LuKaRo via arch-general
Hi everyone,
since yesterday's update, my Xorg crashes with a segmentation fault as soon as I start an electron application and have a second monitor activated. This is reproducible, a minimum working example looks like this:
$ rocketchat-desktop $ xrandr --output DP-1-1 --auto --above eDP-1 --rotate normal $ rocketchat-desktop --disable-gpu $ rocketchat-desktop
Starting rocketchat-desktop before activating the external screen works without error, as does starting rocketchat-desktop without GPU acceleration after activating the external screen. However, as soon as I start rocketchat-desktop normally (e.g. with GPU acceleration) Xorg immediately crashes, tty1 appears for a moment, and then my lightdm-gtk-greeter returns, allowing me to log in again. The same thing happens for example with signal-desktop or element-desktop. An example Xorg.log ends like this:
[ 16805.080] (II) modeset(0): Allocate new frame buffer 1920x2160 stride [ 16805.230] (II) NVIDIA(G0): Setting mode "DP-1-1: nvidia-auto-select @1920x1080 +0+0 {AllowGSYNC=Off, ViewPortIn=1920x1080, ViewPortOut=1920x1080+0+0}" [ 16808.392] (EE) [ 16808.392] (EE) Backtrace: [ 16808.392] (EE) 0: /usr/lib/Xorg (xorg_backtrace+0x5b) [0x55c4bab6457b] [ 16808.392] (EE) 1: /usr/lib/Xorg (0x55c4baa1e000+0x1511d5) [0x55c4bab6f1d5] [ 16808.392] (EE) 2: /usr/lib/libc.so.6 (0x7f8962b82000+0x3cda0) [0x7f8962bbeda0] [ 16808.392] (EE) 3: /usr/lib/Xorg (0x55c4baa1e000+0xd81f2) [0x55c4baaf61f2] [ 16808.392] (EE) 4: /usr/lib/Xorg (0x55c4baa1e000+0xd9d1c) [0x55c4baaf7d1c] [ 16808.392] (EE) 5: /usr/lib/Xorg (0x55c4baa1e000+0xd8dc6) [0x55c4baaf6dc6] [ 16808.392] (EE) 6: /usr/lib/Xorg (0x55c4baa1e000+0x398ac) [0x55c4baa578ac] [ 16808.392] (EE) 7: /usr/lib/libc.so.6 (__libc_start_main+0xd5) [0x7f8962ba9b25] [ 16808.392] (EE) 8: /usr/lib/Xorg (_start+0x2e) [0x55c4baa57cee] [ 16808.392] (EE) [ 16808.392] (EE) Segmentation fault at address 0x68 [ 16808.393] (EE) Fatal server error: [ 16808.393] (EE) Caught signal 11 (Segmentation fault). Server aborting [ 16808.393] (EE) [ 16808.393] (EE) Please consult the The X.Org Foundation support ???????????????? at http://wiki.x.org ??for help. [ 16808.393] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information. [ 16808.393] (EE) [ 16808.393] (II) AIGLX: Suspending AIGLX clients for VT switch [ 16808.450] (EE) Server terminated with error (1). Closing log file.
Hardware is a Lenovo ThinkPad P53 notebook with dual graphics:
$ lspci|grep -i vga 00:02.0 VGA compatible controller: Intel Corporation CoffeeLake-H GT2 [UHD Graphics 630] 01:00.0 VGA compatible controller: NVIDIA Corporation TU117GLM [Quadro T1000 Mobile] (rev a1)
Kernel is the latest Linux:
$ uname -a Linux 5.12.15-arch1-1 #1 SMP PREEMPT Wed, 07 Jul 2021 23:35:29 +0000 x86_64 GNU/Linux
And I'm using the NVIDIA driver:
$ pacman -Qe|grep nvidia nvidia 465.31-14 nvidia-lts 1:465.31-12 nvidia-prime 1.0-4 nvidia-settings 465.31-1
I also downgraded to the earliest kernel and nvidia versions I had available from late May:
[2021-07-12T18:25:56+0200] [ALPM] downgraded linux-lts (5.10.42-1 -> 5.10.39-1) [2021-07-12T18:25:59+0200] [ALPM] downgraded linux-lts-headers (5.10.42-1 -> 5.10.39-1) [2021-07-12T18:26:00+0200] [ALPM] downgraded linux (5.12.9.arch1-1 -> 5.12.6.arch1-1) [2021-07-12T18:26:02+0200] [ALPM] downgraded linux-headers (5.12.9.arch1-1 -> 5.12.6.arch1-1) [2021-07-12T18:26:03+0200] [ALPM] downgraded linux-hardened (5.12.14.hardened1-1 -> 5.11.22.hardened1-1) [2021-07-12T18:26:06+0200] [ALPM] downgraded linux-hardened-headers (5.12.14.hardened1-1 -> 5.11.22.hardened1-1) [2021-07-12T18:26:06+0200] [ALPM] downgraded nvidia-lts (1:465.31-7 -> 1:465.31-3) [2021-07-12T18:26:06+0200] [ALPM] downgraded nvidia (465.31-7 -> 465.31-4)
However, the issue still persits. Any ideas on how to further diagnose this issue? I can provide more information, if needed. I'll try switching Also, is anyone having a similar issue?
Kind regards,
Lukas
I wonder if this is somewhat similar to my issue, https://lists.archlinux.org/pipermail/arch-general/2021-July/049097.html . Can you: 1. Start electron from a terminal and request electron to run in another desktop. Desktop in the window manager sense. Don't switch to the other desktop yet. Are there any error messages in the terminal? If you kill electron before switching to the desktop it should run in, does the xserver contiues in an apparent normal operation? 2. Does the xserver dumps core? Can you post gdb analyses for the core dump? 3. Do you use electron often enough in order to make an educated guess in which version of the xorg server the problem appeared for the 1st time? ? -- u34