071230 New ck kernels for 5.9

[uelsk8s]

here they are:
http://vectorlinux.osuosl.org/veclinux-5.9/kernels/kernel-2.6.22.15ck2-i586-1vl59.tlz

Is there any proof that these ck-kernels have better performance that standard ones?
I tracked the discussion on the net about Con Kolivas patches and it seemed that benchmarks showed that ck-kernels do not work faster. Maybe I missed something?

I think it depends on what you consider faster. I think in benchmarks they may actually be slower, but what they seem to do is make apps load faster make multitasking seem faster let you do more things at once without mouse "jitters"
So you end up what a desktop that seems faster. Just ask anyone that has tried them

[Witek Mozga]

I think it depends on what you consider faster. I think in benchmarks they may actually be slower, but what they seem to do is make apps load faster make multitasking seem faster let you do more things at once without mouse "jitters"
So you end up what a desktop that seems faster. Just ask anyone that has tried them

I don`t know that`s why I`m asking. As far as I remember benchmarks were comparable for both kernels, none of them slower or faster very much.

Isn`t it ironic: in the linux world where everybody emphasize 'choice, choice, choice uber alles' when it came to make an alternative version of kernel, the discussion on the alternative was cut by kernel gurus and we ended up with 'one kernel to rule them all'. So maybe choice isn`t that good as everybody thinks afterall? Whatever...

[newt]

Regarding CK speed improvements: I certainly did not do any controlled testing/benchmarking of how my system performs under the two kernels (vanilla and ck2), but I did notice applications starting noticably faster.  If the difference had been negligible I would not say it was faster, but in my case it WAS noticably faster.  For instance, an application that normally takes 2 seconds to start was about half that (somewhere in that range).  Now I know we're talking miniscule differences here but if package compile time is affected as greatly then I stand to gain minutes, in some cases.  When it's all compounded it could add up to something much more substantial.  But, again, I did not to any kind of official benchmarking (but it would be fun to do and see the difference - I just wouldn't know how to procede).

Regarding wireless problems:  I have a linksys wpc54gs v1 (broadcom 4306 chipset) wireless PCMCIA card.  The first problem was the modules were not loaded but I got them loaded (modprobe bcm43xx).  I had already configured my wpa_supplicant.conf file and have it set to start at boot in the background/waiting (-Bw), so that was ready.  I had already configured wifi-radar with my wireless information.  I attempted to connect via wifi-radar and it showed a connection but I could not ping the network.  I downed/upped eth1 a couple times; I disconnected/connected wifi-radar a couple times; I tried pinging everything I could think of and nothing worked.  I booted back to vanilla and tried the same with the same results, at which point I remembered that I have to 'down' my eth0 card otherwise my system attempts to use that nic to make it's connection, and since it's not wired into anything it fails.  I'm sure if I would have booted back to the ck2 kernel and downed eth0 I would have had the same success; I just never got around to it.  I'll probably give it a shot tonight and will let you know the outcome.

Question: In vl58, if my wired nic was not connected to anything it would not be 'up' after boot but it vl59 it is; is that normal? and is it possible to make it act like it did in vl58?