[OS] rxOC easy-to-use Linux AMD Mining v_stopgap

[kylewalter]

I'm not sure if anyone is aware of this, but the motherboards that fullzero lists are all intel motherboards.
If anyone was wondering though, I have a MSI 970A SLI Krait Edition AMD motherboard and an Athlon II CPU and I got fed up with ethOS, so I gave rxOC a shot.  I can confirm this works with 4 GPUs.

by the way fullzero, thanks for nvOC and rxOC.  Amazing work!

[1713589434]

1713589434

[1713589434]

1713589434

[1713589434]

1713589434

[xleejohnx]

okay been trying to test this OS
but the love of me i cant get it to stop rebooting
it comes up with a popup says HI m1, you have a file open what would you like to do.... bam reboot

im a moron i left the SRR connected to the motherboard so it kept restarting it because it wasnt getting the signal..

[citronick]

okay been trying to test this OS
but the love of me i cant get it to stop rebooting
it comes up with a popup says HI m1, you have a file open what would you like to do.... bam reboot

im a moron i left the SRR connected to the motherboard so it kept restarting it because it wasnt getting the signal..

LOL...

very customizable this rxOC and nvOC

at the same time, for a Linux newbie,  great way to play and learn Linux

thumbs up to fullzero

[OverEasy]

Ok I am converting all rigs to this now and ran into an issue. All rigs cannot be called m1-desktop!

How can I change the hostname so all your scripts still work?

m2-desktop, etc...?

[xleejohnx]

Ok I am converting all rigs to this now and ran into an issue. All rigs cannot be called m1-desktop!

How can I change the hostname so all your scripts still work?

m2-desktop, etc...?

Why can't the rigs all be called that?
What issue are you having?

[stef_stef]

Just what I was looking for

One question though, does that driver cause any problems with modded bios on the rx 480?

[fullzero]

I'm not sure if anyone is aware of this, but the motherboards that fullzero lists are all intel motherboards.
If anyone was wondering though, I have a MSI 970A SLI Krait Edition AMD motherboard and an Athlon II CPU and I got fed up with ethOS, so I gave rxOC a shot.  I can confirm this works with 4 GPUs.

by the way fullzero, thanks for nvOC and rxOC.  Amazing work!

Thanks kylewalter,

Good to know it works with:

MSI 970A SLI Krait Edition AMD motherboard and an Athlon II CPU

[fullzero]

okay been trying to test this OS
but the love of me i cant get it to stop rebooting
it comes up with a popup says HI m1, you have a file open what would you like to do.... bam reboot

im a moron i left the SRR connected to the motherboard so it kept restarting it because it wasnt getting the signal..

LOL...

very customizable this rxOC and nvOC

at the same time, for a Linux newbie,  great way to play and learn Linux

thumbs up to fullzero

Glad you like them; I still need to get to a bunch of your requests. 

[fullzero]

Just what I was looking for

One question though, does that driver cause any problems with modded bios on the rx 480?

I don't have any 480s so I can't test this; has anyone tried using a custom rom with rx480s + rxOC?

[fullzero]

Ok I am converting all rigs to this now and ran into an issue. All rigs cannot be called m1-desktop!

How can I change the hostname so all your scripts still work?

m2-desktop, etc...?

Why can't the rigs all be called that?
What issue are you having?

What service / program are you using, that I am guessing ( is mapping hostname to IP addresses) ?

If you change the hostname you will need to add Maximus007s change again with the new hostname in place of m1.

Code:

sudo visudo  -f /etc/sudoers.d/90-m1
m1  ALL=(ALL) NOPASSWD:ALL

This is the only thing that stands out off the top of my head.

I am planning on allowing for the modification of host names in a later version.

[OverEasy]

Ok I am converting all rigs to this now and ran into an issue. All rigs cannot be called m1-desktop!

How can I change the hostname so all your scripts still work?

m2-desktop, etc...?

Why can't the rigs all be called that?
What issue are you having?

