| === lestac is now known as lestac_afk | ||
| === lestac_afk is now known as lestac | ||
| lotuspsychje | good morning | 02:19 |
|---|---|---|
| blockiv | Hello, I'm really really hoping to get some help figuring out why I'm unable to get my computer to recognize my nvidia gpu. I'm on 20.04, so figured I should be asking here rather than #ubuntu | 02:46 |
| blockiv | Though, I haven't been able to get it to work on the current release either. | 02:46 |
| lotuspsychje | whats your chipset blockiv | 02:47 |
| blockiv | I'm using an xps 15 9570. With hybrid i7 and a 1050ti | 02:48 |
| blockiv | lotuspsychje: The nvidia module seems to be in use, but `nvidia-smi` returns "no devices were found" | 02:48 |
| lotuspsychje | blockiv: can you pastebin: sudo lshw -C video plz | 02:49 |
| blockiv | lotuspsychje: yep one sec. | 02:49 |
| blockiv | lotuspsychje: https://paste.ubuntu.com/p/NHqJNQKv6f/ | 02:50 |
| lotuspsychje | hm thats pretty weird, driver shows loaded blockiv | 02:51 |
| lotuspsychje | blockiv: can you try from software&sources tab additional drivers to see wich driver is active? | 02:51 |
| blockiv | lotuspsychje: nvidia-driver-440 | 02:52 |
| lotuspsychje | sounds good | 02:52 |
| Thanos | maybe theres just an error in reporting and not in utilization | 02:52 |
| lotuspsychje | blockiv: you boot uefi or legacy? | 02:52 |
| blockiv | and yeah, I really have no clue what's causing it. And uefi lotuspsychje | 02:52 |
| lotuspsychje | blockiv: pastebin your dmesg plz? | 02:53 |
| blockiv | lotuspsychje: https://paste.ubuntu.com/p/tzpDN4gkFF/ | 02:53 |
| blockiv | I'm debating installing windows quick to see if it's a hardware problem. I | 02:53 |
| blockiv | I'm on a pretty new ubuntu install so wouldn't really lose anything data wise | 02:54 |
| lotuspsychje | blockiv: i see a few weird NVRM and cpu temps issues in your dmesg | 02:57 |
| blockiv | What is nvrm? lotuspsychje | 02:58 |
| lotuspsychje | its related to your graphics | 02:58 |
| lotuspsychje | blockiv: did you change things in bios recently? | 02:58 |
| blockiv | lotuspsychje: I reset the bios to factory defaults before installing ubuntu. Then changed RAID to AHCI, disabled secure boot, and disabled TPM | 02:59 |
| blockiv | And set fastboot to "Thourough rather than minimal" | 03:00 |
| lotuspsychje | blockiv: lets try to switch an nvidia driver as a test | 03:00 |
| lotuspsychje | blockiv: try a version lower | 03:01 |
| lotuspsychje | and lets see what your system does | 03:01 |
| blockiv | Alright. 435 or 390? | 03:01 |
| lotuspsychje | if you are not able to enter back your system due a black screen, use this method: | 03:01 |
| lotuspsychje | !nomodeset | 03:01 |
| ubottu | Systems with certain graphics chipsets may not boot properly out of the box. "Temporarily Add a Kernel Boot Parameter for Testing" as discussed at https://wiki.ubuntu.com/Kernel/KernelBootParameters to add the "nomodeset" parameter there. | 03:01 |
| lotuspsychje | blockiv: lets try 435 | 03:01 |
| blockiv | Okay will do. Were the errors you were talking about the RM_init related lines in dmesg? | 03:02 |
| lotuspsychje | [ 9408.670356] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0 | 03:03 |
| lotuspsychje | [ 9444.369429] mce: CPU9: Core temperature above threshold, cpu clock throttled (total events = 30626) | 03:03 |
| lotuspsychje | stuff like this | 03:03 |
| lotuspsychje | we need to find out if its the driver failing, or something bios or kernel related | 03:04 |
| blockiv | Okay, if it nothing seems to change. I'll probably install windows 10 quick to make sure the gpu still even works. I haven't been able to get it working in linux since I had an arch install on some Linux 4.X kernel | 03:05 |
