More
referral
Increase your income with Hive. Invite your friends and earn real cryptocurrency!

AMD RX 6700xt 6800 rig - Can't upgrade HiveOS past 0.6-203

I have not been able to upgrade my AMD 4x6700xt & 3x6800 HiveOS rig past v0.6-203. If I do attempt an upgrade, it completes and boots. However, it seems there’s a driver or OC issue then. The load of the rig goes through the roof and no hashing starts. If I try to do an amd-info the command hangs… When I revert back to 0.6-203 everything works again. I’ve tried wiping OC settings, deleting flight sheets and re-creating with no luck… Looking for some help on this as I’ve tried multiple different versions and the same issue occurs.

Here are some errors I received after upgrading to the latest 0.6-209 -

Sep 01 14:15:03 gpu01 kernel: amdgpu 0000:0e:00.0: amdgpu: message: GetMaxDpmFreq (30) param: 0x00000000 is timeout (no response)
Sep 01 14:15:06 gpu01 kernel: [drm:amdgpu_job_timedout [amdgpu]] ERROR ring sdma1 timeout, signaled seq=1, emitted seq=2
Sep 01 14:15:06 gpu01 kernel: [drm:amdgpu_job_timedout [amdgpu]] ERROR Process information: process pid 0 thread pid 0
Sep 01 14:15:09 gpu01 kernel: amdgpu 0000:05:00.0: amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!
Sep 01 14:15:09 gpu01 kernel: [drm:amdgpu_job_timedout [amdgpu]] ERROR ring sdma0 timeout, signaled seq=2, emitted seq=3
Sep 01 14:15:09 gpu01 kernel: [drm:amdgpu_job_timedout [amdgpu]] ERROR Process information: process pid 0 thread pid 0
Sep 01 14:15:11 gpu01 kernel: amdgpu 0000:0e:00.0: amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!
Sep 01 14:15:16 gpu01 kernel: amdgpu 0000:05:00.0: amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!
Sep 01 14:15:19 gpu01 kernel: amdgpu 0000:0e:00.0: amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!
Sep 01 14:15:24 gpu01 kernel: amdgpu 0000:05:00.0: amdgpu: Msg issuing pre-check failed and SMU may be not in the right state!

Output for OC -

Detected 7 AMD cards

GPU BUS ID : 05 08 0b 0e 14 17 1a
CORE_CLOCK : 1300 0 1300 1350 1350 1350 1350
CORE_VDDC : 950 0 950 950 700 700 700
CORE_STATE : 0 0 0 0 5 5 5
MEM_CLOCK : 1075 0 1075 1075 1070 1070 1070
MEM_STATE :
MVDD : 950 0 950 950 0 0 0
VDDCI :
SOCCLK :
SOCVDDMAX :
REF :
FAN : 60 0 60 60 50 50 50
PL :
AGGRESSIVE =

=== GPU 0, 05:00.0 Radeon RX 6700 XT 12272 MB #0 === 14:18:27
Default Power Play settings from VBIOS for Navi20
CORE Clock max: 2950MHz, Voltage: 912-1200mV, SOC Clock: 480-1200MHz, Voltage: 825-1150mV
MEMORY Clock def/max: 1000/1075 MHz, Voltage: 1250-1350 mV, VDDCI: 675-850mV, TC: 1
POWER PL: 186W OV: -6%/+15%, TDC GFX: 157A, TDC SOC: 31A, TEMP Target: 85C

1 Like

It seems like i’ve resolved this. I had to completely blow away my OC settings since DPM is a hidden value in some views…Once I got rid of that I was able to upgrade and been running for about an hour.

Now onto the issue of why my cards are drawing more power (atleast what hiveos is reporting) after the upgrade.

Core MHz
Absolute core Memory MHz
Absolute memory Core undervolt
Voltage MVDD
Memory voltage MVDDCI
Memory controller voltage
1250 1050 675 1300 800

Hashrate Power Efficiency Earnings ROI
45.59 98 w 0.465 Mh/w $3.46 138 days

Was dpm state the problem ? I have same issue :confused: