Package Details: linux-ck 6.11.10-1

Git Clone URL: https://aur.archlinux.org/linux-ck.git (read-only, click to copy)
Package Base: linux-ck
Description: The Linux kernel and modules with ck's hrtimer patches
Upstream URL: https://wiki.archlinux.org/index.php/Linux-ck
Licenses: GPL-2.0-only
Provides: KSMBD-MODULE, VIRTUALBOX-GUEST-MODULES, WIREGUARD-MODULE
Replaces: virtualbox-guest-modules-arch, wireguard-arch
Submitter: graysky
Maintainer: graysky
Last Packager: graysky
Votes: 459
Popularity: 0.22
First Submitted: 2011-07-22 14:51 (UTC)
Last Updated: 2024-11-26 13:15 (UTC)

Dependencies (14)

Required by (7)

Sources (6)

Latest Comments

« First ‹ Previous 1 .. 107 108 109 110 111 112 113 114 115 116 117 .. 308 Next › Last »

pedrogabriel commented on 2016-10-04 22:00 (UTC)

MuQSS 105 is out now.

pedrogabriel commented on 2016-10-04 13:27 (UTC)

The kernel don't recognize my fans after compile it with MuQSS.

graysky commented on 2016-10-03 18:36 (UTC) (edited on 2016-10-03 20:08 (UTC) by graysky)

@vishwin - Sorry, but not out of date until BFQ for 4.8 is released. I do not see MuQSS104 that patches against the 4.7 code-base (although CK is working on 105 as we speak)... plus, 4.8-ARCH is probably still in the works as it hasn't been build nor hit [testing] just yet.

artafinde commented on 2016-10-03 13:45 (UTC) (edited on 2016-10-03 13:48 (UTC) by artafinde)

I actually won't be enabling this unless it turns out to be stable enough and give something to the <= 8 CPUs. I've enabled it on my laptop and screwed up the BTRFS ctree. Seems like I'm reformatting today. I can't really blame MuQSS since there's a big WARNING on top, but it's a warning to users here. Could be something else but all started after I compiled and booted the MuQSS enabled kernel. The combination was: CPU Haswell, BTRFS single metadata SSD 1 device, BFQ enabled by default, localmodules enabled and MuQSS.

kogone commented on 2016-10-03 13:40 (UTC)

maybe we should create a rubric of tests to run to see the difference of muqss running on cpus with <= 8 cores?

artafinde commented on 2016-10-02 09:26 (UTC)

According to the blog post CK mentions only CPU not cores. I'm not entirely sure how the runqueues are defined (per CPU or per Core) but my understanding from the post is to target servers and make the BFS more comparable to mainline CFS.

zerophase commented on 2016-10-02 09:05 (UTC)

@artafinde Does MuQSS benefit more than 1 cpu die, or multiple cores? Say 8+?

artafinde commented on 2016-10-02 08:58 (UTC)

MuQSS is really targeting multi CPUs. As per the comment from CK "it will make a difference in 16+ CPUs and with high loads". I suppose we can still use it for testing purposes but unlikely we will produce any lockups to help surface any bugs with the desktop/laptop PCs. 103 is out btw.

graysky commented on 2016-10-01 20:05 (UTC)

@SuperIce - Thanks updated to 102 still in 4.7.6-1.

« First ‹ Previous 1 .. 107 108 109 110 111 112 113 114 115 116 117 .. 308 Next › Last »