[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] Fpoll() in GEM programs



Hi,
Yeah, I love such answers.

That's why I wrote that you are the guru who might help here. I expected you could possibly help there.... not to argue. You can say 'it cannot work'. But I would expect because ... some arguments other than that I am not knowledgable enough, you know? To the OS levels mentioned by Phillip.... hmm AES was _not_ designed to be run in multiprocess/multiuser environment and that's why we got all those problems there. Anyway I consider an AES as a kernel level component to be able to provide filedescriptor driven IPC mechanism to an application whatever nature the client application is. And would not call it hack in my eyes.
Btw. where is the big advantage here now against an enhanced evnt_multi except that you need an additional system call to process AES events?

* no more event_multi() rich argument like functions
* already existing libc functions to handle signalling from another app Fselect()/Fpoll(). There are also signals there, right, but I would rather bring those to the discussion as they are not portable at all. Hey guys, ok, if this is not feasible and/or seems like a nonsense to most of you then simply forget it. regards STanda


Frank Naumann writes:
Hello!
You are the guru for open()/read()/write()/select()/close()/socket()/pipe()/... operations. Our intention was not to outline all the implementation details for sure.

But it's a problem that you propose something without thinking about all the consequences your proposal implies. It's not as trivial as it looks like from your point of view as AES program. There are a lot of side effects that need to be evaluated too (but if I ask here I just got the answer this is implementation detail, nothing important).
So the question for you is "Is there any way to get it like this?".

With or without the side effects you don't want to discuss?
discussed before) would be completely normal. The AES would write there the output in a form of a return value of event_multi() call. The client would be woken up from select() and would read the value. It would also call the event_multi_*() to get the rest information it needs.

Btw. where is the big advantage here now against an enhanced evnt_multi except that you need an additional system call to process AES events?
AES should be able to detect whether the client read from the pipe or not and if it is not beeing read than it should rewrite the value (I think this will need some more implementation details, but I think they are not necessary at this moment).

Yeah, I love such answers.

Regards,
Frank
--
ATARI FALCON 060 // MILAN 060
-----------------------------------------
http://www.cs.uni-magdeburg.de/~fnaumann/
e-Mail: fnaumann@freemint.de