Re: a joint letter on low latency and Linux

[Larry McVoy]

> unfair to them. If you want to talk about what specifically is wrong
> with the BeOS or IRIX approach to offering 1-2ms worst-case latencies,

Sure.  Go measure the code paths inside those operating systems and
ask yourself why is it that the same system call on IRIX can take 10-15
times as long as the same system call on Linux.  And of course you'll
say, just as I've heard a zillion times before "my change can't even be
measured, it doesn't add a 15x overhead".  So tell me, why is it that
IRIX is so much slower?  Go look.  I have.  There is no one reason or
even 10 reasons.  There are hundreds of reasons, all of which "didn't
hurt at all" when measured in isolation.  Add them all up and they hurt.

> All we "need" is to lower this upper bound to, say, 5ms. Thats not
> hard real-time in any real sense. Anyone who *was* doing hard real
> time on such a kernel would be crazy. 

Really?  Sure sounds like hard real time to me.  It doesn't matter if
the number is 1ns, 1us, or 1ms, a limit is a limit.

trying to solve your one problem.  If you were really serious, you'd be
looking at doing 32 channels of audio and a couple of channels of video.
All of a sudden, all of your numbers don't work anymore and you need
more hackery in the kernel.

This is really the crucial point.  Suppose Linus lets in some of
the changes - he seems willing to do the ones that look reasonable.
That gets you most of the way there but not all of the way, and not
even most of the way when you try and use a Linux box as a real mixer.
Now what are you going to do?  You'll be forced into RT Linux anyway.