Package Details: linux-ck 6.11-1

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: 0.86
First Submitted: 2011-07-22 14:51 (UTC)
Last Updated: 2024-09-19 13:04 (UTC)

Dependencies (14)

Required by (6)

Sources (6)

Latest Comments

« First ‹ Previous 1 .. 221 222 223 224 225 226 227 228 229 230 231 .. 307 Next › Last »

crondog commented on 2013-03-04 10:02 (UTC)

3.8.2 is now available :)

graysky commented on 2013-03-04 08:49 (UTC)

OK.. I understand now: the path changed. Thank you crondog. Updated but not bumping the pkgver. For others - Plz don't flag out-of-date if a source path changes; it makes me think I missed something upstream or with ARCH.

crondog commented on 2013-03-04 01:49 (UTC)

Just to let you know the ck patch is located at http://ck.kolivas.org/patches/3.0/3.8/3.8-ck1/patch-3.8-ck1.bz2

graysky commented on 2013-03-03 19:19 (UTC)

@Misc - BFS is still superior using a non-latency endpoint, my make-based benchmark tested now on two machines: i7-3770K @ 4.5 GHz (9 threads) i7-2620M @ 2.7 GHz (5 threads) Conclusion: BFS v0.428 _clearly_ differentiates itself from CFS in mainline 3.8.1. Results on machine #1: http://s19.postimage.org/jkil3zx83/3770_K.png Results on machine #2: http://s19.postimage.org/uy54f77qr/2620_M.png These analyses are an ANOVA showing you the time it took to compile (y-axis in sec) each of the 9 runs under each kernel (x-axis). Black dots = time points. Green diamonds = 95 % confidence intervals. The circles to the right are comparison circles. If these are separated, the differences in average times are statistically significant. In other words, BFS is still faster than CFS. This is true for both machines tested. Note that I will expand this to a few other machines in the next few days. Code to benchmark script is 'make_bench' and can be found on my github: https://github.com/graysky2/bin The benchmark simply runs `make -jx bzImage` 9 times and logs the total time to build to a log file for analysis. This the same analysis I did in my study of them a few months ago which is available here: http://repo-ck.com/bench/cpu_schedulers_compared.pdf P.S. Sorry for multiple posts. Cannot edit as you know.

misc commented on 2013-03-03 18:54 (UTC)

Given the improvements to CFS in 3.8, it would be interesting to see how BFS compares to it now.

graysky commented on 2013-03-03 16:40 (UTC)

Bump to v3.8.1-1 LinuxChanges: http://kernelnewbies.org/LinuxChanges_3.8 Changelog: https://lkml.org/lkml/2013/2/28/219 Commit: http://pkgbuild.com/git/aur-mirror.git/commit/linux-ck?id=96b741a6eb56b2a5c2d9dc89cf28181424c0e1a8

graysky commented on 2013-03-01 20:42 (UTC)

@misc - Nothing was actually wrong with the patch, I mistakenly changed the source in the PKGBUILD to point to the patch for the 3.7 tree, not the 3.8 tree. This has been fixed and works with both arches. @radioact - No need, I have done this in the -0 release of linux-ck-3.8. See my comment below (2013-02-25 20:45) for the announcement and link. Still no update from CK by the way.

« First ‹ Previous 1 .. 221 222 223 224 225 226 227 228 229 230 231 .. 307 Next › Last »