| lotuspsychje | blockiv: no need, we can try this from ubuntu aswell | 03:05 |
| Bashing-om | blockiv: lotuspsychje : /var/log/gpu-manager.log might provide some hints. | 03:06 |
| blockiv | great. that's always preferred lol | 03:06 |
| lotuspsychje | blockiv: if you like before driver switch you can provide this log for Bashing-om | 03:06 |
| blockiv | Bashing-om: https://paste.ubuntu.com/p/7wv3xpvW4K/ | 03:07 |
| Bashing-om | blockiv: reading. | 03:08 |
| Bashing-om | blockiv: lotuspsychje : Yukkie: "Is nvidia kernel module available? no". lemme finish reading. | 03:09 |
| lotuspsychje | blockiv: did you also update system to latest? | 03:10 |
| blockiv | yeah, no updates are available after an apt update | 03:10 |
| lotuspsychje | ok great | 03:10 |
| lotuspsychje | if Bashing-om has no other ideas, i would advice driver switch as a test | 03:11 |
| Bashing-om | lotuspsychje: blockiv - looking for what modules are missing. lets start with ' | 03:11 |
| Bashing-om | dpkg -l | grep -i nvidia ' . | 03:12 |
| blockiv | Bashing-om: https://paste.ubuntu.com/p/RKvttcmZzJ/ | 03:13 |
| Bashing-om | blockiv: Whike the 440 version driver is correct: https://www.nvidia.com/Download/driverResults.aspx/159360/en-us, as it is partially installed lets try and purge/reinstall - ok ? | 03:17 |
| blockiv | Bashing-om: alright. Mind letting me know what to run? Been a while since I was using an apt system | 03:18 |
| Bashing-om | blockiv: we also want to make sure that "secure boot" in bios is disabled when we re-install ! | 03:18 |
| nonix4 | ... when running test installation with / on usb, is there much you can do once usb timeouts? As in to recover you'd neet to remount some tmpfs/squashfs as / for a while? | 03:18 |
| Bashing-om | blockiv: sure - run ' sudo apt --purge nvidia*' make sure that the nvidia file(s) in /usr/share/X11/xorg.conf.d have been removed. | 03:19 |
| Bashing-om | blockiv: Also make sure here that 'dkms' is functional ' dkms status '. as we do need that package for Nvidia. | 03:21 |
| blockiv | Bashing-om: hm `sudo apt remove --purge nvidia*` returns "Unable to loacate package nvidia*" | 03:22 |
| blockiv | Maybe cause I just switched from 440 to 435 in the software&updates app? | 03:22 |
| blockiv | And as of right now, `dkms status` returns "nvidia, 435.21: added" | 03:23 |
| lotuspsychje | lets try a reboot first, as 435 has been switched? | 03:24 |
| Bashing-om | blockiv: well we can look. what shows ' lsmod | grep nvidia ; dpkg -l | grep -i nvidia ' . | 03:25 |
| blockiv | Sure. I'll also double check to make sure secure boot is off. I'm on my machine rn so I'll have to disconnect, but I'll send the output of those first Bashing-om lotuspsychje | 03:25 |
| lotuspsychje | sure thing | 03:26 |
| blockiv | lsmod | grep nvidia : https://paste.ubuntu.com/p/zmYWBhMXzz/ | 03:26 |
| blockiv | dpkg -l | grep nvidia : https://paste.ubuntu.com/p/Mcg9Nc4yX7/ | 03:27 |
| blockiv | Okay, I'll reconnect in a few mins | 03:28 |
| lotuspsychje | cross your fingers Bashing-om | 03:28 |
| lotuspsychje | interesting one | 03:28 |
| Bashing-om | lotuspsychje: Looks to me now should workie. | 03:29 |
| blockiv | Bashing-om: lotuspsychje: Okay, secure boot is indeed disabled | 03:32 |
| lotuspsychje | blockiv: check nvidia-smi again now? | 03:32 |
| blockiv | lotuspsychje: just did. new output, but still failing. It says it couldn't communicate with the NVIDIA driver. Maker sure the latest NVIDIA driver is installed and running | 03:33 |
| lotuspsychje | hmm | 03:33 |
| Bashing-om | blockiv: /var/log/gpu-manager.log shows clean now ? | 03:34 |
| blockiv | Also, lspci -k now doesn't show the line " Kernel driver in use: nvidia" | 03:34 |
