Amdgpu seems to have broken somewhere between kernel 6.7.4 and 6.8.5 on Mac Pro 2009-2012 hardware.
Tested with an RX 480 8GB and RX 580 4GB. Nouveau works as expected on the GT 120 (Mac edition) and GTX 1660 Super
During boot the system will repeatedly print a stacktrace from amdgpu until probe failure. At some point during this, the fans will ramp up to max for a few seconds before ramping back down.
10:04:00 kernel: [drm] amdgpu kernel modesetting enabled.
10:04:00 kernel: amdgpu: Virtual CRAT table created for CPU
10:04:00 kernel: amdgpu: Topology: Add CPU node
10:04:00 kernel: amdgpu 0000:05:00.0: enabling device (0006 -> 0007)
10:04:00 kernel: [drm] initializing kernel modesetting (POLARIS10 0x1002:0x67DF 0x1043:0x0505 0xC7).
10:04:00 kernel: [drm] register mmio base: 0x90600000
10:04:00 kernel: [drm] register mmio size: 262144
10:04:00 kernel: [drm] add ip block number 0 <vi_common>
10:04:00 kernel: [drm] add ip block number 1 <gmc_v8_0>
10:04:00 kernel: [drm] add ip block number 2 <tonga_ih>
10:04:00 kernel: [drm] add ip block number 3 <gfx_v8_0>
10:04:00 kernel: [drm] add ip block number 4 <sdma_v3_0>
10:04:00 kernel: [drm] add ip block number 5 <powerplay>
10:04:00 kernel: [drm] add ip block number 6 <dm>
10:04:00 kernel: [drm] add ip block number 7 <uvd_v6_0>
10:04:00 kernel: [drm] add ip block number 8 <vce_v3_0>
10:04:00 kernel: amdgpu 0000:05:00.0: amdgpu: Fetched VBIOS from VFCT
10:04:00 kernel: amdgpu: ATOM BIOS: 115-D000PI0-100
10:04:00 kernel: [drm] UVD is enabled in VM mode
10:04:00 kernel: [drm] UVD ENC is enabled in VM mode
10:04:00 kernel: [drm] VCE enabled in VM mode
10:04:00 kernel: amdgpu 0000:05:00.0: vgaarb: deactivate vga console
10:04:00 kernel: amdgpu 0000:05:00.0: amdgpu: Trusted Memory Zone (TMZ) feature not supported
10:04:00 kernel: amdgpu 0000:05:00.0: amdgpu: PCIE atomic ops is not supported
10:04:00 kernel: [drm] vm size is 128 GB, 2 levels, block size is 10-bit, fragment size is 9-bit
10:04:00 kernel: amdgpu 0000:05:00.0: amdgpu: VRAM: 8192M 0x000000F400000000 - 0x000000F5FFFFFFFF (8192M used)
10:04:00 kernel: amdgpu 0000:05:00.0: amdgpu: GART: 256M 0x000000FF00000000 - 0x000000FF0FFFFFFF
10:04:00 kernel: [drm] Detected VRAM RAM=8192M, BAR=256M
10:04:00 kernel: [drm] RAM width 256bits GDDR5
10:04:00 kernel: [drm] amdgpu: 8192M of VRAM memory ready
10:04:00 kernel: [drm] amdgpu: 12006M of GTT memory ready.
10:04:00 kernel: [drm] GART: num cpu pages 65536, num gpu pages 65536
10:04:00 kernel: [drm] PCIE GART of 256M enabled (table at 0x000000F400000000).
10:04:00 kernel: [drm] Chained IB support enabled!
10:04:00 kernel: amdgpu: hwmgr_sw_init smu backed is polaris10_smu
10:04:00 kernel: [drm] Found UVD firmware Version: 1.130 Family ID: 16
10:04:00 kernel: [drm] Found VCE firmware Version: 53.26 Binary ID: 3
10:04:04 kernel: fbcon: Taking over console
10:04:05 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:09 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:13 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:19 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:25 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:30 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:36 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:40 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:46 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:52 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:04:56 systemd-udevd[386]: 0000:05:00.0: Worker [391] processing SEQNUM=2278 is taking a long time
10:04:59 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:05:05 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:05:11 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:05:15 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:05:17 kernel: amdgpu 0000:05:00.0: [drm] dce110_link_encoder_construct: Failed to get encoder_cap_info from VBIOS with error code 4!