Teamviewer. Now I have 3 m1-desktops. Teamviewer having an issue with that. Or perhaps I am using TeamViewer wrong..

[xleejohnx]

Ok I am converting all rigs to this now and ran into an issue. All rigs cannot be called m1-desktop!

How can I change the hostname so all your scripts still work?

m2-desktop, etc...?

Why can't the rigs all be called that?
What issue are you having?

Teamviewer. Now I have 3 m1-desktops. Teamviewer having an issue with that. Or perhaps I am using TeamViewer wrong..

You can change the hostname of the computer that TeamViewer uses without changing the hostname itself
When you sign in with your account it gives an option to add this computer that's when you change the hostname

[xleejohnx]

Just what I was looking for

One question though, does that driver cause any problems with modded bios on the rx 480?

I don't have any 480s so I can't test this; has anyone tried using a custom rom with rx480s + rxOC?

I don't have 480s anymore I have all 470s
And have no issues with my custom bios

[OverEasy]

Good Morning FullZero -
I apoligize for the long post but I need to get it all in here

My first working mining rxOC AMD rig is up and running.

I want to ask some questions about clock settings.

Question is easy. How do I tune Linux version to get speeds near Windows?

As a base line I am dual mining Claymore ETH/SIA at -dcri 12

My cards are as follows -

GPU #0: Hawaii, 7991 MB available, 40 compute units (R9 390)
GPU #1: Ellesmere, 4050 MB available, 36 compute units (RX 570)
GPU #2: Ellesmere, 4050 MB available, 32 compute units (RX 470)

The 2 RX's have been bios modded and under windows 10 they run app. like this -

ETH: GPU0 29.897 Mh/s, GPU1 27.030 Mh/s, GPU2 27.096 Mh/s

I get those speeds by using Trixx and setting the following -

GPU #0: Hawaii, 7991 MB available, 40 compute units (R9 390) - 1100/1500
GPU #1: Ellesmere, 4050 MB available, 36 compute units (RX 570) - 1140/1900
GPU #2: Ellesmere, 4050 MB available, 32 compute units (RX 470) - 1140/1880

OK thats my stuff and how it runs on Windows...Now lets talk Linux

*****************************************************************************************

As a base line for the rest of this talk.
Using rxOC V12 with Parallax's mod
Dual mining ETH/SIA Claymore 9.5 -dcri 12
ETH_EXTENTION_ARGUMENTS="-dcri 12"
Same cards as above -

MY CURRENT SPEED WITH THE FOLLOWING SETTINGS -

ETH: GPU0 24.910 Mh/s, GPU1 24.086 Mh/s, GPU2 25.116 Mh/s


My oneBASH is currently set as follows -

 
PERFORMANCE_LEVEL="HIGH"        # HIGH LOW
 
# select level: to see supported clocks scroll to the top of the mining process
__CORE_OVERCLOCK_LEVEL=4   # for ETH use lowest without decreasing the hashrate / Highest for ZEC
MEMORY_OVERCLOCK_LEVEL=7   # use highest level
 
