[coyotos-dev] Hi + some stuff I found researching

[Haplo]

> Haplo:
>
> There is a place in the world for a restart mechanism. I'm not arguing
> that. What I'm saying is that the initial applications for Coyotos are
> applications where, if we need to invoke the restart layer, it's  
> already
> too late.

I see.

> Regrettably, it does *not* depend on the implementation. A channel  
> is a
> channel. You can only argue how big. Yes, the implementation has a lot
> to do with how big the channel is.

Similar to an FCRB with an unspecified return data size? It sounds  
vaguely like that, but if you mean to implement it in a way similar  
to synthesis, I can see how it'd be a problem. The data flow between  
applications in synthesis was entirely unprotected (and I can't think  
of a clear way to satisfyingly secure it either).

> ... Performance isn't the primary goal for Coyotos. Robustness is.

True, but I don't think I've ever heard anyone complain about how  
blazingly fast their computer is. On the other hand, the fact that  
this technique specifically holds no value for RISC (the majority of  
embedded systems plus some server/workstation/desktop systems) and  
there isn't any real opportunity to use it in coyotos even on systems  
where it MIGHT work, then it's really not worth the trouble. I mainly  
suggested it in the hopes that there might be some room for  
opportunistic optimization with it, but that turns out not to be the  
case.

> Litmus test question: if you implemented an algorithm like this, how
> soon would you run it on your (personal) defibrilator?

Well, assuming that it ran a general purpose KOS like coyotos, I  
still don't have a personal defibrillator.
Hypothetically, however, most likely never. In a single-purpose  
system like that most kernel paths would be entirely unused, and I  
can't think of anything in such a system
which would utilize that. To be fair, though, coyotos IS a general  
purpose OS kernel, so there will be features exclusively used by one  
application or the other.