Re: [PATCH] scheduler patch, faster still

Felix von Leitner (leitner@math.fu-berlin.de)
Wed, 30 Sep 1998 04:13:14 +0200


Thus spake Larry McVoy (lm@bitmover.com):
> Furthermore, others have argued that long RT run queues will become
> common as more applications become interactive. This is an argument
> that sounds somewhat plausible.

Not to me.
You need root privileges to use RT queues.
Most interactive programs are not run as root.

> The approach of "well, it looks good, let's integrate it" is what kills
> operating systems.

"it looks good" was not my only requirement.
It should also not add code bloat, make the code less readable or
make the common case slower.

Of course, implicitly I also want that the code does something good at
all, which has not been sufficiently established so far.

> To the extent that you care about Linux, you should fight that with
> all of your ability.

I do.

> Remember to step back and consider what would happen to the system if
> everything was changed just a little for some application.

His numbers did not look like it was just one application.
I got the impression that he made the common case faster.
And that is definitely something I want. ;)

Felix

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/