You are not logged in.
If you're using [repo-ck] you need the corresponding skylake headers package. If you're booted into the ck kernel, the compiled modules should just work. Do you have an existing version already loaded? Either reboot or rmmod it.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
graysky, you should consider fixing the server of http://repo-ck.com. The server interrupts every time the download of the kernel file.
I get the same error as cowlick:
:: Retrieving packages...
linux-ck-kx-4.8.15-1... 7,8 MiB 10,6M/s 00:01 [###########################] 12%
(56/56) checking keys in keyring [###########################] 100%
(56/56) checking package integrity [###########################] 100%
error: linux-ck-kx: signature from "graysky (used to sign repo-ck packages) <graysky@archlinux.us>" is invalid
:: File /var/cache/pacman/pkg/linux-ck-kx-4.8.15-1-x86_64.pkg.tar.xz is corrupted (invalid or corrupted package (PGP signature)).
Do you want to delete it? [Y/n] n
error: failed to commit transaction (invalid or corrupted package)
Errors occurred, no packages were upgraded.
Offline
@indianahorst - https://wiki.archlinux.org/index.php/Re … _regularly
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
What would be the recommended way to configure a ck kernel? I have my custom .config file for the vanilla kernel. The PKGBUILD in the AUR manually adds the few configuration options that come with the ck patchset. I'd rather configure the ck kernel via menuconfig so I can browse the options and get consistency checking when I change something.
Offline
@FlowIt, I think you just need to set _makenconfig to anything non-null in the beginning of the package build (line 7) or does that not do what you want it to?
Offline
So, use menuconfig? What do you want? A big fat comment at the top of the PKGBUILD offering to start nconfig for you, or something? Seriously...
Also, the sed'ing that the PKGBUILD does is extremely minor, unquestionably needed (why would you possibly think it is acceptable to NOT use the same sed lines as [core]/linux for setting the right kernel name/version) and utterly impossible to destroy your config, even if the copy-pasted-from-[core]/linux PKGBUILD didn't run `yes '' | make config` a few lines later.
Except for the optional stuff, which is just as optional as the word optional implies. But once again `yes '' | make config` is there to save you from evil...
Managing AUR repos The Right Way -- aurpublish (now a standalone tool)
Offline
Maybe I wasn't clear enough: I want to run menuconfig *before* I run makepkg (and only once for every major release), so that I have a finished config file and all the PKGBUILD has to do is to copy it to the right place.
Now I feel I have two options:
- follow the PKGBUILD and let sed do it's magic. This way some options may be set or unset without my knowledge (like MuQss or BFQ do not play that nicely with cgroups)
- run menuconfig on the patched source to create a .config file myself (but only once, not every time makepkg is invoked)
It's not about preventing my config from getting destroyed but about being able to track all changes.
Offline
That (still) makes no sense.
- menuconfig works on the Linux kernel sources, which don't exist until makepkg has run through until prepare()
- No options are being set or unset without your knowledge. You have to choose to use _localmodconfig, _use_current, _makenconfig, _bfq_enable, et. al.
- Being afraid of MuQSS makes absolutely no sense, why are you using the MuQSS kernel if you don't want MuQSS?
- Why on earth would the PKGBUILD be responsible for knowing where you keep your long-running handcrafted configs? Just use the "_use_current" knob to extract /proc/config.gz instead of the config that ships with the PKGBUILD.
Managing AUR repos The Right Way -- aurpublish (now a standalone tool)
Offline
Just use a patch file on top of the PKGBUILD. Then use a script to pull the new linux-ck package apply your patch and makepkg. Works fine until graysky changes the PKGBUILD structure which doesn't happen often.
Offline
Hi graysky... .. .
Today update:
[2017-01-14 07:34] [ALPM] upgraded linux-ck-core2 (4.8.17-1 -> 4.9.3-1)
[2017-01-14 07:34] [ALPM-SCRIPTLET] >>> Updating module dependencies. Please wait ...
[2017-01-14 07:34] [ALPM] upgraded linux-ck-core2-headers (4.8.17-1 -> 4.9.3-1)
[2017-01-14 07:34] [ALPM] upgraded nvidia-340xx-ck-core2 (340.101-2 -> 340.101-3)
resulted in
[ 6.415] (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the
Downgrade restored system
Al Einstein: "Man soll die Dinge so einfach machen wie möglich ~ aber nicht einfacher." (Things should be as simple as possible ~ but not too simple.) ~ Al (Einstein) war ein Cousin von Albert, "Al" ist die Abkürzung für Aloysius
Offline
Hi graysky... .. .
