On Mon, 2006-03-20 at 10:00 -0500, jamal wrote: > I have to say i am scratching my head - now that i was forced to run the > tests - to see if there is infact a scenario where you could show 2.4 to > be better...
So that is the underlying reason you are resisting - you just can't see that it could be so. Sorry ... it has taken me this long to figure that out. Anyway the analysis has a real life example. > I cant access your page - it is timing out. Jeezz, that pisses me off. What is it with the bloody internet? This isn't the first time this has happened. The page you are accessing is in the US for gods sake. It seems like the internet has walled off islands on occasions. I have mirrored it: http://www.stuart.id.au/tc/hash-analysis http://www.lubemobile.com.au/tc/hash-analysis http://adsl-brisbane.lubemobile.com.au/tc/hash-analysis http://iexec-brisbane.lubemobile.com.au/tc/hash-analysis http://adsl2-sydney.lubemobile.com.au/tc/hash-analysis http://cable-meblourne.lubemobile.com.au/tc/hash-analysis http://adsl-perth.lubemobile.com.au/tc/hash-analysis One of these must work, (surely ?!?). All bar the first one may disappear within the next 24 hours due to cron jobs cleaning up. > The least squares is not bad - only issue is it missed the criteria > i had for looking at the number of buckets selected. I think if we had > no choice on changing the hash table, it would have been perfect. The source a couple of metrics is up on the web site. I find myself using the average lookup time over the least squares, even though I think on a theoretical level least squares is a better metric. This is because the average lookup time measures something "real". > Ok now, give it up Russell ; Yeah, I agree with the sentiment. I am tiring of this too. I have another patch I would like you to look at completely unrelated to u32, and definitely more important. The sticking point seems to be that you don't believe 2.4 can't run faster. It can, and it is in incredibly frustrating to me that I have not been able to put a coherent argument together to demonstrate it. Hopefully the analysis does that. Please look at it. Once I have managed to demonstrate to you that 2.4 can run faster I will accept your decision on the matter without a whimper - I promise. By the way, with the analysis I didn't go out of my way to find a dataset where 2.4 ran faster - it was just the second one I tried. There are pathological "fake" datasets that perform much worse than the one in the analysis, and presumably real ones too. - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
