On Mon, Mar 10, 2014 at 11:27:56AM -0400, Steven Rostedt wrote: > On Mon, 10 Mar 2014 16:17:48 +0100 > Frederic Weisbecker <fweis...@gmail.com> wrote: > > > > On both cases you want to use trace events. So I'd rather suggest you to > > use perf tools > > or trace-cmd. > > > > I believe that trace-cmd unfortunately uses per CPU periodic wake up on live > > recording CPUs so perhaps it's not a good candidate to test isolation. Too > > bad > > because kernelshark is cool for an overview. > > Actually, it's not per-cpu. There's a thread per-cpu, but they do not > need to run on the CPU they monitor. Heck, you can have all threads on > just one CPU. Actually, they are not threads, they are forked children, > but you get the idea.
Ah good to know. > > Now that I have wakeup working (with help of irq_work), I should also > be able to have them only wake if there's something to trace. Perfect! > > > > > Also perf supports some script languages for post processing: python, perl, > > etc... > > I think trace-cmd too but rather for custom display of events. > > Yeah, you can write python plugins that will change the display of the > output. I'm not sure about full integrating with python though. Ok. So to produce a summary of events with a script maybe perf can do better. But kernelshark can offer a good overview as well. So this all depends on what Mats wants to do in the first place: have a tool which tells you about the noise, which, how much, etc... So post analysis. Or have a more visual summary. Thanks. ------------------------------------------------------------------------------ Learn Graph Databases - Download FREE O'Reilly Book "Graph Databases" is the definitive new guide to graph databases and their applications. Written by three acclaimed leaders in the field, this first edition is now available. Download your free book today! http://p.sf.net/sfu/13534_NeoTech _______________________________________________ Ltp-list mailing list Ltp-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ltp-list
