Skip to contentSkip to conversation

Scheduler issue 1, RT tasks ...

RFC Martin Knoblauch 2 participants 2 messages Dec 28, 2001 – Dec 29, 2001

lore
Share & tools
lore
#0 lore raw
Quoted 8 lines

Re: [RFC] Scheduler issue 1, RT tasks ...

Quoted 4 lines

Right, that was my question. George says, in your words, "for better

standards compliancy ..." and I want to know why you guys think

that.

Quoted 8 lines

The thought was that if someone need RT tasks he probably need a very low
latency and so the idea that by applying global preemption decisions would
lead to a better compliancy. But i'll be happy to ear that this is false
anyway ...

without wanting to start a RT flame-fest, what do people really want
when they talk about RT in this [Linux] context:

- very low latency
- deterministic latency ("never to exceed")
- both
- something completely different

Thanks
Martin

Signature · 7 lines
+-----------------------------------------------------+
|Martin Knoblauch                                     |
|-----------------------------------------------------|
|http://www.knobisoft.de/cats                         |
|-----------------------------------------------------|
|e-mail: [email protected]                           |
+-----------------------------------------------------+
#1 lore raw
Quoted 18 lines · Martin Knoblauch
Quoted 9 lines

Re: [RFC] Scheduler issue 1, RT tasks ...

Quoted 4 lines

Right, that was my question. George says, in your words, "for better

standards compliancy ..." and I want to know why you guys think

that.

Quoted 8 lines

The thought was that if someone need RT tasks he probably need a very low
latency and so the idea that by applying global preemption decisions would
lead to a better compliancy. But i'll be happy to ear that this is false
anyway ...

Quoted 8 lines

without wanting to start a RT flame-fest, what do people really want
when they talk about RT in this [Linux] context:

- very low latency
- deterministic latency ("never to exceed")
- both
- something completely different

All of the above from time to time and user to user. That is, some folks want one or more of the above, some folks want more, some less. What is really up? Well they have a job to do that requires certain things. Different jobs require different capabilities. It is hard to say that any given system will do a reasonably complex job with out testing. For example we may have the required latency but find the system fails because, to get the latency, we preempted another task that was (and so still is) in the middle of updating something we need to complete the job.

On the other hand, some things clearly are in the way of doing some real time tasks in a timely fashion. Among these things are long context switch latency, high kernel overhead, and low resolution time keeping/ alarms. So we talk (argue? posture?) most about these. At the same time all the other bullet items of *nix systems are, at least some times, important.

Why Linux? The same reasons it is used any where else. Among these reasons is the desire to have to know and support only one system. Thus the drive to extend it to the more responsive end of the spectrum without loosing other capabilities. And, of course, the standards issue is in here. Standards compliance is important from an investment point of view. It allows the user to move his costly (far more than the hardware) software investment from one kernel/ system to another with little or no rework.

Signature · 3 lines
George           [email protected]
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Real time sched: http://sourceforge.net/projects/rtsched/