I have successfully tested it with my RX 6600 and another user with their RX 7900 XT.

CC: (none) => ezequiel_partida, fri, joselp, mageia, marja11, sysadmin-bugs
Whiteboard: (none) => MGA9TOO

How to test.

Install rocm-amd-opencl package. Required dependencies will follow.

You need then to work with a software benefiting from OpenCL acceleration, like darktable.

So, install darktable.

Then, run:
darktable-cltest

Around the end of the test, you should see:
FINALLY: opencl is AVAILABLE and ENABLED


Also, if you run:
rocm-clinfo

You should see, at the begining:
  Platform Name: AMD Accelerated Parallel Processing
  Number of devices: x

with x >= 1

How to test?

I welcome ideas.

One way is to install Blender
Run blender, go to menu Edit>Preferences>System
You should then be able to see and to see you AMD GPU in the HIP section.

(In reply to christian squidf from comment #5)
> How to test?
> 
> I welcome ideas.
> 
> One way is to install Blender
> Run blender, go to menu Edit>Preferences>System
> You should then be able to see and to see you AMD GPU in the HIP section.

Apologies, ignore the above. It was for https://bugs.mageia.org/show_bug.cgi?id=32646  ...

My congratulations!

This works well for me.

Installed on Mageia 9 Plasma x86_64. I have followed the instructions to test it.

darktable-cltest=FINALLY. opencl is AVAILABLE AND ENABLED

clinfo shows the device as "device: 1".

Reviewed on Slimbook PROX15 AMD Ryzen 7 4800H with Radeon graphics

My question is. So when this package, "rocm-amd-opencl", is released, the previous one will be obsolete and the update will lead to installing this new one?

Good job for Mageia!!

(In reply to Jose Manuel López from comment #7)
> My congratulations!
> 
> This works well for me.
> 
> Installed on Mageia 9 Plasma x86_64. I have followed the instructions to
> test it.
> 
> darktable-cltest=FINALLY. opencl is AVAILABLE AND ENABLED
> 
> clinfo shows the device as "device: 1".
> 
> Reviewed on Slimbook PROX15 AMD Ryzen 7 4800H with Radeon graphics
> 
Thanks for testing! I was not fully expecting it would work on Vega / GCN instruction set.

> My question is. So when this package, "rocm-amd-opencl", is released, the
> previous one will be obsolete and the update will lead to installing this
> new one?
> 
It is tricky. As I just wrote full support to Vega might not be granted.
This package is to complement the others for Navi / RDNA instruction set.
Based on this and my limited knowledge of all configurations (AMD seems quite complex), I would prefer not to force the update of amdgpupro-opencl-pal by this one. Instead, I will add into the description of rocm-amd-opencl that it might work for Vega generation. Up to the users to test on their configurations.

Thank you for the work Christian!
I do not use AMD myself currently.

https://wiki.mageia.org/en/Mageia_9_Release_Notes#Proprietary_AMD_driver

Keywords: (none) => FOR_RELEASENOTES9

I had just uploaded the advisory from comment 4 when I saw there is a newer one in comment 9.
Please do tag old advisories as "obsolete", that'll help my old brain to upload the newest one ;-)

I'll update the one in SVN.

Advisory from comment 9 added to SVN. Please remove the "advisory" keyword if it needs to be changed. It also helps when obsolete advisories are tagged as "obsolete"

Keywords: (none) => advisory

(In reply to Marja Van Waes from comment #12)
> Advisory from comment 9 added to SVN. Please remove the "advisory" keyword
> if it needs to be changed. It also helps when obsolete advisories are tagged
> as "obsolete"

Thanks Marja and, sorry, I have missed / was not aware of this rule to tag the obsolete advisories. Will do it going forward.

Christian I will test but I not expect too much, this is the card I have:
 
AMD Cape Verde XT [Radeon HD 7770/8760 / R7 250X] driver: amdgpu

what I wonder if there is a category of cards that enables this working (as well shown in clinfo) without the need of the other proprietary drivers.

LC_ALL=C urpmi rocm-amd-opencl
To satisfy dependencies, the following packages are going to be installed:
  Package                        Version      Release       Arch    
(medium "QA Testing (64-bit)")
  lib64hsakmt1                   1.0.6        0.5.7.1.2.mg> x86_64  
  lib64rocm-compilersupport5.7.1 5.7.1        1.mga9        x86_64  
  lib64rocm-opencl-runtime-devel 5.7.1        1.1.mga9      x86_64  
  lib64rocm-opencl-runtime5.7    5.7.1        1.1.mga9      x86_64  
  lib64rocm-runtime1             5.7.1        1.mga9        x86_64  
  rocm-amd-opencl                5.7.1        1.1.mga9      x86_64  
  rocm-clinfo                    5.7.1        1.1.mga9      x86_64  (recommended)
  rocm-core                      5.7.1        1.mga9        x86_64  
  rocm-device-libs               5.7.1        1.mga9        x86_64  
  rocminfo                       5.7.1        1.mga9        x86_64  
(medium "Core Release")
  lib64ocl-icd-devel             2.3.1        2.mga9        x86_64  
  lib64opencl-devel              2.3.1        2.mga9        x86_64  
  opencl-headers                 3.0          0.20230206.1> noarch  
183MB of additional disk space will be used.
39MB of packages will be retrieved.
Proceed with the installation of the 13 packages? (Y/n) y


    https://mirrors.kernel.org/mageia/distrib/9/x86_64/media/core/release/lib64ocl-icd-devel-2.3.1-2.mga9.x86_64.rpm
    https://mirrors.kernel.org/mageia/distrib/9/x86_64/media/core/release/lib64opencl-devel-2.3.1-2.mga9.x86_64.rpm                  
    https://mirrors.kernel.org/mageia/distrib/9/x86_64/media/core/release/opencl-headers-3.0-0.20230206.1.mga9.noarch.rpm            
installing //home/katnatek/qa-testing/x86_64/lib64hsakmt1-1.0.6-0.5.7.1.2.mga9.x86_64.rpm                                            
//home/katnatek/qa-testing/x86_64/rocm-device-libs-5.7.1-1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/lib64rocm-opencl-runtime-devel-5.7.1-1.1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/lib64rocm-opencl-runtime5.7-5.7.1-1.1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/rocm-core-5.7.1-1.mga9.x86_64.rpm
/var/cache/urpmi/rpms/lib64ocl-icd-devel-2.3.1-2.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/lib64rocm-compilersupport5.7.1-5.7.1-1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/rocm-clinfo-5.7.1-1.1.mga9.x86_64.rpm
/var/cache/urpmi/rpms/opencl-headers-3.0-0.20230206.1.mga9.noarch.rpm
/var/cache/urpmi/rpms/lib64opencl-devel-2.3.1-2.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/rocminfo-5.7.1-1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/rocm-amd-opencl-5.7.1-1.1.mga9.x86_64.rpm
//home/katnatek/qa-testing/x86_64/lib64rocm-runtime1-5.7.1-1.mga9.x86_64.rpm
Preparing...                     ###################################################################################################
     1/13: rocm-device-libs      ###################################################################################################
     2/13: lib64hsakmt1          ###################################################################################################
     3/13: lib64rocm-runtime1    ###################################################################################################
     4/13: lib64rocm-opencl-runtime5.7
                                 ###################################################################################################
     5/13: rocm-clinfo           ###################################################################################################
     6/13: rocminfo              ###################################################################################################
     7/13: lib64rocm-compilersupport5.7.1
                                 ###################################################################################################
     8/13: lib64opencl-devel     ###################################################################################################
     9/13: opencl-headers        ###################################################################################################
    10/13: lib64ocl-icd-devel    ###################################################################################################
    11/13: lib64rocm-opencl-runtime-devel
                                 ###################################################################################################
    12/13: rocm-core             ###################################################################################################
    13/13: rocm-amd-opencl       ###################################################################################################
Update groups of huesped
/home/katnatek/.bashrc: línea 9: bind: aviso: no se activó la edición de línea
Update groups of katnatek
/home/katnatek/.bashrc: línea 9: bind: aviso: no se activó la edición de línea
/home/katnatek/.bashrc: línea 9: bind: aviso: no se activó la edición de línea
Update groups of live
usermod: user 'live' does not exist
su: user live does not exist or the user entry does not contain all the required fields
/home/katnatek/.bashrc: línea 9: bind: aviso: no se activó la edición de línea
Run ldconfig
----------------------------------------------------------------------
More information on package rocm-amd-opencl-5.7.1-1.1.mga9.x86_64
=============================================================
New users will have to be added to render and video groups.
Already existing users are taken care of during installation.
=============================================================