| lotuspsychje | so 2 drivers failing, sounds like a bios or kernel thing to me | 03:35 |
| blockiv | Here is gpu-manager.log : https://paste.ubuntu.com/p/9T8QFHJJYR/ | 03:36 |
| Bashing-om | reading | 03:36 |
| Bashing-om | blockiv: "Is nvidia loaded? no" - !! | 03:37 |
| Bashing-om | blockiv: Let's try again ' sudo apt remove --purge nvidia-* ; sudo apt update ; sudo apt upgrade ; sudo ubuntu-drivers autoinstall ' with secure boot still disbaled. | 03:38 |
| blockiv | alright | 03:38 |
| blockiv | for some reason the wildcard doesn't seem to be working. | 03:39 |
| blockiv | worked with apt-get | 03:40 |
| Bashing-om | blockiv: Humm --- maybe try as sudo apt remove --purge "nvidia-*" . have the shell expand the quotes. (1st time for everything) | 03:41 |
| Bashing-om | NM | 03:41 |
| blockiv | Just uninstalled. Should I remove 10-nvidia.conf and 11-nvidia-prime.conf before moving forward? | 03:42 |
| Bashing-om | ^^ makes me wonder what has changed in apt in the 20.04 release ?? | 03:42 |
| Bashing-om | blockiv: Yeah remove them - the new install will re-create, the old may be corrupt. | 03:43 |
| blockiv | Bashing-om: Okay, deleted them and ran apt autoremove and ubuntu-drivers autoinstall | 03:46 |
| Bashing-om | blockiv: reboot and let's see the effect :D | 03:46 |
| blockiv | The ubuntu-drivers autoinstall said "dkms: WARNING: Linux headers are missing" | 03:46 |
| blockiv | Any importance you think? | 03:47 |
| Bashing-om | blockiv: Ouch Yes ! .. show ' dpkg -l | grep linux- '. | 03:47 |
| blockiv | https://paste.ubuntu.com/p/2TybRPgTZF/ | 03:48 |
| blockiv | This is the exact line "/etc/kernel/postinst.d/dkms: | 03:48 |
| blockiv | dkms: WARNING: Linux headers are missing, which may explain the above failures. | 03:48 |
| blockiv | please install the linux-headers-5.4.0-24-lowlatency package to fix this. | 03:48 |
| Bashing-om | blockiv: Yup ^^ make it so :D | 03:52 |
| blockiv | Bashing-om: Okay, installed that, re-ran the steps to remove nvidia and reinstall. This time do dkms warnings. I'll give it a reboot. | 03:55 |
| blockiv | *no dkms | 03:55 |
| Bashing-om | blockiv: A check ' sudp apt -f install ; sudo dpkg -C ' . | 03:57 |
| Bashing-om | sudo* | 03:57 |
| blockiv | No output on either Bashing-om | 03:57 |
| blockiv | Well, apt -f install said nothing to be done | 03:57 |
| Bashing-om | great! .. reboot time :D | 03:57 |
| blockiv | Cool, thanks. | 03:58 |
| blockiv | Lol Bashing-om can't seem to boot now. Not even to grub :/ | 04:05 |
| Bashing-om | blockiv: Now that makes no sense - low-latency kernel is this a studio install ? | 04:06 |
| blockiv | No, I just had the default 20.04 lts ISO. | 04:06 |
| blockiv | I have a spare 19.04 iso I can use to either just install that or chroot in to maybe regenerate grub | 04:07 |
| blockiv | I'll check the boot entry first I guess | 04:08 |
| blockiv | Bashing-om: is /EFI/Ubuntu/shimx64.efi the correct path for my boot entry? | 04:09 |
| Bashing-om | blockiv: The only thing I know that uses the low-latency kernel is ubuntu-studio. Might check what you think you installed. | 04:10 |
| Bashing-om | blockiv: /EFI/Ubuntu/shimx64.efi could be correct - different manufactures do things different - there is no EFI standard. | 04:11 |
| blockiv | Okay. Maybe I'll just reinstall and make sure I use the right iso, even tho I'm 95% sure it was a standard Ubuntu 20.04 iso. Seems like my only option now | 04:13 |
| blockiv | I'll do an offline install, then try installing Nvidia drivers from ubuntu-drivers | 04:16 |
| Bashing-om | blockiv: At this point - and time wise - a re-install will be much faster and surer. | 04:17 |
