Re: [Mining OS] SimpleMining.net - Easy to use GPU MINING Operating System

Mining with Claymore and get full HDD (128GB SSD) in half hour.
After checking find /var/log/syslog and kern.log grown up to 52GB each.
System write this 4 messages about 4-5 thousands per second !

Code:

Jul 9 23:58:37 test-rig kernel: [ 1399.017519] VM fault (0x01, vmid 6) at page 135290427, read from 'TC3' (0x54433300) (196)
Jul 9 23:58:37 test-rig kernel: [ 1399.017554] amdgpu 0000:04:00.0: GPU fault detected: 147 0x01c84801
Jul 9 23:58:37 test-rig kernel: [ 1399.017557] amdgpu 0000:04:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x08105E39
Jul 9 23:58:37 test-rig kernel: [ 1399.017559] amdgpu 0000:04:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x0C048001

Jul 9 23:58:37 test-rig kernel: [ 1399.017561] VM fault (0x01, vmid 6) at page 135290425, read from 'TC4' (0x54433400) (72)
Jul 9 23:58:37 test-rig kernel: [ 1399.017567] amdgpu 0000:01:00.0: GPU fault detected: 147 0x01f08801
Jul 9 23:58:37 test-rig kernel: [ 1399.017569] amdgpu 0000:01:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x08105E3E
Jul 9 23:58:37 test-rig kernel: [ 1399.017570] amdgpu 0000:01:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x0C088001

Jul 9 23:58:37 test-rig kernel: [ 1399.017572] VM fault (0x01, vmid 6) at page 135290430, read from 'TC6' (0x54433600) (136)
Jul 9 23:58:37 test-rig kernel: [ 1399.017607] amdgpu 0000:04:00.0: GPU fault detected: 147 0x01e84801
Jul 9 23:58:37 test-rig kernel: [ 1399.017610] amdgpu 0000:04:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x08105E3D
Jul 9 23:58:37 test-rig kernel: [ 1399.017612] amdgpu 0000:04:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x0C048001

Despite this mine worked properly 24.3 MH/s on each card until the disk is full

Someone also stumbled upon a similar problem?

System:
Lubuntu 16.04 with amdgpu-pro_16.30.3

5x RX 480 8GB ref model
Asrock FM2A88X + BTC

Hi, i have the same problem in my new project : http://simplemining.net/
Anyway it doesnt bother mining and about logs you can turn then off:
as root execute this command and you will have no longer such problem

systemctl disable rsyslog.service
then delete big file logs from /var/log/ adn reboot system, then it will work fine.
If you want i will release my http://simplemining.net/ Operating System which will have this fix and many more + feature updates.
If you want make donation BTC: 1ERRVbvV23h51ScH73nQr6TDmaieKNuuVD

Can someone please clarify what Tytanick means here? Is this only for independent Linux users and not Simplemining users?

I am using Simplemining.net and I get these error messages with four out of six RX480s.

Fixes I have tried:
I have the -dbg -1 code in the config text in the global options window.
I have tried each card with risers that I know are stable because they work with two of my cards.
I have tried various versions of drivers and kernels but found no difference with these errors.

I still am thinking that it is a driver/kernel issue but I have run out of options that I know to try.

Please help!

Only Tytanick could think suppressing an actual fairly serious warning is a "feature."

It's a high memclk and/or very tight VBIOS GDDR5 timings. It's causing memory errors.

My memory clock on Simpleminer is set at stock: 2000MHz (8000Mhz GDDR5). So should I learn how to decode the timing values from the HEX code? Or can I just alter the timings in Simplemining?

Have you flashed a diff VBIOS? If not, then Tytanick is probably using AMDGPU firmware and/or a kernel that is older than Jesus.

Damn it. The BIOS page for my card is offline atm. Anything else while I wait (perhaps for quite some time)? I know there was the other well known atikmdag-patcher-1.4.5 which was a kernel issue, but this wouldn't even install without crashes on my system, which no other driver has. Are there any other patches out there?