OVERDRIVE="NO"                  # YES NO  <---LOCKS UP WHEN I TURN THIS ON
 

 
My ROCm Output is as follows -

============================================================================
GPU[0]       : Supported GPU clock frequencies on GPU0
GPU[0]       : 0: 300Mhz
GPU[0]       : 1: 516Mhz
GPU[0]       : 2: 727Mhz
GPU[0]       : 3: 840Mhz
GPU[0]       : 4: 890Mhz *
GPU[0]       : 5: 936Mhz
GPU[0]       : 6: 977Mhz
GPU[0]       : 7: 1015Mhz
GPU[0]       :
GPU[0]       : Supported GPU Memory clock frequencies on GPU0
GPU[0]       : 0: 150Mhz
GPU[0]       : 1: 1500Mhz *
GPU[0]       :
GPU[1]       : Supported GPU clock frequencies on GPU1
GPU[1]       : 0: 300Mhz
GPU[1]       : 1: 608Mhz
GPU[1]       : 2: 910Mhz
GPU[1]       : 3: 1077Mhz *
GPU[1]       : 4: 1145Mhz
GPU[1]       : 5: 1191Mhz
GPU[1]       : 6: 1236Mhz
GPU[1]       : 7: 1266Mhz
GPU[1]       :
GPU[1]       : Supported GPU Memory clock frequencies on GPU1
GPU[1]       : 0: 300Mhz
GPU[1]       : 1: 1750Mhz *
GPU[1]       :
GPU[2]       : Supported GPU clock frequencies on GPU2
GPU[2]       : 0: 300Mhz
GPU[2]       : 1: 588Mhz
GPU[2]       : 2: 952Mhz
GPU[2]       : 3: 1041Mhz
GPU[2]       : 4: 1106Mhz *
GPU[2]       : 5: 1168Mhz
GPU[2]       : 6: 1209Mhz
GPU[2]       : 7: 1264Mhz
GPU[2]       :
GPU[2]       : Supported GPU Memory clock frequencies on GPU2
GPU[2]       : 0: 300Mhz
GPU[2]       : 1: 1000Mhz
GPU[2]       : 2: 1750Mhz *
GPU[2]       :
============================================================================
===================          End of ROCm SMI Log         ===================




===================   ROCm System Management Interface   ===================
GPU[0]       : Successfully reset GPU and Memory clocks
GPU[1]       : Successfully reset GPU and Memory clocks
GPU[2]       : Successfully reset GPU and Memory clocks
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
============================================================================
GPU[0]       : Successfully set current PowerPlay Level to high
GPU[1]       : Successfully set current PowerPlay Level to high
GPU[2]       : Successfully set current PowerPlay Level to high
============================================================================
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
GPU[0]       : Unable to set clock to unsupported Level - Max Level is 1
GPU[1]       : Unable to set clock to unsupported Level - Max Level is 1  <----WHAT'S HAPPENING HERE?
GPU[2]       : Unable to set clock to unsupported Level - Max Level is 2
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
GPU[0]       : Successfully set GPU Clock frequency mask to Level 4
GPU[1]       : Successfully set GPU Clock frequency mask to Level 4
GPU[2]       : Successfully set GPU Clock frequency mask to Level 4
===================          End of ROCm SMI Log         ===================

[Zucca]

I posted this on the other topic, but this is the OS I am using:
I am trying to start using your operating system. Unfortunately I am not able to properly start it.
This is what I am getting at login:

http://imgur.com/a/2U3Wr

Does it work with R9 Fury GPUs?

It should work with R9 fury GPUs.

You have a disk error.  Ubuntu grubloader will attempt to repair the error; if you wait a long time it might repair it and boot; but I would do the following:

Reimage the USB key; like so:

first check and see if your zip has downloaded correctly by following the directions here:

https://bitcointalk.org/index.php?topic=1984302.msg19757170#msg19757170

If your zip produced the same hash it is good.

Maybe you imaged from the zip without extracting?

Ensure that you extract the image before imaging.

After imaging is complete, and you have configured and saved oneBash:

ensure that you properly eject the USB key.

Let me know if this works.

I did it all the process again, downloading, checking, extracting, flashing.
Same error .
I am using an 32Gb Adata USB Key.

What motherboard are you using?

Gygabyte GA-170x-ud3 with disabled audio controller.
I am currently mining with an Ubuntu Server

[B1tUnl0ck3r]

Question is easy. How do I tune Linux version to get speeds near Windows?




===================   ROCm System Management Interface   ===================
GPU[0]       : Unable to set clock to unsupported Level - Max Level is 1
GPU[1]       : Unable to set clock to unsupported Level - Max Level is 1  <----WHAT'S HAPPENING HERE?
GPU[2]       : Unable to set clock to unsupported Level - Max Level is 2
===================          End of ROCm SMI Log         ===================

that's the question that may kill gpu mining. the point of gpus mining was to be able to decentralize mining, actually btc mining is more accessible than gpus mining. There is a solution to easy oc with ubuntu as ethos and this have the functionality. however it's not released publicly. Asics or PoS.

[OverEasy]

Just launched second rxOC rig with 4 AMD's.

Exact same issue!

Doing some testing and editing of oneBash....I'll let ya know how it goes

[fullzero]

I posted this on the other topic, but this is the OS I am using:
I am trying to start using your operating system. Unfortunately I am not able to properly start it.
This is what I am getting at login:

http://imgur.com/a/2U3Wr

Does it work with R9 Fury GPUs?

It should work with R9 fury GPUs.

You have a disk error.  Ubuntu grubloader will attempt to repair the error; if you wait a long time it might repair it and boot; but I would do the following:

Reimage the USB key; like so:

first check and see if your zip has downloaded correctly by following the directions here:

https://bitcointalk.org/index.php?topic=1984302.msg19757170#msg19757170

If your zip produced the same hash it is good.

Maybe you imaged from the zip without extracting?

Ensure that you extract the image before imaging.

After imaging is complete, and you have configured and saved oneBash:

ensure that you properly eject the USB key.

Let me know if this works.

I did it all the process again, downloading, checking, extracting, flashing.
Same error .
I am using an 32Gb Adata USB Key.

What motherboard are you using?

Gygabyte GA-170x-ud3 with disabled audio controller.
I am currently mining with an Ubuntu Server

Maybe this will work:

Ensure the monitor is connected to the primary GPU ( the one in the 16x slot closest to the CPU )

Disconnect the USB or SSD/HHD from the rig.

Fully power off everything: including the PSU.

Press the power button several times to clear any remaining power in the mobo.

Turn the PSU powerswitch back to | "on".

power on (without the USB attached)

See if the bios posts; if you get nothing in 20 seconds; press ctrl + alt + del repeatedly until the system reboots.

Wait and see if the bios posts.

If the bios posts attach the USB key to a USB 2.0 port and press ctrl + alt + delete.


if it boots; stop the mining process before it starts mining:

then go to the top left and click the ubuntu button

type u

and click on software updater

run updates

reboot

Let me know if this works.

[fullzero]

Good Morning FullZero -
I apoligize for the long post but I need to get it all in here

My first working mining rxOC AMD rig is up and running.

I want to ask some questions about clock settings.

Question is easy. How do I tune Linux version to get speeds near Windows?

As a base line I am dual mining Claymore ETH/SIA at -dcri 12

My cards are as follows -

GPU #0: Hawaii, 7991 MB available, 40 compute units (R9 390)
GPU #1: Ellesmere, 4050 MB available, 36 compute units (RX 570)
GPU #2: Ellesmere, 4050 MB available, 32 compute units (RX 470)

The 2 RX's have been bios modded and under windows 10 they run app. like this -

ETH: GPU0 29.897 Mh/s, GPU1 27.030 Mh/s, GPU2 27.096 Mh/s

I get those speeds by using Trixx and setting the following -

GPU #0: Hawaii, 7991 MB available, 40 compute units (R9 390) - 1100/1500
GPU #1: Ellesmere, 4050 MB available, 36 compute units (RX 570) - 1140/1900
GPU #2: Ellesmere, 4050 MB available, 32 compute units (RX 470) - 1140/1880

OK thats my stuff and how it runs on Windows...Now lets talk Linux

*****************************************************************************************

As a base line for the rest of this talk.
Using rxOC V12 with Parallax's mod
Dual mining ETH/SIA Claymore 9.5 -dcri 12
ETH_EXTENTION_ARGUMENTS="-dcri 12"
Same cards as above -

MY CURRENT SPEED WITH THE FOLLOWING SETTINGS -

ETH: GPU0 24.910 Mh/s, GPU1 24.086 Mh/s, GPU2 25.116 Mh/s


My oneBASH is currently set as follows -

 
PERFORMANCE_LEVEL="HIGH"        # HIGH LOW
 
# select level: to see supported clocks scroll to the top of the mining process
__CORE_OVERCLOCK_LEVEL=4   # for ETH use lowest without decreasing the hashrate / Highest for ZEC
MEMORY_OVERCLOCK_LEVEL=7   # use highest level
 
OVERDRIVE="NO"                  # YES NO  <---LOCKS UP WHEN I TURN THIS ON
 

 
My ROCm Output is as follows -

============================================================================
GPU[0]       : Supported GPU clock frequencies on GPU0
GPU[0]       : 0: 300Mhz
GPU[0]       : 1: 516Mhz
GPU[0]       : 2: 727Mhz
GPU[0]       : 3: 840Mhz
GPU[0]       : 4: 890Mhz *
GPU[0]       : 5: 936Mhz
GPU[0]       : 6: 977Mhz
GPU[0]       : 7: 1015Mhz
GPU[0]       :
GPU[0]       : Supported GPU Memory clock frequencies on GPU0
GPU[0]       : 0: 150Mhz
GPU[0]       : 1: 1500Mhz *
GPU[0]       :
GPU[1]       : Supported GPU clock frequencies on GPU1
GPU[1]       : 0: 300Mhz
GPU[1]       : 1: 608Mhz
GPU[1]       : 2: 910Mhz
GPU[1]       : 3: 1077Mhz *
GPU[1]       : 4: 1145Mhz
GPU[1]       : 5: 1191Mhz
GPU[1]       : 6: 1236Mhz
GPU[1]       : 7: 1266Mhz
GPU[1]       :
GPU[1]       : Supported GPU Memory clock frequencies on GPU1
GPU[1]       : 0: 300Mhz
GPU[1]       : 1: 1750Mhz *
GPU[1]       :
GPU[2]       : Supported GPU clock frequencies on GPU2
GPU[2]       : 0: 300Mhz
GPU[2]       : 1: 588Mhz
GPU[2]       : 2: 952Mhz
GPU[2]       : 3: 1041Mhz
GPU[2]       : 4: 1106Mhz *
GPU[2]       : 5: 1168Mhz
GPU[2]       : 6: 1209Mhz
GPU[2]       : 7: 1264Mhz
GPU[2]       :
GPU[2]       : Supported GPU Memory clock frequencies on GPU2
GPU[2]       : 0: 300Mhz
GPU[2]       : 1: 1000Mhz
GPU[2]       : 2: 1750Mhz *
GPU[2]       :
============================================================================
===================          End of ROCm SMI Log         ===================




===================   ROCm System Management Interface   ===================
GPU[0]       : Successfully reset GPU and Memory clocks
GPU[1]       : Successfully reset GPU and Memory clocks
GPU[2]       : Successfully reset GPU and Memory clocks
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
============================================================================
GPU[0]       : Successfully set current PowerPlay Level to high
GPU[1]       : Successfully set current PowerPlay Level to high
GPU[2]       : Successfully set current PowerPlay Level to high
============================================================================
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
GPU[0]       : Unable to set clock to unsupported Level - Max Level is 1
GPU[1]       : Unable to set clock to unsupported Level - Max Level is 1  <----WHAT'S HAPPENING HERE?
GPU[2]       : Unable to set clock to unsupported Level - Max Level is 2
===================          End of ROCm SMI Log         ===================



===================   ROCm System Management Interface   ===================
GPU[0]       : Successfully set GPU Clock frequency mask to Level 4
GPU[1]       : Successfully set GPU Clock frequency mask to Level 4
GPU[2]       : Successfully set GPU Clock frequency mask to Level 4
===================          End of ROCm SMI Log         ===================

With a custom rom I don't know why you would use normal OC settings; the cards should already be hardcoded to the clocks you want other than fanspeed.

you can't try to set these to a value that is outside any of the GPUs range:

__CORE_OVERCLOCK_LEVEL=4   
MEMORY_OVERCLOCK_LEVEL=7   

7 is outside all of the ranges use 1

looking at your windows clocks (and as you are dual mining you want to use higher core clocks) a 7 for core OC

so use:

Code:

__CORE_OVERCLOCK_LEVEL=7   
MEMORY_OVERCLOCK_LEVEL=1 

OVERDRIVE="NO"                  # YES NO  <---LOCKS UP WHEN I TURN THIS ON

sometimes custom roms can't handle any additional pushing (if they are optimized then they should crash with any overdrive at all)

If you want to use -dcri 12:

remove the hard coded -dcri 70 (or just change it to -dcri 12) on line 1185:

Code:

until $HCD -epool $ETH_POOL -ewal $ETHADDR -epsw x -allpools 1 -dpool $SC_POOL -dwal $SCADDR -dpsw x -dcoin sc -dcri 70 -dbg -1 $ETH_EXTENTION_ARGUMENTS

as I think you will have better results only using OC on the GPU that hasn't been bios modded (assuming you editied the clocks in your bios mod) you can add:

Code:

-d 

as and argument to each of the OC implementation lines starting at line 229 change this:

Code:

if [ $OVERDRIVE == "YES" ]
then
sudo '/home/m1/rocm-smi' --setoverdrive $OVERDRIVE_PERCENT
fi

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

sudo '/home/m1/rocm-smi' --setmclk $MEMORY_OVERCLOCK_LEVEL

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

sudo '/home/m1/rocm-smi' --setsclk $__CORE_OVERCLOCK_LEVEL

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

to:

Code:

if [ $OVERDRIVE == "YES" ]
then
sudo '/home/m1/rocm-smi' -d 0 --setoverdrive $OVERDRIVE_PERCENT
fi

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

sudo '/home/m1/rocm-smi' -d 0 --setmclk $MEMORY_OVERCLOCK_LEVEL

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

sudo '/home/m1/rocm-smi' -d 0 --setsclk $__CORE_OVERCLOCK_LEVEL

if [ $SLOW_USB_KEY_MODE == "YES" ]
then
sleep 6
fi

and change

Code:

OVERDRIVE="YES"

In general Claymore spends way more time Optimizing windows; so you should always have slightly lower results on linux when using claymore (but closer than your results reported above)

let me know how this goes

[OverEasy]

Why not modify oneBash to be like the one for Linux?

Have individual card settings?

I don't anybody that mines with AMD that hasn't modified the straps in the bios and then uses some OC tool to set the clocks.

So instead of

# select level: to see supported clocks scroll to the top of the mining process
__CORE_OVERCLOCK_LEVEL=7   # for ETH use lowest without decreasing the hashrate / Highest for ZEC
MEMORY_OVERCLOCK_LEVEL=2   # use highest level

How about -

INDIVIDUAL_CLOCKS="YES"         # YES NO
 
# Set individual clocks here if INDIVIDUAL_CLOCKS="YES"
__CORE_OVERCLOCK_0=100
MEMORY_OVERCLOCK_0=1000
 
__CORE_OVERCLOCK_1=100
MEMORY_OVERCLOCK_1=350
 
__CORE_OVERCLOCK_2=100
MEMORY_OVERCLOCK_2=100
 
__CORE_OVERCLOCK_3=100
MEMORY_OVERCLOCK_3=100
 
__CORE_OVERCLOCK_4=100
MEMORY_OVERCLOCK_4=100
 
__CORE_OVERCLOCK_5=100
MEMORY_OVERCLOCK_5=100
 
__CORE_OVERCLOCK_6=100
MEMORY_OVERCLOCK_6=100
 
__CORE_OVERCLOCK_7=100
MEMORY_OVERCLOCK_7=100
 
__CORE_OVERCLOCK_8=100
MEMORY_OVERCLOCK_8=100
 
__CORE_OVERCLOCK_9=100

Just asking