----------------------------------------------------------------------
[root@phoenix amdgpupro-opencl-orca]# clinfo -i
ignoring unknown command-line parameter -i
Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.1 AMD-APP (3590.0)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_event_callback 
  Platform Extensions function suffix             AMD
  Platform Host timer resolution                  1ns

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 0

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  AMD Accelerated Parallel Processing
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   No devices found in platform [AMD Accelerated Parallel Processing?]
  clCreateContext(NULL, ...) [default]            No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  No devices found in platform

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loaderns
  ICD loader Vendor                               OCL Icd free softwarens
  ICD loader Version                              2.3.1ns
  ICD loader Profile                              OpenCL 3.0ns

As I suspect, this card is not compatible

Do not add all users to a group. Just include a README.urpmi explaining
the need, and a reminder that logout/in or reboot is required for group
changes to take effect.

It is not possible to find out all users who can login by parsing /home
or /etc/passwd, as the admin is free to create directories in /home that
are not for users, and is free to use non-standard id numbers in /etc/passwd.
Even if that worked, the admin wouldn't know that it's needed for users who
are created in the future.

Qa testing should also ensure that the change doesn't interfere for people
who use ssh with X forwarding to a system with a compatible gpu, if possible.

CC: (none) => davidwhodgins
Keywords: (none) => feedback

(In reply to Dave Hodgins from comment #17)
> Do not add all users to a group. Just include a README.urpmi explaining
> the need, and a reminder that logout/in or reboot is required for group
> changes to take effect.
> 
> It is not possible to find out all users who can login by parsing /home
> or /etc/passwd, as the admin is free to create directories in /home that
> are not for users, and is free to use non-standard id numbers in /etc/passwd.
> Even if that worked, the admin wouldn't know that it's needed for users who
> are created in the future.
> 
> Qa testing should also ensure that the change doesn't interfere for people
> who use ssh with X forwarding to a system with a compatible gpu, if possible.

Hi. There is already a README.urpmi informing new users will have to be added to the relevant groups.
That being said, I followed your advice and removed the automatic addition of existing users. Instead, the README.urpmi explains the need to add users (current and future) to the render and video groups.

 @katnatek, your GPU is indeed too old.

This package is for Navi x at first and it might work for Vega and Polaris GPU.
In other words, any RX card should benefit from it, together with Vega iGPU.

I suggest a writeup of this and other recent additions to AMD proprietary drivers in

https://wiki.mageia.org/en/Setup_the_graphical_server#AMD_proprietary_drivers

and then link to there from release notes.

Status comment: (none) => FOR_WIKI

I'm installing this now for testing.

One thing I suggest is to had "task" to the name of the package "rocm-amd-opencl" to make it easier to tell what should be installed, following the trend with other package sets in Mageia (e.g. task-c++-devel).

I removed all mesa OpenCL stuff to start with a clean slate and avoid issues.


System: Mageia 9, x86_64, Plasma DE, AMD Ryzen 5 5600G with Radeon Graphics using amdgpu driver.


This is a test for the iGPU of the AMD Ryzen 5 5600G. Will test the Radeon RX 6500 XT in a later post.

The darktable-cltest produced the following messages:
"""
[opencl_init] OpenCL successfully initialized. Internal numbers and names of available devices:
[opencl_init]           0       'AMD Accelerated Parallel Processing gfx90c:xnack-'
[opencl_init] FINALLY: opencl is AVAILABLE and ENABLED.
"""

Also tested some random OpenCL examples found of github and gitlab. So it seems ROCM's OpenCL is working, at least for darktable and a few small examples.

Next the the full length output of the various commands.

# ##### BEFORE installation I get this:
# uname -a
Linux jupiter 6.5.13-desktop-6.mga9 #1 SMP PREEMPT_DYNAMIC Sun Dec 17 22:42:25 UTC 2023 x86_64 GNU/Linux
$ LANGUAGE=C lscpu | grep name
Model name:                         AMD Ryzen 5 5600G with Radeon Graphics
$ lspci | grep VGA
03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 24 [Radeon RX 6400/6500 XT/6500M] (rev c1)
0c:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Cezanne [Radeon Vega Series / Radeon Vega Mobile Series] (rev c9)
$ lspcidrake  | grep VGA
Card:ATI Volcanic Islands and later (amdgpu): Advanced Micro Devices, Inc. [AMD/ATI]|Cezanne [Radeon Vega Series / Radeon Vega Mobile Series] [DISPLAY_VGA] (rev: c9)
Card:AMD Southern Islands and later (amdgpu): Advanced Micro Devices, Inc. [AMD/ATI]|Navi 24 [Radeon RX 6400/6500 XT/6500M] [DISPLAY_VGA] (rev: c1)
# clinfo
Number of platforms                               0

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.3.1
  ICD loader Profile                              OpenCL 3.0

# ##### Installed rocm-amd-opencl with no issues.
# LANGUAGE=C urpmi rocm-amd-opencl
To satisfy dependencies, the following packages are going to be installed:
  Package                        Version      Release       Arch    
(medium "QA Testing (64-bit)")
  lib64hsakmt1                   1.0.6        0.5.7.1.2.mg> x86_64  
  lib64rocm-compilersupport5.7.1 5.7.1        1.mga9        x86_64  
  lib64rocm-opencl-runtime-devel 5.7.1        2.mga9        x86_64  
  lib64rocm-opencl-runtime5.7    5.7.1        2.mga9        x86_64  
  lib64rocm-runtime1             5.7.1        1.mga9        x86_64  
  rocm-amd-opencl                5.7.1        2.mga9        x86_64  
  rocm-clinfo                    5.7.1        2.mga9        x86_64  (recommended)
  rocm-core                      5.7.1        1.mga9        x86_64  
  rocm-device-libs               5.7.1        1.mga9        x86_64  
  rocminfo                       5.7.1        1.mga9        x86_64  
(medium "Core Release")
  lib64ocl-icd-devel             2.3.1        2.mga9        x86_64  
  lib64opencl-devel              2.3.1        2.mga9        x86_64  
  opencl-headers                 3.0          0.20230206.1> noarch  
183MB of additional disk space will be used.
39MB of packages will be retrieved.
Proceed with the installation of the 13 packages? (Y/n) 

