ScaredySquirrel | hrm how would I get the latest firefox version 72 in focal? | 04:47 |
---|---|---|
ScaredySquirrel | it's at version 72 in Ubuntu stable eoan | 04:48 |
Bashing-om | !info firefox focal | ScaredySquirrel | 04:53 |
ubottu | ScaredySquirrel: firefox (source: firefox): Safe and easy web browser from Mozilla. In component main, is optional. Version 71.0+build5-0ubuntu1 (focal), package size 49179 kB, installed size 192698 kB | 04:53 |
guiverc | ScaredySquirrel, https://snapcraft.io/firefox shows ver 72.0.1-1 available from Mozilla via snap | 06:55 |
kblin | hi folks | 10:11 |
kblin | I'm currently playing with the 20.04 nightly on an HP Elitebook x360 with an Intel UHD 620 card, and I'm seeing some really bad screen tearing, especially when scrolling | 10:12 |
kblin | I don't have that issue in a Debian 10 live image, but that in turn doesn't like the WiFi | 10:12 |
kblin | looking at the output from "inxi -G", it seems like on the Debian live image, I'm using the modesetting driver for X11, and on focal I'm using the i915 driver | 10:14 |
tomreyn | kblin: have you consdiered 19.10 (20.04 is in development still) | 10:14 |
kblin | I have the same issue in 19.10, and tried 20.04 because some forum posts suggested that the problem might have been fixed on a more recent kernel | 10:15 |
tomreyn | you'll be using i915 on both | 10:15 |
kblin | funnily enough, I can actually make the problem appear on a Debian 10 install if I switch to a 5.x series kernel :) | 10:15 |
kblin | in any case I'm not ready to use this machine for production anyway, so I figured that any fix I found would also be applicable to 19.10 | 10:16 |
tomreyn | this was discussed on #ubuntu yesterday - could be related, but a workaround is in anything but focal: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1845584 | 10:16 |
ubottu | Launchpad bug 1845584 in linux (Ubuntu Focal) "intel-lpss driver conflicts with write-combining MTRR region" [Undecided,In progress] | 10:16 |
tomreyn | i'm not sure this is the problem you're seeing, it *could* be | 10:17 |
kblin | that looks like a "fails to boot" issue, my system is booting just fine | 10:18 |
kblin | I'm not seeing the issue if I use the live system in "safe graphics" mode | 10:18 |
tomreyn | good point, yes. | 10:18 |
kblin | so to me this really sounds like an X issue more than a kernel thing | 10:19 |
tomreyn | is anything logged? | 10:19 |
kblin | I've tried uninstalling the xserver-xorg-video-intel package, but somehow I'm still on the i915 driver | 10:19 |
tomreyn | xserver-xorg-video-intel is the 2d graphics driver for Xorg, you need it. i915 is the linux kernel driver for supporting intel graphics in the first place, you need it | 10:20 |
kblin | where's the X log file these days? shouldn't it be in /var/log somewhere? | 10:21 |
tomreyn | actually seems like i'm misinformed, sorry. you don't need xserver-xorg-video-intel | 10:21 |
kblin | I mainly uninstalled that because the package description said "you don't need this on hardware from 2017 and newer" | 10:22 |
kblin | I'm mainly noticing how clueless I've gotten on X11 configuration, because all my stuff has just been working fine for so many years :) | 10:23 |
tomreyn | X logs to syslog / the systemd journal, it may addiitonally log to either ~/.local/share/xorg/ or /var/log/Xorg/ | 10:23 |
tomreyn | if you use lspci -k | grep -A3 VGA you'll notice the hardware itself is actually driven by i915 | 10:24 |
tomreyn | so this you won't be able to replace | 10:25 |
kblin | yeah, both the kernel line and the X line claim i915 | 10:25 |
tomreyn | your only option there is to fiddle with kernel parameters or to try different kernel versions | 10:25 |
kblin | I mainly want to try getting xorg to use the modesetting driver | 10:26 |
tomreyn | if you'd like to post your system journal for the current boot i can take a look to see if i spot anything unusual. | 10:26 |
tomreyn | journalctl -b | nc termbin.com 9999 | 10:27 |
tomreyn | X should be using modesetting on i915 by default for a good while already | 10:27 |
tomreyn | definitely with anything including and past ubuntu 18.04, more likely since 14.04 or 16.04 | 10:28 |
tomreyn | actually earlier, intel was amongst the first to do it | 10:28 |
kblin | so what's the "safe graphics" mode for the live image doing, because as I mentioned that works just fine | 10:29 |
tomreyn | it disables kernel mode setting | 10:29 |
tomreyn | you can always press 'e' at the grub menu to inspect the "linux" line and which parameters are passed to it | 10:30 |
tomreyn | the "safe graphics" boot option adds the "nomodeset" kernel parameter (on top of the other parameters also used by the default grub menu option) | 10:32 |
kblin | I'll give that a try, for giggles | 10:36 |
kblin | also because the screen refresh issue makes any terminal use super annoying :/ | 10:36 |
kblin | ok, the "nomodeset" kernel option "fixes" it, but of course at the cost of not having a proper graphics driver | 10:44 |
kblin | funny, now I've set a custom config to try the uxa renderer for the intel driver, and X11 seems to have decided to use the fbdev driver instead | 11:07 |
kblin | but it looks like the uxa renderer is enabled and fixes the issues I'm seeing | 11:11 |
tomreyn | there's the TearFree option to X(org), but i think such issues usually point to the graphics drrievr rather | 11:26 |
tomreyn | *driver | 11:26 |
tomreyn | btw. there are several models in this hp elitebook x360 series, with different hardware: https://i.imgur.com/YA4zz5n.png | 11:32 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!