There also exist a few on-CPU implementations of OpenCL. These come as drivers provided by INTEL or AMD. We observed that they do not give us any speed gain versus our hand-optimized CPU code. Therefore we simply discard these devices.
Does this include GPU on the same die as CPU? My AMD A12 (id Carrizo) comes with 8 compute units according to clinfo, but is not usable?
Again - I have a separate Radeon R5230 (id Caicos) with 2 compute units but this is not used either.
darktable-cltest output pasted below:
Performance using just my CPU (and with 16GB DDR4 RAM) isn’t bad; but it would be nice to have the ability to use the silicon that could make the performance fantastic.
darktable-cltest
[opencl_init] opencl related configuration options:
[opencl_init]
[opencl_init] opencl: 1
[opencl_init] opencl_library: ‘’
[opencl_init] opencl_memory_requirement: 768
[opencl_init] opencl_memory_headroom: 300
[opencl_init] opencl_device_priority: ‘/!0,//’
[opencl_init] opencl_mandatory_timeout: 200
[opencl_init] opencl_size_roundup: 16
[opencl_init] opencl_async_pixelpipe: 0
[opencl_init] opencl_synch_cache: 0
[opencl_init] opencl_number_event_handles: 25
[opencl_init] opencl_micro_nap: 1000
[opencl_init] opencl_use_pinned_memory: 0
[opencl_init] opencl_use_cpu_devices: 0
[opencl_init] opencl_avoid_atomics: 0
[opencl_init]
[opencl_init] found opencl runtime library ‘libOpenCL’
[opencl_init] opencl library ‘libOpenCL’ found on your system and loaded
amdgpu_parse_asic_ids: Cannot parse ASIC IDs: Resource temporarily unavailable
[opencl_init] found 1 platform
[opencl_init] found 2 devices
[opencl_init] discarding device 0 AMD CAICOS (DRM 2.50.0 / 4.13.0-21-lowlatency, LLVM 6.0.0)' due to missing image support. [opencl_init] discarding device 1 AMD CARRIZO (DRM 3.18.0 / 4.13.0-21-lowlatency, LLVM 6.0.0)’ due to missing image support.
[opencl_init] no suitable devices found.
[opencl_init] FINALLY: opencl is NOT AVAILABLE on this system.
[opencl_init] initial status of opencl enabled flag is OFF.
[opencl_init] opencl related configuration options:
[opencl_init]
[opencl_init] opencl: 1
[opencl_init] opencl_library: ‘’
[opencl_init] opencl_memory_requirement: 768
[opencl_init] opencl_memory_headroom: 300
[opencl_init] opencl_device_priority: ‘/!0,//’
[opencl_init] opencl_mandatory_timeout: 200
[opencl_init] opencl_size_roundup: 16
[opencl_init] opencl_async_pixelpipe: 0
[opencl_init] opencl_synch_cache: 0
[opencl_init] opencl_number_event_handles: 25
[opencl_init] opencl_micro_nap: 1000
[opencl_init] opencl_use_pinned_memory: 0
[opencl_init] opencl_use_cpu_devices: 0
[opencl_init] opencl_avoid_atomics: 0
[opencl_init]
[opencl_init] could not find opencl runtime library ‘libOpenCL’
[opencl_init] could not find opencl runtime library ‘libOpenCL.so’
[opencl_init] found opencl runtime library ‘libOpenCL.so.1’
[opencl_init] opencl library ‘libOpenCL.so.1’ found on your system and loaded
[opencl_init] could not get platforms: -1001
[opencl_init] FINALLY: opencl is NOT AVAILABLE on this system.
[opencl_init] initial status of opencl enabled flag is OFF.
wait time 0.167082s
wait time 0.108670s
I’m using A10-7870K APU with built in R7 graphics. Does this mean OpenCL support is not available?
My system is using the Open Source AMD GPU driver. However for darktable I use the closed source opencl compiler. I’ve just extracted the required individual files from the closed source driver packages (RPM) for that:
I trashed my driver installation trying to set up amdgpu-pro and had to reinstall my OS files, and wish to avoid going through that pain again.
Did you use the AMD supplied script, or (as I think you imply) selectively install just some of the .deb packages?
Or did you not install .deb packages but merely extract the files you list in /opt/amdgpu-pro/lib64 fom the packages?