| blockiv | Yeah, I didn't have any personal data anyways. Either way, I really appreciate the help you've given me so far, Bashing-om lotuspsychje | 04:18 |
| lotuspsychje | 20.04 should have all the nvidia drivers on the install media now | 04:19 |
| lotuspsychje | i suspect a deeper issue here | 04:19 |
| Bashing-om | vlBoot directory I see /boot/efi/ubuntu/grubx64.efi as most common. | 04:19 |
| blockiv | Two weird things to at least note. Not sure about their impacts. When I first installed Ubuntu 20.04, everything went fine, and it was an offline install. I installed it again later, I forget the reason why, but towards the end of the install, it stopped with an error while upgrading new packages, since it was an online install. It then took me to a bug report where it was already documented, talking about there being | 04:43 |
| blockiv | an issue with online installs on Nvidia systems. I'd be glad to test it again, but it seems it's already a documented issue | 04:43 |
| blockiv | The install didn't complete, but it did at least get far enough where there was something to boot into | 04:44 |
| Bashing-om | blockiv: Boots a functional system ? | 05:41 |
| blockiv | Yeah | 05:44 |
| Bashing-om | blockiv: Good deal :) .. Our work here is done ? | 05:51 |
| blockiv | Nvidia-smi still isn't working, but im gonna stay were I'm at for now, gotta get some sleep soon | 05:56 |
| blockiv | Where* | 05:56 |
| Bashing-om | blockiv: same - sleep calleth :D ,, we can pick this up later :D | 06:00 |
| wingedrhino | Does the 20.04 Ubuntu come with NetworkManager support for WireGuard? I only see OpenVPN and Point-to-Point Tunneling Protocol when I go to Network > Settings. | 06:21 |
| wingedrhino | Also, if I install one of the nightlies on a laptop, can I assume that it'll be somewhat stable in a few days? | 06:23 |
| housecat | dunno about the first question. the answer to the second question is yes: | 06:23 |
| housecat | ubottu: final | 06:23 |
| ubottu | If you install a development version of Ubuntu Focal and keep up with package updates, then you will be upgraded to the official release of 20.04 when it comes out. To make sure, type « sudo apt update && sudo apt full-upgrade » in a terminal. | 06:23 |
| housecat | according to google, wireguard is supported in networkmanager >= 1.16, which is available in disco onwards | 06:25 |
| housecat | so if it's not showing up, either that info's wrong or there's a missing UI bit | 06:25 |
| wingedrhino | I'm going to google for images on how people have configured wireguard in the UI then. If it's been around since Disco, there's little chance the UI doesn't already have this! | 06:27 |
| wingedrhino | Aah from the looks of it the support for WireGuard is only present in NetworkManager itself. But not in any of the UIs - including nm-cli. | 06:40 |
| housecat | oh. well that's not very helpful :( | 06:40 |
| housecat | kinda hoping this sort of thing improves now that wireguard is in-kernel and 1.0 | 06:41 |
| wingedrhino | From the looks of it, KDE has GUIs for WireGuard already. | 07:23 |
| TheSilentLink | Hi anyone running Ubuntu 20.04 on the raspberry pi 4? For me none of the USB ports work making you unable to login in | 10:56 |
| Jordan_U | TheSilentLink: See this thread: https://www.raspberrypi.org/forums/viewtopic.php?t=270854&p=1644427 the work around seems to be limit yourself to only being able to use 3 of your 4 GiB of RAM, in exchange for working USB. | 10:57 |
| Jordan_U | TheSilentLink: It's hard for me to tell, but the fixed package may already have been released. So if you use the work around, then "sudo apt update && sudo apt upgrade" you might be able to remove the work around and reboot to a system with 4 usable GiB of RAM *and* working USB :) | 11:00 |
| TheSilentLink | Jordan_U: thanks. Also is rpi-eeprom-update available so we can update the eeprom without having to use raspbian? | 11:05 |
| Jordan_U | TheSilentLink: https://packages.ubuntu.com/ doesn't show a package by that name. | 11:08 |
| TheSilentLink | Oh ok also is cloud init required? | 11:09 |
| tomreyn | probably not | 13:27 |
| === TheSilentLink_ is now known as TheSilentLink | ||
| amitprakash | Why is ubuntu recognizing only 8 out of 16G ram? lshw outcome https://paste.ubuntu.com/p/sntz8hhfYh/ | free -m -> Mem: 7886 2024 4874 158 987 5420 | dmesg https://paste.ubuntu.com/p/vDShNDB4qZ/ | 16:15 |
| amitprakash | This is specific to Ubuntu (gentoo/windows work fine) | 16:16 |
| mfilipe[m] | i'm using terminal with 20.04 and i wanna change the tab title but gnome-terminal doesn't show this option up. when i use fedora 32, it shows (same gnome-terminal version 3.36). do you guys know what is wrong? | 17:09 |
| ChmEarl | whew, cat /etc/issue.net -> Ubuntu 20.04 LTS | 17:10 |
| oerheks | gnome-terminal --tab -e top -t "this is my funky tab title" | 17:11 |
| mfilipe[m] | oerheks: thanks is odd because this option exists in gnome-terminal but somehow ubuntu removes it | 17:12 |
| mfilipe[m] | that's* | 17:12 |
| oerheks | yes, it was removed in 3.12 something IIRC | 17:12 |
| mfilipe[m] | this reason makes the situation more strange... once fedora offers a vanilla version of gnome, so ubuntu hasn't this option because the ubuntu devs removed it? | 17:15 |
| mfilipe[m] | i use tmux-cssh a lot, if i maintain the dynamic title it is going to show strange titles | 17:16 |
| MikeRL | Anyone here have a recent enough Yubikey? I've just started tinkering with the 5 I just bought. Documentation is vague about pam config file locations, and I don't want to insert the text into the wrong one. | 17:44 |
| MikeRL | Following this: https://developers.yubico.com/yubico-pam/ | 17:44 |
| MikeRL | And also this: https://support.yubico.com/support/solutions/articles/15000011356-ubuntu-linux-login-guide-u2f | 17:45 |
| oerheks | add the ppa ppa:yubico/stable and install libpam-yubico yubikey-manager ? https://launchpad.net/~yubico/+archive/ubuntu/stable | 17:47 |
| oerheks | it has Focal packages already | 17:48 |
| MikeRL | Where I'm stuck at on the first URL is finding the right file in /etc/pam.d/ (for the first box under configuration) and I also don't know where PAM modules are stored on 20.04. | 17:49 |
| MikeRL | Already have PPA, thanks though. | 17:49 |
| MikeRL | For the second URL, I'm stuck where it says to add " authfile=/etc/Yubico/u2f_keys " to the PAM file's config. I don't know where the pam file is, since there appears to be a folder with pam files in /etc/pam.d/ | 17:50 |
| MikeRL | It does have focal packages. But I only saw one. | 17:51 |
| MikeRL | Most tools work with eoan or focal, but some do not, such as the configuration GUI. | 17:52 |
| MikeRL | It uses Qt4 which is gone from 20.04 default repos. | 17:53 |
| MikeRL | And it must be built from source for Linux. | 17:53 |
| MikeRL | But I'm not concerned about the GUI config tool as much as those two URLs and relevant questions. I can survive without the GUI. | 17:54 |
| amitprakash | resovled the ram issue, had to reseat | 18:29 |
| amitprakash | Journal is spammed w/ Apr 19 23:44:38 harran gnome-shell[1341]: failed to bind to /tmp/.X11-unix/X3684: No such file or directory | 18:29 |
| amitprakash | What gives? | 18:29 |
| Guma | Hello. I been running 20.04 with latest updates in vbox (6.1.6 r137129) with 3D acceleration enabled. Everything works pretty smooth except when I click on "Show Applications) in lower left correct of my screen then is draws very slow | 18:31 |
