I don't think it's random.Īfter a lot more analysis, it appears D3cold is problematic for many PCIe bridge/device combo. The nvidia power state failure is actually already reported over here: system76/firmware-open#160. Enjoy max spinning fans + on the next restart the logs which says the exact identical same: nvidia power state error followed by Xid 79 followed by snd_hda_intel again throwing the same garbage about power state.Quickly Lock the GNOME session while you hear the headset saying "Power off".Close the app which was running on the dGPU.Run a load on the dGPU (ex run crappy GL code (triangle/etc) on dGPU, even MC or Steam whatever you want that will use the dGPU).Connect bluetooth headset, make sure to play some continuous sound over the headset.Resume from sleep by typing enter key (still no crash).Press Fn + F12 to put computer to sleep.Make sure you're booted in Hybrid graphics.Here are the actions I did (this is only reproducible on the OryxPro system): Made a lot of tests and I could reproduce the crash one time. I'm gonna do some tests to attempt to reproduce the crash. I don't see any power state issue in the logs on that reddit. I'm not sure if your issue is the same as this one. Jun 28 15:23:30 oryp7 kernel: snd_hda_intel 0000:01:00.1: can't change power state from D3cold to D0 (config space inaccessible) Jun 28 15:23:30 oryp7 kernel: snd_hda_codec_hdmi hdaudioC2D0: Unable to sync register 0x4f0800. Jun 28 15:23:30 oryp7 kernel: snd_hda_codec_hdmi hdaudioC2D0: HDMI: invalid ELD buf size -1 This is a follow up of my older issue which is still unsolved.ĮDIT: I forgot to mention that according to logs that audio crashed with the exact same power mode error at the exact same time than NVIDIA. May occur after resume from suspend, may occur after locking GNOME session, may occur after a load on the GPU, may even occur while running a high load on the GPU. No particular steps produces this, it's absolutely random. Unfortunatly the workstation is running an ASUS P9X79 board. There are forums which also mentions a bug with some ASUS motherboards. On my workstation only the like which says Xid 79 GPU fallen off the bus appears.Īccording to NVIDIA Xid 79 could be temps, power, gpu or PCI hardware defect or even BIOS/firmware issue. Jun 28 15:23:30 oryp7 kernel: nvidia 0000:01:00.0: can't change power state from D3cold to D0 (config space inaccessible)Īfter this you get an infinite loop of "kernel: nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c67d:0 2:0:4048:4040". Logs from Oryx Pro which includes the weird additional log before reaching Xid 79 like my workstation: So perhaps Oryx is a different bug but at least now we know the reason why GTX 650 is crashing. And guess what: Oryx Pro is also affected by Xid 79 however the error it produces includes something weird about power state. So the crash is in fact NVIDIA related and is a Xid 79 error the famous "GPU has fallen off the bus.". Without sudo you won't get any information from NVIDIA driver. It turns out in order to get proper logs you must use sudo journalctl -b -1 -r. I have finally figured out what exactly is the random crash on my desktop workstation which uses a GTX 650 Ti. Related Application and/or Package Version (run apt policy $PACKAGE NAME):
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |