Public bug reported: Ubuntu 26.04 LTS on a Lenovo IdeaPad 320-15IKB consistently hits a kernel panic when entering suspend while the proprietary NVIDIA 580 driver stack is loaded. System: - Ubuntu 26.04 LTS (resolute) - Kernel: 7.0.0-30-generic, package version 7.0.0-30.30 - CPU: Intel Core i7-7500U (Kaby Lake) - Integrated GPU: Intel HD Graphics 620 [8086:5916], driver i915 - Discrete GPU: NVIDIA GM108M [GeForce 920MX] [10de:134f] - NVIDIA driver: 580.173.02-0ubuntu0.26.04.1 - linux-modules-nvidia-580-7.0.0-30-generic: 7.0.0-30.30 - Secure Boot enabled - Suspend mode: deep (S3) Problem: With the NVIDIA driver stack loaded, running: sudo systemctl suspend causes the graphical session to switch briefly to a text/console screen and then the system enters a kernel panic. The screen displays: KERNEL PANIC! Please reboot your computer. System is deadlocked on memory The Caps Lock LED flashes continuously, networking is lost, SSH becomes unreachable, and the machine requires a forced power-off. Closing the laptop lid also reproduced an unrecoverable suspend/resume failure when the NVIDIA stack was active. Expected result: The laptop should enter S3/deep suspend and resume normally when a keyboard key is pressed or the lid is reopened. Actual result: With the NVIDIA stack active, suspend can result in a kernel panic / unrecoverable system state. Important A/B test: I performed a controlled test using exactly the same Ubuntu installation and kernel 7.0.0-30, but completely blocked the NVIDIA modules at kernel boot using: module_blacklist=nvidia,nvidia_drm,nvidia_modeset,nvidia_uvm,nouveau After reboot I confirmed: - no nvidia or nouveau modules were loaded; - Intel HD 620 / i915 was the active graphics driver; - the NVIDIA 920MX was present on PCI but had no "Kernel driver in use". With NVIDIA completely disabled: 1. `sudo systemctl suspend` - entered deep suspend successfully; - resumed immediately after pressing Enter; - display returned normally; - login screen returned; - fan behaviour was normal; - Wi-Fi/networking and SSH recovered normally; - no panic occurred. 2. Lid-close suspend - `systemd-logind` detected Lid closed; - kernel entered `PM: suspend entry (deep)`; - reopening the lid resumed the system normally; - kernel logged `PM: suspend exit`; - networking and display recovered normally; - no panic occurred. Relevant successful-suspend log excerpts with NVIDIA blacklisted: Module nvidia is blacklisted PM: suspend entry (deep) PM: suspend exit and for lid suspend: systemd-logind: Lid closed. PM: suspend entry (deep) systemd-logind: Lid opened. PM: suspend exit This strongly indicates that the failure depends on the NVIDIA driver stack being active, rather than on ACPI lid detection or generic S3 suspend support. Additional diagnostics already performed: - ACPI lid switch correctly reports both `state: closed` and `state: open`. - Kernel PM debug `pm_test=freezer` completed successfully. - Direct PM device testing showed the expected NVIDIA error when bypassing the NVIDIA procfs suspend interface with PreserveVideoMemoryAllocations enabled; this was a diagnostic test and is distinct from the systemctl suspend panic. - NVIDIA suspend/resume systemd services are installed. - `PreserveVideoMemoryAllocations: 1`. - kdump is installed and loaded. - The original panic did not produce a vmcore. kdump memory reservation was subsequently increased according to the kdump-tools estimator, but no further panic has intentionally been triggered. - No