# ##### AFTER installation I get this:
# clinfo 
Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.1 AMD-APP (3590.0)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_event_callback 
  Platform Extensions function suffix             AMD
  Platform Host timer resolution                  1ns

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 1
  Device Name                                     gfx90c:xnack-
  Device Vendor                                   Advanced Micro Devices, Inc.
  Device Vendor ID                                0x1002
  Device Version                                  OpenCL 2.0 
  Driver Version                                  3590.0 (HSA1.1,LC)
  Device OpenCL C Version                         OpenCL C 2.0 
  Device Type                                     GPU
  Device Board Name (AMD)                         AMD Radeon Graphics
  Device PCI-e ID (AMD)                           0x1638
  Device Topology (AMD)                           PCI-E, 0000:0c:00.0
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               7
  SIMD per compute unit (AMD)                     4
  SIMD width (AMD)                                16
  SIMD instruction width (AMD)                    1
  Max clock frequency                             1900MHz
  Graphics IP (AMD)                               9.0
  Device Partition                                (core)
    Max number of sub-devices                     7
    Supported partition types                     None
    Supported affinity domains                    (n/a)
  Max work item dimensions                        3
  Max work item sizes                             1024x1024x1024
  Max work group size                             256
  Preferred work group size (AMD)                 256
  Max work group size (AMD)                       1024
  Preferred work group size multiple (kernel)     64
  Wavefront width (AMD)                           64
  Preferred / native vector sizes                 
    char                                                 4 / 4       
    short                                                2 / 2       
    int                                                  1 / 1       
    long                                                 1 / 1       
    half                                                 1 / 1        (cl_khr_fp16)
    float                                                1 / 1       
    double                                               1 / 1        (cl_khr_fp64)
  Half-precision Floating-point support           (cl_khr_fp16)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Single-precision Floating-point support         (core)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  Yes
  Double-precision Floating-point support         (cl_khr_fp64)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Address bits                                    64, Little-Endian
  Global memory size                              536870912 (512MiB)
  Global free memory (AMD)                        464896 (454MiB) 464896 (454MiB)
  Global memory channels (AMD)                    4
  Global memory banks per channel (AMD)           4
  Global memory bank width (AMD)                  256 bytes
  Error Correction support                        No
  Max memory allocation                           402653184 (384MiB)
  Unified memory for Host and Device              No
  Shared Virtual Memory (SVM) capabilities        (core)
    Coarse-grained buffer sharing                 Yes
    Fine-grained buffer sharing                   Yes
    Fine-grained system sharing                   No
    Atomics                                       No
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Preferred alignment for atomics                 
    SVM                                           0 bytes
    Global                                        0 bytes
    Local                                         0 bytes
  Max size for global variable                    402653184 (384MiB)
  Preferred total size of global vars             536870912 (512MiB)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        16384 (16KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   Yes
    Max number of samplers per kernel             16
    Max size for 1D images from buffer            134217728 pixels
    Max 1D or 2D image array size                 8192 images
    Base address alignment for 2D image buffers   256 bytes
    Pitch alignment for 2D image buffers          256 pixels
    Max 2D image size                             16384x16384 pixels
    Max 3D image size                             16384x16384x8192 pixels
    Max number of read image args                 128
    Max number of write image args                8
    Max number of read/write image args           64
  Max number of pipe args                         16
  Max active pipe reservations                    16
  Max pipe packet size                            402653184 (384MiB)
  Local memory type                               Local
  Local memory size                               65536 (64KiB)
  Local memory size per CU (AMD)                  65536 (64KiB)
  Local memory banks (AMD)                        32
  Max number of constant args                     8
  Max constant buffer size                        402653184 (384MiB)
  Preferred constant buffer size (AMD)            16384 (16KiB)
  Max size of kernel argument                     1024
  Queue properties (on host)                      
    Out-of-order execution                        No
    Profiling                                     Yes
  Queue properties (on device)                    
    Out-of-order execution                        Yes
    Profiling                                     Yes
    Preferred size                                262144 (256KiB)
    Max size                                      8388608 (8MiB)
  Max queues on device                            1
  Max events on device                            1024
  Prefer user sync for interop                    Yes
  Number of P2P devices (AMD)                     0
  Profiling timer resolution                      1ns
  Profiling timer offset since Epoch (AMD)        0ns (Thu Jan  1 01:00:00 1970)
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
    Thread trace supported (AMD)                  No
    Number of async queues (AMD)                  8
    Max real-time compute queues (AMD)            8
    Max real-time compute units (AMD)             7
  printf() buffer size                            4194304 (4MiB)
  Built-in kernels                                (n/a)
  Device Extensions                               cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_media_ops cl_amd_media_ops2 cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program 

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  AMD Accelerated Parallel Processing
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [AMD]
  clCreateContext(NULL, ...) [default]            Success [AMD]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx90c:xnack-
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx90c:xnack-
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx90c:xnack-

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.3.1
  ICD loader Profile                              OpenCL 3.0

# rocm-clinfo 
Number of platforms:                             1
  Platform Profile:                              FULL_PROFILE
  Platform Version:                              OpenCL 2.1 AMD-APP (3590.0)
  Platform Name:                                 AMD Accelerated Parallel Processing
  Platform Vendor:                               Advanced Micro Devices, Inc.
  Platform Extensions:                           cl_khr_icd cl_amd_event_callback 


  Platform Name:                                 AMD Accelerated Parallel Processing
Number of devices:                               1
  Device Type:                                   CL_DEVICE_TYPE_GPU
  Vendor ID:                                     1002h
  Board name:                                    AMD Radeon Graphics
  Device Topology:                               PCI[ B#12, D#0, F#0 ]
  Max compute units:                             7
  Max work items dimensions:                     3
    Max work items[0]:                           1024
    Max work items[1]:                           1024
    Max work items[2]:                           1024
  Max work group size:                           256
  Preferred vector width char:                   4
  Preferred vector width short:                  2
  Preferred vector width int:                    1
  Preferred vector width long:                   1
  Preferred vector width float:                  1
  Preferred vector width double:                 1
  Native vector width char:                      4
  Native vector width short:                     2
  Native vector width int:                       1
  Native vector width long:                      1
  Native vector width float:                     1
  Native vector width double:                    1
  Max clock frequency:                           1900Mhz
  Address bits:                                  64
  Max memory allocation:                         402653184
  Image support:                                 Yes
  Max number of images read arguments:           128
  Max number of images write arguments:          8
  Max image 2D width:                            16384
  Max image 2D height:                           16384
  Max image 3D width:                            16384
  Max image 3D height:                           16384
  Max image 3D depth:                            8192
  Max samplers within kernel:                    16
  Max size of kernel argument:                   1024
  Alignment (bits) of base address:              1024
  Minimum alignment (bytes) for any datatype:    128
  Single precision floating point capability
    Denorms:                                     Yes
    Quiet NaNs:                                  Yes
    Round to nearest even:                       Yes
    Round to zero:                               Yes
    Round to +ve and infinity:                   Yes
    IEEE754-2008 fused multiply-add:             Yes
  Cache type:                                    Read/Write
  Cache line size:                               64
  Cache size:                                    16384
  Global memory size:                            536870912
  Constant buffer size:                          402653184
  Max number of constant args:                   8
  Local memory type:                             Scratchpad
  Local memory size:                             65536
  Max pipe arguments:                            16
  Max pipe active reservations:                  16
  Max pipe packet size:                          402653184
  Max global variable size:                      402653184
  Max global variable preferred total size:      536870912
  Max read/write image args:                     64
  Max on device events:                          1024
  Queue on device max size:                      8388608
  Max on device queues:                          1
  Queue on device preferred size:                262144
  SVM capabilities:                              
    Coarse grain buffer:                         Yes
    Fine grain buffer:                           Yes
    Fine grain system:                           No
    Atomics:                                     No
  Preferred platform atomic alignment:           0
  Preferred global atomic alignment:             0
  Preferred local atomic alignment:              0
  Kernel Preferred work group size multiple:     64
  Error correction support:                      0
  Unified memory for Host and Device:            0
  Profiling timer resolution:                    1
  Device endianess:                              Little
  Available:                                     Yes
  Compiler available:                            Yes
  Execution capabilities:                                
    Execute OpenCL kernels:                      Yes
    Execute native function:                     No
  Queue on Host properties:                              
    Out-of-Order:                                No
    Profiling :                                  Yes
  Queue on Device properties:                            
    Out-of-Order:                                Yes
    Profiling :                                  Yes
  Platform ID:                                   0x7f4dee51de90
  Name:                                          gfx90c:xnack-
  Vendor:                                        Advanced Micro Devices, Inc.
  Device OpenCL C version:                       OpenCL C 2.0 
  Driver version:                                3590.0 (HSA1.1,LC)
  Profile:                                       FULL_PROFILE
  Version:                                       OpenCL 2.0 
  Extensions:                                    cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_media_ops cl_amd_media_ops2 cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program 


# rocminfo 
ROCk module is loaded
=====================    
HSA System Attributes    
=====================    
Runtime Version:         1.1
System Timestamp Freq.:  1000.000000MHz
Sig. Max Wait Duration:  18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model:           LARGE                              
System Endianness:       LITTLE                             
Mwaitx:                  DISABLED
DMAbuf Support:          YES

==========               
HSA Agents               
==========               
*******                  
Agent 1                  
*******                  
  Name:                    AMD Ryzen 5 5600G with Radeon Graphics
  Uuid:                    CPU-XX                             
  Marketing Name:          AMD Ryzen 5 5600G with Radeon Graphics
  Vendor Name:             CPU                                
  Feature:                 None specified                     
  Profile:                 FULL_PROFILE                       
  Float Round Mode:        NEAR                               
  Max Queue Number:        0(0x0)                             
  Queue Min Size:          0(0x0)                             
  Queue Max Size:          0(0x0)                             
  Queue Type:              MULTI                              
  Node:                    0                                  
  Device Type:             CPU                                
  Cache Info:              
    L1:                      32768(0x8000) KB                   
  Chip ID:                 0(0x0)                             
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   4464                               
  BDFID:                   0                                  
  Internal Node ID:        0                                  
  Compute Unit:            12                                 
  SIMDs per CU:            0                                  
  Shader Engines:          0                                  
  Shader Arrs. per Eng.:   0                                  
  WatchPts on Addr. Ranges:1                                  
  Features:                None
  Pool Info:               
    Pool 1                   
      Segment:                 GLOBAL; FLAGS: FINE GRAINED        
      Size:                    32144280(0x1ea7b98) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 2                   
      Segment:                 GLOBAL; FLAGS: KERNARG, FINE GRAINED
      Size:                    32144280(0x1ea7b98) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 3                   
      Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
      Size:                    32144280(0x1ea7b98) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
  ISA Info:                
*******                  
Agent 2                  
*******                  
  Name:                    gfx90c                             
  Uuid:                    GPU-XX                             
  Marketing Name:          AMD Radeon Graphics                
  Vendor Name:             AMD                                
  Feature:                 KERNEL_DISPATCH                    
  Profile:                 BASE_PROFILE                       
  Float Round Mode:        NEAR                               
  Max Queue Number:        128(0x80)                          
  Queue Min Size:          64(0x40)                           
  Queue Max Size:          131072(0x20000)                    
  Queue Type:              MULTI                              
  Node:                    1                                  
  Device Type:             GPU                                
  Cache Info:              
    L1:                      16(0x10) KB                        
    L2:                      1024(0x400) KB                     
  Chip ID:                 5688(0x1638)                       
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   1900                               
  BDFID:                   3072                               
  Internal Node ID:        1                                  
  Compute Unit:            7                                  
  SIMDs per CU:            4                                  
  Shader Engines:          1                                  
  Shader Arrs. per Eng.:   1                                  
  WatchPts on Addr. Ranges:4                                  
  Features:                KERNEL_DISPATCH 
  Fast F16 Operation:      TRUE                               
  Wavefront Size:          64(0x40)                           
  Workgroup Max Size:      1024(0x400)                        
  Workgroup Max Size per Dimension:
    x                        1024(0x400)                        
    y                        1024(0x400)                        
    z                        1024(0x400)                        
  Max Waves Per CU:        40(0x28)                           
  Max Work-item Per CU:    2560(0xa00)                        
  Grid Max Size:           4294967295(0xffffffff)             
  Grid Max Size per Dimension:
    x                        4294967295(0xffffffff)             
    y                        4294967295(0xffffffff)             
    z                        4294967295(0xffffffff)             
  Max fbarriers/Workgrp:   32                                 
  Packet Processor uCode:: 464                                
  SDMA engine uCode::      40                                 
  IOMMU Support::          None                               
  Pool Info:               
    Pool 1                   
      Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
      Size:                    524288(0x80000) KB                 
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       FALSE                              
    Pool 2                   
      Segment:                 GLOBAL; FLAGS:                     
      Size:                    524288(0x80000) KB                 
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       FALSE                              
    Pool 3                   
      Segment:                 GROUP                              
      Size:                    64(0x40) KB                        
      Allocatable:             FALSE                              
      Alloc Granule:           0KB                                
      Alloc Alignment:         0KB                                
      Accessible by all:       FALSE                              
  ISA Info:                
    ISA 1                    
      Name:                    amdgcn-amd-amdhsa--gfx90c:xnack-   
      Machine Models:          HSA_MACHINE_MODEL_LARGE            
      Profiles:                HSA_PROFILE_BASE                   
      Default Rounding Mode:   NEAR                               
      Default Rounding Mode:   NEAR                               
      Fast f16:                TRUE                               
      Workgroup Max Size:      1024(0x400)                        
      Workgroup Max Size per Dimension:
        x                        1024(0x400)                        
        y                        1024(0x400)                        
        z                        1024(0x400)                        
      Grid Max Size:           4294967295(0xffffffff)             
      Grid Max Size per Dimension:
        x                        4294967295(0xffffffff)             
        y                        4294967295(0xffffffff)             
        z                        4294967295(0xffffffff)             
      FBarrier Max Size:       32                                 
*** Done ***             

