Warning: Undefined variable $s_watching_topic_img in /usr/home/zenrat/public_html/tech/forums/viewtopic.php on line 677
|
[RESOLVED] System freeze with kernel 7.2.3-3-liquorix-amd64
Hello,
I would like to report an issue that occurred after upgrading the Liquorix kernel from: 7.2.3-2-liquorix-amd64 -> 7.2.3-3-liquorix-amd64. After booting with the new kernel, the system completely freezes. The system becomes unresponsive, and I am unable to interact normally with the graphical environment or switch to a virtual console. It also freezes when starting a test server in terminal mode. But It works fine on virtual machines. I had to revert to the previous kernel: 7.2.3-2-liquorix-amd64. With this version, the system works correctly. Thanks, Antonio --- System information: Operating System: Debian GNU/Linux forky/sid KDE Plasma Version: 6.7.4 KDE Frameworks Version: 6.28.0 Qt Version: 6.10.2 Kernel Version: 7.2.3-2-liquorix-amd64 (64-bit) Graphics Platform: Wayland Processors: 32 × Intel® Core™ i9-14900T Memory: 128 GiB of RAM (125.5 GiB usable) Graphics Processor: Intel® Graphics Manufacturer: ASRock Product Name: Z790M-ITX WiFi cmdline: audit=0 intel_pstate=disable amd_pstate=disable split_lock_detect=off initrd=\EFI\Linux\initrd.img root=LABEL=DEBIAN quiet nosplash vga=791 resume=LABEL=SWAP selinux=0 lsm=landlock,capability,yama,bpf Back to top |
|||||
|
With this kernel just updated I get a kernel panic instead and rebooting to previous kernel 7.2.3-2 works again.
System info: :: Code :: $ inxi -Fxxz
System: Kernel: 7.2.3-2-liquorix-amd64 arch: x86_64 bits: 64 compiler: gcc v: 14.2.0 Desktop: KDE Plasma v: 6.3.6 tk: Qt v: N/A wm: kwin_wayland dm: SDDM Distro: MX-25.2_KDE_x64 Infinity May 24 2026 base: Debian GNU/Linux 13 (trixie) Machine: Type: Laptop System: HP product: HP ProBook 640 G2 v: N/A serial: <superuser required> Chassis: type: 10 serial: <superuser required> Mobo: HP model: 80FD v: KBC Version 22.74 serial: <superuser required> part-nu: V1P74UT#ABA Firmware: UEFI vendor: HP v: N76 Ver. 01.62 date: 03/16/2024 Battery: ID-1: BAT0 charge: 30.9 Wh (80.6%) condition: 38.4/48 Wh (79.9%) volts: 12.33 min: 11.4 model: Hewlett-Packard Primary serial: <filter> charging: status: not charging cycles: 24 Device-1: hidpp_battery_0 model: Logitech ERGO M575 Trackball serial: <filter> charge: 100% status: discharging CPU: Info: dual core model: Intel Core i7-6600U bits: 64 type: MT MCP arch: Skylake rev: 3 cache: L1: 128 KiB L2: 512 KiB L3: 4 MiB Speed (MHz): avg: 2600 min/max: 400/2701 boost: enabled cores: 1: 2600 2: 2600 3: 2600 4: 2600 bogomips: 22399 Flags-basic: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx Graphics: Device-1: Intel Skylake GT2 [HD Graphics 520] vendor: Hewlett-Packard driver: i915 v: kernel arch: Gen-9 ports: active: DP-4 off: eDP-1 empty: DP-1, DP-2, DP-3, DP-5, HDMI-A-1, HDMI-A-2 bus-ID: 00:02.0 chip-ID: 8086:1916 Device-2: Chicony HP HD Camera driver: uvcvideo type: USB rev: 2.0 speed: 480 Mb/s lanes: 1 bus-ID: 1-9:5 chip-ID: 04f2:b51c Display: wayland server: X.org v: 1.21.1.16 with: Xwayland v: 24.1.9 compositor: kwin_wayland driver: X: loaded: modesetting unloaded: fbdev,vesa dri: iris gpu: i915 display-ID: 0 Monitor-1: DP-4 model: HP 24uh res: 1920x1080 hz: 60 dpi: 92 diag: 609mm (24") Monitor-2: eDP-1 model: AU Optronics 0x123d res: 1920x1080 dpi: 158 diag: 354mm (13.9") API: EGL v: 1.5 platforms: device: 0 drv: iris device: 1 drv: swrast gbm: drv: iris surfaceless: drv: iris wayland: drv: iris x11: drv: iris API: OpenGL v: 4.6 vendor: intel mesa v: 26.1.4-1~mx25ahs glx-v: 1.4 direct-render: yes renderer: Mesa Intel HD Graphics 520 (SKL GT2) device-ID: 8086:1916 display-ID: :1.0 API: Vulkan v: 1.4.309 surfaces: xcb,xlib,wayland device: 0 type: integrated-gpu driver: mesa intel device-ID: 8086:1916 device: 1 type: cpu driver: mesa llvmpipe device-ID: 10005:0000 Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo de: kscreen-console,kscreen-doctor wl: wayland-info x11: xdriinfo, xdpyinfo, xprop, xrandr Audio: Device-1: Intel Sunrise Point-LP HD Audio vendor: Hewlett-Packard driver: snd_hda_intel v: kernel bus-ID: 00:1f.3 chip-ID: 8086:9d70 API: ALSA v: k7.2.3-2-liquorix-amd64 status: kernel-api Server-1: PipeWire v: 1.6.8 status: active with: 1: pipewire-pulse status: active 2: wireplumber status: active 3: pipewire-alsa type: plugin 4: pw-jack type: plugin Network: Device-1: Intel Ethernet I219-LM vendor: Hewlett-Packard driver: e1000e v: kernel port: N/A bus-ID: 00:1f.6 chip-ID: 8086:156f IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <filter> Device-2: Intel Wireless 8260 driver: iwlwifi v: kernel pcie: speed: 2.5 GT/s lanes: 1 bus-ID: 02:00.0 chip-ID: 8086:24f3 IF: wlan0 state: down mac: <filter> Bluetooth: Device-1: Broadcom BCM20702A0 Bluetooth 4.0 driver: btusb v: 0.8 type: USB rev: 2.0 speed: 12 Mb/s lanes: 1 bus-ID: 1-4.4.1:7 chip-ID: 0a5c:21e8 Report: hciconfig ID: hci1 rfk-id: 2 state: up address: <filter> bt-v: 4.0 lmp-v: 6 sub-v: 220e Device-2: Intel Bluetooth wireless interface driver: btusb v: 0.8 type: USB rev: 2.0 speed: 12 Mb/s lanes: 1 bus-ID: 1-7:3 chip-ID: 8087:0a2b Report: ID: hci0 rfk-id: 1 state: up address: <filter> bt-v: 4.2 lmp-v: 8 sub-v: 100 Drives: Local Storage: total: 1.85 TiB used: 43.52 GiB (2.3%) ID-1: /dev/mmcblk0 model: SR32G size: 29.72 GiB type: Removable serial: <filter> ID-2: /dev/nvme0n1 vendor: Western Digital model: WDS100T3X0C-00SJG0 size: 931.51 GiB speed: 31.6 Gb/s lanes: 4 serial: <filter> temp: 49.9 C ID-3: /dev/sda vendor: Western Digital model: WD10SPZX-60Z10T0 size: 931.51 GiB speed: 6.0 Gb/s serial: <filter> Partition: ID-1: / size: 113.91 GiB used: 22.24 GiB (19.5%) fs: ext4 dev: /dev/nvme0n1p5 ID-2: /boot/efi size: 196 MiB used: 40.5 MiB (20.6%) fs: vfat dev: /dev/nvme0n1p1 ID-3: /home size: 245.02 GiB used: 21.24 GiB (8.7%) fs: ext4 dev: /dev/nvme0n1p6 Swap: ID-1: swap-1 type: partition size: 23.44 GiB used: 0 KiB (0.0%) priority: -1 dev: /dev/nvme0n1p7 Sensors: System Temperatures: cpu: 44.0 C pch: 38.5 C mobo: N/A Fan Speeds (rpm): N/A Info: Memory: total: 16 GiB note: est. available: 14.88 GiB used: 3.47 GiB (23.3%) Processes: 324 Power: uptime: 2m wakeups: 0 Init: systemd v: 257 default: graphical Packages: 3612 pm: dpkg pkgs: 3596 pm: flatpak pkgs: 16 Compilers: gcc: 14.2.0 Shell: Bash v: 5.2.37 running-in: konsole inxi: 3.3.41 Back to top |
|||||
|
The kernel parameters used from WORKING kernel (Note that the same parameters were used for the broken 7.2.3-3 kernel):
:: Code :: $ cat /proc/cmdline
audit=0 intel_pstate=disable amd_pstate=disable split_lock_detect=off BOOT_IMAGE=/boot/vmlinuz-7.2.3-2-liquorix-amd64 root=UUID=09add0aa-0130-46a4-a29d-e5809ed68234 ro quiet apparmor=1 security=apparmor Back to top |
|||||
|
[Regression] Liquorix 7.2.3-3 freezes / kernel panic – 7.2.3-2 works, tested with NVIDIA 610.57.04 proprietary and open modules
I am seeing reproducible freezes / kernel panics with Liquorix 7.2.3-3 (7.2-5ubuntu1~resolute) on Kubuntu 26.04.1.
The previous kernel 7.2.3-2 (7.2-4ubuntu1~resolute) works on the same system. My system: Kubuntu 26.04.1 KDE Plasma / Wayland Fujitsu D3348-B1 Intel Xeon E5-2666 v3 NVIDIA GeForce RTX 3050 8 GB NVIDIA driver 610.57.04 Initially I was using the proprietary NVIDIA 610.57.04 kernel module. With Liquorix 7.2.3-3 I experienced hard freezes and kernel panics. To rule out the proprietary NVIDIA kernel module as the cause, I switched the complete NVIDIA stack to the NVIDIA Open Kernel Modules, while keeping exactly the same driver version: 610.57.04. I verified the Open driver installation successfully: $ modinfo -F license nvidia Dual MIT/GPL $ modinfo -F version nvidia 610.57.04 DKMS successfully built NVIDIA 610.57.04 for both kernels, including: nvidia/610.57.04, 7.2.3-2-liquorix-amd64, x86_64: installed nvidia/610.57.04, 7.2.3-3-liquorix-amd64, x86_64: installed dpkg --audit is clean. Booting 7.2.3-2 with the NVIDIA Open Kernel Module works: $ uname -r 7.2.3-2-liquorix-amd64 $ cat /proc/driver/nvidia/version NVRM version: NVIDIA UNIX Open Kernel Module for x86_64 610.57.04 $ nvidia-smi --query-gpu=name,driver_version --format=csv,noheader NVIDIA GeForce RTX 3050, 610.57.04 I then booted 7.2.3-3 again with the NVIDIA Open Kernel Module. The problem still occurs. PStore captured a new crash from the 7.2.3-3 boot. Before the panic there is already an RCU stall: rcu: INFO: rcu_preempt detected expedited stalls on CPUs/tasks RIP: sha512_transform_rorx Call Trace: sha512_blocks_avx2 __sha512_update sha512 crypto_sha512_digest crypto_shash_digest pkcs7_digest pkcs7_verify verify_pkcs7_signature mod_verify_sig module_sig_check load_module A little later the kernel crashes in the scheduler/timer path: BUG: unable to handle page fault for address: ffffffffab574ac0 RIP: _raw_spin_trylock+0x10/0x50 Call Trace: sched_tick+0x45c/0x5d0 update_process_times+0xb6/0xe0 tick_nohz_handler+0x8b/0x1d0 __hrtimer_run_queues+0x131/0x270 hrtimer_interrupt+0xf3/0x2d0 __sysvec_apic_timer_interrupt+0x58/0x110 sysvec_apic_timer_interrupt+0x38/0x90 asm_sysvec_apic_timer_interrupt+0x1a/0x20 followed by: Kernel panic - not syncing: Fatal exception in interrupt The kernel reports: 7.2.3-3-liquorix-amd64 liquorix 7.2-5ubuntu1~resolute The important part is that I have now reproduced the 7.2.3-3 failure with both NVIDIA kernel module variants: 7.2.3-2 + NVIDIA 610.57.04 proprietary -> works 7.2.3-2 + NVIDIA 610.57.04 open -> works 7.2.3-3 + NVIDIA 610.57.04 proprietary -> freezes / kernel panic 7.2.3-3 + NVIDIA 610.57.04 open -> freezes / kernel panic So switching from the proprietary NVIDIA module to the Open Kernel Module does not change the result. Since 7.2.3-2 works on the exact same hardware and NVIDIA driver version, while 7.2.3-3 reproduces the scheduler/timer panic with both NVIDIA module variants, this looks like a regression introduced in Liquorix 7.2.3-3 / package 7.2-5, rather than an NVIDIA proprietary-vs-open issue. I also noticed that other users in the forum are reporting freezes after installing 7.2.3-3, so this does not appear to be limited to my system. Back to top |
|||||
|
Re: [Regression] Liquorix 7.2.3-3 freezes / kernel panic – 7.2.3-2 works, tested with NVIDIA 610.57.04 proprietary and open modules
:: Procus wrote :: I am seeing reproducible freezes / kernel panics with Liquorix 7.2.3-3 (7.2-5ubuntu1~resolute) on Kubuntu 26.04.1.
The previous kernel 7.2.3-2 (7.2-4ubuntu1~resolute) works on the same system. My system: Kubuntu 26.04.1 KDE Plasma / Wayland Fujitsu D3348-B1 Intel Xeon E5-2666 v3 NVIDIA GeForce RTX 3050 8 GB NVIDIA driver 610.57.04 Initially I was using the proprietary NVIDIA 610.57.04 kernel module. With Liquorix 7.2.3-3 I experienced hard freezes and kernel panics. To rule out the proprietary NVIDIA kernel module as the cause, I switched the complete NVIDIA stack to the NVIDIA Open Kernel Modules, while keeping exactly the same driver version: 610.57.04. I verified the Open driver installation successfully: $ modinfo -F license nvidia Dual MIT/GPL $ modinfo -F version nvidia 610.57.04 DKMS successfully built NVIDIA 610.57.04 for both kernels, including: nvidia/610.57.04, 7.2.3-2-liquorix-amd64, x86_64: installed nvidia/610.57.04, 7.2.3-3-liquorix-amd64, x86_64: installed dpkg --audit is clean. Booting 7.2.3-2 with the NVIDIA Open Kernel Module works: $ uname -r 7.2.3-2-liquorix-amd64 $ cat /proc/driver/nvidia/version NVRM version: NVIDIA UNIX Open Kernel Module for x86_64 610.57.04 $ nvidia-smi --query-gpu=name,driver_version --format=csv,noheader NVIDIA GeForce RTX 3050, 610.57.04 I then booted 7.2.3-3 again with the NVIDIA Open Kernel Module. The problem still occurs. PStore captured a new crash from the 7.2.3-3 boot. Before the panic there is already an RCU stall: rcu: INFO: rcu_preempt detected expedited stalls on CPUs/tasks RIP: sha512_transform_rorx Call Trace: sha512_blocks_avx2 __sha512_update sha512 crypto_sha512_digest crypto_shash_digest pkcs7_digest pkcs7_verify verify_pkcs7_signature mod_verify_sig module_sig_check load_module A little later the kernel crashes in the scheduler/timer path: BUG: unable to handle page fault for address: ffffffffab574ac0 RIP: _raw_spin_trylock+0x10/0x50 Call Trace: sched_tick+0x45c/0x5d0 update_process_times+0xb6/0xe0 tick_nohz_handler+0x8b/0x1d0 __hrtimer_run_queues+0x131/0x270 hrtimer_interrupt+0xf3/0x2d0 __sysvec_apic_timer_interrupt+0x58/0x110 sysvec_apic_timer_interrupt+0x38/0x90 asm_sysvec_apic_timer_interrupt+0x1a/0x20 followed by: Kernel panic - not syncing: Fatal exception in interrupt The kernel reports: 7.2.3-3-liquorix-amd64 liquorix 7.2-5ubuntu1~resolute The important part is that I have now reproduced the 7.2.3-3 failure with both NVIDIA kernel module variants: 7.2.3-2 + NVIDIA 610.57.04 proprietary -> works 7.2.3-2 + NVIDIA 610.57.04 open -> works 7.2.3-3 + NVIDIA 610.57.04 proprietary -> freezes / kernel panic 7.2.3-3 + NVIDIA 610.57.04 open -> freezes / kernel panic So switching from the proprietary NVIDIA module to the Open Kernel Module does not change the result. Since 7.2.3-2 works on the exact same hardware and NVIDIA driver version, while 7.2.3-3 reproduces the scheduler/timer panic with both NVIDIA module variants, this looks like a regression introduced in Liquorix 7.2.3-3 / package 7.2-5, rather than an NVIDIA proprietary-vs-open issue. I also noticed that other users in the forum are reporting freezes after installing 7.2.3-3, so this does not appear to be limited to my system. Update: Liquorix 7.2.3-4 (7.2-6ubuntu1~resolute) is now working on the same system. According to the official Liquorix changelog, 7.2-6 reverts the Project-C patches from 7.2-5. The broken 7.2.3-3 build corresponds to Liquorix 7.2-5, which introduced additional sched/alt SMT scheduling changes. This matches the PStore crash I reported in _raw_spin_trylock -> sched_tick -> tick_nohz_handler. With 7.2.3-4 and NVIDIA 610.57.04 Open Kernel Modules, the system boots and runs normally again. Back to top |
|||||
|
Liquorix kernel 7.2.3-4 came in and was able to boot to desktop on my laptop with Debian Stable successfully with no lockups or anything.
@Procus, try this kernel see if it resolves your own issue. Back to top |
|||||
|
So then it's that the scheduler broke again with those patches for 7.2.3-3 and reverted by -4. Hopefully the scheduler changes could be reworked to avoid crashes.
Back to top |
|||||
|
Yes, 7.2.3-3 had a bug in an optimization for the scheduler. I pushed out a fixed kernel same day by reverting those patches.
I'm working on testing a rewrite of the optimizations before pushing it out to the broader audience. I'll mark this thread as resolved. Back to top |
|||||
|
All times are GMT - 8 Hours
|
|||||