evidence currently suggests a physical GPU failure. There are repeated i915 HPD polling workqueue warnings during some boots, but these are not considered proven causal and successful suspend/resume with i915 alone demonstrates that the Intel GPU can suspend correctly on this machine. A similar Ubuntu 26.04 / Linux 7.0 / NVIDIA 580 issue has been reported where suspend produces: "Kernel panic - not syncing: System is deadlocked on memory". There are also reports involving NVIDIA GM108M hardware and the NVIDIA suspend path on Linux 7.0.x. I am currently keeping the NVIDIA GPU disabled as a workaround because the system suspends and resumes reliably in that configuration. I can perform additional controlled tests if requested by Ubuntu kernel developers, including testing an older Ubuntu 7.0.0-14 kernel with the NVIDIA 580 modules available from the official resolute repositories. ProblemType: Bug DistroRelease: Ubuntu 26.04 Package: linux-image-7.0.0-30-generic 7.0.0-30.30 ProcVersionSignature: Ubuntu 7.0.0-30.30-generic 7.0.12 Uname: Linux 7.0.0-30-generic x86_64 ApportVersion: 2.34.1-0ubuntu0.1 Architecture: amd64 AudioDevicesInUse: USER PID ACCESS COMMAND /dev/snd/controlC0: diogo 3661 F.... pipewire diogo 3678 F.... wireplumber /dev/snd/pcmC0D0p: diogo 3661 F...m pipewire /dev/snd/seq: diogo 3661 F.... pipewire CasperMD5CheckResult: pass CurrentDesktop: ubuntu:GNOME Date: Mon Aug 24 20:00:16 2026 InstallationDate: Installed on 2026-08-20 (4 days ago) InstallationMedia: Ubuntu 26.04 "Resolute Raccoon" - Release amd64 (20260423.1) Lsusb: Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 001 Device 002: ID 8087:0a2a Intel Corp. Bluetooth wireless interface Bus 001 Device 003: ID 5986:210f Bison Electronics Inc. EasyCamera Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub MachineType: LENOVO 80XL ProcEnviron: LANG=pt_PT.UTF-8 PATH=(custom, no user) SHELL=/bin/bash TERM=xterm-256color XDG_RUNTIME_DIR=<set> ProcFB: 0 i915drmfb ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-7.0.0-30-generic root=/dev/mapper/ubuntu--vg-ubuntu--lv ro quiet splash crashkernel=1200M module_blacklist=nvidia,nvidia_drm,nvidia_modeset,nvidia_uvm,nouveau SourcePackage: linux UpgradeStatus: No upgrade log present (probably fresh install) dmi.bios.date: 06/30/2020 dmi.bios.release: 1.47 dmi.bios.vendor: LENOVO dmi.bios.version: 4WCN47WW dmi.board.asset.tag: NO Asset Tag dmi.board.name: LNVNB161216 dmi.board.vendor: LENOVO dmi.board.version: SDK0J40709 WIN dmi.chassis.asset.tag: NO Asset Tag dmi.chassis.type: 10 dmi.chassis.vendor: LENOVO dmi.chassis.version: Lenovo ideapad 320-15IKB dmi.ec.firmware.release: 1.47 dmi.modalias: dmi:bvnLENOVO:bvr4WCN47WW:bd06/30/2020:br1.47:efr1.47:svnLENOVO:pn80XL:pvrLenovoideapad320-15IKB:rvnLENOVO:rnLNVNB161216:rvrSDK0J40709WIN:cvnLENOVO:ct10:cvrLenovoideapad320-15IKB:skuLENOVO_MT_80XL_BU_idea_FM_ideapad320-15IKB:pfaideapad320-15IKB: dmi.product.family: ideapad 320-15IKB dmi.product.name: 80XL dmi.product.sku: LENOVO_MT_80XL_BU_idea_FM_ideapad 320-15IKB dmi.product.version: Lenovo ideapad 320-15IKB dmi.sys.vendor: LENOVO ** Affects: linux (Ubuntu) Importance: Undecided Status: New ** Tags: gm108m hybrid-graphics kernel-panic nvidia resume suspend ** Attachment added: "Photo of the kernel panic displayed immediately after sudo systemctl suspend with NVIDIA 580 active." https://bugs.launchpad.net/bugs/2164938/+attachment/5994922/+files/Kernel%20Panic.jpg -- You received this bug notification because you are subscribed to linux in Ubuntu. Matching subscriptions: Bgg, Bmail, Nb https://bugs.launchpad.net/bugs/2164938 Title: Kernel panic "System is deadlocked on memory" during suspend with NVIDIA 580 on GeForce 920MX (GM108M), Ubuntu 26.04 kernel 7.0.0-30 Status in linux package in Ubuntu: New Bug description: Ubuntu 26.04 LTS on a Lenovo IdeaPad 320-15IKB consistently hits a kernel panic when entering suspend while the proprietary NVIDIA 580 driver stack is loaded. System: - Ubuntu 26.04 LTS (resolute) - Kernel: 7.0.0-30-generic, package version 7.0.0-30.30 - CPU: Intel Core i7-7500U (Kaby Lake) - Integrated GPU: Intel HD Graphics 620 [8086:5916], driver i915 - Discrete GPU: NVIDIA GM108M [GeForce 920MX] [10de:134f] - NVIDIA driver: 580.173.02-0ubuntu0.26.04.1 - linux-modules-nvidia-580-7.0.0-30-generic: 7.0.0-30.30 - Secure Boot enabled - Suspend mode: deep (S3) Problem: With the NVIDIA driver stack loaded, running: sudo systemctl suspend causes the graphical session to switch briefly to a text/console screen and then the system enters a kernel panic. The screen displays: KERNEL PANIC! Please reboot your computer. System is deadlocked on memory The Caps Lock LED flashes continuously, networking is lost, SSH becomes unreachable, and the machine requires a forced power-off. Closing the laptop lid also reproduced an unrecoverable suspend/resume failure when the NVIDIA stack was active. Expected result: The laptop should enter S3/deep suspend and resume normally when a keyboard key is pressed or the lid is reopened. Actual result: With the NVIDIA stack active, suspend can result in a kernel panic / unrecoverable system state. Important A/B test: I performed a controlled test using exactly the same Ubuntu installation and kernel 7.0.0-30, but completely blocked the NVIDIA modules at kernel boot using: module_blacklist=nvidia,nvidia_drm,nvidia_modeset,nvidia_uvm,nouveau After reboot I confirmed: - no nvidia or nouveau modules were loaded; - Intel HD 620 / i915 was the active graphics driver; - the NVIDIA 920MX was present on PCI but had no "Kernel driver in use". With NVIDIA completely disabled: 1. `sudo systemctl suspend` - entered deep suspend successfully; - resumed immediately after pressing Enter; - display returned normally; - login screen returned; - fan behaviour was normal; - Wi-Fi/networking and SSH recovered normally; - no panic occurred. 2. Lid-close suspend - `systemd-logind` detected Lid closed; - kernel entered `PM: suspend entry (deep)`; - reopening the lid resumed the system normally; - kernel logged `PM: suspend exit`; - networking and display recovered normally; - no panic occurred. Relevant successful-suspend log excerpts with NVIDIA blacklisted: Module nvidia is blacklisted PM: suspend entry (deep) PM: suspend exit and for lid suspend: systemd-logind: Lid closed. PM: suspend entry (deep) systemd-logind: Lid opened. PM: suspend exit This strongly indicates that the failure depends on the NVIDIA driver stack being active, rather than on ACPI lid detection or generic S3 suspend support. Additional diagnostics already performed: - ACPI lid switch correctly reports both `state: closed` and `state: open`. - Kernel PM debug `pm_test=freezer` completed successfully. - Direct PM device testing showed the expected NVIDIA error when bypassing the NVIDIA procfs suspend interface with PreserveVideoMemoryAllocations enabled; this was a diagnostic test and is distinct from the systemctl suspend panic. - NVIDIA suspend/resume systemd services