[En-Nut-Discussion] Real-ly the right time? (was: Nut/OS goes Real-Time and USART on SPEED)

[Matthias Ringwald]

Hi Michael,

  Thanks for your opinions on what to expect from an embedded OS.
I'm quite in-line with your writing but like to clarifiy that
our discussion mainly focussed on two issues:
- improving USART speed ( 2 bytes fifo read, compile time IRQ  
registration, blocking read)
     without loosing generality
- reducing Critical Secions/Blocked IRQs which is a general benefit  
to all apps

> ..
> If I'm developing a application that requires a stable well  
> featured serial
> interface that can provide functionality with the smallest amount  
> of work
> overhead I would look at the OS and expect it to deliver just that.
>
> But what if my requirements take the hardware - and that is what we  
> are
> actually looking at in the moment - to the limit?
I'm stating that because of Critical sections and non-optimal USART  
handling,
we didn't reach the limit of the hardware and like to reach them.
Then, as you say, everything else should be custom-designed.

> So my credo is to keep Nut/OS clean - optimize it yes but don't try  
> to turn
> it into something that it can't be.
That's fine. The USART discussion is following that goal. My initial  
mail on a
much more responsive OS actually didn't receive much comments. Maybe  
lets boost
the USARt a bit without changing the rest of the system.

matthias