Using old amd gpu. Black screen after install. PLEASE HELP!!!

goodwinmcd · 2025-10-31 20:16:33

Hey guys, originally posted this after trying to install endeavour. Thought it was something to do with that so I wiped it out and started with a fresh arch install.

I can plug my hdmi into my motherboard and everything works fine. But I have an old amd gpu. Part of the AMD Radeon HD 7000 series "Southern Islands" nickname. So I got the basics working. Network, drives partitioned, bootup, ssh, package manager, etc. Then I went to install the mesa package to get the amdgpu driver. The docs say this should support all southern island gpus. It should just be plug and play from here? But not the case.

I've tried installing lib32-mesa.

Set kernal module params for southern island support in amdgpu and disable in radeon.

I've edited mkinitcpio.conf so that MODULES=(amdgpu radeon)

I've gone ahead and installed and launched a windows manager thinking it might not display anything unless I tell it to? (desperate)

I got a dvi to hdmi cord to try the dvi port on the gpu instead.

I've no clue what to do from here.......... Please if any body has any ideas of how to get to the bottom of this. Thinking of buying a newer gpu.... But I hate leaving things unsolved.

Some command logs

[gm@archgm ~]$ lspci -k -d ::03xx
04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Pitcairn XT [Radeon HD 7870 GHz Edition]
	Subsystem: XFX Pine Group Inc. Device 3250
	Kernel driver in use: amdgpu
	Kernel modules: radeon, amdgpu

[gm@archgm ~]$ sudo dmesg | grep -i amd
[    0.008239] RAMDISK: [mem 0xb2e35000-0xb58b5fff]
[    5.430981] [drm] amdgpu kernel modesetting enabled.
[    5.431126] amdgpu: Virtual CRAT table created for CPU
[    5.431138] amdgpu: Topology: Add CPU node
[    5.431300] amdgpu 0000:04:00.0: amdgpu: initializing kernel modesetting (PITCAIRN 0x1002:0x6818 0x1682:0x3250 0x00).
[    5.431331] amdgpu 0000:04:00.0: amdgpu: register mmio base: 0xEF900000
[    5.431332] amdgpu 0000:04:00.0: amdgpu: register mmio size: 262144
[    5.431418] amdgpu 0000:04:00.0: amdgpu: detected ip block number 0 <si_common>
[    5.431438] amdgpu 0000:04:00.0: amdgpu: detected ip block number 1 <gmc_v6_0>
[    5.431439] amdgpu 0000:04:00.0: amdgpu: detected ip block number 2 <si_ih>
[    5.431441] amdgpu 0000:04:00.0: amdgpu: detected ip block number 3 <gfx_v6_0>
[    5.431442] amdgpu 0000:04:00.0: amdgpu: detected ip block number 4 <si_dma>
[    5.431443] amdgpu 0000:04:00.0: amdgpu: detected ip block number 5 <si_dpm>
[    5.431444] amdgpu 0000:04:00.0: amdgpu: detected ip block number 6 <dce_v6_0>
[    5.431445] amdgpu 0000:04:00.0: amdgpu: detected ip block number 7 <uvd_v3_1>
[    5.449235] amdgpu 0000:04:00.0: amdgpu: Fetched VBIOS from ROM BAR
[    5.449238] amdgpu: ATOM BIOS: 113-787CDF83_113-AD42800-100-XF
[    5.449247] kfd kfd: amdgpu: PITCAIRN  not supported in kfd
[    5.465594] amdgpu 0000:04:00.0: vgaarb: deactivate vga console
[    5.465597] amdgpu 0000:04:00.0: amdgpu: Trusted Memory Zone (TMZ) feature not supported
[    5.465600] amdgpu 0000:04:00.0: amdgpu: PCIE atomic ops is not supported
[    5.465773] amdgpu 0000:04:00.0: amdgpu: vm size is 128 GB, 2 levels, block size is 10-bit, fragment size is 9-bit
[    5.465975] amdgpu 0000:04:00.0: amdgpu: VRAM: 2048M 0x000000F400000000 - 0x000000F47FFFFFFF (2048M used)
[    5.465977] amdgpu 0000:04:00.0: amdgpu: GART: 1024M 0x000000FF00000000 - 0x000000FF3FFFFFFF
[    5.466215] amdgpu 0000:04:00.0: amdgpu: amdgpu: 2048M of VRAM memory ready
[    5.466217] amdgpu 0000:04:00.0: amdgpu: amdgpu: 16021M of GTT memory ready.
[    5.466862] amdgpu 0000:04:00.0: amdgpu: PCIE GART of 1024M enabled (table at 0x000000F400400000).
[    5.467868] amdgpu 0000:04:00.0: amdgpu: [drm] Internal thermal controller with fan control
[    5.467879] [drm] amdgpu: dpm initialized
[    5.467914] [drm] AMDGPU Display Connectors
[    5.913711] amdgpu 0000:04:00.0: amdgpu: SE 2, SH per SE 2, CU per SH 5, active_cu_number 20
[    6.316103] amdgpu 0000:04:00.0: amdgpu: Runtime PM not available
[    6.316491] amdgpu 0000:04:00.0: [drm] Registered 6 planes with drm panic
[    6.316493] [drm] Initialized amdgpu 3.64.0 for 0000:04:00.0 on minor 1
[    6.447214] fbcon: amdgpudrmfb (fb0) is primary device
[    6.447220] amdgpu 0000:04:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[gm@archgm ~]$

ewaller · 2025-11-01 00:05:50

Please edit your first post and update your title to remove the all caps redundant plea for help.

seth · 2025-11-01 08:47:36

Set kernal module params for

https://en.wikipedia.org/wiki/Kernel *sigh*

I've gone ahead and installed and launched a windows manager thinking it might not display anything unless I tell it to? (desperate)

So I *guess* you don't have any output on the console?
Do you if you add "nomodeset" to the kernel parameters?
(You won't be able to start a display server, this is just to test the principal output behavior)

The dmesg grep is from when using the onboard graphics?
Please post your Xorg log, https://wiki.archlinux.org/title/Xorg#General

https://raw.githubusercontent.com/torva … /fbcon.rst
See the "fbcon=map:<0123>" paragraph

cryptearth · 2025-11-01 12:33:52

i had a 7870 myself (asus du2 oc) - but that was about 10 years ago long before i learnrd about radeon vs amdgpu with my r290x years later - so, sanity check: does it work with radeon (and maybe mesa-amber)?

seth · 2025-11-01 13:23:12

You can try radeon (though it tends to perform worse for being neglected) but mesa-amber only supports r100 and r200 - radeonsi is in mesa.

Arch Linux

#1 2025-10-31 20:16:33

Using old amd gpu. Black screen after install. PLEASE HELP!!!

#2 2025-11-01 00:05:50

Re: Using old amd gpu. Black screen after install. PLEASE HELP!!!

#3 2025-11-01 08:47:36

Re: Using old amd gpu. Black screen after install. PLEASE HELP!!!

#4 2025-11-01 12:33:52

Re: Using old amd gpu. Black screen after install. PLEASE HELP!!!

#5 2025-11-01 13:23:12

Re: Using old amd gpu. Black screen after install. PLEASE HELP!!!

Board footer