Your last line suggests having to call darktable from a terminal. Is it possible to set LD_LIBRARY_PATH in my envionment, or does it need to be set as a commandline option?
Hi All. I’m having issues with getting darktable to use opencl on my system. After having reinstalled my nvidia-387 driver.
I’m getting the following output:
➜ ~ darktable -d opencl
[opencl_init] opencl related configuration options:
[opencl_init]
[opencl_init] opencl: 1
[opencl_init] opencl_library: ''
[opencl_init] opencl_memory_requirement: 768
[opencl_init] opencl_memory_headroom: 4000
[opencl_init] opencl_device_priority: '*/!0,*/*/*'
[opencl_init] opencl_mandatory_timeout: 200
[opencl_init] opencl_size_roundup: 16
[opencl_init] opencl_async_pixelpipe: 0
[opencl_init] opencl_synch_cache: 0
[opencl_init] opencl_number_event_handles: 25
[opencl_init] opencl_micro_nap: 1000
[opencl_init] opencl_use_pinned_memory: 0
[opencl_init] opencl_use_cpu_devices: 0
[opencl_init] opencl_avoid_atomics: 0
[opencl_init]
[opencl_init] found opencl runtime library 'libOpenCL'
[opencl_init] opencl library 'libOpenCL' found on your system and loaded
[opencl_init] found 1 platform
[opencl_init] could not get device id size: -1
[opencl_init] found 0 device
[opencl_init] FINALLY: opencl is NOT AVAILABLE on this system.
[opencl_init] initial status of opencl enabled flag is OFF.
The device it’s not detecting is a nvidia GeForce 1070 GTX. The driver is functional in so far as it’s running cuda code correctly.
The system is linuxmint 18.2.
Any advice in how to debug this problem is greatly appreciated.
Finally managed to get opencl support in darktable;
@asn, although I made progress following your tips, ultimately I got stalled on version number of the DRM.
I bit the bullet and installed amdgpu-pro version 17.50 using the install script, which built a new kernel and ramdisk image.
I followed the directions on this page
The speed-up factor in use on darktable is only slight over 2, however, which is considerably less than the factor I achieved using my corei7 CPU laptop with opencl. I don’t know the exact factor, but I do know that I was exporting a jpg for a 4000x6000 image in about 8 seconds. Unfortunately, exports using my new computer are still taking nearly a minute each.
Interestingly my stand-alone video card (yes a pretty low-spec Radeon R5 230 was not detected as a usable device at all. Given it has but two GPU units (versus 8 on the CPU die - I have an AMD A12) means this is a moot point.
Mileage _will vary, but maybe this tale will help someone with similar hardware & OS version (Ubuntu Studio 17.10)
I have a Radeon RX 470, so the GPU is pretty strong. I needed to set ‘OpenCL scheduling profile: Very fast GPU’. With that everything is working much faster
Thanks for the tip. This has given me an extra 2x speedup.
But when I set “multiple GPUs” I get much better … another 4x speedup.
Very happy with the results now!!!
did you download and install the amdgpu-pro package?
It is a bit of a “heart in your mouth” moment, because for me there is the fear of making my system unusable, and hard to return to its previous state.
However I have had success in Ubuntu and Manjaro with the -pro package. It is, unfortunately, not FOSS.
I’m thinking about trying the latest (update as at 20 March 2018) here because I do have some flicker issues that - once they start (sometimes after half an hour, sometimes never) - only a reboot solves.
Radeon Pro Software Enterprise Edition 18.Q1.1 for Linux Highlights
Radeon™ Pro Software Enterprise Edition 18.Q1.1 for Linux delivers enterprise-level support for RHEL 7.4 and CentOS 7.4.
Download Full Release Notes
Radeon Pro Software Adrenalin Edition 17.12.1 for Linux Highlights
Radeon™ Pro Software Adrenalin Edition 17.12.1 for Linux delivers amdgpu-pro and amdgpu-open stack using the same packaging infrastructure and introduces support for RHEL 7.4 and CentOS 7.4.