are installed. - `PreserveVideoMemoryAllocations: 1`. - kdump is installed and loaded. - The original panic did not produce a vmcore. kdump memory reservation was subsequently increased according to the kdump-tools estimator, but no further panic has intentionally been triggered. - No evidence currently suggests a physical GPU failure. There are repeated i915 HPD polling workqueue warnings during some boots, but these are not considered proven causal and successful suspend/resume with i915 alone demonstrates that the Intel GPU can suspend correctly on this machine. A similar Ubuntu 26.04 / Linux 7.0 / NVIDIA 580 issue has been reported where suspend produces: "Kernel panic - not syncing: System is deadlocked on memory". There are also reports involving NVIDIA GM108M hardware and the NVIDIA suspend path on Linux 7.0.x. I am currently keeping the NVIDIA GPU disabled as a workaround because the system suspends and resumes reliably in that configuration. I can perform additional controlled tests if requested by Ubuntu kernel developers, including testing an older Ubuntu 7.0.0-14 kernel with the NVIDIA 580 modules available from the official resolute repositories. ProblemType: Bug DistroRelease: Ubuntu 26.04 Package: linux-image-7.0.0-30-generic 7.0.0-30.30 ProcVersionSignature: Ubuntu 7.0.0-30.30-generic 7.0.12 Uname: Linux 7.0.0-30-generic x86_64 ApportVersion: 2.34.1-0ubuntu0.1 Architecture: amd64 AudioDevicesInUse: USER PID ACCESS COMMAND /dev/snd/controlC0: diogo 3661 F.... pipewire diogo 3678 F.... wireplumber /dev/snd/pcmC0D0p: diogo 3661 F...m pipewire /dev/snd/seq: diogo 3661 F.... pipewire CasperMD5CheckResult: pass CurrentDesktop: ubuntu:GNOME Date: Mon Aug 24 20:00:16 2026 InstallationDate: Installed on 2026-08-20 (4 days ago) InstallationMedia: Ubuntu 26.04 "Resolute Raccoon" - Release amd64 (20260423.1) Lsusb: Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 001 Device 002: ID 8087:0a2a Intel Corp. Bluetooth wireless interface Bus 001 Device 003: ID 5986:210f Bison Electronics Inc. EasyCamera Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub MachineType: LENOVO 80XL ProcEnviron: LANG=pt_PT.UTF-8 PATH=(custom, no user) SHELL=/bin/bash TERM=xterm-256color XDG_RUNTIME_DIR=<set> ProcFB: 0 i915drmfb ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-7.0.0-30-generic root=/dev/mapper/ubuntu--vg-ubuntu--lv ro quiet splash crashkernel=1200M module_blacklist=nvidia,nvidia_drm,nvidia_modeset,nvidia_uvm,nouveau SourcePackage: linux UpgradeStatus: No upgrade log present (probably fresh install) dmi.bios.date: 06/30/2020 dmi.bios.release: 1.47 dmi.bios.vendor: LENOVO dmi.bios.version: 4WCN47WW dmi.board.asset.tag: NO Asset Tag dmi.board.name: LNVNB161216 dmi.board.vendor: LENOVO dmi.board.version: SDK0J40709 WIN dmi.chassis.asset.tag: NO Asset Tag dmi.chassis.type: 10 dmi.chassis.vendor: LENOVO dmi.chassis.version: Lenovo ideapad 320-15IKB dmi.ec.firmware.release: 1.47 dmi.modalias: dmi:bvnLENOVO:bvr4WCN47WW:bd06/30/2020:br1.47:efr1.47:svnLENOVO:pn80XL:pvrLenovoideapad320-15IKB:rvnLENOVO:rnLNVNB161216:rvrSDK0J40709WIN:cvnLENOVO:ct10:cvrLenovoideapad320-15IKB:skuLENOVO_MT_80XL_BU_idea_FM_ideapad320-15IKB:pfaideapad320-15IKB: dmi.product.family: ideapad 320-15IKB dmi.product.name: 80XL dmi.product.sku: LENOVO_MT_80XL_BU_idea_FM_ideapad 320-15IKB dmi.product.version: Lenovo ideapad 320-15IKB dmi.sys.vendor: LENOVO To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2164938/+subscriptions
Комментариев нет:
Отправить комментарий