Messages in this thread | | | From | Ronny Meeus <> | Date | Thu, 7 May 2015 22:25:19 +0200 | Subject | Re: Issue in PI boosting code in __sched_setscheduler |
| |
On Thu, May 7, 2015 at 10:22 PM, Steven Rostedt <rostedt@goodmis.org> wrote: > On Thu, 7 May 2015 22:17:48 +0200 > Ronny Meeus <ronny.meeus@gmail.com> wrote: > >> > View on what? >> >> Whether the real-time priority threads are used a lot. >> I have the impression that there are severe issues in the Linux eco >> system (Kernel+glibc) when it comes to real-time priority handling. >> > > It matters what you want from the thread. Real-time is about > determinism, not throughput. Most actions are about throughput, so it > would not surprise me that they do not make them RT. >
OK I understand. I will try to test the patch of Thomas tomorrow.
Thanks Steve and Thomas for solving this issue.
> -- Steve
| |