Download Full Release Notes
I cannot understand if I must use “Radeon Pro Software Enterprise Edition 18.Q1.1” or “Radeon Pro Software Adrenalin Edition 17.12.1”, they do not explain what is each for. I actually trying installing both, the first makes my system unusable, the second works but doesn’t make openocl works…
Paolo, it seems to me that there is rebranding going on. “Radeon Pro Software Enterprise Edition 18.Q1.1” is obviously the more recent iteration. (The other is referencing December 2017). As I mentioned, the most recent update is 20 March this year. It’s obviously disappointing that neither give you what you want - a stable, usable display output and provision of opencl using your AMD GPU(s).
You may find that using Andreas’ method works for you; it’s a case of your mileage may vary because of the huge number of variables: which exact hardware, which Linux distro, and sometimes the way you hold your mouth
Good luck, and please do share any success stories.
Good luck, and please do share any success stories.
Heeding my own request, I am hapy to report that while the mid-April release of the Radeon Pro software resulted in instability and much flickering on my Kubuntu 18.04 official release installation, the May 3 release (definitively after the April 26 official Ubuntu 18.04 and flavours release date) - installed flawlessly and I have only noticed one glitch, that have not since been able to reproduce - a slowing to a crawl of darktable (but no other running applications).
So, I am happy to recommend downloading and installing from here the amdgpu-pro-18.20-579836 version.
I’m running AMD A12-9800 (APU with Radeon R7 Graphics - 8 graphics cores). Your mileage may vary.
Hi to all! I’m planning to buy new PC. What graphic card should I choose to have easy opencl setup and effective use in DT in ubuntu based distribution? Now I select nvidia gt1030 with 1Gb Ram. Am I right?
As I mentioned in another thread, if you’re after OpenCL performance, the “best” card you can buy is the one that doesn’t break your wallet. The more compute power you can throw at OpenCL, the faster it’ll go, generally.
If you want to go the nVidia router, and don’t mind the proprietary driver, have a look at using the Ubuntu Graphics Drivers PPA, I’m using it on my KDE neon computer get the the freshest drivers. It has worked very well for me on this computer.
However, if you want an out-of-the-box experience, I believe that AMD will soon be the better route.
If you want libre drivers, AMD is the only route.
As always, research carefully and make sure the driver you want to use supports the hardware you’re going to buy.
Its the free nVidia drivers that are not great, because nVidia does not help the people who write the free driver, while AMD at least answers questions about their hardware for those writing the free driver.
Pharonix actually has good coverage of all the video card driver stuff.
The problem with low bugdet nvidia cards I had in the past was there weak performance with the equalizier and the profiled denoise. Please correct me if I’m wrong, but OpenCl is still not fully supported by the opensource AMD drivers. There are several descriptions howto extract the OpenCl libraries from the proprietary drivers and use them with the opensource AMD drivers. So, for both vendors you need at least some parts (for AMD) of the proprietary drivers.
Since 2003 proprietary nvidia drivers are working for me. Whereas AMD proprietary drivers dropped support in AMDgpu-pro two years after I bought a R9 270X. And frglx is not working on newer LInux distributions. You will find some discussions about this “issues” on the darktable mailing lists. My R9 270X performed very well with mentioned moduls, which made it even more annoying.
The new way to use OpenCL with AMD cards on the open source drivers is ROCm, but it only supports recent CPUs (needs PCIe atomics) and certain versions of Linux. I’m not sure when it’s going to be mainlined…
Searching the web, I noticed that a number of people had problems implementing OpenCL with their AMD card. Many prefer the display drivers that come with their Linux distribution (in my case Mint), but would like to add OpenCL support through the proprietory AMD drivers provided on the AMD support page https://www.amd.com/en/support .
There is a way to do this for your card (in my case an AMD RX 570, but I am sure it will also work for AMD RX 580, 590, Vega and many others). After unpacking the drivers, open a terminal in that directory and type (or copy from here):
OpenCL installation will start after you approve through sudo (enter password).
The --headless option will keep the display drivers of your Linux distribution, but still add OpenCL. When I rebooted and started Darktable the next time, OpenCL support was automatically activated. I found the AMD drivers to be very reliable. Speed improvements are really nice, especially with ‘fast GPU’ on.
I hope this helps a few AMD graphics card users out there.
That’s how I installed OpenCL when I was running an AMD RX580 GPU. Sadly I found it unstable in other software which used heavy OpenCL calculations, so I eventually returned it and bought an Nvidia GTX 1660!
