Package Details: linux-pf 6.11.pf4-1

Git Clone URL: https://aur.archlinux.org/linux-pf.git (read-only, click to copy)
Package Base: linux-pf
Description: The pf-kernel and modules
Upstream URL: https://pfkernel.natalenko.name
Keywords: bbr bbr3 kernel ksm linux linux-pf pf-kernel uksm uksmd v4l2loopback zstd
Licenses: GPL-2.0-only
Provides: KSMBD-MODULE, linux-pf, NTFS3-MODULE, UKSMD-BUILTIN, V4L2LOOPBACK-MODULE, VIRTUALBOX-GUEST-MODULES, WIREGUARD-MODULE
Replaces: virtualbox-guest-modules-arch, wireguard-arch
Submitter: nous
Maintainer: post-factum
Last Packager: post-factum
Votes: 209
Popularity: 0.22
First Submitted: 2011-07-24 12:01 (UTC)
Last Updated: 2024-11-06 15:49 (UTC)

Dependencies (17)

Required by (14)

Sources (2)

Pinned Comments

post-factum commented on 2023-09-25 20:30 (UTC) (edited on 2024-10-08 14:21 (UTC) by post-factum)

Official binary builds for various x86_64 ISA levels are available here.

Latest Comments

« First ‹ Previous 1 .. 44 45 46 47 48 49 50 51 52 53 54 .. 106 Next › Last »

jnbek commented on 2012-12-31 15:21 (UTC)

yea, I been meaning to mention this for a few days, and keep forgetting for one reason or a thousand. I have yet to be prompted by the PKGBUILD if I want to set _SUBLEVEL=y or not.. I go through the questions, and it just goes straight to the build... defaulting to no sublevel... Not sire what the issue is, I've just gone and set it to Y manually in the PKGBUILD to accomodate my sometimes seedy desires. But now that I remember, I can now live my life to it's fullest knowing that I have finally reported my _SUBLEVEL issue here :D

jahiy commented on 2012-12-31 11:59 (UTC)

Please set _SUBLEVEL to "y"(default), or nvidia cannot be complied correctly without SUBLEVEL!!

nous commented on 2012-12-22 11:52 (UTC)

"Core2 is one that the kernel devs include in the source." Thank you.

graysky commented on 2012-12-22 11:01 (UTC)

@nous - Core2 vs generic was only 88 ms faster. Core2 is one that the kernel devs include in the source. If you are rejecting the others based on your argument about practical significance, then why are you offering any optimized builds? Based on these data, they all offer, in you option, no practical significance. Do you see the disconnect here?

nous commented on 2012-12-22 10:32 (UTC)

@Thaodan: The gcc-4.7 patch is there, the optimizations are enabled but I won't pass them onto the package name beyond the in-kernel maximum (i.e. core2 and k8). To put an end to this, there's already provision for extreme optimization enthusiasts in the PKGBUILD: uncomment the 'export KCFLAGS="-march=native -Ofast"' line just before make bzImage and one gets the fastest kernel in the universe one's CPU can run (I do, for my core2 laptop). That been said, I won't add optimizations into the PKGBUILD other than those from the upstream kernel. @graysky: The same obvious fact is stated in lkml; statistically significant does not mean practically significant. As I've written elsewhere, I believe the kernel developers know better than me when to include a CPU optimization into the kernel configuration. Please, if it's not a great hassle, run a comparison between core2 and avx-i kernels. We've spent more time on this topic than all our CPU's combined would've saved running coreavx kernels for several thousand years.

Thaodan commented on 2012-12-21 23:51 (UTC)

BTW: nvidia-pf detecs all builds that I added fine without any issues.

Thaodan commented on 2012-12-21 23:50 (UTC)

nous: I don't say build all aviable optimisations, I only say inclue it in the PKGBUILD nothing more.

jnbek commented on 2012-12-21 15:01 (UTC)

egads, aur spam!!! lol, woke up this morning to an inbox full of notices from this pkg... :D Love a well used package.

graysky commented on 2012-12-21 14:25 (UTC)

@nous - You're missing the point. Upstream already includes optimization options in the nconfig that give the same order of decrease (hundreds of milliseconds) vs. generic. The options the patch introduces just expands the existing menu. The real point is that there are statistically significant gains by using them -- at least on the machines I looked at... for more, see https://lkml.org/lkml/2012/12/9/47 and the reply.

« First ‹ Previous 1 .. 44 45 46 47 48 49 50 51 52 53 54 .. 106 Next › Last »