[Software] blank screen after installing a DE in a Minimal Install

[davidtmyers]

I performed a minimal install of Debian 12 Bookworm a few weeks ago, and it's been working great. I skipped installing a DE at the time and have been using the machine entirely headless via SSH without issue.

I just installed a GPU, and attempted to install a DE, but I can't seem to get anything to actually render to that GPU. I've confirmed both the GPU and monitor work on other installs. I believe I'm just missing a key step in moving from a minimal install to a working Desktop Environment.

The Issue
Whenever I boot, the monitor plugged into my discrete GPU stays in its power-save state. Once the kernel loads, the monitor exits its power-save state, turns its backlight on, but the screen remains completely blank (no text cursor, no mouse cursor, and certainly no greeter/login-screen or DE). I am, however, able to SSH into the machine just fine.

The Question
How do I get KDE up and running on this machine with this AMD Radeon X300 SE GPU?

What I've Done

1. Started with a Minimal Install from the netinst .ISO
2. Ensured /etc/apt/sources.list contains the correct repositories (i.e., main, contrib, and non-free-firmware):

   deb http://deb.debian.org/debian bookworm main contrib non-free-firmware
   deb-src http://deb.debian.org/debian bookworm main contrib non-free-firmware

3. Installed AMD-related firmware, drivers, packages, etc.:

   apt-get install firmware-amd-graphics amd64-microcode xserver-xorg-video-radeon

4. Installed KDE:

   apt-get install plasma-desktop plasma-nm

Potentially Helpful Info
In the spirit of trying to simplify the debug process, here's a few relevant outputs:

The card is correctly showing up on the PCIe bus (see "e1:00" in output below)

Code: Select all

root@workstation:~# lspci
[...]
e1:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV370 [Radeon X300]
e1:00.1 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] RV370 [Radeon X300 SE]
e2:00.0 PCI bridge: ASPEED Technology, Inc. AST1150 PCI-to-PCI Bridge (rev 06)
e3:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 52)
e4:00.0 Non-Essential Instrumentation [1300]: Advanced Micro Devices, Inc. [AMD] Device 14ac (rev 01)
e4:00.4 USB controller: Advanced Micro Devices, Inc. [AMD] Device 14c9 (rev da)
[...]

The kernel appears to be able to correctly detect and probe the card (see "Device-1" in output below).
The kernel is also correctly detecting the monitor plugged into the card (see "Monitor-3" in output below).

Code: Select all

root@workstation:~# inxi -GSaz --vs
inxi 3.3.26-00 (2023-03-28)
System:
  Kernel: 6.1.0-18-amd64 arch: x86_64 bits: 64 compiler: gcc v: 12.2.0
    parameters: BOOT_IMAGE=/BOOT/debian@/vmlinuz-6.1.0-18-amd64 root=ZFS=/ROOT/debian ro
    root=ZFS=workstation-rpool/ROOT/debian quiet
  Console: pty pts/0 Distro: Debian GNU/Linux 12 (bookworm)
Graphics:
  Device-1: AMD RV370 [Radeon X300] driver: radeon v: kernel alternate: radeonfb,amdgpu
    arch: Rage-9 code: Radeon IGP process: TSMC 110nm built: 2003-08 pcie: gen: 1 speed: 2.5 GT/s
    lanes: 16 ports: active: VGA-2 empty: DVI-I-1 bus-ID: e1:00.0 chip-ID: 1002:5b60
    class-ID: 0300
  Device-2: ASPEED Graphics Family driver: ast v: kernel ports: active: DP-1,VGA-1 empty: none
    bus-ID: e3:00.0 chip-ID: 1a03:2000 class-ID: 0300
  Display: server: X.org v: 1.21.1.7 with: Xwayland v: 22.1.9 driver: gpu: ast,radeon
    note: X driver n/a tty: 318x98
  Monitor-1: DP-1 size-res: N/A in console modes: max: 640x480 min: 800x600
  Monitor-2: VGA-1 size-res: N/A in console modes: max: 1024x768 min: 640x480
  Monitor-3: VGA-2 model: Dell 2007FP serial: <filter> built: 2007 res: 1600x1200 dpi: 99
    gamma: 1.2 size: 367x275mm (14.45x10.83") diag: 514mm (20.2") ratio: 4:3, 5:4 modes:
    max: 1600x1200 min: 720x400
  API: OpenGL Message: GL data unavailable in console for root.

Similarly, we can also see quite a bit more info about the card (see "*-display:0" in output below):

Code: Select all

root@workstation:~# lshw -C display
  *-display:0               
       description: VGA compatible controller
       product: RV370 [Radeon X300]
       vendor: Advanced Micro Devices, Inc. [AMD/ATI]
       physical id: 0
       bus info: pci@0000:e1:00.0
       logical name: /dev/fb1
       version: 00
       width: 32 bits
       clock: 33MHz
       capabilities: pm pciexpress msi vga_controller bus_master cap_list rom fb
       configuration: depth=32 driver=radeon latency=0 resolution=1600,1200
       resources: irq:272 memory:c0000000-cfffffff ioport:f000(size=256) memory:d8330000-d833ffff memory:d8300000-d831ffff
  *-display:1 UNCLAIMED
       description: Display controller
       product: RV370 [Radeon X300 SE]
       vendor: Advanced Micro Devices, Inc. [AMD/ATI]
       physical id: 0.1
       bus info: pci@0000:e1:00.1
       version: 00
       width: 32 bits
       clock: 33MHz
       capabilities: pm pciexpress cap_list
       configuration: latency=0
       resources: memory:d8320000-d832ffff
  *-display
       description: VGA compatible controller
       product: ASPEED Graphics Family
       vendor: ASPEED Technology, Inc.
       physical id: 0
       bus info: pci@0000:e3:00.0
       logical name: /dev/fb0
       version: 52
       width: 32 bits
       clock: 33MHz
       capabilities: pm msi vga_controller cap_list fb
       configuration: depth=32 driver=ast latency=0 resolution=1024,768
       resources: irq:24 memory:d4000000-d7ffffff memory:d8000000-d803ffff ioport:e000(size=128)

Lastly, the card also appears to be detected properly based on relevant DRM output from dmesg:

Code: Select all

root@workstation:~# dmesg | grep "drm"
[   13.155799] systemd[1]: Starting modprobe@drm.service - Load Kernel Module drm...
[   13.255966] ACPI: bus type drm_connector registered
[   13.257382] systemd[1]: modprobe@drm.service: Deactivated successfully.
[   13.257772] systemd[1]: Finished modprobe@drm.service - Load Kernel Module drm.
[   14.215102] ast 0000:e3:00.0: [drm] P2A bridge disabled, using default configuration
[   14.215109] ast 0000:e3:00.0: [drm] AST 2600 detected
[   14.215119] ast 0000:e3:00.0: [drm] Using analog VGA
[   14.215124] ast 0000:e3:00.0: [drm] dram MCLK=396 Mhz type=1 bus_width=16
[   14.216038] [drm] Initialized ast 0.1.0 20120228 for 0000:e3:00.0 on minor 0
[   14.287738] fbcon: astdrmfb (fb0) is primary device
[   14.313540] ast 0000:e3:00.0: [drm] fb0: astdrmfb frame buffer device
[   14.616300] [drm] radeon kernel modesetting enabled.
[   14.616759] [drm] initializing kernel modesetting (RV380 0x1002:0x5B60 0x1002:0x3000 0x00).
[   14.682912] [drm] GPU not posted. posting now...
[   14.796694] [drm] Generation 2 PCI interface, using max accessible memory
[   14.796750] [drm] Detected VRAM RAM=256M, BAR=256M
[   14.796753] [drm] RAM width 128bits DDR
[   14.796766] [drm] radeon: 256M of VRAM memory ready
[   14.796769] [drm] radeon: 512M of GTT memory ready.
[   14.796797] [drm] GART: num cpu pages 131072, num gpu pages 131072
[   14.797938] [drm] radeon: 1 quad pipes, 1 Z pipes initialized
[   14.802047] [drm] PCIE GART of 512M enabled (table at 0x00000000C0040000).
[   14.802477] [drm] radeon: irq initialized.
[   14.802499] [drm] Loading R300 Microcode
[   14.804870] [drm] radeon: ring at 0x00000000A0001000
[   14.804893] [drm] ring test succeeded in 1 usecs
[   14.805162] [drm] ib test succeeded in 0 usecs
[   14.805645] [drm] Radeon Display Connectors
[   14.805648] [drm] Connector 0:
[   14.805649] [drm]   VGA-2
[   14.805651] [drm]   DDC: 0x60 0x60 0x60 0x60 0x60 0x60 0x60 0x60
[   14.805656] [drm]   Encoders:
[   14.805657] [drm]     CRT1: INTERNAL_DAC1
[   14.805660] [drm] Connector 1:
[   14.805662] [drm]   DVI-I-1
[   14.805663] [drm]   HPD1
[   14.805665] [drm]   DDC: 0x64 0x64 0x64 0x64 0x64 0x64 0x64 0x64
[   14.805670] [drm]   Encoders:
[   14.805671] [drm]     CRT2: INTERNAL_DAC2
[   14.805674] [drm]     DFP1: INTERNAL_TMDS1
[   14.819974] radeon 0000:e1:00.0: [drm] Cannot find any crtc or sizes
[   14.820162] [drm] Initialized radeon 2.50.0 20080528 for 0000:e1:00.0 on minor 1
[   15.848487] radeon 0000:e1:00.0: [drm] Cannot find any crtc or sizes
[   17.149767] [drm] amdgpu kernel modesetting enabled.

Any guidance on how I can get KDE up and running with GPU on top of a properly-running minimal install would be welcome.

[Aki]

Hello,

According to your log, the Radeon graphics card do not recognize the connected monitor:

Code: Select all

 
[   14.820162] [drm] Initialized radeon 2.50.0 20080528 for 0000:e1:00.0 on minor 1
[   15.848487] radeon 0000:e1:00.0: [drm] Cannot find any crtc or sizes

Could it be caused by the cable or the monitor ?

Is the ast graphics card working properly ?

[davidtmyers]

Apologies, that output must have come from when I was swapping VGA cables as part of my initial troubleshooting!

Could it be caused by the cable or the monitor ?
Good suggestion, but I don't think so. I can confirm the monitor, the VGA cables, and the GPU all work. Further, I can confirm that the connected monitor is recognized (see the "Monitor-3: VGA-2 model: Dell 2007FP serial: <filter> built: 2007 res: 1600x1200 dpi: 99" output from the "inxi -GSaz --vs" command above). Lastly, I was able to get this monitor to briefly work when I was trying things out on the Debian GNOME 12.5 LiveCD, so I'm certain the entire display chain from GPU -> cable -> monitor is functioning.

Rather than go in and edit the dmesg part of the original post, I'll provide the correct output below. (You'll notice there's no mention of "Cannot find any crtc or sizes".)

Code: Select all

root@workstation:~# dmesg | grep "drm"
[   18.752590] systemd[1]: Starting modprobe@drm.service - Load Kernel Module drm...
[   18.769951] ACPI: bus type drm_connector registered
[   18.770566] systemd[1]: modprobe@drm.service: Deactivated successfully.
[   18.770625] systemd[1]: Finished modprobe@drm.service - Load Kernel Module drm.
[   19.066537] ast 0000:e3:00.0: [drm] P2A bridge disabled, using default configuration
[   19.066539] ast 0000:e3:00.0: [drm] AST 2600 detected
[   19.066544] ast 0000:e3:00.0: [drm] Using analog VGA
[   19.066545] ast 0000:e3:00.0: [drm] dram MCLK=396 Mhz type=1 bus_width=16
[   19.066838] [drm] Initialized ast 0.1.0 20120228 for 0000:e3:00.0 on minor 0
[   19.069353] fbcon: astdrmfb (fb0) is primary device
[   19.092079] ast 0000:e3:00.0: [drm] fb0: astdrmfb frame buffer device
[   19.108911] [drm] radeon kernel modesetting enabled.
[   19.111395] [drm] initializing kernel modesetting (RV380 0x1002:0x5B60 0x1002:0x3000 0x00).
[   19.178823] [drm] GPU not posted. posting now...
[   19.290331] [drm] Generation 2 PCI interface, using max accessible memory
[   19.290358] [drm] Detected VRAM RAM=256M, BAR=256M
[   19.290359] [drm] RAM width 128bits DDR
[   19.290368] [drm] radeon: 256M of VRAM memory ready
[   19.290369] [drm] radeon: 512M of GTT memory ready.
[   19.290379] [drm] GART: num cpu pages 131072, num gpu pages 131072
[   19.290577] [drm] radeon: 1 quad pipes, 1 Z pipes initialized
[   19.291378] [drm] PCIE GART of 512M enabled (table at 0x00000000C0040000).
[   19.291558] [drm] radeon: irq initialized.
[   19.291563] [drm] Loading R300 Microcode
[   19.294760] [drm] radeon: ring at 0x00000000A0001000
[   19.294779] [drm] ring test succeeded in 1 usecs
[   19.294849] [drm] ib test succeeded in 0 usecs
[   19.294939] [drm] Radeon Display Connectors
[   19.294940] [drm] Connector 0:
[   19.294940] [drm]   VGA-2
[   19.294940] [drm]   DDC: 0x60 0x60 0x60 0x60 0x60 0x60 0x60 0x60
[   19.294941] [drm]   Encoders:
[   19.294941] [drm]     CRT1: INTERNAL_DAC1
[   19.294942] [drm] Connector 1:
[   19.294942] [drm]   DVI-I-1
[   19.294942] [drm]   HPD1
[   19.294942] [drm]   DDC: 0x64 0x64 0x64 0x64 0x64 0x64 0x64 0x64
[   19.294943] [drm]   Encoders:
[   19.294943] [drm]     CRT2: INTERNAL_DAC2
[   19.294943] [drm]     DFP1: INTERNAL_TMDS1
[   19.347766] [drm] fb mappable at 0xC00C0000
[   19.347780] [drm] vram apper at 0xC0000000
[   19.347781] [drm] size 7680000
[   19.347781] [drm] fb depth is 24
[   19.347781] [drm]    pitch is 6400
[   19.347826] radeon 0000:e1:00.0: [drm] fb1: radeondrmfb frame buffer device
[   19.347830] [drm] Initialized radeon 2.50.0 20080528 for 0000:e1:00.0 on minor 1
[   19.475545] [drm] amdgpu kernel modesetting enabled.

Is the ast graphics card working properly ?
Yes, it's working great. I get a text console/text login screen on that VGA port without any issues. FYI: The ast graphics card is the ASPEED GPU mentioned in the command outputs in my first post. It is small/low-powered on-board VGA chip that is part of the motherboard's BMC/IPMI management (the motherboard is an ASUS Pro WS WRX90E-SAGE SE). It can do text consoles and basic graphics, but that's about it. For a proper DE, a discrete GPU would run circles around it.

[CwF]

You need to visit the bios and disable the AST.
You may need to enable the bios rom for the slot in the MB's bios.
Then the boot screen will not appear on the AST and should on the AMD.

Or you can give it a X a hint and create a file to boot on the ast, still active, and switch to the amd when the os loads, but why? For wayland, not sure.

[davidtmyers]

Hi CwF - thank you for your suggestions.

Regarding disabling the AST
I'd like to avoid disabling this feature if at all possible: the AST is part of the motherboard's BMC/IPMI management functionality, and if I disable it, I'll be unable to use the web-based KVM that comes with the motherboard's BMC/IPMI solution. This feature is critical when you need remote access to the machine, but for whatever reason, SSH is unavailable or the machine hasn't yet booted into an OS capable of SSH access (often the case when in the UEFI BIOS, or when flashing firmware to other hardware (like other PCIe cards) via FreeDOS or some other basic UEFI environment).

Regarding enabling the BIOS ROM for the slot in the MB's BIOS
I took a look at the ASUS Pro WS WRX90E-SAGE SE's User Manual and there doesn't seem to be any concept of BIOS ROMs with this motherboard. I know exactly what you're talking about though since I've seen similar concepts with Supermicro boards. Perhaps ASUS's high-end workstation motherboards don't include this capability.

Regarding giving X a hint to switch from the AST to the AMD GPU when the OS loads
This would actually be ideal. Keeping the AST active would be useful when I need to KVM into the machine remotely using the motherboard's web-based IPMI functionality. The only wrinkle is that I'd prefer to find a Wayland-based solution given everyone's transitioning to it.

[CwF]

The X solution is a minimal file. Bios and grub will be on the ast
determine your BusID, etc
/usr/share/X11/xorg.0.conf/52-whatever.conf or /etc/X11/xorg.conf.d/

Code: Select all

Section "Screen"
    Identifier  "Default Screen"
    Device      "Card 0"
EndSection

Section "Device"
	Identifier  "Card0" 
	Driver      "radeon"
	BusID       "PCI:1:0:0"
EndSection

[davidtmyers]

Hmm, having the UEFI BIOS and GRUB on the AST VGA port wouldn't be too bad of a compromise/workaround. But if I go with an X-based solution, wouldn't that then prevent me from using a Wayland-based greeter and desktop environment like KDE?

(Apologies if this is a silly question: it's been ~20 years since I last used a desktop environment - the vast majority of my use/work can all be accomplished via SSH.)

[CwF]

As far as I know yes, only X.

[davidtmyers]

Got it. Well, in the spirit of at least giving it a shot (your suggestions have been very welcome and I'd hate not to at least fully explore what you're recommending), I attempted to do as suggested:

First, I created a new entry entitled /usr/share/X11/xorg.0.conf/52-whatever.conf with the following contents:

Code: Select all

root@workstation:~# cat /usr/share/X11/xorg.conf.d/52-whatever.conf 
Section "Screen"
    Identifier  "Default Screen"
    Device      "Card 0"
EndSection

Section "Device"
	Identifier  "Card0" 
	Driver      "radeon"
	BusID       "PCI:225:0:0"
EndSection

Based on this post from StackOverflow, I believe I need to use a BusID of "225:0:0" since lspci -s 0:e1:0 returns the right VGA:

Code: Select all

root@workstation:~# lspci -s 0:e1:0
e1:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV370 [Radeon X300]
e1:00.1 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] RV370 [Radeon X300 SE]

And because X doesn't accept hex values in its BusID's (it expects decimal), I have to convert "e1" to decimal (i.e., "225"), yielding a BusID of "PCI:225:0:0".

Then, I restarted. The UEFI BIOS and GRUB outputs went to the iKVM (the "AST" GPU that is on the BMC/IPMI), as expected. Unfortunately, when the kernel started loading, I'm getting the same behavior: the monitor exits its power-save state, turns its backlight on, but the screen remains completely blank (no text cursor, no mouse cursor, and certainly no greeter/login-screen or DE).

[CwF]

The Driver could be amdgpu?

Not sure on the address.

?

Code: Select all

 lspci -vnkd 1002:

on edit I read above, some info there...
while this happens, what is displayed on the ast?

...and yes, this works on my supermicro's!

[davidtmyers]

Here is the output of the lspci command you suggested. I'm reasonably confident my hex->decimal conversion math is correct, so I believe the BusID is correct as well (e1 in hex = 225 in decimal).

Code: Select all

root@workstation:~# lspci -vnkd 1002:
e1:00.0 0300: 1002:5b60 (prog-if 00 [VGA controller])
	Subsystem: 1002:3000
	Flags: bus master, fast devsel, latency 0, IRQ 272, IOMMU group 9
	Memory at c0000000 (32-bit, prefetchable) [size=256M]
	I/O ports at f000 [size=256]
	Memory at d8330000 (32-bit, non-prefetchable) [size=64K]
	Expansion ROM at d8300000 [disabled] [size=128K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Capabilities: [80] MSI: Enable+ Count=1/1 Maskable- 64bit+
	Capabilities: [100] Advanced Error Reporting
	Kernel driver in use: radeon
	Kernel modules: radeonfb, radeon, amdgpu

e1:00.1 0380: 1002:5b70
	Subsystem: 1002:3001
	Flags: fast devsel, IOMMU group 9
	Memory at d8320000 (32-bit, non-prefetchable) [disabled] [size=64K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Kernel modules: amdgpu

Here is the behavior I get with the following file:

Code: Select all

root@workstation:~# cat /usr/share/X11/xorg.conf.d/52-whatever.conf 
Section "Screen"
    Identifier  "Default Screen"
    Device      "Card 0"
EndSection

Section "Device"
	Identifier  "Card0" 
	Driver      "radeon"
	BusID       "PCI:225:0:0"
EndSection

The monitor plugged into my discrete AMD GPU is off/in a power-save state. On my laptop, I log into my motherboard's BMC/IPMI web-ui portal. I go to the iKVM within the web-ui. I remotely power on the system. I see the UEFI BIOS screen and the GRUB menu shows up with a countdown on the default kernel. The kernel starts to boot. Mid-way through the boot process, the monitor plugged into the discrete AMD GPU turns on/leaves the power-save state, but remains blank (i.e., the power button goes from amber to green, and the backlight comes on, but the screen is still empty with no text or mouse cursor or anything). The machine fully boots up, and on the iKVM on the web-ui that I'm viewing from my laptop, I'm presented with a text console to log in. The monitor plugged into the discrete GPU is now "on" but again - still blank, no text or mouse cursor.

Let me know if you'd like screenshots? Another advantage of using the iKVM via the BMC/IPMI is that I'm able to record both images and video of the boot process.

I then tried swapping the driver to "amdgpu" in /usr/share/X11/xorg.conf.d/52-whatever.conf, but this did not seem to change anything. Same behavior as I've described above.

It really does feel like Debian is correctly turning this GPU on and making it available to use, we just need to get something to display on it. Here's an idea: I don't know how to do this (guidance welcome), but perhaps there's some way we could spin up another text console on that display?

[CwF]

spin up another text console on that display?

Ctrl+Alt+F2 and F3, F4

Something is not right

e1:00.1 0380: 1002:5b70

this should not use amdgpu, it's the hdmi sound, usually snd_hda_ intel I'd expect?
and I'd probably use radeon in case amdgpu requires an extra declaration to be prioritized.

...and do you have other slot choices?

[davidtmyers]

Ctrl+Alt+F2, Ctrl+Alt+F3, Ctrl+Alt+F4 do indeed swap to the other TTYs, but those are all still presented on the same VGA output as the first one: the iKVM. I'm wondering if there's some way to say "start this TTY, on this GPU" (where "this GPU" is the discrete AMD one, and not the iKVM one where all the other TTYs are located).

I believe you when you say something isn't right, but I'm not sure I can provide any greater insight on that front. I double checked, and that output was indeed copied correctly.

I will swap /usr/share/X11/xorg.conf.d/52-whatever.conf back to using the "radeon" driver (it was previously switched to "amdgpu" for testing purposes per the post #11).

Lastly, I have many slot choices! The ASUS Pro WS WRX90E-SAGE SE motherboard I'm using has 7 PCIe 5.0 x16 slots (actually 7 physical PCIe 5.0 x16 slots, 6 of which are x16 electrical and 1 of which is x8 electrical). I'm currently plugged into the top one (labeled "PCIEX16(G5)_1" in the User Manual). I will move the GPU down to Slot 2 and report back.

[davidtmyers]

Alright, I've swapped the discrete GPU into Slot 2. This is evidenced by the new PCIe Bus ID assigned to the card (now "c1" instead of "e1"):

Code: Select all

root@workstation:~# lspci -vnkd 1002:
c1:00.0 0300: 1002:5b60 (prog-if 00 [VGA controller])
	Subsystem: 1002:3000
	Flags: bus master, fast devsel, latency 0, IRQ 272, IOMMU group 17
	Memory at c0000000 (32-bit, prefetchable) [size=256M]
	I/O ports at b000 [size=256]
	Memory at d0130000 (32-bit, non-prefetchable) [size=64K]
	Expansion ROM at d0100000 [disabled] [size=128K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Capabilities: [80] MSI: Enable+ Count=1/1 Maskable- 64bit+
	Capabilities: [100] Advanced Error Reporting
	Kernel driver in use: radeon
	Kernel modules: radeonfb, radeon, amdgpu

c1:00.1 0380: 1002:5b70
	Subsystem: 1002:3001
	Flags: fast devsel, IOMMU group 17
	Memory at d0120000 (32-bit, non-prefetchable) [disabled] [size=64K]
	Capabilities: [50] Power Management version 2
	Capabilities: [58] Express Endpoint, MSI 00
	Kernel modules: amdgpu

Unfortunately, there doesn't seem to be any change in behavior:

The monitor that is plugged into the discrete AMD GPU is off/in a power-save state. After logging into my motherboard's BMC/IPMI web-ui portal, I go to the iKVM within the web-ui and then remotely power on the system. I see the UEFI BIOS screen and the GRUB menu show up in the iKVM. Once the GRUB countdown expires, the kernel starts to boot. Mid-way through the boot process, the monitor plugged into the discrete AMD GPU turns on/leaves the power-save state, but remains blank (i.e., the power button on the monitor goes from amber to green, and the backlight comes on, but the screen is still empty with no text or mouse cursor or anything).

The machine fully boots up, and on the iKVM (which I can see from the web ui), I'm presented with a text console to log in. The monitor plugged into the discrete GPU is now "on" but again - still blank, no text or mouse cursor.

[CwF]

strange

I'd try slot 4
I'd also cycle the cable after it is booted and settle. Disconnect, pause , and reconnect the DP or HDMI cable. I've had one monitor do this on an amd, and it was the monitor not the card. I'd get a bios, then grub, then blank. The disconnect/reconnect worked for that one monitor, a different monitor worked fine.

[CwF]

One more thought,
Are you logged in still on the iKVM while trying and thus holding an x session open on the ast?
You would need to logoff for the cycle to move to the slot gpu.

[davidtmyers]

No problem, I can definitely give Slot 4 a shot.

Interesting... I swapped the GPU down to Slot 4 and... I'm now getting different behavior!

The machine seems to be stuck in a constant boot loop whenever this GPU is in Slot 4. I'm watching the Q-Code (those little diagnostic codes that at spelled out by an on-board 8-segment alphanumeric LED on the motherboard) cycle though, and it seems having this GPU in Slot 4 causes the machine to boot up for a few seconds, then reboot.

I'm going to re-seat the GPU back into Slot 1 or 2 to see if that fixes things... yep it did. Interesting.

Out of curiosity, what made you suggest I try Slot 4? The block diagram on Page "A-1" (PDF Page 55) of the User Manual doesn't seem to indicate anything special about this PCIe slot compared to Slots 1-3. (Slots 5, 6, and 7 come with signal re-drivers, while Slot 6 is also the x8 electrical in an x16 physical slot.)

[davidtmyers]

Are you logged in still on the iKVM while trying and thus holding an x session open on the ast? You would need to logoff for the cycle to move to the slot gpu.

Ah, interesting theory. Just to make sure I'm following you here: when you say "while trying" are you referring to when I press Ctrl+Alt+F2, Ctrl+Alt+F3, or Ctrl+Alt+F4 from a few posts back?

I initially would have thought that the iKVM is always holding an X session open regardless of whether anyone is logged into the web UI and actually viewing it, but perhaps not... This is a good idea. I just stopped the iKVM, logged out, and closed the window. I then tried Ctrl+Alt+F2 and Ctrl+Alt+F3. No change to the monitor plugged into the discrete GPU: still "on"/not in it's power-save state, but still a blank screen (backlight is on, but screen is black with no text, no cursor).

Logging back into the iKVM, I see that the virtual console is now on TTY3 (consistent with my last chord of Ctrl+Alt+F3).

[davidtmyers]

If anyone else has any suggestions, I'm all ears. Currently at a loss as to how to fix this and get KDE up and running on this GPU after installing a minimal environment.

[CwF]

Last idea-
not going in the direction you want to go, but I would disconnect the ast and test that something else is not amiss. You should remove the xorg file first and rely on detection. Then, add the ast back in working backwards.