Created attachment 1515511 [details] /var/log/messages Description of problem: While 0.rc5.git3.1.vanilla.knurd.1.fc29 is working correctly, with the later kernel versions (rc6, rc7...) I experience the issue. It shows a splash screen with ASUS logo after the kernel is selected in GRUB2; and nothing else happens. Version-Release number of selected component (if applicable): 0.rc6.git01.vanilla.knurd.1.fc29 and above How reproducible: boot the computer Steps to Reproduce: boot my computer Actual results: stops boot with a splash screen. Expected results: after the splash screen it should display the login screen. Additional info: please, advice what log to include here.... I am attaching my /var/log/messages
> Dec 19 08:46:47 localhost kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-4.20.0-0.rc5.git3.1.vanilla.knurd.1.fc29.x86_64 root=UUID=532c1a72-d32a-4bca-a06c-199eb6bd3c3b ro rhgb quiet LANG=hu_HU.UTF-8 Try removing "rhgb quiet" from the kernel command-line: Method 1: Edit the kernel command-line from grub. Method 2: Edit /etc/default/grub and then run: # grub2-mkconfig -o /boot/grub2/grub.cfg # it is a good idea to make a backup copy of grub.cfg first > Dec 19 07:58:04 localhost dnf[2590]: RPM Fusion for Fedora 29 - Free - Updates 7.6 kB/s | 8.9 kB 00:01 Also, please remove any RPM Fusion packages before testing, so you are only reporting bugs against Fedora packages. > please, advice what log to include here.... I am attaching my /var/log/messages $ journalctl -b # log for current boot $ journalctl -b -1 # log for previous boot Documentation: $ info grub2 $ man journalctl
> Dec 19 07:46:51 localhost kernel: Generic PHY r8169-600:00: attached PHY driver [Generic PHY] (mii_bus:phy_addr=r8169-600:00, irq=IGNORE) > Dec 19 08:46:16 localhost kernel: smpboot: CPU0: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz (family: 0x6, model: 0x5e, stepping: 0x3) The Intel i7-6700K is a Skylake processor. If you are getting a boot hang, you could be seeing: Bug 1660649 - r8169: hang or reboot when networking is enabled with kernel 4.19.9 on Intel Skylake system See if you can reproduce the hang using the method described in that bug report.
Thank you for your quick reply. @ I am following advices from "2018-12-19 08:56:00 UTC": > cp /boot/efi/EFI/fedora/grub.cfg /boot/efi/EFI/fedora/grub.cfg_my_orig > sudo featherpad /etc/default/grub # commented out the options > sudo grub2-mkconfig -o /boot/efi/EFI/fedora/grub.cfg Reboot... Now I see the messages :-) Thanks for the tips! This is the last message, then loading stops: dec 20 01:19:57 localhost.localdomain kernel: fb0: switching to amdgpudrmfb from EFI VGA Then I pressed CTRL+ALT+DEL to restart. So I am attaching the logs here now: > sudo journalctl -b -1 > /home/andrew/Letöltések/msgs_fail.txt > sudo journalctl -b > /home/andrew/Letöltések/msgs_goodload.txt But I found "amdgpu 0000:01:00.0: Fatal error during GPU init" in the log :S Fusion should not had affected the boot... > yum list installed | grep fusion compat-ffmpeg28.x86_64 2.8.15-2.fc29 @rpmfusion-free faac.x86_64 1.29.9.2-4.fc29 @rpmfusion-nonfree faad2-libs.x86_64 1:2.8.8-2.fc29 @rpmfusion-free fdk-aac.x86_64 0.1.6-3.fc29 @rpmfusion-nonfree ffmpeg.x86_64 4.0.3-1.fc29 @rpmfusion-free-updates ffmpeg-libs.x86_64 4.0.3-1.fc29 @rpmfusion-free-updates ffmpegthumbs.x86_64 18.08.1-1.fc29 @rpmfusion-free gstreamer1-plugins-bad-freeworld.x86_64 1.14.4-1.fc29 @rpmfusion-free gstreamer1-plugins-bad-nonfree.x86_64 1.14.4-1.fc29 @rpmfusion-nonfree gstreamer1-plugins-ugly.x86_64 1.14.4-2.fc29 @rpmfusion-free-updates libavdevice.x86_64 4.0.3-1.fc29 @rpmfusion-free-updates libdca.x86_64 0.0.6-2.fc29 @rpmfusion-free libde265.x86_64 1.0.3-2.fc29 @rpmfusion-free libmms.x86_64 0.6.4-7.fc29 @rpmfusion-free libmpeg2.x86_64 0.5.1-16.fc29 @rpmfusion-free librtmp.x86_64 2.4-12.20160224.gitfa8646d.fc29 @rpmfusion-free live555.x86_64 2018.10.17-1.fc29 @rpmfusion-free mjpegtools-libs.x86_64 2.1.0-13.fc29 @rpmfusion-free opencore-amr.x86_64 0.1.5-5.fc29 @rpmfusion-free python3-vlc.noarch 1.1.2-8.20161001git5d389c7.fc29 @rpmfusion-free rpmfusion-free-appstream-data.noarch 29-4.20181021.fc29 @rpmfusion-free rpmfusion-free-release.noarch 29-1 @rpmfusion-free-updates rpmfusion-nonfree-appstream-data.noarch 29-3.20181021.fc29 @rpmfusion-nonfree rpmfusion-nonfree-release.noarch 29-1 @rpmfusion-nonfree-updates vcdimager.x86_64 2.0.1-4.fc29 @rpmfusion-free vcdimager-libs.x86_64 2.0.1-4.fc29 @rpmfusion-free vlc.x86_64 3.0.5-4.fc29 @rpmfusion-free vlc-core.x86_64 3.0.5-4.fc29 @rpmfusion-free vlc-extras.x86_64 3.0.5-4.fc29 @rpmfusion-free vo-amrwbenc.x86_64 0.1.3-8.fc29 @rpmfusion-free x264-libs.x86_64 0.155-2.20180806git0a84d98.fc29 @rpmfusion-free-updates x265-libs.x86_64 2.8-1.fc29 @rpmfusion-free xvidcore.x86_64 1.3.5-3.fc29 @rpmfusion-free @ I am checking your advices provided at 2018-12-19 09:19:02 UTC... I do not see timeout in the log... - I am not sure if it is the same....
Created attachment 1515778 [details] log when failure with new kernel
Created attachment 1515779 [details] log when good with old kernel
> But I found "amdgpu 0000:01:00.0: Fatal error during GPU init" in the log :S Thanks for pointing that out. This could be the problem: dec 20 01:19:57 localhost.localdomain kernel: mc: Failed to load firmware "amdgpu/polaris11_k_mc.bin" See: Bug 1520682 - amdgpu module fails to load firmware The firmware should be here: $ rpm -ql linux-firmware | grep polaris11 The "lsinitrd" command will show what is in the initramfs: # lsinitrd -k 4.20.0-0.rc7.git0.1.vanilla.knurd.1.fc29.x86_64 | grep polaris11
Try updating linux-firmware from here: Information for build linux-firmware-20181219-89.git0f22c852.fc29 https://koji.fedoraproject.org/koji/buildinfo?buildID=1174570 Specifically: # dnf update https://kojipkgs.fedoraproject.org//packages/linux-firmware/20181219/89.git0f22c852.fc29/noarch/linux-firmware-20181219-89.git0f22c852.fc29.noarch.rpm Tested in an F29 VM: $ rpm -q linux-firmware linux-firmware-20181219-89.git0f22c852.fc29.noarch $ rpm -ql linux-firmware | grep polaris11_k_mc.bin /usr/lib/firmware/amdgpu/polaris11_k_mc.bin Note that the package is not signed, because it hasn't been released: $ rpm -qi linux-firmware | grep Signature Signature : (none)
It seems this bug was reported 2018-12-16 17:21 UTC: Bug 1659810 - linux-firmware: amdgpu fatal error during gpu init (In reply to Steve from comment #7) > Try updating linux-firmware from here: ... Anyway, I actually started here: Fedora Packages Search https://apps.fedoraproject.org/packages/ Although I could have started here: Fedora Updates System https://bodhi.fedoraproject.org/ See, also: Fedora Project Packages GIT repositories https://src.fedoraproject.org/cgit/ I recommend bookmarking all of those ...
According to a comment on Bodhi, the initramfs needs to be rebuilt: bugfix update in Fedora 29 for linux-firmware https://bodhi.fedoraproject.org/updates/FEDORA-2018-ba224c644f There aren't any examples in the comment, but try something like this: # dracut --kver 4.20.0-0.rc7.git0.1.vanilla.knurd.1.fc29.x86_64 --force Documentation: $ man dracut
Hi All, Thank you for your answer! I updated linux-firmware as Steve suggested, but it did not make any difference. I also updated my system and it installed kernel 4.20.0-0.rc7.git1.1.vanilla.knurd.1.fc29.x86_64. - with this, it loads fine again :-). Yours, András
With the dracut tool on the 4.20.0-0.rc7.git0.1.vanilla.knurd.1.fc29.x86_64 version; it is now also working fine. Thank you!
Thanks for your reports. If you feel like it, you could post karma here: bugfix update in Fedora 29 for linux-firmware https://bodhi.fedoraproject.org/updates/FEDORA-2018-ba224c644f
*** This bug has been marked as a duplicate of bug 1659810 ***