Tell me more ×
Unix & Linux Stack Exchange is a question and answer site for users of Linux, FreeBSD and other Un*x-like operating systems.. It's 100% free, no registration required.
up vote 3 down vote favorite

Does the PREEMPT_RT patch (real-time kernel) have any benefit for regular desktop users?

share|improve this question

2 Answers

up vote 2 down vote accepted

I don't think so. The patch seems to provide real-time scheduling which is very important for some enviroments (planes, nuclear reactors etc.) but overkill for regular desktop. The current kernels however seems to be enough "real-time" and "preemptive" for regular desktop users[1].

It may be useful if you work with high quality audio recording and playing in which even small amount of time may dramatically reduce the quality.

[1] Technically both are 0/1 features but I guess it is clear what I mean ;)

share|improve this answer
can we change kernel scheduler on the fly, just like cpu_governor? – uray Sep 7 '10 at 12:51
you cannot change kernel scheduler like CFS or BFS. However you CAN change scheduling properties by schedtool on the fly (priority, make process as interactive/batch etc.) – Maciej Piechotka Sep 7 '10 at 17:30
up vote 2 down vote

I'm guessing you misunderstand the concept of 'real-timeness'. If not, sorry, but it happens a lot, and I thought I'd throw a little clarification in here.

The main point of a real-time kernel is to serve requests within a predictable deadline. That does not necessarily mean faster than a 'normal' kernel. So for desktop systems, a preemptive kernel is good, a real-time kernel much less so.

share|improve this answer
I do understand a little bit about real-time application, and I did developed soft-realtime applications for embedded systems, and I understood that it won't give "faster" experience, but in theory it should give more "responsive" user interactions, isn't that true? – uray Sep 13 '10 at 15:36
Not necessarily. As far as I understand it, it will give you applications with a predictable response time. – wzzrd Sep 13 '10 at 19:06
And since the total processing power of a CPU is constant (simplifying a bit), and context switches cost CPU power, it's likely that some things could become slower (but still responsive to a certain response time). – Alexios Jun 7 '12 at 9:12

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.