The Linux Desktop Responsiveness Patches Are Feeling Good

[damentz]

Damentz (he's one of the zen devs) has some experience with backporting patches so he might take a crack at backporting all 7 patches to 2.6.35 (and maybe 2.6.34)

I backported them two days ago -> http://git.zen-kernel.org/?p=kernel/...eads/mm-2.6.34

You'll want to merge this branch into your 2.6.34 kernel tree.

[DeepDayze]

I backported them two days ago -> http://git.zen-kernel.org/?p=kernel/...eads/mm-2.6.34

You'll want to merge this branch into your 2.6.34 kernel tree.

Great!

You put them all into your latest 2.6.35 kernel right?

[damentz]

Great!

You put them all into your latest 2.6.35 kernel right?

Yes, 2.6.35 already had the necessary bits, no extra backporting is required.

[RealNC]

Where are the patches for 2.6.35?

[rewind]

Are you for real?
You don't protect your machine by setting the right limits and you expect, what? That the kernel will magically transform your ultra-slow HD into a RAM like speed-daemon? Maybe it should kill your process with an EIDIOTPROTECTION error? Come-on!

- Gilboa

I meant that swapping always makes the problem with responsiveness to come up. If you open a file larger than the RAM /~10 GB *.txt/ with let's say UltraEdit /found it in the Ubuntu Software Center/ the OS will not crawl, because there will be no excessive swapping. Actually I had more than 2 gb spare to start a virtualbox and try the changes to the file.
Every intense use of the hard disks makes it visible. I have three drives and I always try to balance the load when large amounts of data have to be transferred or extracted or etc. and even then it's perceivable.
Greetings.

[DeepDayze]

I meant that swapping always makes the problem with responsiveness to come up. If you open a file larger than the RAM /~10 GB *.txt/ with let's say UltraEdit /found it in the Ubuntu Software Center/ the OS will not crawl, because there will be no excessive swapping. Actually I had more than 2 gb spare to start a virtualbox and try the changes to the file.
Every intense use of the hard disks makes it visible. I have three drives and I always try to balance the load when large amounts of data have to be transferred or extracted or etc. and even then it's perceivable.
Greetings.

Obviously that the system will start swapping when you open something that takes up more than the physical RAM, but with these patches the swapping is supposed to be done in a way that does not excessively slow down the system such as swapping out things that are not actively in use.
For example if your web browser is active, the system won't try to swap it out

[kernelOfTruth]

Obviously that the system will start swapping when you open something that takes up more than the physical RAM, but with these patches the swapping is supposed to be done in a way that does not excessively slow down the system such as swapping out things that are not actively in use.
For example if your web browser is active, the system won't try to swap it out

even if you don't use Con's BFS cpu scheduler, give his ck1 patchset on top of 2.6.35 a try !

combine them with this patch and you'll have a way more responsive system

if you want to improve on top of that: use BFQ (included in the zen-kernel) and BFS

[RealNC]

BFQ makes the problem worse for me.

[rewind]

Obviously that the system will start swapping when you open something that takes up more than the physical RAM, but with these patches the swapping is supposed to be done in a way that does not excessively slow down the system such as swapping out things that are not actively in use.
For example if your web browser is active, the system won't try to swap it out

Which is just brilliant! I'll definitely gonna try them. Thanks for the explanation.

[RealNC]

I'm still looking for the full set of those patches for 2.6.35. Where are they?