Package Details: opencl-amd 1:6.3.2-1

Git Clone URL: https://aur.archlinux.org/opencl-amd.git (read-only, click to copy)
Package Base: opencl-amd
Description: ROCm components repackaged from AMD's Ubuntu releases (ROCr runtime, ROCm runtime, HIP runtime) - This package is intended to work along with the free amdgpu stack.
Upstream URL: http://www.amd.com
Keywords: amd amdgpu computing gpgpu opencl radeon
Licenses: custom:AMD
Conflicts: amd-smi-lib, comgr, hip, hip-dev, hip-doc, hip-runtime-amd, hip-samples, hipcc, hsa-amd-aqlprofile, hsa-rocr, hsa-rocr-dev, libdrm-amdgpu-amdgpu1, openmp-extras-runtime, rocdecode, rocdecode-dev, rocjpeg, rocjpeg-dev, rocm-cmake, rocm-core, rocm-dbgapi, rocm-debug-agent, rocm-device-libs, rocm-gdb, rocm-hip-runtime, rocm-language-runtime, rocm-ocl-icd, rocm-opencl, rocm-opencl-dev, rocm-opencl-icd-loader, rocm-opencl-runtime, rocm-smi-lib, rocm-utils, rocminfo, rocprofiler, rocprofiler-dev, rocprofiler-plugins, rocprofiler-register, roctracer, roctracer-dev
Provides: amd-smi-lib, comgr, hip, hip-dev, hip-doc, hip-runtime-amd, hip-samples, hipcc, hsa-amd-aqlprofile, hsa-rocr, hsa-rocr-dev, libdrm-amdgpu-amdgpu1, opencl-driver, openmp-extras-runtime, rocdecode, rocdecode-dev, rocjpeg, rocjpeg-dev, rocm-cmake, rocm-core, rocm-dbgapi, rocm-debug-agent, rocm-device-libs, rocm-gdb, rocm-hip-runtime, rocm-language-runtime, rocm-ocl-icd, rocm-opencl, rocm-opencl-dev, rocm-opencl-icd-loader, rocm-opencl-runtime, rocm-smi-lib, rocm-utils, rocminfo, rocprofiler, rocprofiler-dev, rocprofiler-plugins, rocprofiler-register, roctracer, roctracer-dev
Submitter: grmat
Maintainer: sperg512 (luciddream)
Last Packager: luciddream
Votes: 132
Popularity: 0.21
First Submitted: 2016-12-01 03:45 (UTC)
Last Updated: 2025-01-29 20:12 (UTC)

Required by (128)

Sources (38)

Pinned Comments

nho1ix commented on 2023-12-29 08:43 (UTC) (edited on 2024-02-10 07:13 (UTC) by nho1ix)

Note for anyone who has a Polaris GPU (Radeon RX 5xx) debugging issues with this package; Packages that use OpenCL like clinfo or davinci-resolve-studio will need you to downgrade opencl-amd to 1:5.7.1-1 as well as amdgpu-pro-oglp to 23.10_1620044-1 to avoid coredumps & segfaults.

DVR would not open unless these 2 packages were downgraded (along with their dependencies). Had to figure it out the hard way after hours using valgrind and rebooting over and over. Hopefully someone else will not have to pull their hair out trying to resolve their issue.

luciddream commented on 2021-12-26 15:14 (UTC) (edited on 2025-01-29 20:13 (UTC) by luciddream)

Current release is for ROCm 6.3.2 opencl-amd package includes only OpenCL / HIP runtime. You also need to use opencl-amd-dev package for ROCm LLVM compiler, OpenCL and HIP SDK. Please relog / reboot after installing so your PATH gets updated

There are now official packages available: rocm-opencl-sdk for OpenCL and rocm-hip-sdk for HIP - You might have better luck with these packages depending on your GPU.

Latest Comments

« First ‹ Previous 1 .. 15 16 17 18 19 20 21 22 23 24 25 .. 79 Next › Last »

br_shadow commented on 2022-04-16 10:26 (UTC)

I get the same problem as @limsandy in Manjaro Linux


==> Validating source files with sha256sums...
    ncurses-6.3.tar.gz ... Passed
    ncurses-6.3.tar.gz.sig ... Skipped
==> Verifying source file signatures with gpg...
    ncurses-6.3.tar.gz ... cat: write error: Broken pipe
FAILED
==> ERROR: One or more PGP signatures could not be verified!
Finished with result: exit-code
Main processes terminated with: code=exited/status=1
Service runtime: 14.754s
CPU time consumed: 1.946s
Error: Failed to build ncurses5-compat-libs

luciddream commented on 2022-04-14 21:19 (UTC) (edited on 2022-04-14 21:23 (UTC) by luciddream)

I tried Hashcat with HIP support which was added today and for some reason benchmark took 5 minutes to complete, and GPU fans were silent while it was running. So I guess that's similar to what @limsandy is saying. But the benchmark numbers are fine so I'm not sure why it took so long. The only thing I noticed in the first run is that gnome tracker indexing was using the CPU (about 8%) and Hashcat was also using the CPU (about 7%).

After a couple of hours I run Hashcat OpenCL benchmark and it took 2 minutes to complete. Then ran Hashcat HIP benchmark for a second time and it also took 2 minutes to complete, while fans were spinning a lot and GPU usage was 99%

The benchmark results are the same in all 3 runs, for some reason. At least I have a software I can use now to make tests with HIP and ROCm 5.1 :)

limsandy commented on 2022-04-10 17:15 (UTC)

Yeah, I don't understand why opencl performance under Manjaro is slower than in Windows 10. Even slower than Ubuntu 20.04. I had been able to change the Radeon profile to "profile_peak" which will make the GPU clock stay at max, but will throttle once a certain temperature is reached.

It doesn't make any sense, but I benchmarked VDF times both in Manjaro and Ubuntu, and Ubuntu sets the record fastest VDF time and consistently lower VDF times.

luciddream commented on 2022-04-09 12:37 (UTC)

I updated the packages to 22.10.1 and 5.1.1 - I don't see any release notes or any significant changes (blender hip still crashes, AMF still not working, opencl performance is still a bit lower than 5.0)

limsandy commented on 2022-04-01 16:52 (UTC)

So I'm starting to understand how this works.... Linux will try to read every file with the extension .rules in the folder /etc/udev/rules.d/

You can rename it with whatever you want, just put the extension .rules afterwards. Mine is obviously renamed to radeon-vega-7.rules

luciddream commented on 2022-04-01 16:28 (UTC)

Cool stuff @limsandy, I will give it a try later and see if it affects the scores on my PC.

limsandy commented on 2022-04-01 16:09 (UTC)

Oh I'll be damned.... I put this line in /etc/udev/rules.d/30-radeon-pm.rules

KERNEL=="card0", SUBSYSTEM=="drm", DRIVERS=="amdgpu", ATTR{device/power_dpm_force_performance_level}="high"

Everything in single line, that's a space before the ATTR.

Now power_dpm_force_performance_level is set to high after reboot. Still you've got to make sure that power_dpm_state is set to performance. Notice that the drivers is now set to "amdgpu"?

limsandy commented on 2022-04-01 12:22 (UTC)

Nope, still doesn't work. I think it's got to do with this file name: /etc/udev/rules.d/30-radeon-pm.rules

Like the udev doesn't read this file unless it's got the right file name.

I'll work something out later, but on a closing remark.... AMD's still gotta work on how the Radeon drivers is not clocking up high/fast enough when there is a work load being placed in the queue. Even after setting the performance level to 'high', I see the Vddgfx downvolting to 0.574V, which means it's allowed to downclock. On the positive side, my APU stays cool at 35-36C which is very near to my ambient temp.

limsandy commented on 2022-04-01 12:14 (UTC)

@luciddream,

Thank you for that link. I tried but it didn't work. This file didn't even exist: /etc/udev/rules.d/30-radeon-pm.rules

So I manually create the file and paste this: KERNEL=="dri/card0", SUBSYSTEM=="drm", DRIVERS=="radeon", ATTR{device/power_dpm_force_performance_level}="high"

It turns out my KERNEL name is wrong. On my sytem, it's just "card0". I've made the change accordingly, and restarting my computer now....

luciddream commented on 2022-04-01 11:32 (UTC) (edited on 2022-04-01 11:33 (UTC) by luciddream)

I've never done this but I guess https://wiki.archlinux.org/title/ATI#Persistent_configuration should work? In any case post your findings, it might be helpful for other people.

« First ‹ Previous 1 .. 15 16 17 18 19 20 21 22 23 24 25 .. 79 Next › Last »