Re: My 66000 and High word facility

On Mon, 19 Aug 2024 18:22:27 +0000
mitchalsup@aol.com (MitchAlsup1) wrote:

On Mon, 19 Aug 2024 16:05:22 +0000, Stefan Monnier wrote:
 

Task swapping time is way down in the noise. It’s reloading the L1
and L2
cache that swamps the time. 64 registers is nothing compared to
32k or megabytes. 

>
Depends on the kind of swap.  If you're thinking of time-sharing
preemption, then indeed context switch time is not important. 

 

But when considering communication between processes, then very fast
context switch times allow for finer grain divisions, like
micro-kernels. 

 
MicroKernels failed due to the excessive overhead of context
switching. Whether is was control delivery delay, TLB reloads, Cache
reloads, register file loads and stores, ... it doesn't really mater
as each delay adds up. When there is too much delay the system is
sluggish and unacceptable en-the-large.
>

I don't believe that failure of uKernels to take over the world of
OSes is related to the factors, you mentioned.
It failed because relatively to monolithic kernel it is less convenient
way to structure the OS software. Various parts of the OS are
more dependent on each other logically, esp. in read-only manner, than
proponents of uKernels are admitting. Every change takes more
developer's time and causes touching more places in code than with
monolithic.

Historically, these things have never really materialized,
admittedly. 

 
Pigs don't win the 100 yard dash at the Olympics, either.
 

>
        Stefan