[Info-vax] Volatile, was: Re: yet another sys$qiow question

[Johnny Billquist]

On 2015-08-29 03:08, David Froble wrote:
> JF Mezei wrote:
>> On 15-08-28 09:53, Stephen Hoffman wrote:
>>
>>> As for why?  Maybe lack of time, knowledge, staff, money or interest?
>>
>> When I dabbed in X, I found the need for a totally different programming
>> paradigm because of the "event loop" which is forced on me. Would have
>> much prefered if X events were delivered as ASTs.
>>
>> And when discussing with people who deal in real time, they are all
>> adament that it must all be an event loop, no ASTs allowed. So while I
>> never had to do this, I think it is possible that some types of apps
>> might actually do need to poll the IOSB as part of their real time event
>> loop. (if it is more efficient than checking event flags).
>
> Caviet!  I've never done a serious device driver.

Device drivers do not use ASTs. They use interrupts.

> Might some devices have their own processors for some of the on-board
> stuff?

Many do. Not that it is relevant from this point of view. A device do 
some work, and then indicate to the main CPU that it is done. Normally 
you have both some status register/information and also an interrupt.

> Tying up a modern CPU, or CPUs, with tight loops seems conta-indicated
> to me.

That is essentially done only under one circumstance, and that is for 
spin-locks. Sometimes still the best way of dealing with multiprocessor 
issues. But nothing else.

	Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol