You are not logged in.
@snack: In the mean time I also asked about the issue on ck's blog and got a clear response from him. Apparently PSI is a debugging feature and according to ck there's not much point in enabling it. You can disable PSI by adding 'psi=0' kernel parameter and get rid of the message at boot.
@graysky: Maybe you could change the default here by setting CONFIG_PSI_DEFAULT_DISABLED, so linux-ck users won't have to set any specific kernel parameter to fix this. Just a suggestion, I do realize this would be extra work for you.
Offline
Let me see where that got enabled and fix.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
The CONFIG_PSI=y option is set in the default Arch config that I mirror for linux-ck. It can be disabled easily enough... I am wondering what in the CK patchset is triggering it and is it a false positive? 4.20.0-2-ck1 has this setting disabled, but it seems cosmetic unless the warning has a functional implication.
Last edited by graysky (2019-01-04 19:06:56)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
@graysky: IMHO it would be easier to leave CONFIG_PSY=y as is and change # CONFIG_PSI_DEFAULT_DISABLED is not set instead. Setting that as CONFIG_PSI_DEFAULT_DISABLED=y will save users who want to disable the PSI feature from making any changes. Only users who want to keep the PSI feature would have to use kernel parameter 'psi=1' to override.
Offline
@glitsj16 - Good idea. See 4.20.0-3-ck1
https://wiki.archlinux.org/index.php?ti … did=561757
Last edited by graysky (2019-01-04 20:58:13)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
-deleted-
Last edited by fsyyy (2019-06-01 19:18:25)
Offline
Docker seems to have a Problem with 4.20.1-1-ck-skylake #1 SMP PREEMPT Wed Jan 9 18:00:42 EST 2019 x86_64 GNU/Linux
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126220330+01:00" level=warning msg="Your kernel does not support cgroup cpu shares"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126231724+01:00" level=warning msg="Your kernel does not support cgroup cfs period"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126237747+01:00" level=warning msg="Your kernel does not support cgroup cfs quotas"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126242981+01:00" level=warning msg="Your kernel does not support cgroup rt period"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126248545+01:00" level=warning msg="Your kernel does not support cgroup rt runtime"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.126698932+01:00" level=info msg="Loading containers: start."
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.357216135+01:00" level=info msg="Default bridge (docker0) is assigned with an IP address 172.18.0.0/16. Daemon option --bip can be used to set a preferred IP address"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.447599031+01:00" level=info msg="Loading containers: done."
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.535639159+01:00" level=warning msg="Not using native diff for overlay2, this may cause degraded performance for building images: kernel has CONFIG_OVERLAY_FS_REDIRECT_DIR enabled" storage-driver=overlay2
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.535789616+01:00" level=info msg="Docker daemon" commit=4c52b901c6 graphdriver(s)=overlay2 version=18.09.1-ce
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.535944419+01:00" level=info msg="Daemon has completed initialization"
Jan 10 11:38:06 xps15 dockerd[7896]: time="2019-01-10T11:38:06.543938472+01:00" level=info msg="API listen on /var/run/docker.sock"
Jan 10 11:38:06 xps15 systemd[1]: Started Docker Application Container Engine.
this means that a start of my Containers failed with:
Jan 10 11:39:41 xps15 dockerd[7896]: time="2019-01-10T11:39:41.476848997+01:00" level=info msg="shim containerd-shim started" address="/containerd-shim/moby/5bc3c88bd9c7dd50fe669dc7b81d680f4f96cda7e8e980c1d919ac96717d9cb1/shim.sock" debug=false pid=13866
Jan 10 11:39:41 xps15 dockerd[7896]: time="2019-01-10T11:39:41.815105370+01:00" level=info msg="shim reaped" id=5bc3c88bd9c7dd50fe669dc7b81d680f4f96cda7e8e980c1d919ac96717d9cb1
Jan 10 11:39:42 xps15 dockerd[7896]: time="2019-01-10T11:39:42.247848503+01:00" level=error msg="5bc3c88bd9c7dd50fe669dc7b81d680f4f96cda7e8e980c1d919ac96717d9cb1 cleanup: failed to delete container from containerd: no such container"
Jan 10 11:39:42 xps15 dockerd[7896]: time="2019-01-10T11:39:42.247874941+01:00" level=error msg="Handler for POST /v1.39/containers/wikisql/start returned error: unable to find \"cpuacct\" in controller set: unknown"
with standard archlinux kernel the problem doesn't exist.
Offline
@bigfreak - It would appear the ck1 patchset does this, see here for a diff of the Arch config before and after simply patching with CK1: https://github.com/graysky2/4.20.1_temp … 6faa187273
Recommend you post to CK's blog referencing that diff and your problem. Might be that he did not mean to do that if I am correct.
Last edited by graysky (2019-01-10 20:28:03)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Recommend you post to CK's blog referencing that diff and your problem. Might be that he did not mean to do that if I am correct.
https://github.com/ckolivas/linux/commi … ebd1cR1148
The change has been in every release of MUQSS.
Offline
Yes, but someone can still ask CK if it is intentional and needed particularly if it affects functionality of docker.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
@bigfreak - It would appear the ck1 patchset does this, see here for a diff of the Arch config before and after simply patching with CK1: https://github.com/graysky2/4.20.1_temp … 6faa187273
Recommend you post to CK's blog referencing that diff and your problem. Might be that he did not mean to do that if I am correct.
done and the response from ck
No, docker and containers that use CPU scheduler cgroups in general do not work at all with MuQSS. There is no 'containment' as such, and the cgroups are only there to allow systems to run that mandate their existence. from https://ck-hack.blogspot.com/2018/12/li … 7335602087
Last edited by bigfreak (2019-01-11 10:59:12)
Offline
Yeah, I can confirm that, unfortunately. I think it'll be wise to drop a line in the wiki (linux-ck and docker, perhaps?) in case someone gets stuck in docker wondering what may have happened ... It's a pity, though
"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." -- Richard P. Feynman
Offline
@vltr - good idea, go for it... it is a wiki after all
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
I will, since I had to change the output of pacman -V to match a prior version so I can pass on the wiki registration riddle
"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." -- Richard P. Feynman
Offline
Offline
Not sure to follow: is there a fix or workaround for using both `docker build` and linux-ck ?
Offline
Not sure to follow: is there a fix or workaround for using both `docker build` and linux-ck ?
I saw some reports where people were able to run docker with newest linux-ck kernel versions (4.20.x), but had to downgrade systemd (to 239 I think?) in order to achieve that. I'm not sure about if this is a workaround or fix since I've not tested myself nor want to pin my systemd version.
"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." -- Richard P. Feynman
Offline
update to nvidia-utils again
installing nvidia-utils (418.43-1) breaks dependency 'nvidia-utils=415.27' required by nvidia-ck-zen
Offline
Building now, thanks. In the future, just flag the affected package out-of-date in the AUR.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Hello there,
error: failed retrieving file 'linux-ck-zen-4.20.11-2-x86_64.pkg.tar.xz' from repo-ck.com : The requested URL returned error: 404
Just dropping a line to let you know there's a problem. Seems the package is missing from the repo?
Regards
Offline
Thank all good
Offline
unable to satisfy dependency 'nvidia-utils=418.43' required by nvidia-ck-haswell
Offline
unable to satisfy dependency 'nvidia-utils=418.43' required by nvidia-ck-haswell
Works for me...
# pacman -S nvidia-ck
resolving dependencies...
looking for conflicting packages...
warning: dependency cycle detected:
warning: xorg-server will be installed before its nvidia-utils dependency
Packages (14) egl-wayland-1.1.2-1 eglexternalplatform-1.0+3+g7c8f8e2-1 libepoxy-1.5.3-1 libevdev-1.6.0-1 libgudev-232-1
libinput-1.12.6-1 libwacom-0.32-1 mtdev-1.1.5-2 nvidia-utils-418.43-2 xf86-input-libinput-0.28.2-1
xorg-server-1.20.4-1 xorg-server-common-1.20.4-1 xorg-setxkbmap-1.3.1-2 nvidia-ck-2:418.43-4
Total Download Size: 50.97 MiB
Total Installed Size: 216.81 MiB
:: Proceed with installation? [Y/n]
:: Retrieving packages...
nvidia-utils-418.43-2-x86_64 39.4 MiB 15.7M/s 00:03 [############################################] 100%
nvidia-ck-2:418.43-4-x86_64 11.6 MiB 4.54M/s 00:03 [############################################] 100%
(14/14) checking keys in keyring [############################################] 100%
(14/14) checking package integrity [############################################] 100%
(14/14) loading package files [############################################] 100%
(14/14) checking for file conflicts [############################################] 100%
:: Processing package changes...
( 1/14) installing libepoxy [############################################] 100%
( 2/14) installing xorg-setxkbmap [############################################] 100%
( 3/14) installing xorg-server-common [############################################] 100%
( 4/14) installing mtdev [############################################] 100%
( 5/14) installing libevdev [############################################] 100%
( 6/14) installing libgudev [############################################] 100%
( 7/14) installing libwacom [############################################] 100%
( 8/14) installing libinput [############################################] 100%
Optional dependencies for libinput
gtk3: libinput debug-gui
python-pyudev: libinput measure
python-evdev: libinput measure
( 9/14) installing xf86-input-libinput [############################################] 100%
(10/14) installing xorg-server [############################################] 100%
>>> xorg-server has now the ability to run without root rights with
the help of systemd-logind. xserver will fail to run if not launched
from the same virtual terminal as was used to log in.
Without root rights, log files will be in ~/.local/share/xorg/ directory.
Old behavior can be restored through Xorg.wrap config file.
See Xorg.wrap man page (man xorg.wrap).
(11/14) installing eglexternalplatform [############################################] 100%
(12/14) installing egl-wayland [############################################] 100%
(13/14) installing nvidia-utils [############################################] 100%
If you run into trouble with CUDA not being available, run nvidia-modprobe first.
Optional dependencies for nvidia-utils
nvidia-settings: configuration tool
xorg-server-devel: nvidia-xconfig
opencl-nvidia: OpenCL support
(14/14) installing nvidia-ck [############################################] 100%
:: Running post-transaction hooks...
(1/5) Updating linux-ck module dependencies...
(2/5) Reloading system manager configuration...
(3/5) Creating system user accounts...
(4/5) Reloading device manager configuration...
(5/5) Arming ConditionNeedsUpdate...
Last edited by graysky (2019-02-28 23:08:27)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
sudo pacman -S nvidia-ck-haswell
resolving dependencies...
warning: cannot resolve "nvidia-utils=418.43", a dependency of "nvidia-ck-haswell"
:: The following package cannot be upgraded due to unresolvable dependencies:
nvidia-ck-haswell
:: Do you want to skip the above package for this upgrade? [y/N]
Offline
Offline