Advertisement
Front, Kernel, Security, Distributions, Development. See your byline here on LWN.net.
| |||||||||||
Fixing interactive response in 2.6Fixing interactive response in 2.6Posted Jul 31, 2003 3:45 UTC (Thu) by Peter (guest, #1127)Parent article: Fixing interactive response in 2.6
His most recent patch is 011int. Everyone seems to make this mistake. It's O11int, not 011int. I don't know the significance of the big-oh (we're not doing complexity analysis, after all), but that's what Con has been calling these things.
(Log in to post comments)
Fixing interactive response in 2.6 Posted Jul 31, 2003 17:08 UTC (Thu) by cpeterso (guest, #305) [Link] I have no idea why Con is calling his patch series "-O#int". I guess int stands for interactive, but what about the O? He used to maintain a "-ck#" patch series and that naming scheme seems a lot simpler to me.
Big oh Posted Aug 1, 2003 16:23 UTC (Fri) by giraffedata (subscriber, #1954) [Link] I don't know the significance of the big-oh (we're not doing complexity analysis, after all)Yes, I believe we're doing complexity analysis. My understanding is that O(1) refers to the fact that the time it takes the scheduler to choose a process to run does not vary with the number of processes there are. By contrast, an O(n) scheduler might run down the list of processes looking for the best one to run.
Big oh Posted Aug 3, 2003 6:43 UTC (Sun) by Peter (guest, #1127) [Link] Yes, I believe we're doing complexity analysis. My understanding is that O(1) refers to Yes, I know what an O(1) scheduler is. (Sheesh, give me a little credit.) What I don't know is why an O11int scheduler would be called an O11int scheduler (other than the obvious fact that it came after the O10int scheduler). The big-oh here is clearly not referring to an asymptotic worst-case runtime. Ingo Molnar uses a combination of letter + number to designate different versions of his patches. Perhaps Con is doing the same thing. But, maybe I'm just not paying attention, but I never saw the A through N series of interactivity patches.
Fixing interactive response in 2.6 Posted Aug 4, 2003 23:28 UTC (Mon) by daniel (subscriber, #3181) [Link] "I don't know the significance of the big-oh (we're not doing complexity analysis, after all), but that's what Con has been calling these things."Though I never asked Con, it's evident that the O1 refers to the 2.5/6 scheduler, and "int" refers to addressing its interactivity shortcomings.
|
|||||||||||