RE: [MiNT] DATE/TIME cookies

["Julian Reschke" <reschke@medicaldataservice.de>]

> From: owner-mint@fishpool.com [mailto:owner-mint@fishpool.com]On Behalf
> Of Petr Stehlik
> Sent: Friday, February 26, 1999 4:51 PM
> To: Julian Reschke
> Cc: mint
> Subject: RE: [MiNT] DATE/TIME cookies
>
>
> > Either the DATE/TIME cookie contains the date and time values
> > themselves. In
> > this case they need to be updated regularly by somebody -- either
> > the kernel
> > or a background process. This does not break memory protection, but is a
> > waste of time, especially if the internal time format is not DOS based
> > anymore.
>
> Yes, this is what I mean because I understood that introducing
> anything that
> would break memory protection is not possible.
> As for the waste of time, Guido said it currently wouldn't mean any
> performance loss. I don't know the date/time code well so I can't judge. I
> believe Guido is right.

This is correct. But hopefully you agree that one should not introduce
services into the kernel which will BECOME costly once an already planned
change has been made?

> > how do you access the screen from within the interrupt?
>
> it seems you're trying to convince me the whole corner clock TSR is a BAD
> IDEA ;-)

Correct.

> OK, so here's the ultimate answer: I do access the screen by
> 'move.b (a0)+,
> (a1)+'. Surprised? :-)
>
> BTW, if this thread is going to be even more about Clocky's internals I
> suggest to move off the MiNT list, to either private mails or to a Clocky
> mailing list I would try to get created ;-)

OK, let me summarize: your clock TSR accesses the screen in a dirty way,
because it runs from the interrupt. It also likes to retrieve the system way
by a new (for MiNT) method, because it runs from within an interrupt. Why
don't you just fix your program instead of trying to get new features into
the kernel?

Regards, Julian