Categories
- 17.3K All Categories
- 9.5K Mining
- 574 Pool Discussion
- 378 Promotional
- 1.4K General Project Discussion (non-technical)
- 485 Education
- 811 Protocol and Client discussion
- 170 web3-js
- 29 Whisper
- 16 Swarm
- 3 RLP
- 303 IoT & Hardware
- 1.2K Smart Contracts and Dapps
- 28 Serpent
- 359 Solidity
- 664 Projects
- 1.2K Reference clients code and builds
- 249 Eth & AlethZero- Cpp Implementation
- 471 Geth - Go Implementation
- 242 Mist
- 15 Node.js Implementation
- 36 Python Implementation
- 49 Mix
- 36 Other Implementations
- 170 Meetups
- 40 Other Events
- 226 Jobs & Skills
- 281 Press and Articles
- 75 Audio/Video
- 296 Ether Sale
- 1.2K Other Languages
- 96 Chinese
- 255 German
- 33 Italian
- 111 French
- 3 Hebrew
- 42 Japanese
- 75 Portugese
- 46 Romanian
- 185 Russian
- 231 Spanish
- 47 Turkish
- 125 Watercooler
GPU mining with ethminer on AMD
No matter what I do, it seems that ethminer only ever picks up the integrated graphics(Radeon R5) instead of the Radeon R9, even though it's properly installed:
Thanks!!
It doesn't matter which opencl device I set, it's always choosing the same... Any one knows what could be wrong?
$ lspci | grep -i VGA
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Curacao XT [Radeon R9 270X]
$ ./ethminer -G -M --opencl-device 0
Benchmarking on platform: { "platform": "AMD Accelerated Parallel Processing", "device": "AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G", "version": "OpenCL 1.2 AMD-APP (1411.4)" }
Preparing DAG...
ℹ 02:56:12|ethminer Loading from libethash...
ℹ 02:56:12|ethminer Done loading.
ℹ 02:56:12|gpuminer0 workLoop 0 #00000000… #00000000…
Warming up...
Using platform: AMD Accelerated Parallel Processing
Using device: AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G(OpenCL 1.2 AMD-APP (1411.4))
Segmentation fault (core dumped)
[email protected]:~/git/cpp-ethereum/build/ethminer$
$ ./ethminer -G -M --opencl-device 1
Benchmarking on platform: { "platform": "AMD Accelerated Parallel Processing", "device": "AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G", "version": "OpenCL 1.2 AMD-APP (1411.4)" }
Preparing DAG...
ℹ 02:56:30|ethminer Loading from libethash...
ℹ 02:56:30|ethminer Done loading.
ℹ 02:56:30|gpuminer0 workLoop 0 #00000000… #00000000…
Warming up...
Using platform: AMD Accelerated Parallel Processing
Using device: AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G(OpenCL 1.2 AMD-APP (1411.4))
Segmentation fault (core dumped)
$ lspci | grep -i VGA
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Curacao XT [Radeon R9 270X]
[email protected]:~/git/cpp-ethereum/build/ethminer$ ./ethminer -G -M --opencl-device 10
Benchmarking on platform: { "platform": "AMD Accelerated Parallel Processing", "device": "AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G", "version": "OpenCL 1.2 AMD-APP (1411.4)" }
Preparing DAG...
ℹ 02:58:01|ethminer Loading from libethash...
ℹ 02:58:01|ethminer Done loading.
ℹ 02:58:01|gpuminer0 workLoop 0 #00000000… #00000000…
Warming up...
Using platform: AMD Accelerated Parallel Processing
Using device: AMD A6-7400K Radeon R5, 6 Compute Cores 2C+4G(OpenCL 1.2 AMD-APP (1411.4))
Segmentation fault (core dumped)
Thanks!!
Comments
Make sure fglrx driver is installed for gpu cards.
I have A version of fglrx installed but it was hectic, do you have a recommended way to go about it on Ubuntu 14.04.1 (it appears 14.04.2 has different requirements).
Thanks!
I'm on my n^th wipe on 14.04 trying to just get 1 AMD R9 290 to talk to ./eth via OpenCL but most of my setups are just going to a black screen and I'm having to wipe again if I can't get to command prompt to purge that attempt of fglrx
Thinking of switching my focus back over to the NVIDIA machine. I've actually considered running windows for the AMD machine.
In my case though it's clear that it's not properly installed: It only finds my integrated AMD (R5), not my actual one (R9)
I tried pointing it to the specific locations but no dice either (neither 2.9 nor 3.0Beta): Basically I think eth/ethminer won't work until this shows it as detected...
What's odd is that the R9 card seems to be fine:
1) wipe buntu (sorry, but we have to make sure). Insure you are running 14.04.02
2) sudo-apt get update then sudo-apt get upgrade (standard first update)
3) click system settings in the toolbar
4) go to 'software and updates'
5) go to the 'updates' tab
6) enable the 'pre released updates (trusty proposed)' option
7) reboot
8) go back to 'software and updates', click the 'additional drivers' tab
9) in the list, you'll see the 'open' drivers are selected. Switch to fglrx-updates (proprietary)
10) reboot
11) should work
If things still fail, especially manifested by a 'hung' ethminer, this is probably a build issue, not a buntu issue. Go to my mining FAQ on this forums, i posted the commit that works with that exact same setup.
Note that any other workaround will fail (and i'm aware of the 'cheese' trick etc... - tried them all until 330am on Friday night hahah). The above is the only way for now, it's a known buntu bug: https://bugs.launchpad.net/ubuntu/+source/fglrx-installer/+bug/1424491
I'll report back.
One thing I find strange is that even though I downloaded 14.04.2, I still get: And yet this is the file I used (downloaded it today): (EDIT: I did do the update then upgrade)
And using @Genoil 's trick from https://forum.ethereum.org/discussion/comment/8696/#Comment_8696, it also shows that it's not picking up on it: Still in spite of: I believed I followed all instructions above to the letter, including the commits :-/ Maybe it's a problem with my card specifically... But I'll do it once more to be safe
I'm currently attempting a fresh build based on @StephanTual 's May 20 suggestion to use fglrx-updates from the app repository rather than the direct-from-AMD drivers, which I had been trying to get to work (unsuccessfully) for days. My only problem is that I'm not about to stop successful mining to run tests.
I'm actually in a good position - my stuff works (based on a cloned drive). It does drive me a bit nuts that I can't explain it or replicate it.