Messages in this thread | | | Date | Tue, 29 Sep 1998 11:00:47 +0100 | From | Jamie Lokier <> | Subject | Re: [PATCH] scheduler patch, faster still |
| |
Rik van Riel wrote: > Right now, we have an overly long code path because > each action done checks for RT-ness of the task at > hand ('harmless' actions are performed regardless).
Aha! A good point.
Felix von Leitner wrote: > I don't care if it fixes his problem or not. If it looks sensible and > actually speeds something up without slowing the common case down, > it's a good idea to integrate it. [...] > Make the common case fast. [...] > If the effects of his optimization is not measurable for typical > workloads, who cares? As long as it does not make the code obfuscated > or bloated or slower in the common case, it doesn't hurt.
Actually, the separate run queue may make the common case *faster*, because then there's no need for the all the RT-specific tests, branches etc.
-- Jamie
- 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/
| |