# ##### Test using darktable

$ groups
pclx video pulseaudio render libvirt
$ darktable-cltest
     0,0150 [dt_get_sysresource_level] switched to 1 as `default'
     0,0151   total mem:       31390MB
     0,0151   mipmap cache:    3923MB
     0,0151   available mem:   15695MB
     0,0151   singlebuff:      245MB
     0,0151   OpenCL tune mem: OFF
     0,0151   OpenCL pinned:   OFF
[opencl_init] opencl related configuration options:
[opencl_init] opencl: ON
[opencl_init] opencl_scheduling_profile: 'default'
[opencl_init] opencl_library: 'default path'
[opencl_init] opencl_device_priority: '*/!0,*/*/*/!0,*'
[opencl_init] opencl_mandatory_timeout: 400
[opencl_init] opencl library 'libOpenCL' found on your system and loaded
[opencl_init] found 1 platform
[opencl_init] found 1 device

[dt_opencl_device_init]
   DEVICE:                   0: 'gfx90c:xnack-', NEW
   PLATFORM NAME & VENDOR:   AMD Accelerated Parallel Processing, Advanced Micro Devices, Inc.
   CANONICAL NAME:           amdacceleratedparallelprocessinggfx90cxnack
   DRIVER VERSION:           3590.0 (HSA1.1,LC)
   DEVICE VERSION:           OpenCL 2.0 
   DEVICE_TYPE:              GPU
   GLOBAL MEM SIZE:          512 MB
   MAX MEM ALLOC:            384 MB
   MAX IMAGE SIZE:           16384 x 16384
   MAX WORK GROUP SIZE:      256
   MAX WORK ITEM DIMENSIONS: 3
   MAX WORK ITEM SIZES:      [ 1024 1024 1024 ]
   ASYNC PIXELPIPE:          NO
   PINNED MEMORY TRANSFER:   NO
   MEMORY TUNING:            NO
   FORCED HEADROOM:          400
   AVOID ATOMICS:            NO
   MICRO NAP:                250
   ROUNDUP WIDTH:            16
   ROUNDUP HEIGHT:           16
   CHECK EVENT HANDLES:      128
   TILING ADVANTAGE:         0.000
   DEFAULT DEVICE:           NO
   KERNEL BUILD DIRECTORY:   /usr/share/darktable/kernels
   KERNEL DIRECTORY:         /home/pclx/.cache/darktable/cached_v1_kernels_for_AMDAcceleratedParallelProcessinggfx90cxnack_35900HSA11LC
   CL COMPILER OPTION:       -cl-fast-relaxed-math
   KERNEL LOADING TIME:       9.6639 sec
[opencl_init] OpenCL successfully initialized. Internal numbers and names of available devices:
[opencl_init]           0       'AMD Accelerated Parallel Processing gfx90c:xnack-'
[opencl_init] FINALLY: opencl is AVAILABLE and ENABLED.
[opencl_init] set scheduling profile to default.
[dt_opencl_update_priorities] these are your device priorities:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       -1      0       0       -1
[dt_opencl_update_priorities] show if opencl use is mandatory for a given pixelpipe:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       0       0       0       0
[opencl_synchronization_timeout] synchronization timeout set to 200
[dt_opencl_update_priorities] these are your device priorities:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       -1      0       0       -1
[dt_opencl_update_priorities] show if opencl use is mandatory for a given pixelpipe:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       0       0       0       0
[opencl_synchronization_timeout] synchronization timeout set to 200

Thanks PC LX for this thorough test. It shows it works for Vega.
Indeed, @Joselp, amdgpupro-opencl-pal could be removed automatically during this new package install.

@PC LX: I don't understand why calling this package task-rocm-amd-opencl would help the user. But, if most of you are for it, I can update the package name. It will complicate a bit the spec because I would need to obsolete rocm-amd-opencl and the package name will be too long for my taste (but I will cope with that :) )

(In reply to christian squidf from comment #22)
> @PC LX: I don't understand why calling this package task-rocm-amd-opencl
> would help the user. But, if most of you are for it, I can update the
> package name. It will complicate a bit the spec because I would need to
> obsolete rocm-amd-opencl and the package name will be too long for my taste
> (but I will cope with that :) )

When I first saw the list of packages, I was uncertain what I should install and I though that a task package would help here (like with task-c++-devel or task-c-devel). Maybe it is just a me thing and since it requires extra work then just ignore my comment.

(In reply to PC LX from comment #23)
> (In reply to christian squidf from comment #22)
> > @PC LX: I don't understand why calling this package task-rocm-amd-opencl
> > would help the user. But, if most of you are for it, I can update the
> > package name. It will complicate a bit the spec because I would need to
> > obsolete rocm-amd-opencl and the package name will be too long for my taste
> > (but I will cope with that :) )
> 
> When I first saw the list of packages, I was uncertain what I should install
> and I though that a task package would help here (like with task-c++-devel
> or task-c-devel). Maybe it is just a me thing and since it requires extra
> work then just ignore my comment.

Hi. As the package is still in Testing, it requires dependencies which are not yet available. It is why I listed all of them.
However, there is only one package to install: rocm-amd-opencl. All the others will follow automatically. Thanks again for all your tests.

Installed and tested without issues on a AMD Radeon RX 6500 XT GPU.
Previous comment 21 has tests for a AMD Ryzen 5 5600G iGPU. Also no issues.
This gets two OK from me.


System: Mageia 9, x86_64, Plasma DE, AMD Ryzen 5 5600G, AMD Radeon RX 6500 XT using amdgpu driver.


This is a test for a Radeon RX 6500 XT (gfx1034) with 4GiB VDRAM.

The darktable-cltest produced the following messages:
"""
[opencl_init] OpenCL successfully initialized. Internal numbers and names of available devices:
[opencl_init]           0       'AMD Accelerated Parallel Processing gfx1034'
[opencl_init] FINALLY: opencl is AVAILABLE and ENABLED.
"""

Also tested some OpenCL examples found of github and gitlab. ROCM's OpenCL seems to be working with no issues.



Next the the full length output of the various commands.

# ##### BEFORE installation I get this:
# uname -a
Linux jupiter-vm-mageia-9-jogos 6.5.13-desktop-6.mga9 #1 SMP PREEMPT_DYNAMIC Sun Dec 17 22:42:25 UTC 2023 x86_64 GNU/Linux
# LANGUAGE=C lscpu | grep name
Model name:                         AMD Ryzen 5 5600G with Radeon Graphics
BIOS Model name:                    pc-q35-5.2  CPU @ 2.0GHz
# lspci | grep VGA
0c:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 24 [Radeon RX 6400/6500 XT/6500M] (rev c1)
# clinfo
Number of platforms                               0

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.3.1
  ICD loader Profile                              OpenCL 3.0

# ##### Installed rocm-amd-opencl with no issues.
# urpmi rocm-amd-opencl
To satisfy dependencies, the following packages are going to be installed:
  Package                        Version      Release       Arch    
(medium "QA Testing (64-bit)")
  lib64hsakmt1                   1.0.6        0.5.7.1.2.mg> x86_64  
  lib64rocm-compilersupport5.7.1 5.7.1        1.mga9        x86_64  
  lib64rocm-opencl-runtime-devel 5.7.1        3.mga9        x86_64  
  lib64rocm-opencl-runtime5.7    5.7.1        3.mga9        x86_64  
  lib64rocm-runtime1             5.7.1        1.mga9        x86_64  
  rocm-amd-opencl                5.7.1        3.mga9        x86_64  
  rocm-clinfo                    5.7.1        3.mga9        x86_64  (recommended)
  rocm-core                      5.7.1        1.mga9        x86_64  
  rocm-device-libs               5.7.1        1.mga9        x86_64  
  rocminfo                       5.7.1        1.mga9        x86_64  
(medium "Core Release")
  lib64ocl-icd-devel             2.3.1        2.mga9        x86_64  
  lib64opencl-devel              2.3.1        2.mga9        x86_64  
  opencl-headers                 3.0          0.20230206.1> noarch  
183MB of additional disk space will be used.
39MB of packages will be retrieved.
Proceed with the installation of the 13 packages? (Y/n) 

