Ubuntu 16.04 one gpu low hashrate

[kobazik]

Hi,

I'm running Ubuntu 16.04 with claymore 9.3 and amdgpu-pro drivers 17.10. There are six XFX RX480 8GB cards in the system with low power bios 29mh/s. Five out of six cards are mining fine but GPU0 is only doing 4mh/s.

ETH: 05/01/17-10:47:41 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 147.562 Mh/s, Total Shares: 18, Rejected: 0, Time: 00:05
ETH: GPU0 4.168 Mh/s, GPU1 28.694 Mh/s, GPU2 28.676 Mh/s, GPU3 28.688 Mh/s, GPU4 28.681 Mh/s, GPU5 28.656 Mh/s
GPU0 t=43C fan=56%, GPU1 t=60C fan=56%, GPU2 t=56C fan=56%, GPU3 t=57C fan=56%, GPU4 t=59C fan=56%, GPU5 t=54C fan=56%

I had the same setup running Windows previously and all cards were mining with ~29mh/s.

# lspci |grep VGA
00:02.0 VGA compatible controller: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller (rev 06)
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)
02:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)
03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)
04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)
05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)
06:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Device 67df (rev c7)

=================== ROCm System Management Interface ===================
============================================================================
GPU[0] : GPU ID: 0x0402
GPU[1] : GPU ID: 0x67df
GPU[2] : GPU ID: 0x67df
GPU[3] : GPU ID: 0x67df
GPU[4] : GPU ID: 0x67df
GPU[5] : GPU ID: 0x67df
GPU[6] : GPU ID: 0x67df
============================================================================
=================== End of ROCm SMI Log ===================


For some reason roc-smi cannot detect clocks on the problematic card:


# ./rocm-smi -d 1 -a

=================== ROCm System Management Interface ===================
============================================================================
GPU[1] : GPU ID: 0x67df
============================================================================
============================================================================
GPU[1] : Temperature: 44.0c
============================================================================
============================================================================
GPU[1] : Unable to determine current clocks. Check dmesg or GPU temperature
============================================================================
GPU[1] : Fan Level: 145 (56.86)%
============================================================================
============================================================================
GPU[1] : Current PowerPlay Level: auto
============================================================================
============================================================================
GPU[1] : Current OverDrive value: 0%
============================================================================
============================================================================
GPU[1] : Compute Power Profile not supported
============================================================================
============================================================================
GPU[1] : Supported GPU clock frequencies on GPU1
GPU[1] : 0: 300Mhz
GPU[1] : 1: 515Mhz
GPU[1] : 2: 775Mhz
GPU[1] : 3: 915Mhz
GPU[1] : 4: 940Mhz
GPU[1] : 5: 975Mhz
GPU[1] : 6: 1000Mhz
GPU[1] : 7: 1095Mhz
GPU[1] :
GPU[1] : Supported GPU Memory clock frequencies on GPU1
GPU[1] : 0: 300Mhz
GPU[1] : 1: 2080Mhz
GPU[1] :
============================================================================
=================== End of ROCm SMI Log ===================

Other cards e.g GPU2 is fine

#./rocm-smi -d 2 -a


=================== ROCm System Management Interface ===================
============================================================================
GPU[2] : GPU ID: 0x67df
============================================================================
============================================================================
GPU[2] : Temperature: 59.0c
============================================================================
============================================================================
GPU[2] : GPU Clock Level: 7 (1095Mhz)
GPU[2] : GPU Memory Clock Level: 1 (2080Mhz)
============================================================================
============================================================================
GPU[2] : Fan Level: 145 (56.86)%
============================================================================
============================================================================
GPU[2] : Current PowerPlay Level: auto
============================================================================
============================================================================
GPU[2] : Current OverDrive value: 0%
============================================================================
============================================================================
GPU[2] : Compute Power Profile not supported
============================================================================
============================================================================
GPU[2] : Supported GPU clock frequencies on GPU2
GPU[2] : 0: 300Mhz
GPU[2] : 1: 515Mhz
GPU[2] : 2: 775Mhz
GPU[2] : 3: 915Mhz
GPU[2] : 4: 940Mhz
GPU[2] : 5: 975Mhz
GPU[2] : 6: 1000Mhz
GPU[2] : 7: 1095Mhz *
GPU[2] :
GPU[2] : Supported GPU Memory clock frequencies on GPU2
GPU[2] : 0: 300Mhz
GPU[2] : 1: 2080Mhz *
GPU[2] :
============================================================================
=================== End of ROCm SMI Log ===================

I can also see amdgpdu powerplay and voltage errors in dmesg:

[ 9.135564] amdgpu: [powerplay] amdgpu: powerplay sw initialized
[ 9.135998] amdgpu 0000:06:00.0: fence driver on ring 0 use gpu addr 0x0000000200000008, cpu addr 0xffff8802baefd008
[ 9.136556] amdgpu 0000:06:00.0: fence driver on ring 1 use gpu addr 0x0000000200000018, cpu addr 0xffff8802baefd018
[ 9.137110] amdgpu 0000:06:00.0: fence driver on ring 2 use gpu addr 0x0000000200000028, cpu addr 0xffff8802baefd028
[ 9.137597] amdgpu 0000:06:00.0: fence driver on ring 3 use gpu addr 0x0000000200000038, cpu addr 0xffff8802baefd038
[ 9.138125] amdgpu 0000:06:00.0: fence driver on ring 4 use gpu addr 0x0000000200000048, cpu addr 0xffff8802baefd048
[ 9.138558] amdgpu 0000:06:00.0: fence driver on ring 5 use gpu addr 0x0000000200000058, cpu addr 0xffff8802baefd058
[ 9.138979] amdgpu 0000:06:00.0: fence driver on ring 6 use gpu addr 0x0000000200000068, cpu addr 0xffff8802baefd068
[ 9.139446] amdgpu 0000:06:00.0: fence driver on ring 7 use gpu addr 0x0000000200000078, cpu addr 0xffff8802baefd078
[ 9.139960] amdgpu 0000:06:00.0: fence driver on ring 8 use gpu addr 0x0000000200000088, cpu addr 0xffff8802baefd088
[ 9.140408] amdgpu 0000:06:00.0: fence driver on ring 9 use gpu addr 0x0000000200000098, cpu addr 0xffff8802baefd098
[ 9.140826] amdgpu 0000:06:00.0: fence driver on ring 10 use gpu addr 0x00000002000000a8, cpu addr 0xffff8802baefd0a8
[ 9.141243] amdgpu 0000:06:00.0: fence driver on ring 11 use gpu addr 0x00000002000000b8, cpu addr 0xffff8802baefd0b8
[ 9.145606] amdgpu 0000:06:00.0: fence driver on ring 12 use gpu addr 0x0000000001165420, cpu addr 0xffffc9000c25a420
[ 9.146956] amdgpu 0000:06:00.0: fence driver on ring 13 use gpu addr 0x00000002000000d8, cpu addr 0xffff8802baefd0d8
[ 9.147444] amdgpu 0000:06:00.0: fence driver on ring 14 use gpu addr 0x00000002000000e8, cpu addr 0xffff8802baefd0e8
[ 9.187800] amdgpu: [powerplay] [AVFS] Something is broken. See log!
[ 9.190613] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.191538] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.192014] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.192491] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.192965] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.193427] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.193875] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.194307] amdgpu: [powerplay] Can't find requested voltage id in vdd_dep_on_sclk table!
[ 9.219785] [drm] amdgpu: freesync_module init done ffff8802bae10c60.
[ 9.452036] amdgpu 0000:06:00.0: fb6: amdgpudrmfb frame buffer device

[1713889859]

1713889859

[1713889859]

1713889859

[Metalspawn]

Hi.

I have 580s and I am experiencing the same issue. Did you solve it ?

Running on an ASUS H270 Prime Pro with Gigabyte RX580 gaming 8G

[mugenkami]

Had issues with hardware before, check risers, PCIE ports... 

[Metalspawn]

I got it solved by playing around with the settings in the bios. Forcing everything to GEn 1 caused the issue - Turned on Auto and it worked. 4G decoding was still enabled.