| Guma | See slow partial redraws. But everything finishes correct. | 18:33 |
| MikeRL | Think I figured it out by taking an educated guess. I have a backup if everything goes south. | 18:35 |
| MikeRL | Ugh. Still not working. | 18:40 |
| MikeRL | Luckily I saved the terminal window and edited it with sudo active. I simply removed the line. But problems persist. | 18:41 |
| tomreyn | Guma: that's most likely not an ubuntu issue, but one with (imperfect by design) shared graphics acceleration of Desktop virtualization. | 19:10 |
| theshagg | Should I expect weird issues with AMD cpu/gpus in 20.04 beta? | 19:32 |
| theshagg | or am I just unlucky? :P | 19:33 |
| tomreyn | theshagg: can you explain "weird issues"? were you running other linux versions previously? if so, did they occur there, too, or not? which hardware exactly? | 19:35 |
| theshagg | On shutdown I get soft lockups | 19:35 |
| tomreyn | are those logged? | 19:35 |
| theshagg | I think not | 19:36 |
| tomreyn | also, 2 more questions to answer ;) | 19:36 |
| theshagg | It actually might all be related to the amdgpu driver | 19:36 |
| theshagg | I filed a bug - there is a known issue in the kernel that is fixed in a newer kernel | 19:37 |
| tomreyn | do you think sharing the bug id might improve your support experience here? | 19:37 |
| theshagg | https://bugs.launchpad.net/bugs/1873547 | 19:37 |
| ubottu | Launchpad bug 1873547 in linux (Ubuntu) "amdgpu: [powerplay] Failed to send message errors" [Undecided,Confirmed] | 19:37 |
| theshagg | That one links to a thread | 19:38 |
| tomreyn | so you have those powerplay errors on your logs? | 19:38 |
| theshagg | when I connect more than 1 monitor, yes | 19:39 |
| theshagg | Supposedly this is fixed in 5.6-rc2 | 19:39 |
| tomreyn | does the system shutdown fine with just one mointor connected then? | 19:39 |
| theshagg | And I get corresponding graphics lockup for many seconds after login, and horrid graphics lockups after the monitor goes to sleep and wakes back up | 19:39 |
| theshagg | Right now I am trying to figure out my system shutdown issues. I am having some kind of shutdown lockup but it was being hidden by the "quiet splash" | 19:40 |
| theshagg | I'm currently trying to reproduce that error with "quite splash" disabled | 19:40 |
| theshagg | quiet | 19:40 |
| theshagg | I'm not seeing any shutdown lockups with "systemctl set-default multi-user.target" | 19:42 |
| theshagg | trying graphical.target with quiet splash disabled | 19:43 |
| tomreyn | is this board really new? i can't seem to convince the (very slow) MSI website to display any downloads for thie board, including manuals | 19:46 |
| theshagg | OK so I got it to hang with a softlockup on one of the CPUs | 19:46 |
| tomreyn | so it's a multi-cpu system? | 19:47 |
| theshagg | top of the stack trace is amdgpu_cgs_read_register+0x14/0x20 [amdgpu] | 19:47 |
| theshagg | Yes this is an AMD 2700x | 19:47 |
| theshagg | well, multi-core, single socket | 19:47 |
| theshagg | basically it's the same is this other bug I reported: | 19:48 |
| theshagg | https://bugs.launchpad.net/bugs/1873548 | 19:48 |
| ubottu | Launchpad bug 1873548 in linux (Ubuntu) "[amdgpu] hangs at shutdown" [Undecided,Confirmed] | 19:48 |
| woenx | Hi | 19:49 |
| woenx | One question. If i download and install the ubuntu 20.04 beta today, and I keep it updated, will it be the same thing as having the final version once it releases? | 19:49 |
| theshagg | I have been told yes, you will want to do an apt-get upgrade | 19:50 |
| theshagg | I'm not sure if you will get it during the point release, (months later), or immediately | 19:50 |
| woenx | Ok, but if I do that, it will be the same, right? | 19:50 |
| woenx | Ok | 19:50 |
| theshagg | I am a novice, but I'm in the same shoes as you | 19:50 |
| theshagg | I think you will have all of the same packages, etc. | 19:51 |
| CarlFK | woenx: there is no simple answer. if you are worried about it, you should wait. If you are up for helping, then do it. which might mean you are still helping after it is released. | 19:51 |
| tomreyn | theshagg: MSI.com just keeps returning "504 Gateway Time-out" messages to me, so i can't check whether you bios is current. if it isn't, i recommend upgrading it first of all. | 19:54 |
| theshagg | tomreyn - its up to date | 20:06 |
| theshagg | reasonably | 20:06 |
| theshagg | i had to update it to get virtualization / gpu passthrough working | 20:07 |
| theshagg | maybe ill try ubuntu 19 as a baseline | 20:11 |
| tomreyn | you mean ubuntu 19.10 ? that's what i'd try, yes. | 20:12 |
| tomreyn | your bios version is from november last year, but it seems there is nothing newer | 20:13 |
| tomreyn | which surprises me for an x470 | 20:13 |
| tomreyn | it's hard to check, though https://www.reddit.com/r/MSI_Gaming/comments/g49wmo/support_site_broken/ | 20:14 |
| theshagg | Yeah I updated it within the last month | 20:14 |
| theshagg | FWIW I had similar issues on fedora | 20:14 |
| theshagg | But I think the powerplay bug needs to get fixed | 20:15 |
| theshagg | before the official release | 20:15 |
| tomreyn | if the bug is upstream then that's not too likely to happen | 20:16 |
| tomreyn | there have been powerplay bugs with some of the newer GPUs for some months, though | 20:17 |
| theshagg | I don't know how ubuntu can have an LTS that isn't compatible with multiple monitors on AMD gpus | 20:20 |
| tomreyn | i'm not yet sure this is correct as a general statement. also, there'll be point releases, HWE kernels etc. | 20:22 |
| === halvors1 is now known as halvors | ||
| theshagg | tomreyn the bug is supposedly fixed in 5.6-rc2 | 21:09 |
| theshagg | So I assume the fix would just need to be backported? | 21:09 |
| lestac | o/ when 20.04 is released, we need to format or can update from current version? (18.04 in that case) | 21:12 |
| oerheks | lestac, prepare a fresh usb with the iso, then upgrade, but the upgrade path will be available when 20.04.1 is out | 21:14 |
| lestac | well! u rock | 21:15 |
| Synx_hm | Anybody experiencing issues installing vagrant on 20.04. Ive got virtualbox installed and vagrant installed via apt and I cannot vagrant-up | 21:15 |
| oerheks | i just upgraded a laptop to 20.04 with the -d option | 21:15 |
| oerheks | no problemo | 21:15 |
| Synx_hm | its erroring with /var/run/libvirt/libvirt-sock no such file or directory | 21:16 |
| theshagg | oerheks does it have AMD graphics? | 21:23 |
| oerheks | no, nvidia | 21:24 |
| howarth | Has anyone noticed that single user or rescue mode doesn't seem to work under the nvidia drivers? | 21:51 |
| howarth | On my machine, for both nvidia 340 and 440, I see the loading ramdisk line but it never goes past that | 21:52 |
| mfilipe[m] | how do i use screenshot on the fly with gnome 3.36? | 21:54 |
| tomreyn | theshagg: maybe bring it up in #ubuntu-kernel if you are convinced it affects a wide range of hardware and configurations. but i bet it's way too late for 20.04.0 | 22:18 |
| tomreyn | kernel patching can take place any time after release, though. | 22:19 |
| tomreyn | theshagg: i don't know how narvi and your two monitors will work without it, but you could try booting with kernel command line parameter amdgpu.powerplay=0 | 22:44 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!