# ##### AFTER installation I get this:
# clinfo 
Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.1 AMD-APP (3590.0)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_event_callback 
  Platform Extensions function suffix             AMD
  Platform Host timer resolution                  1ns

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 1
  Device Name                                     gfx1034
  Device Vendor                                   Advanced Micro Devices, Inc.
  Device Vendor ID                                0x1002
  Device Version                                  OpenCL 2.0 
  Driver Version                                  3590.0 (HSA1.1,LC)
  Device OpenCL C Version                         OpenCL C 2.0 
  Device Type                                     GPU
  Device Board Name (AMD)                         AMD Radeon RX 6500 XT
  Device PCI-e ID (AMD)                           0x743f
  Device Topology (AMD)                           PCI-E, 0000:0c:00.0
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               8
  SIMD per compute unit (AMD)                     4
  SIMD width (AMD)                                32
  SIMD instruction width (AMD)                    1
  Max clock frequency                             2975MHz
  Graphics IP (AMD)                               10.3
  Device Partition                                (core)
    Max number of sub-devices                     8
    Supported partition types                     None
    Supported affinity domains                    (n/a)
  Max work item dimensions                        3
  Max work item sizes                             1024x1024x1024
  Max work group size                             256
  Preferred work group size (AMD)                 256
  Max work group size (AMD)                       1024
  Preferred work group size multiple (kernel)     32
  Wavefront width (AMD)                           32
  Preferred / native vector sizes                 
    char                                                 4 / 4       
    short                                                2 / 2       
    int                                                  1 / 1       
    long                                                 1 / 1       
    half                                                 1 / 1        (cl_khr_fp16)
    float                                                1 / 1       
    double                                               1 / 1        (cl_khr_fp64)
  Half-precision Floating-point support           (cl_khr_fp16)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Single-precision Floating-point support         (core)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  Yes
  Double-precision Floating-point support         (cl_khr_fp64)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Address bits                                    64, Little-Endian
  Global memory size                              4278190080 (3.984GiB)
  Global free memory (AMD)                        3997696 (3.812GiB) 3997696 (3.812GiB)
  Global memory channels (AMD)                    2
  Global memory banks per channel (AMD)           4
  Global memory bank width (AMD)                  256 bytes
  Error Correction support                        No
  Max memory allocation                           3636461568 (3.387GiB)
  Unified memory for Host and Device              No
  Shared Virtual Memory (SVM) capabilities        (core)
    Coarse-grained buffer sharing                 Yes
    Fine-grained buffer sharing                   Yes
    Fine-grained system sharing                   No
    Atomics                                       No
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Preferred alignment for atomics                 
    SVM                                           0 bytes
    Global                                        0 bytes
    Local                                         0 bytes
  Max size for global variable                    3636461568 (3.387GiB)
  Preferred total size of global vars             4278190080 (3.984GiB)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        16384 (16KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   Yes
    Max number of samplers per kernel             16
    Max size for 1D images from buffer            134217728 pixels
    Max 1D or 2D image array size                 8192 images
    Base address alignment for 2D image buffers   256 bytes
    Pitch alignment for 2D image buffers          256 pixels
    Max 2D image size                             16384x16384 pixels
    Max 3D image size                             16384x16384x8192 pixels
    Max number of read image args                 128
    Max number of write image args                8
    Max number of read/write image args           64
  Max number of pipe args                         16
  Max active pipe reservations                    16
  Max pipe packet size                            3636461568 (3.387GiB)
  Local memory type                               Local
  Local memory size                               65536 (64KiB)
  Local memory size per CU (AMD)                  65536 (64KiB)
  Local memory banks (AMD)                        32
  Max number of constant args                     8
  Max constant buffer size                        3636461568 (3.387GiB)
  Preferred constant buffer size (AMD)            16384 (16KiB)
  Max size of kernel argument                     1024
  Queue properties (on host)                      
    Out-of-order execution                        No
    Profiling                                     Yes
  Queue properties (on device)                    
    Out-of-order execution                        Yes
    Profiling                                     Yes
    Preferred size                                262144 (256KiB)
    Max size                                      8388608 (8MiB)
  Max queues on device                            1
  Max events on device                            1024
  Prefer user sync for interop                    Yes
  Number of P2P devices (AMD)                     0
  Profiling timer resolution                      1ns
  Profiling timer offset since Epoch (AMD)        0ns (Thu Jan  1 01:00:00 1970)
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
    Thread trace supported (AMD)                  No
    Number of async queues (AMD)                  8
    Max real-time compute queues (AMD)            8
    Max real-time compute units (AMD)             8
  printf() buffer size                            4194304 (4MiB)
  Built-in kernels                                (n/a)
  Device Extensions                               cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_media_ops cl_amd_media_ops2 cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program 

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  AMD Accelerated Parallel Processing
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [AMD]
  clCreateContext(NULL, ...) [default]            Success [AMD]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx1034
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx1034
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   gfx1034

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.3.1
  ICD loader Profile                              OpenCL 3.0


# rocminfo
ROCk module is loaded
=====================    
HSA System Attributes    
=====================    
Runtime Version:         1.1
System Timestamp Freq.:  1000.000000MHz
Sig. Max Wait Duration:  18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model:           LARGE                              
System Endianness:       LITTLE                             
Mwaitx:                  DISABLED
DMAbuf Support:          YES

