=> need-resched || / _---=> hardirq/softirq ||| / _--=> preempt-depth |||| / ||||| delay cmd pid ||||| time | caller \ / ||||| \ | / trace-it-4751 0D... 0us : __next_cpu (user_trace_start) it's very easy to interpret: it traces all the function calls the kernel executes, and puts the symbolic function name (and its parent function) into the trace, time ordered. So it's a proper execution trace. Normally you can ignore the 'DN' type of flags - what matters in this case is the observed 130 msecs latency and the functions that were called while that latency happened.
Ingo - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |