You are not logged in.
Аble to boot only with 'nomodeset' parameter in GRUB, otherwise getting black screen with 'No signal'.
Intel XE graphics (NUC), standard setup with systemd, X, mesa, GDM, Gnome, Xfce4. Followed the installation guide, used only standard and recommended packages.
Been working for years with this or similar Arch config or various machines (1+ year work this particular one), including laptops, using native and external monitors. In fact, my laptop boots fine after upgrade, with same config...
(Dual) boot to Windows works fine, so it's not a hardware/cable problem. What freaks me out, occasionally Linux also boots, even after the trouble had appeared...
Tried (with nomodeset param) to rebuild init with mkinicpio, seemed to cure the problem, but then failed at next boot.
Would someone kindly help diagnose and fix the problem, or how to quick fix it until the working package is issued? Maybe to clear some static settings so they would be automatically renewed?
Offline
Enable REISUB: https://wiki.archlinux.org/title/Keyboa … el_(SysRq) boot into the faulty kernel, use the REISUB sequence to safely reboot, boot with nomodeset and post the journal of the failing boot with
sudo journalctl -b-1https://wiki.archlinux.org/title/List_o … n_services
Consider installing the LTS kernel so you can continue having a working graphical setup.
FWIW potentially related: https://bbs.archlinux.org/viewtopic.php?id=293976
Offline
here it is...
Offline
sudo journalctl -b-1
Please have a look.
Offline
Consider installing the LTS kernel so you can continue having a working graphical setup.
Should have used it as a rule.
Thanks for your attention to my question.
Offline
бер 20 18:24:20 aPanther kernel: ------------[ cut here ]------------
бер 20 18:24:20 aPanther kernel: i915 0000:00:02.0: drm_WARN_ON(!connector->dp.dsc_decompression_aux || !connector->dp.dsc_decompression_enabled)
бер 20 18:24:20 aPanther kernel: WARNING: CPU: 7 PID: 163 at drivers/gpu/drm/i915/display/intel_dp.c:3193 intel_dp_sink_disable_decompression+0x63/0xf0 [i915]
бер 20 18:24:20 aPanther kernel: Modules linked in: i915(+) i2c_algo_bit sdhci_pci drm_buddy ttm cqhci nvme intel_gtt sdhci drm_display_helper spi_intel_pci nvme_core crc32c_intel mmc_core xhci_pci spi_intel nvme_auth cec xhci_pci_renesas video wmi
бер 20 18:24:20 aPanther kernel: CPU: 7 PID: 163 Comm: (udev-worker) Not tainted 6.8.1-arch1-1 #1 52f97d9bb37be6168651745a1a9f8f7240d21ce5
бер 20 18:24:20 aPanther kernel: Hardware name: Intel(R) Client Systems NUC11PAHi5/NUC11PABi5, BIOS PATGL357.0048.2022.0801.1148 08/01/2022
бер 20 18:24:20 aPanther kernel: RIP: 0010:intel_dp_sink_disable_decompression+0x63/0xf0 [i915]https://bbs.archlinux.org/viewtopic.php … 0#p2158430
Re-plug the cable.
Offline
Re-plug the cable.
Maybe that's some connectivity test that fails with new kernel.
Faced same trouble at another connection spot (monitor/cable).
Here's what current boot journal says, with linux-lts.
Also, Windows boot runs smoothly...
Offline
Maybe not.
Is it HDMI?
There's a hyper-fragile handshake involved that doesn't tolerate any delay during the handshake and there's a bogus MST related test/warning during the init.
HDMI/DP "cables" aren't just some copper, there's a minicomputer in each of the plugs.
1. does the LTS kernel work?
2. does re-plugging the cable work?
Offline
Maybe not.
Is it HDMI?
There's a hyper-fragile handshake involved that doesn't tolerate any delay during the handshake and there's a bogus MST related test/warning during the init.
HDMI/DP "cables" aren't just some copper, there's a minicomputer in each of the plugs.1. does the LTS kernel work?
2. does re-plugging the cable work?
1. Yes, it works just fine.
2. I use this machine at two different spots, plugging and replugging. 'No signal' on both of them after last kernel upgrade.
I use HDMI cables provided with the monitors.
After downgrade to linux-lts, all works fine.
Boot logs:
6.6.22-1-lts
6.8.1-arch1-1
If it helps the community to trace the problem to avoid it in future releases.
Offline
This here is the warning that *maybe* tumbles hdcp
бер 20 18:24:20 aPanther kernel: WARNING: CPU: 7 PID: 163 at drivers/gpu/drm/i915/display/intel_dp.c:3193 intel_dp_sink_disable_decompression+0x63/0xf0 [i915]but the journal is only 28s and doesn't seem to contain a re-plug event?
After re-plugging the cable, does the output show up in xrandr or
for OUT in /sys/class/drm/card*; do echo $OUT; edid-decode $OUT/edid; echo "================="; doneYou'll need https://aur.archlinux.org/packages/edid-decode-git
Edit: you btw. do not want to use the xf86-video-intel driver w/ a tigerlake chip, but if you'd insist still set "DRI" to "iris"
бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (EE) AIGLX error: dlopen of /usr/lib/dri/i965_dri.so failed (/usr/lib/dri/i965_dri.so: cannot open shared object file: No such file or directory)
бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (EE) AIGLX error: unable to load driver i965
бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (II) IGLX: Loaded and initialized swrast
бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (II) GLX: Initialized DRISWRAST GL provider for screen 0https://wiki.archlinux.org/title/Intel_ … ecent_GPUs
Last edited by seth (2024-03-22 07:49:30)
Offline
do not want to use the xf86-video-intel driver w/ a tigerlake chip, but if you'd insist still set "DRI" to "iris"
бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (EE) AIGLX error: dlopen of /usr/lib/dri/i965_dri.so failed (/usr/lib/dri/i965_dri.so: cannot open shared object file: No such file or directory) бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (EE) AIGLX error: unable to load driver i965 бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (II) IGLX: Loaded and initialized swrast бер 20 20:28:22 aPanther /usr/lib/gdm-x-session[1006]: (II) GLX: Initialized DRISWRAST GL provider for screen 0
This something out of my control currently.
Will dive into it )
Offline
/etc/X11/xorg.conf.d/20-intel.conf
Section "Device"
Identifier "Intel Graphics"
Driver "intel"
Option "DRI" "iris"
EndSectionhttps://pastebin.com/BdWujmT4
Visually didn't see any difference, I believe something good happened ![]()
Last edited by sheraff (2024-03-23 10:25:37)
Offline
This here is the warning that *maybe* tumbles hdcp
бер 20 18:24:20 aPanther kernel: WARNING: CPU: 7 PID: 163 at drivers/gpu/drm/i915/display/intel_dp.c:3193 intel_dp_sink_disable_decompression+0x63/0xf0 [i915]but the journal is only 28s and doesn't seem to contain a re-plug event?
After re-plugging the cable, does the output show up in xrandr or
for OUT in /sys/class/drm/card*; do echo $OUT; edid-decode $OUT/edid; echo "================="; doneYou'll need https://aur.archlinux.org/packages/edid-decode-git
you btw. do not want to use the xf86-video-intel driver w/ a tigerlake chip
Offline