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

Re: [MiNT] DATE/TIME cookies

On Mon, Feb 22, 1999 at 05:44:11PM +0100, Konrad Kokoszkiewicz wrote:
> The point is, that even if the code exists, it can be executed only
> if someone calls Tgettime() (for example), and not _constantly_. Once
> per second? There is no interrupt in MiNT that occurs once per second,
> which means, that you would have to introduce new code to the 200 Hz
> timer. I know Afterburner is fast, but not yet everyone has a 68040.

There is no one-second-interrupt but a one-second-timeout (see timeout.c,
function checkalarms).  The timestamps have to be updated once a second
(actually only every 2nd second) because file system drivers (notably
the kernel's process and pipe file system) rely on that behavior.
Currently Petr's ideas wouldn't mean any performance loss.



What if everything is an illusion and nothing exists?  In that case, I
definitely overpaid for my carpet.
		-- Woody Allen, "Without Feathers"