10.0 not booting with amdgpu

[ssartor <ssartor%bellatlantic.net@localhost>]

Hi,

I just installed 10.0 (from 9.3) using sysupgade and installed the gpufw.tar.xz set using sysinst.  I’m running a SFF box that, according to dmesg, has an AMD Ryzen Embedded V1605B with Radeon Vega Gfx.  lspci shows "04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raven Ridge [Radeon Vega Series / Radeon Vega Mobile Series] (rev 83)”.  From what I’ve read this should be compatible with amdgpu.  So I compiled a kernel with amdgpu (but left out radeon) and booted.  Before the box locked up I saw the following:

[drm] add ip block number 6 ‹powerplay>
[drm] add ip block number 7 ‹dm>
[drm] add ip block number 8 <vcn_vl_0>
ATOM BIOS: 113-RAVEN-107
[drm] VCN decode is enabled in VM mode
[drm] VCN encode is enabled in VM mode
[drm] JPEG decode is enabled in VM mode
[drm] vm size is 262144 GB, 4 levels, block size is 9-bit, fragment size is 9-bit
amdgpu0: VRAM: 256M 0x000000F400000000 - 0x000000F40FFFFFFF (256M used)
amdgpu0: GART: 1024M 0x0000000000000000 - 0x000000003FFFFFFF 
amdgpu0: AGP: 267419648M 0x000000F800000000 - 0x0000FFFFFFFFFFFF
[drm] Detected VRAM RAM=256M, BAR=256M
[drm] RAM width 128bits DDR4
Zone kernel: Available graphics memory: 9007199253816026 KiB
Zone dnm32: Available graphics memory: 2097152 KiB
[drm] amdgpu: 256M of VRAM memory ready
[drm] amdgpu: 3072M of GTT memory ready.
[drm] GART: num cpu pages 262144, num gpu pages 262144
[drm] PCIE GART of 1024M enabled (table at 0x000000F400900000) .
amdgpu0: interrupting at msi8 vec 0 (amdgpu0)
[drm] use_doorbell being set to: [true]
hwmgr_sw_init smu backed is smu10_smu
powerplay sw init successfully
[drm] Found VCN Firmuare Version ENC: 1.12 DEC: 2 VEP: 0 Revision: 9
[drm] PSP loading VCN firmware
[drm] reserve 0x400000 from Oxf40f800000 for PSP TMR
warning: [drm] psp command (0x5) failed and response status is (0x0)
{drm:netbsd:psp_hw_start+0x337} «ERROR* PSP load tmr failed!
{drm:netbsd:psp_hw_init+0x126} *ERROR PSP firmware loading failed
{drm:netbsd:amdgpu_device_fw_loading+0xfa} *ERROR* hw_init of IP block ‹psp> failed -22
amdgpu0: error: amdgpu_device_ip_init failed
amdgpu0: error: Fatal error during GPU init
[drm] amdgpu: finishing device.
warning: /usr/src/sys/external/bsd/drm2/dist/drm/drm_mm.c: 1095: Memory manager not clean dur
warning: /us/src/sys/external/bsd/drm2/dist/drm/drm_mm.c: 1095: Memory manager not clean dur
Delayed destroy list was clean
Swap list 0 was clean
Swap list 1 was clean
Swap list 2 was clean
Swap list 3 was clean

At this point the box hangs…

Any ideas as to what the issue might be or is this system just not compatible with amdgpu?

thanks,

Steve