On 29/06/18 15:56, Stephen Pelc wrote:
> On Fri, 29 Jun 2018 14:16:45 GMT, Pabst Blue Ribbon
>  wrote:
>
>> They doing it in an attempt to catch up with languages such as C and C++ in
>> terms of speed. Java, for example, have no problems with byte code (which,
>> in essence, just another name for threaded code.) If  I understand
>> correctly, it still should be possible to use code optimizations with
>> threaded code or even generate mixed threaded and native code
>> automatically.
>
> What would the benefits be of mixing threaded and native code?
>
> Especially in embedded systems, speed is always an advantage.
> Among other things, efficient code generation reduces power
> consumption. For the last 15 years or so, I have not written
> an ARM/Cortex interrupt handler in assembler - I just let
> the compiler do its job. For reasons like this, our TCP/IP
> stack contains no assembler (unless you want it) and runs
> on a range of CPU architectures.
>

So how does ARM handle peripheral hardware?

On INTEL you MUST have some assembler in a library at least to access
I/O registers, because I/O is not part of Cs understanding of architecture..



> Stephen
>


--
"The great thing about Glasgow is that if there's a nuclear attack it'll
look exactly the same afterwards."

Billy Connolly

--- SoupGate-Win32 v1.05