Created attachment 946890 [details] Hopefully, the relevant output from journalctl Description of problem: When booting with the new kernel, modesetting fails and my monitor receives no input Version-Release number of selected component (if applicable): kernel-3.17.0-301.fc21.x86_64 xorg-x11-drv-nouveau-1.0.11-1.fc21.x86_64 How reproducible: 100% Steps to Reproduce: 1. select kernel 3.17.0-301.fc21.x86_64 on boot Actual results: When modesetting should kick in, screen goes black, blue led on front bezel blinks, indicating no signal Expected results: Normal boot (text in something higher resolution than 80x25) Additional info: 1. System works properly with kernel-3.16.3-300.fc21.x86_64 2. Relevant lspci output: 02:00.0 VGA compatible controller: NVIDIA Corporation C78 [GeForce 9100] (rev a2) 3. This seems to happen with every upgrade from kernel 3.x to 3.(x+1)
same problem with kernel-3.17.1-300.fc21.x86_64
I can add that this also affects optimus systems - though in a different way. There are some ibus related errors during boot concerning NOUVEAU - I'll see if I can snag them at some point. I am able to boot successfully and get to the login screen, however several kerneloops timeouts are detected by ABRT and eventually the whole system locks up. I'm also having issues with the wireless card starting up reliably. Some times it works, other times it detects no networks (and I'm fairly sure the antennae is not actually on). There is another boot error that happens with this. However, not having a network connection is making it difficult to report. I'll have to get set up with an ethernet connection at some point.
Note: This also affects a Live ISO boot. Standard boot results in kernel crashes and eventual lockups. Simple graphics boot functions fine (though the wifi card is not always properly on).
I get a complete lockup with no keyboard response when modesetting kicks in. Can't even do a blind login. Adding nomodeset to the kernel line in grub lets me boot, but in 1024x768 mode. My monitor's native mode is 1680x1050, and the monitor does not emit a proper EDID.
Adding a blacklist nouveau in modeprobe.d allows me to boot and function as normal running off the intel card. Obviously not a good workaround for nvidia-only systems. I may try running proprietary nvidia drivers later to see what happens.
WTF? I just installed kernel 3.18.0-0.rc0.git9.4.fc22.x86_64. Lo and behold, my system boots properly, and the nouveau driver works as it should.
I manually updated to the 3.17.1-302 kernel on bodhi, and it seems like it may be fixed. That patch references a couple kernel oops bugs, so give it a try.
Tried kernel-3.17.1-302. It did not work for me
Must have been a fluke - just booted again and it didn't work. I think it may not be properly turning on components, as it seems I also fail to get a working wifi card quite often - and eventually a hung system. I'll give the 3.18.0 kernel a shot - with a package rebuild - and see how it goes.
I've been testing a few kernels. Rebuilding 3.18.0.0.rc0.git9.5 for f21 at least gives me a reasonably stable system - though there are issues already listed in Bug#1154771 . On the 3.17.1-303 kernel, these nouveau bugs persist. I will attach a dmesg from the afflicted kernel shortly. Problem being, when I boot into the 3.17.x system - the nouveau issues seem to also cascade over to my wifi card, causing the antennae to fail and not turn on. It requires a total shut down and boot into a different kernel to turn it back on. And finally - the system may freeze up at any time and is totally unable to shut down or restart properly. (Though this also appears be to true for the 3.18.x kernel.)
Created attachment 950411 [details] dmesg for kernel-3.17.1-303 You can see the NOUVEAU failures toward the bottom - and I think the last couple lines show the wifi card failing. I should also mention that while running on the 3.17.x kernels, my nvidia card appears to be constantly on - as I hear the GPU fan running loudly while booted into that system. I can provide any other abrt file if needed. I'll go ahead and attach the backtrace for this report as well.
Created attachment 950412 [details] kernel-3.17.1-303 backtrace
I'm no longer seeing nouveau problems on 3.17.2 - have you tried it yet?
*** Bug 1154276 has been marked as a duplicate of this bug. ***
I hit similar problem using all kernel 3.17 series. Using 3.18 sorta helps but will randomly freeze after certain times. My desktop uses Geforce GTX 460 v2.
Tried kernel-3.17.2; it does not work for me. I'm going to stay with 3.18, as it seems to work just fine for me.
Just booted Nov 7 (or 8) F21 nightly w/ kernel-3.17.2 from USB stick on my mother-in-law's computer which has the same onboard video controller (NVIDIA Corporation C78 [GeForce 9100]) as my computer. Works fine. I suggest the problem I'm experiencing on my home computer is due to my LCD not emitting a proper EDID.
My bug https://bugzilla.redhat.com/show_bug.cgi?id=1164351 is probably a duplicate of this one. My system works with the 3.18 Rawhide kernels, but I can make it halt by using Google Maps in Firefox.
This message is a reminder that Fedora 21 is nearing its end of life. Approximately 4 (four) weeks from now Fedora will stop maintaining and issuing updates for Fedora 21. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '21'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 21 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
Bumped to Fedora 23. Will test if the issue still exists.
Closing this bug after testing with F22 Workstation. Kernel in that version runs smoothly aside know issue related to graphic driver (mesa).