10:05:17 kernel: amdgpu 0000:05:00.0: [drm] dce110_link_encoder_construct: Failed to get encoder_cap_info from VBIOS with error code 4!
10:05:17 kernel: amdgpu 0000:05:00.0: [drm] dce110_link_encoder_construct: Failed to get encoder_cap_info from VBIOS with error code 4!
10:05:17 kernel: [drm] Display Core v3.2.266 initialized on DCE 11.2
10:05:17 kernel: amdgpu 0000:05:00.0: [drm] *ERROR* No EDID read.
10:05:20 kernel: amdgpu 0000:05:00.0: amdgpu:
last message was failed ret is 0
10:05:22 kernel: amdgpu 0000:05:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring uvd_enc0 test failed (-110)
10:05:22 kernel: [drm:amdgpu_device_init [amdgpu]] *ERROR* hw_init of IP block <uvd_v6_0> failed -110
10:05:22 kernel: amdgpu 0000:05:00.0: amdgpu: amdgpu_device_ip_init failed
10:05:22 kernel: amdgpu 0000:05:00.0: amdgpu: Fatal error during GPU init
10:05:22 kernel: amdgpu 0000:05:00.0: amdgpu: amdgpu: finishing device.
10:05:22 kernel: ------------[ cut here ]------------
10:05:22 kernel: WARNING: CPU: 1 PID: 391 at drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c:622 amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: Modules linked in: amdgpu(+) video wmi amdxcp i2c_algo_bit drm_ttm_helper ttm drm_exec gpu_sched drm_suballoc_helper drm_buddy crc32c_intel sha512_ssse3 drm_display_helper sha256_ssse3 firewire_ohci e1000e sha1_ssse3 firewire_core cec crc_itu_t ip6_tables ip_tables fuse
10:05:22 kernel: CPU: 1 PID: 391 Comm: (udev-worker) Tainted: G I 6.8.4-100.fc38.x86_64 #1
10:05:22 kernel: Hardware name: Apple Inc. MacPro5,1/Mac-F221BEC8, BIOS MP51.88Z.007F.B03.1010071432 10/07/10
10:05:22 kernel: RIP: 0010:amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: Code: c0 74 33 48 8b 4e 10 48 83 39 00 74 29 89 d1 48 8d 04 88 8b 08 85 c9 74 11 f0 ff 08 74 07 31 c0 c3 cc cc cc cc e9 5a fd ff ff <0f> 0b b8 ea ff ff ff c3 cc cc cc cc b8 ea ff ff ff c3 cc cc cc cc
10:05:22 kernel: RSP: 0018:ffffb710007f7b30 EFLAGS: 00010246
10:05:22 kernel: RAX: ffff9300417b3940 RBX: ffff9300524187a8 RCX: 0000000000000000
10:05:22 kernel: RDX: 0000000000000000 RSI: ffff930052424db8 RDI: ffff930052400000
10:05:22 kernel: RBP: ffff9300524101e8 R08: 0000000000000000 R09: 0000000000000006
10:05:22 kernel: R10: 0000000000000000 R11: ffff930052400010 R12: ffff9300524105c8
10:05:22 kernel: R13: ffff930052400010 R14: ffff930052400000 R15: ffff930052424db8
10:05:22 kernel: FS: 00007f5851a35940(0000) GS:ffff9305a7840000(0000) knlGS:0000000000000000
10:05:22 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
10:05:22 kernel: CR2: 000055d14ad10ba8 CR3: 000000010146a000 CR4: 00000000000006f0
10:05:22 kernel: Call Trace:
10:05:22 kernel: <TASK>
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: ? __warn+0x81/0x130
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: ? report_bug+0x171/0x1a0
10:05:22 kernel: ? handle_bug+0x3c/0x80
10:05:22 kernel: ? exc_invalid_op+0x17/0x70
10:05:22 kernel: ? asm_exc_invalid_op+0x1a/0x20
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: amdgpu_fence_driver_hw_fini+0xfe/0x130 [amdgpu]
10:05:22 kernel: amdgpu_device_fini_hw+0xa6/0x400 [amdgpu]
10:05:22 kernel: ? blocking_notifier_chain_unregister+0x36/0x50
10:05:22 kernel: amdgpu_driver_load_kms+0xec/0x190 [amdgpu]
10:05:22 kernel: amdgpu_pci_probe+0x18b/0x510 [amdgpu]
10:05:22 kernel: local_pci_probe+0x45/0xa0
10:05:22 kernel: pci_device_probe+0xc7/0x240
10:05:22 kernel: really_probe+0x19e/0x3e0
10:05:22 kernel: ? __pfx___driver_attach+0x10/0x10
10:05:22 kernel: __driver_probe_device+0x78/0x160
10:05:22 kernel: driver_probe_device+0x1f/0x90
10:05:22 kernel: __driver_attach+0xd2/0x1c0
10:05:22 kernel: bus_for_each_dev+0x88/0xd0
10:05:22 kernel: bus_add_driver+0x116/0x220
10:05:22 kernel: driver_register+0x59/0x100
10:05:22 kernel: ? __pfx_amdgpu_init+0x10/0x10 [amdgpu]
10:05:22 kernel: do_one_initcall+0x5b/0x320
10:05:22 kernel: do_init_module+0x60/0x240
10:05:22 kernel: __do_sys_init_module+0x17f/0x1b0
10:05:22 kernel: ? syscall_exit_to_user_mode+0x80/0x230
10:05:22 kernel: do_syscall_64+0x89/0x170
10:05:22 kernel: ? exc_page_fault+0x7f/0x180
10:05:22 kernel: entry_SYSCALL_64_after_hwframe+0x73/0x7b
10:05:22 kernel: RIP: 0033:0x7f5852360aae
10:05:22 kernel: Code: 48 8b 0d 5d 03 0c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 49 89 ca b8 af 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 2a 03 0c 00 f7 d8 64 89 01 48
10:05:22 kernel: RSP: 002b:00007fffb6b52b28 EFLAGS: 00000246 ORIG_RAX: 00000000000000af
10:05:22 kernel: RAX: ffffffffffffffda RBX: 000055d14abe2250 RCX: 00007f5852360aae
10:05:22 kernel: RDX: 00007f585249d07d RSI: 00000000019de376 RDI: 00007f584f200010
10:05:22 kernel: RBP: 00007fffb6b52be0 R08: 000055d14abe01d0 R09: 00000000019deff0
10:05:22 kernel: R10: 000055d417a95310 R11: 0000000000000246 R12: 00007f585249d07d
10:05:22 kernel: R13: 0000000000020000 R14: 000055d14abddaa0 R15: 000055d14abe3e30
10:05:22 kernel: </TASK>
10:05:22 kernel: ---[ end trace 0000000000000000 ]---
10:05:22 kernel: ------------[ cut here ]------------
10:05:22 kernel: WARNING: CPU: 1 PID: 391 at drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c:622 amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: Modules linked in: amdgpu(+) video wmi amdxcp i2c_algo_bit drm_ttm_helper ttm drm_exec gpu_sched drm_suballoc_helper drm_buddy crc32c_intel sha512_ssse3 drm_display_helper sha256_ssse3 firewire_ohci e1000e sha1_ssse3 firewire_core cec crc_itu_t ip6_tables ip_tables fuse
10:05:22 kernel: CPU: 1 PID: 391 Comm: (udev-worker) Tainted: G W I 6.8.4-100.fc38.x86_64 #1
10:05:22 kernel: Hardware name: Apple Inc. MacPro5,1/Mac-F221BEC8, BIOS MP51.88Z.007F.B03.1010071432 10/07/10
10:05:22 kernel: RIP: 0010:amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: Code: c0 74 33 48 8b 4e 10 48 83 39 00 74 29 89 d1 48 8d 04 88 8b 08 85 c9 74 11 f0 ff 08 74 07 31 c0 c3 cc cc cc cc e9 5a fd ff ff <0f> 0b b8 ea ff ff ff c3 cc cc cc cc b8 ea ff ff ff c3 cc cc cc cc
10:05:22 kernel: RSP: 0018:ffffb710007f7b30 EFLAGS: 00010246
10:05:22 kernel: RAX: ffff9300417b3948 RBX: ffff930052418db0 RCX: 0000000000000000
10:05:22 kernel: RDX: 0000000000000002 RSI: ffff930052424db8 RDI: ffff930052400000
10:05:22 kernel: RBP: ffff9300524101f0 R08: 0000000000000000 R09: 0000000000000006
10:05:22 kernel: R10: 0000000000000000 R11: ffff930052400010 R12: ffff9300524105c8
10:05:22 kernel: R13: ffff930052400010 R14: ffff930052400000 R15: ffff930052424db8
10:05:22 kernel: FS: 00007f5851a35940(0000) GS:ffff9305a7840000(0000) knlGS:0000000000000000
10:05:22 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
10:05:22 kernel: CR2: 000055d14ad10ba8 CR3: 000000010146a000 CR4: 00000000000006f0
10:05:22 kernel: Call Trace:
10:05:22 kernel: <TASK>
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: ? __warn+0x81/0x130
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: ? report_bug+0x171/0x1a0
10:05:22 kernel: ? handle_bug+0x3c/0x80
10:05:22 kernel: ? exc_invalid_op+0x17/0x70
10:05:22 kernel: ? asm_exc_invalid_op+0x1a/0x20
10:05:22 kernel: ? amdgpu_irq_put+0x46/0x70 [amdgpu]
10:05:22 kernel: amdgpu_fence_driver_hw_fini+0xfe/0x130 [amdgpu]
10:05:22 kernel: amdgpu_device_fini_hw+0xa6/0x400 [amdgpu]
10:05:22 kernel: ? blocking_notifier_chain_unregister+0x36/0x50
10:05:22 kernel: amdgpu_driver_load_kms+0xec/0x190 [amdgpu]
10:05:22 kernel: amdgpu_pci_probe+0x18b/0x510 [amdgpu]
10:05:22 kernel: local_pci_probe+0x45/0xa0
10:05:22 kernel: pci_device_probe+0xc7/0x240
10:05:22 kernel: really_probe+0x19e/0x3e0
10:05:22 kernel: ? __pfx___driver_attach+0x10/0x10
10:05:22 kernel: __driver_probe_device+0x78/0x160
10:05:22 kernel: driver_probe_device+0x1f/0x90
10:05:22 kernel: __driver_attach+0xd2/0x1c0
10:05:22 kernel: bus_for_each_dev+0x88/0xd0
10:05:22 kernel: bus_add_driver+0x116/0x220
10:05:22 kernel: driver_register+0x59/0x100
10:05:22 kernel: ? __pfx_amdgpu_init+0x10/0x10 [amdgpu]
10:05:22 kernel: do_one_initcall+0x5b/0x320
10:05:22 kernel: do_init_module+0x60/0x240
10:05:22 kernel: __do_sys_init_module+0x17f/0x1b0
10:05:22 kernel: ? syscall_exit_to_user_mode+0x80/0x230
10:05:22 kernel: do_syscall_64+0x89/0x170
10:05:22 kernel: ? exc_page_fault+0x7f/0x180
10:05:22 kernel: entry_SYSCALL_64_after_hwframe+0x73/0x7b
10:05:22 kernel: RIP: 0033:0x7f5852360aae
10:05:22 kernel: Code: 48 8b 0d 5d 03 0c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 49 89 ca b8 af 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 2a 03 0c 00 f7 d8 64 89 01 48
10:05:22 kernel: RSP: 002b:00007fffb6b52b28 EFLAGS: 00000246 ORIG_RAX: 00000000000000af
10:05:22 kernel: RAX: ffffffffffffffda RBX: 000055d14abe2250 RCX: 00007f5852360aae
10:05:22 kernel: RDX: 00007f585249d07d RSI: 00000000019de376 RDI: 00007f584f200010
10:05:22 kernel: RBP: 00007fffb6b52be0 R08: 000055d14abe01d0 R09: 00000000019deff0
10:05:22 kernel: R10: 000055d417a95310 R11: 0000000000000246 R12: 00007f585249d07d
10:05:22 kernel: R13: 0000000000020000 R14: 000055d14abddaa0 R15: 000055d14abe3e30
10:05:22 kernel: </TASK>
10:05:22 kernel: ---[ end trace 0000000000000000 ]---
10:05:22 kernel: ------------[ cut here ]------------
This issue effects Silverblue F39 as the same behaviour will manifest once the system has applied an updated image.
Not sure of the exact kernel update that introduced this regression, as the system isn’t updated often.