I am using BFS for few days (zen kernel -rc7/8-zen1 - not AUR though but git), better system responsiveness under load than CFS,
zen has some extra perks (Sched_iso X - for X server) and one can select both BFS and BFQ for nice/experimantal CPU and I/O scheduling.This is still experimental so while .31-rc7-zen1 works without any issues with .31-rc8-zen1, I lost suspend to RAM. Still this is a small price for improved responsiveness.
I hope that Con Kolivas will stay and we will have again good responsive desktop linux kernel.
you are quite behind zen: recent stable is 2.6.32-zen5. It may depend on your config too but suspend to RAM works for me (HP dv5000)
]]>Now, I know, why it is faster than my other kernel.
]]>I am using BFS for few days (zen kernel -rc7/8-zen1 - not AUR though but git), better system responsiveness under load than CFS, zen has some extra perks (Sched_iso X - for X server)
anyone interested, patch is here: http://ck.kolivas.org/patches/bfs/autoiso-xorg.patch
]]>[OT]
Primoz wrote:A bit off-topic but, does anyone know where to get stock rc.sysinit so I can compare and resolve this problem?
if you're using [testing] : http://projects.archlinux.org/initscrip … =2009.11-1
if you're using [current] : http://projects.archlinux.org/initscrip … =2009.08-1[/OT]
Thanks!
]]>A bit off-topic but, does anyone know where to get stock rc.sysinit so I can compare and resolve this problem?
if you're using [testing] : http://projects.archlinux.org/initscrip … =2009.11-1
if you're using [current] : http://projects.archlinux.org/initscrip … =2009.08-1
[/OT]
]]>A bit off-topic but, does anyone know where to get stock rc.sysinit so I can compare and resolve this problem?
]]>Also the most of the changes aren't really measurable, if it helps at all, it just makes it feel better.
]]>definitely not placebo, there have been quite a few benchmarks on LKML demonstrating BFS efficiency. even without benchmarking, i can easily witness this by playing 1080p videos with a 965GM, which is a near-death-experience with cfs.
I misread that as ck
]]>definitely not placebo, there have been quite a few benchmarks on LKML demonstrating BFS efficiency. even without benchmarking, i can easily witness this by playing 1080p videos with a 965GM, which is a near-death-experience with cfs.
I have seen benchmarks, but i really wasn't impressed. It seems for some people it works, but on my 2 machines it wasn't an improvement.
]]>Primoz wrote:After so much trouble I'm saddened to say that my ck kernel is worse than normal arch kernel...
I don't know what exactly it is, I suspect the scheduler or something. As it caches more memory and it also uses more CPU...
But I think this was done by me in kernel configuration...Those things are good. What are you really complaining about? More RAM in use = good. Is your computer more or less responsive with the ck kernel? That's what it's about.
You're right. But I don't see any big difference. It's not any faster... I think it's the problem that bangkok_manuel described...
Also with all that it really doesn't feel anything more responsive, if anything it feels slower... (But I have to make a real test...)
After so much trouble I'm saddened to say that my ck kernel is worse than normal arch kernel...
I don't know what exactly it is, I suspect the scheduler or something. As it caches more memory and it also uses more CPU...
But I think this was done by me in kernel configuration...
Those things are good. What are you really complaining about? More RAM in use = good. Is your computer more or less responsive with the ck kernel? That's what it's about.
]]>