You are not logged in.
The NVIDIA driver package still supports only v4.16. Is everything OK?
No idea what you're talking about or how the question is related to [repo-ck].
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Tharbad wrote:The NVIDIA driver package still supports only v4.16. Is everything OK?
No idea what you're talking about or how the question is related to [repo-ck].
nvidia-ck-ivybridge-2:396.24-1-x86_64.pkg.tar.xz is part of the repo:
http://repo-ck.com/x86_64/
Offline
nvidia-ck-ivybridge-2:396.24-1-x86_64.pkg.tar.xz is part of the repo:
http://repo-ck.com/x86_64/
What am I missing?
# Generated by makepkg 5.1.0
# using fakeroot version 1.22
pkgname = nvidia-ck-ivybridge
pkgbase = nvidia-ck-ivybridge
pkgver = 2:396.24-1
pkgdesc = NVIDIA drivers for linux-ck. Third Gen Intel Core i3/i5/i7 optimized.
url = http://www.nvidia.com/
builddate = 1530137031
packager = graysky
size = 9526272
arch = x86_64
license = custom
group = ck-ivybridge
conflict = nvidia-340xx-ck-ivybridge
conflict = nvidia-390xx-ck-ivybridge
depend = linux-ck-ivybridge>=4.17
depend = linux-ck-ivybridge<4.18
depend = libgl
depend = nvidia-utils=396.24
makedepend = nvidia-utils=396.24
makedepend = libglvnd
makedepend = linux-ck-ivybridge-headers>=4.17
makedepend = linux-ck-ivybridge-headers<4.18
Last edited by graysky (2018-07-03 21:27:45)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
no clue:
installing linux-ck-ivybridge (4.17.3-1) breaks dependency 'linux-ck-ivybridge<4.17' required by nvidia-ck-ivybridge
That's what I have installed:
Name : nvidia-ck-ivybridge
Version : 2:396.24-2
Description : NVIDIA drivers for linux-ck. Third Gen Intel Core i3/i5/i7 optimized.
Architecture : x86_64
Last edited by Tharbad (2018-07-03 22:21:26)
Offline
EDIT: You're on 2:396.24-2? I haven't released -2.
Last edited by graysky (2018-07-03 23:13:56)
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
EDIT: You're on 2:396.24-2? I haven't released -2.
It was there when I updated the system in 12.5.2018
EDIT: I installed the one in the repo. Now it's compatible.
EDIT2: The package .PKGINFO
# Generated by makepkg 5.0.2
# using fakeroot version 1.22
# Sat May 12 19:34:38 UTC 2018
pkgname = nvidia-ck-ivybridge
pkgver = 2:396.24-2
pkgdesc = NVIDIA drivers for linux-ck. Third Gen Intel Core i3/i5/i7 optimized.
url = http://www.nvidia.com/
builddate = 1526153678
packager = graysky <graysky@archlinux.us>
size = 9525248
arch = x86_64
license = custom
group = ck-ivybridge
conflict = nvidia-340xx-ck-ivybridge
conflict = nvidia-390xx-ck-ivybridge
depend = linux-ck-ivybridge>=4.16
depend = linux-ck-ivybridge<4.17
depend = libgl
depend = nvidia-utils=396.24
makedepend = linux-ck-ivybridge-headers>=4.16
makedepend = linux-ck-ivybridge-headers<4.17
makedepend = nvidia-libgl
Was that package an error?
Last edited by Tharbad (2018-07-04 01:24:35)
Offline
graysky wrote:EDIT: You're on 2:396.24-2? I haven't released -2.
It was there when I updated the system in 12.5.2018
EDIT: I installed the one in the repo. Now it's compatible.
EDIT2: The package .PKGINFO
# Generated by makepkg 5.0.2 # using fakeroot version 1.22 # Sat May 12 19:34:38 UTC 2018 pkgname = nvidia-ck-ivybridge pkgver = 2:396.24-2 pkgdesc = NVIDIA drivers for linux-ck. Third Gen Intel Core i3/i5/i7 optimized. url = http://www.nvidia.com/ builddate = 1526153678 packager = graysky <graysky@archlinux.us> size = 9525248 arch = x86_64 license = custom group = ck-ivybridge conflict = nvidia-340xx-ck-ivybridge conflict = nvidia-390xx-ck-ivybridge depend = linux-ck-ivybridge>=4.16 depend = linux-ck-ivybridge<4.17 depend = libgl depend = nvidia-utils=396.24 makedepend = linux-ck-ivybridge-headers>=4.16 makedepend = linux-ck-ivybridge-headers<4.17 makedepend = nvidia-libgl
Was that package an error?
Some times this happens with me too. Everytime is happens, I remove the nvidia-ck package and then install it again
not sure why it happens though
Offline
resolving dependencies...
warning: cannot resolve "nvidia-utils=396.24", a dependency of "nvidia-ck-zen"
:: The following package cannot be upgraded due to unresolvable dependencies:
nvidia-ck-zen
Offline
I just noticed the instability of the CK kernel server, The Wiki says "Graysky is using Go Daddy as his web host. Some of the transfers from their poorly implemented server end in an incomplete transfer." and one of Grey's own comments earlier in this thread says "That is the quality service godaddy provides.". So it may seem like a silly question but, why would you possibly stay with the same host when it's so completely unstable? (is it free or something?)
Offline
@john29 - I updated. The problem is that the nvidia-utils version in [extra] was bumped but no one flagged the AUR package out-of-date so I didn't know about it.
@Enverex - Yes, the cost is insanely low.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Hello graysky,
I noticed that upgrading to nvidia-ck-ivybridge 2:396.45-1 (from 396.24) breaks CUDA functionality (with device enumeration failing in CUDA applications). On the NVIDIA forums it is suggested that 396.45 needs CONFIG_NUMA for CUDA to work.
Kind regards,
E.
Last edited by eigengrau (2018-08-02 06:09:42)
Offline
@eigengrau - Yes, see: https://bbs.archlinux.org/viewtopic.php … 2#p1800302
For [repo-ck] I will continue shipping the NUMA disabled packages, but individual CUDA users can compile from the AUR with that option set to null.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
@eigengrau - Yes, see: https://bbs.archlinux.org/viewtopic.php … 2#p1800302
Thank you kindly for the pointer!
Offline
Hi, and thanks for your work.
I would like to report a bug.. I think.
For a while now i have had problem with my box that run a N3540. It has been crashing and crashing.. So i tried baytrail kernel and then ck-silvermont.. nothing helped. At last i found out i had two memory faults, one at 640mb and one at 6784Mb.
So I figured out the memmap=1M$640M memmap=1M$6784M on the kernel line..
The first was swallowed fine but the second memmap refused to go down.. on ck-silvermont. Then i got the idea to try the arch mainline.. and glory and behold.. it worked!
I would really like to use the silvermont kernel.. but with my two memory faults and all.. so I just wanted to report and hope this is easy to fix. And that you care. Thanks
Offline
@jirido - Does the standard kernel (Arch) work without the memmap lines? If so, does the generic ck kernel work without the lines?
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
No both crash after a while. But arch mainline could boot with both memmap'ings on the line while ck-silvermont can't boot if the second mapping where on the line.
Offline
No both crash after a while. But arch mainline could boot with both memmap'ings on the line while ck-silvermont can't boot if the second mapping where on the line.
OK, so the memlines are required on the standard Arch kernel. How about the generic ck kernel (not ck-silvermont)? Does it run with the memlines ok? I want to see if the problem is the silventmout code or ck1 itself. Further, if you're remapping or limiting memory with those switches, you'll mostly likely just buying time until more memory goes bad (in my experience). Finally, post the output of:
gcc -c -Q -march=native --help=target | grep march
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Ok so I installed ck-generic and will now reboot and run it with this kernel line..
options cryptdevice=UUID=6af25e01-222b-4a39-b9c4-8cf002ce9f5b:archlv root=/dev/mapper/archvg-root rw 3 memmap=1M$640M memmap=1M$6784M and theeze HOOKS=(base udev autodetect modconf block keyboard keymap encrypt lvm2 filesystems fsck) in mkinitcpio.conf
The result of: gcc -c -Q -march=native --help=target | grep march
Was without big suprizes: -march= silvermont
Maybe you cod get space for your repo on arch servers. The download broke 3 times.. Not that it was any danger as pacman resumed fine this time (It did not with silvermount that i had to dnload 2x). It seams like arch community is around you and surely they would not mind the space. But I guess you know that and like it like this
Offline
Ok so ck-generic worked well with that kernel line. ck-silvermont did not (I tried again).
Say just if you need any more testings.
Offline
Well, there's something with the silvermont code might be a conclusion. Not related to ck1. What makes it more difficult though is that your hardware has know bad memory.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
sddm crashed with nvidia 396.51. Had to revert back to 396.45. error is:
Failed to read display number from pipe
Display server failed to start. Exiting
Stack trace of thread 870:
#0 0x00007f6c59a29b5f raise (libc.so.6)
#1 0x00007f6c59a14452 abort (libc.so.6)
#2 0x00007f6c59de17fc _ZNK14QMessageLogger5fatalEPKcz (libQt5Core.so.5)
#3 0x000055bd8b3ac1f3 n/a (sddm)
#4 0x000055bd8b3ebc2e _ZN4SDDM4Seat13createDisplayEi (sddm)
#5 0x000055bd8b3ebe72 _ZN4SDDM4SeatC2ERK7QStringP7QObject (sddm)
#6 0x000055bd8b3ede41 _ZN4SDDM11SeatManager10createSeatERK7QString (sddm)
#7 0x000055bd8b3eec62 n/a (sddm)
#8 0x00007f6c59ffd0b0 _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
#9 0x000055bd8b3eceb0 _ZN4SDDM10LogindSeat19canGraphicalChangedEb (sddm)
#10 0x000055bd8b3ed1c3 n/a (sddm)
#11 0x00007f6c59ffd0b0 _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
#12 0x00007f6c5ad04900 _ZN23QDBusPendingCallWatcher8finishedEPS_ (libQt5DBus.so.5)
#13 0x00007f6c5ad04a01 n/a (libQt5DBus.so.5)
#14 0x00007f6c59ffd832 _ZN7QObject5eventEP6QEvent (libQt5Core.so.5)
#15 0x00007f6c59fd30f7 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5)
#16 0x00007f6c59fd61ac _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5)
#17 0x00007f6c5a026ec4 n/a (libQt5Core.so.5)
#18 0x00007f6c579a7368 g_main_context_dispatch (libglib-2.0.so.0)
#19 0x00007f6c579a75b1 n/a (libglib-2.0.so.0)
#20 0x00007f6c579a763e g_main_context_iteration (libglib-2.0.so.0)
#21 0x00007f6c5a0264a9 _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5)
#22 0x00007f6c59fd1dac _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5)
#23 0x00007f6c59fda0a6 _ZN16QCoreApplication4execEv (libQt5Core.so.5)
#24 0x000055bd8b3afb0a main (sddm)
#25 0x00007f6c59a16003 __libc_start_main (libc.so.6)
#26 0x000055bd8b3afe0a _start (sddm)
Stack trace of thread 871:
#0 0x00007f6c59ae2991 __poll (libc.so.6)
#1 0x00007f6c579a7523 n/a (libglib-2.0.so.0)
#2 0x00007f6c579a763e g_main_context_iteration (libglib-2.0.so.0)
#3 0x00007f6c5a0264a9 _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5)
#4 0x00007f6c59fd1dac _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5)
#5 0x00007f6c59e1af59 _ZN7QThread4execEv (libQt5Core.so.5)
#6 0x00007f6c5aca8ba6 n/a (libQt5DBus.so.5)
#7 0x00007f6c59e25005 n/a (libQt5Core.so.5)
#8 0x00007f6c599d8a8d start_thread (libpthread.so.0)
#9 0x00007f6c59aed823 __clone (libc.so.6)
Checked the Arch bugs report, nothing like this on the latest and lts drivers.
Offline
initcpio image fails to build with nvidia-ck-broadwell 2:396.51-1, also tried generic nvidia-ck 2:396.51-1. The error line says that modules nvidia nvidia_modset nvidia_uvm nvidia_drm are missing. The initcpio image builds fine with extra\nvidia though. Wasn't a problem with nvidia-ck-broadwell 2:396.45.
Error output:-
==> Building image from preset: /etc/mkinitcpio.d/linux-ck-broadwell.preset: 'default'
-> -k /boot/vmlinuz-linux-ck-broadwell -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-ck-broadwell.img
==> Starting build: 4.17.11-1-ck-broadwell
-> Running build hook: [base]
-> Running build hook: [udev]
-> Running build hook: [autodetect]
-> Running build hook: [modconf]
-> Running build hook: [block]
-> Running build hook: [keyboard]
-> Running build hook: [encrypt]
-> Running build hook: [openswap]
-> Running build hook: [resume]
-> Running build hook: [shutdown]
-> Running build hook: [suspend]
-> Running build hook: [filesystems]
==> ERROR: module not found: `nvidia'
==> ERROR: module not found: `nvidia_modeset'
==> ERROR: module not found: `nvidia_uvm'
==> ERROR: module not found: `nvidia_drm'
==> Generating module dependencies
==> Creating xz-compressed initcpio image: /boot/initramfs-linux-ck-broadwell.img
==> WARNING: errors were encountered during the build. The image may not be complete.
/etc/mkinitcpio.conf
MODULES=(nvidia nvidia_modeset nvidia_uvm nvidia_drm lz4 lz4_compress)
Last edited by lll2yu (2018-08-14 17:07:36)
Offline
Well if you fix it I'll use it Best of luck's
Offline
initcpio image fails to build with nvidia-ck-broadwell 2:396.51-1, also tried generic nvidia-ck 2:396.51-1. The error line says that modules nvidia nvidia_modset nvidia_uvm nvidia_drm are missing. The initcpio image builds fine with extra\nvidia though. Wasn't a problem nvidia-ck-broadwell 2:396.45.
/etc/mkinitcpio.conf
MODULES=(nvidia nvidia_modeset nvidia_uvm nvidia_drm lz4 lz4_compress)
Why do u use those modules? I don't have them everything is ok (I have nvidia card too).
Last edited by Tharbad (2018-08-14 09:54:05)
Offline
lll2yu wrote:initcpio image fails to build with nvidia-ck-broadwell 2:396.51-1, also tried generic nvidia-ck 2:396.51-1. The error line says that modules nvidia nvidia_modset nvidia_uvm nvidia_drm are missing. The initcpio image builds fine with extra\nvidia though. Wasn't a problem nvidia-ck-broadwell 2:396.45.
/etc/mkinitcpio.conf
MODULES=(nvidia nvidia_modeset nvidia_uvm nvidia_drm lz4 lz4_compress)
Why do u use those modules? I don't have them everything is ok (I have nvidia card too).
I have the same issue and I have this
MODULES=(nvidia nvidia_modeset nvidia_uvm nvidia_drm)
The reason, IIRC, was that I enabled Kernel mode setting. I got it from here https://wiki.archlinux.org/index.php/NV … de_setting
Last edited by raggerv8 (2018-08-14 14:23:54)
Brottweiler@#archlinux
Offline