[LITMUS^RT] high miss ratio for low task utilization

[Björn Brandenburg]

> On 21 Nov 2014, at 17:53, jp walters <mailinglistjp39 at gmail.com> wrote:
> 
> Ah, okay, I misunderstood a little.  I was starting rtspin with the -w option, then starting st_trace.  When I start st_trace first, then I get all of the appropriate names, costs, and periods.  So then I’m guessing that the issue I was having earlier is from the experiment scripts and not Litmus itself.

I'm glad the issue has been resolved. If you figure out what's happening with the other scripts, please post it here as well. I not familiar with them, but I'm sure others will run into the same issue eventually. 

>  Can you confirm that it's reasonable to be using the tools from:
> 
> https://github.com/brandenburg/sched-trace-tools
> 
> to pull the data out of the binary logs?  

Yes, those are the tools that I use. I guess I should stick in a little README explaining how to run them...

If you run into any other issues with LITMUS^RT we're happy to help, but as Glenn pointed out run LITMUS^RT 128 cores might be problematic at the moment.

- Björn