Package Details: linux-ck 6.12.1-1

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: 0.24
First Submitted: 2011-07-22 14:51 (UTC)
Last Updated: 2024-11-23 13:58 (UTC)

Dependencies (14)

Required by (6)

Sources (6)

Latest Comments

« First ‹ Previous 1 .. 258 259 260 261 262 263 264 265 266 267 268 .. 308 Next › Last »

graysky commented on 2012-05-14 21:09 (UTC)

Updated v3.3.6-1 Changelog: Removed optional uksm since there is already a linux-ck-uksm package which is maintained by a member of the uksm dev team. No point in me duplicating this in linux-ck. PKG Commit: http://pkgbuild.com/git/aur-mirror.git/commit/linux-ck?id=c2412cde953b772b20fee22ea84a2a62e24d2916

Lawliet commented on 2012-05-14 02:18 (UTC)

@graysk: not different from the linux-ck, but the linux-uksm-ck package main major is UKSM patch, and defalut enable. we are UKSM dev and support team, so the package unlike linux-ck, it is BFS main major, not UKSM. and UKSM have new patch "0002" the fix libsandbox crash with libc at Gentoo(maybe some case will happen with Arch..) http://kerneldedup.org/download/uksm/0.1.1.1/patches/0002-UKSM-add-VM_MERGEABLE-to-vm_flags-before-vma_merge.patch

dkaylor commented on 2012-05-13 18:57 (UTC)

Actually, I think you've earned our trust in terms of the repo packages and the PKGBUILD on AUR, just to clarify. There's something about UKSM that makes me nervous....

dkaylor commented on 2012-05-13 18:54 (UTC)

@graksky - I get your point, and I trust your judgement in regards to the repo package. But I would urge real caution when it comes to UKSM, atleast for now.

<deleted-account> commented on 2012-05-13 17:53 (UTC)

FWIW I did try this for a week and noticed _NO_ improvement in mem usage but the 'uksm' process was always in the top 5 cpu usage, almost constantly running at 2% - 5%. Just my unscientific 2 pence.

graysky commented on 2012-05-13 17:29 (UTC)

No worries, the default is NOT compiled in as you can see in the PKGBUILD. As I have previously stated, the repo packages will NOT get this due to the former. Arch is about choice, so go ahead and enable and compile is you wish.

dkaylor commented on 2012-05-13 16:34 (UTC)

"I don't know enough about that patch to know if it is 100 % dormant without `echo 1 >/sys/kernel/mm/uksm/run` since the page and forums are in Chinese." And therein lies a big problem. That, combined with the responses on LKML almost 15 months ago, combined with the lack of follow up by xianai here or on LKML since then makes me think it's not worth the maintenance hassle or the risk. There are those, me formerly being one of them, who will assume "hey BFS and BFQ are performance related (in the sense of interactive responsiveness on a contemporary desktop) so UKSM must be also, therefore I will compile it in". And if graysky himself can't really understand the interactions, due to no documentation in the dominant kernel development language, then I wouldn't touch it, not even as an option.

graysky commented on 2012-05-13 10:49 (UTC)

@kyak - I don't know enough about that patch to know if it is 100 % dormant without `echo 1 >/sys/kernel/mm/uksm/run` since the page and forums are in Chinese. I therefore do not want to force it on all the repo users. BFQ by comparison is 100 % dormant unless the user chooses to active it. Can you point me to documentation?

« First ‹ Previous 1 .. 258 259 260 261 262 263 264 265 266 267 268 .. 308 Next › Last »