==========               
HSA Agents               
==========               
*******                  
Agent 1                  
*******                  
  Name:                    AMD Ryzen 5 5600G with Radeon Graphics
  Uuid:                    CPU-XX                             
  Marketing Name:          AMD Ryzen 5 5600G with Radeon Graphics
  Vendor Name:             CPU                                
  Feature:                 None specified                     
  Profile:                 FULL_PROFILE                       
  Float Round Mode:        NEAR                               
  Max Queue Number:        0(0x0)                             
  Queue Min Size:          0(0x0)                             
  Queue Max Size:          0(0x0)                             
  Queue Type:              MULTI                              
  Node:                    0                                  
  Device Type:             CPU                                
  Cache Info:              
    L1:                      65536(0x10000) KB                  
  Chip ID:                 0(0x0)                             
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   0                                  
  BDFID:                   0                                  
  Internal Node ID:        0                                  
  Compute Unit:            12                                 
  SIMDs per CU:            0                                  
  Shader Engines:          0                                  
  Shader Arrs. per Eng.:   0                                  
  WatchPts on Addr. Ranges:1                                  
  Features:                None
  Pool Info:               
    Pool 1                   
      Segment:                 GLOBAL; FLAGS: FINE GRAINED        
      Size:                    16384548(0xfa0224) KB              
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 2                   
      Segment:                 GLOBAL; FLAGS: KERNARG, FINE GRAINED
      Size:                    16384548(0xfa0224) KB              
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 3                   
      Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
      Size:                    16384548(0xfa0224) KB              
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
  ISA Info:                
*******                  
Agent 2                  
*******                  
  Name:                    gfx1034                            
  Uuid:                    GPU-XX                             
  Marketing Name:          AMD Radeon RX 6500 XT              
  Vendor Name:             AMD                                
  Feature:                 KERNEL_DISPATCH                    
  Profile:                 BASE_PROFILE                       
  Float Round Mode:        NEAR                               
  Max Queue Number:        128(0x80)                          
  Queue Min Size:          64(0x40)                           
  Queue Max Size:          131072(0x20000)                    
  Queue Type:              MULTI                              
  Node:                    1                                  
  Device Type:             GPU                                
  Cache Info:              
    L1:                      16(0x10) KB                        
    L2:                      1024(0x400) KB                     
    L3:                      16384(0x4000) KB                   
  Chip ID:                 29759(0x743f)                      
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   2975                               
  BDFID:                   3072                               
  Internal Node ID:        1                                  
  Compute Unit:            16                                 
  SIMDs per CU:            2                                  
  Shader Engines:          1                                  
  Shader Arrs. per Eng.:   2                                  
  WatchPts on Addr. Ranges:4                                  
  Features:                KERNEL_DISPATCH 
  Fast F16 Operation:      TRUE                               
  Wavefront Size:          32(0x20)                           
  Workgroup Max Size:      1024(0x400)                        
  Workgroup Max Size per Dimension:
    x                        1024(0x400)                        
    y                        1024(0x400)                        
    z                        1024(0x400)                        
  Max Waves Per CU:        32(0x20)                           
  Max Work-item Per CU:    1024(0x400)                        
  Grid Max Size:           4294967295(0xffffffff)             
  Grid Max Size per Dimension:
    x                        4294967295(0xffffffff)             
    y                        4294967295(0xffffffff)             
    z                        4294967295(0xffffffff)             
  Max fbarriers/Workgrp:   32                                 
  Packet Processor uCode:: 113                                
  SDMA engine uCode::      34                                 
  IOMMU Support::          None                               
  Pool Info:               
    Pool 1                   
      Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
      Size:                    4177920(0x3fc000) KB               
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       FALSE                              
    Pool 2                   
      Segment:                 GLOBAL; FLAGS:                     
      Size:                    4177920(0x3fc000) KB               
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       FALSE                              
    Pool 3                   
      Segment:                 GROUP                              
      Size:                    64(0x40) KB                        
      Allocatable:             FALSE                              
      Alloc Granule:           0KB                                
      Alloc Alignment:         0KB                                
      Accessible by all:       FALSE                              
  ISA Info:                
    ISA 1                    
      Name:                    amdgcn-amd-amdhsa--gfx1034         
      Machine Models:          HSA_MACHINE_MODEL_LARGE            
      Profiles:                HSA_PROFILE_BASE                   
      Default Rounding Mode:   NEAR                               
      Default Rounding Mode:   NEAR                               
      Fast f16:                TRUE                               
      Workgroup Max Size:      1024(0x400)                        
      Workgroup Max Size per Dimension:
        x                        1024(0x400)                        
        y                        1024(0x400)                        
        z                        1024(0x400)                        
      Grid Max Size:           4294967295(0xffffffff)             
      Grid Max Size per Dimension:
        x                        4294967295(0xffffffff)             
        y                        4294967295(0xffffffff)             
        z                        4294967295(0xffffffff)             
      FBarrier Max Size:       32                                 
*** Done ***             


# ##### Test using darktable
$ groups
pclx video render
$ darktable-cltest
     0,0148 [dt_get_sysresource_level] switched to 1 as `default'
     0,0148   total mem:       16000MB
     0,0148   mipmap cache:    2000MB
     0,0148   available mem:   8000MB
     0,0148   singlebuff:      125MB
     0,0148   OpenCL tune mem: OFF
     0,0148   OpenCL pinned:   OFF
[opencl_init] opencl related configuration options:
[opencl_init] opencl: ON
[opencl_init] opencl_scheduling_profile: 'default'
[opencl_init] opencl_library: 'default path'
[opencl_init] opencl_device_priority: '*/!0,*/*/*/!0,*'
[opencl_init] opencl_mandatory_timeout: 400
[opencl_init] opencl library 'libOpenCL' found on your system and loaded
[opencl_init] found 1 platform
[opencl_init] found 1 device

[dt_opencl_device_init]
   DEVICE:                   0: 'gfx1034', NEW
   PLATFORM NAME & VENDOR:   AMD Accelerated Parallel Processing, Advanced Micro Devices, Inc.
   CANONICAL NAME:           amdacceleratedparallelprocessinggfx1034
   DRIVER VERSION:           3590.0 (HSA1.1,LC)
   DEVICE VERSION:           OpenCL 2.0 
   DEVICE_TYPE:              GPU
   GLOBAL MEM SIZE:          4080 MB
   MAX MEM ALLOC:            3468 MB
   MAX IMAGE SIZE:           16384 x 16384
   MAX WORK GROUP SIZE:      256
   MAX WORK ITEM DIMENSIONS: 3
   MAX WORK ITEM SIZES:      [ 1024 1024 1024 ]
   ASYNC PIXELPIPE:          NO
   PINNED MEMORY TRANSFER:   NO
   MEMORY TUNING:            NO
   FORCED HEADROOM:          400
   AVOID ATOMICS:            NO
   MICRO NAP:                250
   ROUNDUP WIDTH:            16
   ROUNDUP HEIGHT:           16
   CHECK EVENT HANDLES:      128
   TILING ADVANTAGE:         0.000
   DEFAULT DEVICE:           NO
   KERNEL BUILD DIRECTORY:   /usr/share/darktable/kernels
   KERNEL DIRECTORY:         /home/pclx/.cache/darktable/cached_v1_kernels_for_AMDAcceleratedParallelProcessinggfx1034_35900HSA11LC
   CL COMPILER OPTION:       -cl-fast-relaxed-math
   KERNEL LOADING TIME:       10.0428 sec
[opencl_init] OpenCL successfully initialized. Internal numbers and names of available devices:
[opencl_init]           0       'AMD Accelerated Parallel Processing gfx1034'
[opencl_init] FINALLY: opencl is AVAILABLE and ENABLED.
[opencl_init] set scheduling profile to default.
[dt_opencl_update_priorities] these are your device priorities:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       -1      0       0       -1
[dt_opencl_update_priorities] show if opencl use is mandatory for a given pixelpipe:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       0       0       0       0
[opencl_synchronization_timeout] synchronization timeout set to 200
[dt_opencl_update_priorities] these are your device priorities:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       -1      0       0       -1
[dt_opencl_update_priorities] show if opencl use is mandatory for a given pixelpipe:
[dt_opencl_update_priorities]           image   preview export  thumbs  preview2
[dt_opencl_update_priorities]           0       0       0       0       0
[opencl_synchronization_timeout] synchronization timeout set to 200

Great and thanks again PC LX

@dave: is it okay to validate it?
I don't expect side effect on ssh with X forwarding, as OpenCL is for computational purpose. What did you have in mind more specifically?

