Sounds like the system is up (multi-user.target, NM up, logind seeing your VT hotkeys) but nothing ever makes it to a visible VT—classic display stack handoff/KMS issue (plymouth ↔ simpledrm ↔ real DRM driver ↔ display manager). Here’s a tight, do-now checklist from a LiveCD chroot that usually pins it down fast.

1) Prove it’s graphics/KMS

From the chroot, edit the kernel cmdline in /etc/default/grub and regenerate the GRUB config:

• Test A (turn off plymouth + verbosity) Add: plymouth.enable=0 rd.plymouth=0 noplymouth noquiet Remove: rhgb quiet Rebuild GRUB:

  • EFI: grub2-mkconfig -o /boot/efi/EFI/fedora/grub.cfg
  • BIOS: grub2-mkconfig -o /boot/grub2/grub.cfg

• Test B (force no KMS) If Test A still black, temporarily add also: nomodeset If you now get visible VTs or at least kernel boot spew, you’ve confirmed it’s the DRM driver / modeset handoff.

(Keep nomodeset only for diagnosis; it’ll kill acceleration.)

2) Check what driver stack you think you’re on

In the last boot’s journal:

journalctl -b -1 | sed -n '1,200p' | head -n 200 >/tmp/bootstart.txt
journalctl -b -1 -k | grep -Ei 'drm|amdgpu|i915|nouveau|nvidia|simpledrm|efifb|fbcon'
journalctl -b -1 -p err..alert
journalctl -b -1 -u gdm -u sddm -u lightdm

Red flags:

• Stuck on simpledrm/efifb with no amdgpu/i915/nvidia takeover.
• amdgpu probes but fails firmware load.
• NVIDIA present but nvidia_drm not modesetting (or Secure Boot blocks it).
• Wayland crashes GDM repeatedly.

3) If you’re on AMD/Intel (FOSS drivers)

Common fix is missing firmware or a busted initramfs:

dnf reinstall -y kernel-core kernel-modules linux-firmware
dracut --regenerate-all --force

For AMD that needs DC early, seed drivers in initramfs:

• Create /etc/dracut.conf.d/amdgpu.conf with: add_drivers+=" amdgpu "
• Then dracut -f.

If you saw DC/DP link-training errors, you can try a temporary kernel param to boot once:

• AMD: amdgpu.dc=0 (just to get in; remove after driver update)

4) If you’re on NVIDIA (proprietary)

Typical black-screen causes: Secure Boot not enrolling the module key, nvidia_drm not in modeset, or akmods not built for your current kernel.

From chroot:

mokutil --sb-state                  # If Secure Boot enabled
rpm -qa | grep -E 'akmod|kmod-nvidia|nvidia'
akmods --force
dracut --regenerate-all --force

Kernel params:

• Ensure: nvidia_drm.modeset=1
• Ensure nouveau is blacklisted (package removed or rd.driver.blacklist=nouveau modprobe.blacklist=nouveau), and /usr/lib/modprobe.d/nvidia*.conf exists.

If Secure Boot is enabled, either disable it in firmware for a quick test or enroll the akmods MOK so the NVIDIA modules load.

5) Rule out GDM/Wayland specifically

Even if you can’t see a VT, you can boot straight to multi-user to see if the console appears without the display manager:

• Add kernel param: systemd.unit=multi-user.target If you now see the login prompt (without GDM), the issue is definitely in the graphical stack (GDM/Wayland/mutter vs driver).

Then, in chroot, force Xorg for GDM as a workaround:

# /etc/gdm/custom.conf
[daemon]
WaylandEnable=false

or temporarily disable the DM to test:

systemctl disable gdm sddm lightdm
systemctl set-default multi-user.target

Boot → confirm VTs are visible → then re-enable the DM after driver fixes.

6) Plymouth handoff gotcha

If logs show everything fine but screen stays black right after the splash should disappear, it’s often Plymouth not handing off. Keeping plymouth.enable=0 rd.plymouth=0 permanently resolves it for some setups. Also ensure vt.handoff isn’t pinned to a wrong TTY (you can just omit it).

7) Clean out conflicting X configs

Stray Xorg snippets can force a dead driver:

mkdir -p /root/xorg-backup
mv /etc/X11/xorg.conf /root/xorg-backup/ 2>/dev/null || true
mv /etc/X11/xorg.conf.d /root/xorg-backup/ 2>/dev/null || true

(If you actually need one, add it back after you get display.)

8) Kernel regression / quick escape hatches

• Try an older kernel from GRUB’s “Advanced options” (or set default via grubby from chroot).
• Or just fully update everything, rebuild initramfs, and reboot:

dnf upgrade --refresh -y
dracut --regenerate-all --force

9) What to capture if it’s still black

After trying the above, grab these (from the failed boot -b -1):

journalctl -b -1 -k > /var/tmp/boot-kernel.log
journalctl -b -1 -u gdm > /var/tmp/gdm.log
journalctl -b -1 | grep -i -E 'drm|kms|plymouth|gdm|mutter|Xorg|simpledrm|efifb' -n > /var/tmp/display-grep.log

Plus:

lsinitrd /boot/initramfs-$(uname -r).img | grep -E 'amdgpu|i915|nvidia|firmware'

Post (or check) for: missing firmware blobs, no vendor DRM in initramfs, Secure Boot blocks, or DM loop.

Typical “fast fixes” by scenario

• AMD GPU after update → black screen: reinstall linux-firmware, add add_drivers+=" amdgpu " to dracut, rebuild initramfs, disable plymouth.
• NVIDIA proprietary: set nvidia_drm.modeset=1, ensure akmods built for current kernel, fix Secure Boot (enroll MOK or disable), blacklist nouveau, rebuild initramfs.
• All GPUs, Plymouth weirdness: drop rhgb quiet, set plymouth.enable=0 rd.plymouth=0.

If you tell me your GPU (amdgpu/i915/nvidia) and whether Secure Boot is on, I can give you a minimal, exact command sequence for that stack.