Today update:
[2017-01-14 07:34] [ALPM] upgraded linux-ck-core2 (4.8.17-1 -> 4.9.3-1) [2017-01-14 07:34] [ALPM-SCRIPTLET] >>> Updating module dependencies. Please wait ... [2017-01-14 07:34] [ALPM] upgraded linux-ck-core2-headers (4.8.17-1 -> 4.9.3-1) [2017-01-14 07:34] [ALPM] upgraded nvidia-340xx-ck-core2 (340.101-2 -> 340.101-3)
resulted in
[ 6.415] (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the
Downgrade restored system
Ditto, a problem with nvidia-utils, which works with linux,linux-lts,linux-ck.
I tried downgrading all of these only to have problems with the xorg-server arrangement with a problem of ABI support for xorg.
Not getting anywhere with that pacman -Su back to newest, and linux-lts is the only thing working right now.
Linux-lts was using a newer version of one of the nvidia libraries than linux or linux-ck.
Last edited by nomorewindows (2017-01-14 17:05:17)
I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.
Offline
So problems with nvidis-340xx-ck and nvidia-ck? I am running nvidia-ck-core2 1:375.26-3 on a system without issue.
Is the nvidia module loaded?
lsmod | grep nv
nvidia_modeset 782336 1 nvidia_drm
nvidia 11870208 1 nvidia_modeset
drm_kms_helper 126976 1 nvidia_drm
drm 294912 3 nvidia_drm,drm_kms_helper
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
It says something like Nvidia already registered, and then lsmod | grep nvidia shows nothing. Modprobe nvidia gives the first message...
I may have to CONSOLE you about your usage of ridiculously easy graphical interfaces...
Look ma, no mouse.
Offline
repo-ck/linux-ck-core2 4.9.3-1 doesn't boot. It gets stuck at the second line about loading initramfs.
[CPU] AMD Ryzen 5 2400G
[iGPU] AMD RX Vega 11
[Kernel] linux-zen
[sway] • [zsh] • Arch user since [2014-09-01 02:09]
Offline
There are seriously problems with Akonadi with every linux 4.9.x including linux-ck. Akonadiserver often crashing, and - sometimes - linux 4.9.x breaks akonadi base.
Offline
Hi, I have softlockup's with ck 4.9.x at any moment, 4.8.x works ok .
Jan 15 18:17:53 arch kernel: NMI watchdog: BUG: soft lockup - CPU#7 stuck for 22s! [python:588]
Jan 15 18:17:53 arch kernel: Modules linked in: sch_htb cls_u32 sch_ingress act_mirred ifb ctr ccm fuse nls_iso8859_1 nls_cp437 vfat fat intel_rapl x
Jan 15 18:17:53 arch kernel: dm_mod ie31200_edac edac_core battery ac wmi mac_hid evdev vboxpci(O) fjes mei_me mei i2c_i801 fan lpc_ich i2c_smbus vb
Jan 15 18:17:53 arch kernel: CPU: 7 PID: 588 Comm: python Tainted: G U O 4.9.4-1-ck-haswell #1
Jan 15 18:17:53 arch kernel: Hardware name: Micro-Star International Co., Ltd. GS60 2PC Ghost/MS-16H2, BIOS E16H2IMS.112 05/05/2015
Jan 15 18:17:53 arch kernel: task: ffff880419958000 task.stack: ffffc90000c10000
Jan 15 18:17:53 arch kernel: RIP: 0010:[<ffffffff810e8498>] [<ffffffff810e8498>] smp_call_function_single+0xe8/0x150
Jan 15 18:17:53 arch kernel: RSP: 0018:ffffc90000c13c50 EFLAGS: 00000202
Jan 15 18:17:53 arch kernel: RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000830
Jan 15 18:17:53 arch kernel: RDX: 0000000000000003 RSI: 00000000000008fb RDI: 0000000000000830
Jan 15 18:17:53 arch kernel: RBP: ffffc90000c13c98 R08: ffff88040ffc2000 R09: ffff88040ffc2000
Jan 15 18:17:53 arch kernel: R10: 00007f222c1f59c0 R11: ffff880419958000 R12: ffffffff8131e510
Jan 15 18:17:53 arch kernel: R13: ffff88040ffc2000 R14: ffff8803f2f40600 R15: ffff88040ffc2000
Jan 15 18:17:53 arch kernel: FS: 00007f22269b3700(0000) GS:ffff88042fbc0000(0000) knlGS:0000000000000000
Jan 15 18:17:53 arch kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 15 18:17:53 arch kernel: CR2: 00002ec66d7dcaf0 CR3: 0000000419476000 CR4: 00000000001406e0
Jan 15 18:17:53 arch kernel: Stack:
Jan 15 18:17:53 arch kernel: ffffc90000c13c98 ffffffff81267005 0000000000000000 ffffffff8131e510
Jan 15 18:17:53 arch kernel: ffffc90000c13ca8 0000000000000003 00000000b064086c ffffc90000c13cf4
Jan 15 18:17:53 arch kernel: ffffc90000c13cf0 ffffc90000c13ce0 ffffffff8131e68d 00000000000001b1
Jan 15 18:17:53 arch kernel: Call Trace:
Jan 15 18:17:53 arch kernel: [<ffffffff81267005>] ? kernfs_fop_open+0x135/0x380
Jan 15 18:17:53 arch kernel: [<ffffffff8131e510>] ? __extract_hwseed+0x39/0x39
Jan 15 18:17:53 arch kernel: [<ffffffff8131e68d>] rdmsr_on_cpu+0x5d/0x90
Jan 15 18:17:53 arch kernel: [<ffffffffa0235ae6>] show_temp+0xa6/0xd0 [coretemp]
Jan 15 18:17:53 arch kernel: [<ffffffff8142fd60>] dev_attr_show+0x20/0x50
Jan 15 18:17:53 arch kernel: [<ffffffff8126788c>] sysfs_kf_seq_show+0xbc/0x140
Jan 15 18:17:53 arch kernel: [<ffffffff81266343>] kernfs_seq_show+0x23/0x30
Last edited by gnox (2017-01-16 01:08:58)
Offline
Well, after today's update I'm in the "module nvidia is already loaded, abort" club: linux-ck-k10 4.9.4-1 / nvidia-340xx-ck-k10 340.101-3.
Actually, xorg-server update broke the mainline kernel for me as well, but while downgrading xorg-server helped me at least boot into my system with the mainline kernel 4.8.13, k10-ck-4.9.4 still won't boot.
Currently running downgraded linux-ck-k10 4.8.15-2/ nvidia-340xx-ck-k10 340.101-1.
Last edited by Soukyuu (2017-01-20 17:21:40)
[ Arch x86_64 | linux | Framework 13 | AMD Ryzen™ 5 7640U | 32GB RAM | KDE Plasma Wayland ]
Offline
Same issue here with module already registered for nvidia 340xx and kernel 4.9. There is a bug report with patch that solves the issue.
Last edited by remaros (2017-01-21 02:10:36)
Offline
Same issue here with module already registered for nvidia 340xx and kernel 4.9. There is a patch that solves the issue.
I updated nvidia-340xx-ck with the patch mentioned in FS#52243. Please try it (AUR) or update as I pushed to [repo-ck].
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Works as advertised Thank You.
Offline
Anyone experienced wifi bridge performance drop after upgrading to 4.9.5?
Offline
Does linux-ck use BBR? I saw the config option in the build files, but I'm not all too familiar with this stuff. Thanks in advance.
Offline
Does linux-ck use BBR?
The new TCP congestion control?
linux-ck's configuration is based on arch's linux kernel configuration, therefore if BBR is enabled in arch kernel, it should be enabled in linux-ck as well.
If you look at the kernel .config files, there's "CONFIG_TCP_CONG_BBR=m", which means BBR is built as a module, so you can use it if you want.
Offline
Thanks for your clarification. I created /etc/sysctl.d/51-net.conf with net.ipv4.tcp_congestion_control=bbr, that should do the trick.
Last edited by protake (2017-01-22 15:45:10)
Offline
remaros wrote:Same issue here with module already registered for nvidia 340xx and kernel 4.9. There is a patch that solves the issue.
I updated nvidia-340xx-ck with the patch mentioned in FS#52243. Please try it (AUR) or update as I pushed to [repo-ck].
With
repo-ck/linux-ck-kx 4.9.5-1 (ck-kx)
repo-ck/nvidia-304xx-ck-kx 304.134-2 (ck-kx)
my system still doesn't boot. The error message hasn't changed: "nvidia driver already registered".
Offline