Keywords: feedback => (none)

Advisory in SVN updated to the version from comment 25

Keywords: (none) => advisory

(In reply to christian squidf from comment #29)
> Apologies @Marja. I had to rebuild a few packages because I noticed some
> issues to update them while working on the new 6.0.0 release.
> I took the opportunity to review the description to be more precise on GPU
> series compatibility.

No problem, but:

> ADVISORY NOTICE PROPOSAL

<snip>
> 
> 
> SRPMS
> 9/core
>   rocm-amd-opencl-5.7.1-3.mga9.src.rpm

papoteur's mga-advisory script saw that that SRPM has subrel 1, and indeed, the changelog mail says:
[changelog] [RPM] 9 core/updates_testing rocm-amd-opencl-5.7.1-3.1.mga9

I'll put rocm-amd-opencl-5.7.1-3.1.mga9 in the advisory. For the rest, I have updated the advisory with the data from comment 29.

Keywords: (none) => advisory

Thanks @Marja
I paid extra attention to the section for QA but missed the section above...
sorry for that and thank you for covering my back.

Actually, the Provided packages section had some misses as well.
It might not be a big deal (is this section required in addition to the QA one?).

Here is the final version.

ADVISORY NOTICE PROPOSAL (Final)
========================
ROCm OpenCL stack 5.7.1 for AMD GPU


Description
ROCm OpenCL language runtime for AMD GPU, Navi series (RDNA instruction set) and most of the Vega and Polaris series (GCN instruction set), as long as OpenCL 2.0 and beyond is supported.
Old GPU based on GCN architecture might be supported, but it is not granted; it is safe to give it a try and to uninstall later, if it doesn't work.

                
References
https://bugs.mageia.org/show_bug.cgi?id=32580
https://docs.amd.com/
https://github.com/ROCm-Developer-Tools/clr
https://github.com/RadeonOpenCompute/rocminfo
https://github.com/RadeonOpenCompute/ROCR-Runtime
https://github.com/RadeonOpenCompute/ROCm-CompilerSupport
https://github.com/RadeonOpenCompute/ROCm-Device-Libs
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface
https://github.com/RadeonOpenCompute/rocm-cmake
https://github.com/RadeonOpenCompute/llvm-project


SRPMS
9/core
  rocm-amd-opencl-5.7.1-3.1.mga9.src.rpm
  rocminfo-5.7.1-1.1.mga9.src.rpm
  rocm-runtime-5.7.1-1.1.mga9.src.rpm
  rocm-compilersupport-5.7.1-1.mga9.src.rpm
  rocm-device-libs-5.7.1-1.mga9.src.rpm
  hsakmt-1.0.6-0.5.7.1.2.mga9.src.rpm
  rocm-core-5.7.1-1.mga9.src.rpm
  rocm-cmake-5.7.1-1.mga9.src.rpm
  rocm-llvm-5.7.1-1.mga9.src.rpm


PROVIDED PACKAGES:

  rocm-amd-opencl-5.7.1-3.1.mga9
  rocm-clinfo-5.7.1-3.1.mga9
  lib64rocm-opencl-runtime-devel-5.7.1-3.1.mga9
  lib64rocm-opencl-runtime5.7-5.7.1-3.1.mga9
  rocminfo-5.7.1-1.1.mga9
  lib64rocm-runtime-devel-5.7.1-1.1.mga9
  lib64rocm-runtime1-5.7.1-1.1.mga9
  lib64rocm-compilersupport-devel-5.7.1-1.mga9
  lib64rocm-compilersupport5.7.1-5.7.1-1.mga9
  rocm-device-libs-5.7.1-1.mga9
  lib64hsakmt-devel-1.0.6-0.5.7.1.2.mga9
  lib64hsakmt1-1.0.6-0.5.7.1.2.mga9
  rocm-core-5.7.1-1.mga9
  rocm-cmake-5.7.1-1.mga9
  rocm-llvm-tools-extra-5.7.1-1.mga9
  rocm-llvm-analyzer-5.7.1-1.mga9
  lib64rocm-llvm-devel-5.7.1-1.mga9
  lib64rocm-llvm17-5.7.1-1.mga9
  rocm-llvm-5.7.1-1.mga9
  rocm-llvm-static-5.7.1-1.mga9

    
PACKAGES FOR QA TESTING
=======================
x86_64:

rocm-amd-opencl-5.7.1-3.1.mga9.x86_64.rpm
rocm-clinfo-5.7.1-3.1.mga9.x86_64.rpm
lib64rocm-opencl-runtime-devel-5.7.1-3.1.mga9.x86_64.rpm
lib64rocm-opencl-runtime5.7-5.7.1-3.1.mga9.x86_64.rpm
rocminfo-5.7.1-1.1.mga9.x86_64.rpm
lib64rocm-runtime-devel-5.7.1-1.1.mga9.x86_64.rpm
lib64rocm-runtime1-5.7.1-1.1.mga9.x86_64.rpm
lib64rocm-compilersupport-devel-5.7.1-1.mga9.x86_64.rpm
lib64rocm-compilersupport5.7.1-5.7.1-1.mga9.x86_64.rpm
rocm-device-libs-5.7.1-1.mga9.x86_64.rpm
lib64hsakmt-devel-1.0.6-0.5.7.1.2.mga9.x86_64.rpm
lib64hsakmt1-1.0.6-0.5.7.1.2.mga9.x86_64.rpm
rocm-core-5.7.1-1.mga9.x86_64.rpm
rocm-cmake-5.7.1-1.mga9.x86_64.rpm
rocm-llvm-tools-extra-5.7.1-1.mga9.x86_64.rpm
rocm-llvm-analyzer-5.7.1-1.mga9.noarch.rpm
lib64rocm-llvm-devel-5.7.1-1.mga9.x86_64.rpm
lib64rocm-llvm17-5.7.1-1.mga9.x86_64.rpm
rocm-llvm-5.7.1-1.mga9.x86_64.rpm
rocm-llvm-static-5.7.1-1.mga9.x86_64.rpm

(In reply to christian squidf from comment #31)

> 
> Actually, the Provided packages section had some misses as well.
> It might not be a big deal (is this section required in addition to the QA
> one?).
> 

For the advisory in SVN, only the SRPMs are needed. For QA only the RPMs with their arches, like you listed for them. So no, the Provided packages section is not needed ;-)

Successfully tested on:

- MSI Modern 15 B7M, iGPU AMD Barcelo vendor: Micro-Star MSI driver: amdgpu v: kernel arch: GCN-5 code: Vega
- Desktop computer with Ryzen 9 5900X, GPU AMD Navi 23 [Radeon RX 6600/6600 XT/6600M] vendor: Sapphire driver: amdgpu v: kernel arch: RDNA-2 code: Navi-2x

@Dave: what other tests are still required?

MGA9-64 Plasma Wayland on HP Pavillion
No  installation issues.

As this is an all Intel Celeron laptop, all I can say is  that this update does not harm the current working.
Ref to Comment 26:
# clinfo
Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.1 AMD-APP (3590.0)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_event_callback 
  Platform Extensions function suffix             AMD
  Platform Host timer resolution                  1ns

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 0

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  AMD Accelerated Parallel Processing
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   No devices found in platform [AMD Accelerated Parallel Processing?]
  clCreateContext(NULL, ...) [default]            No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  No devices found in platform

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loaderns
  ICD loader Vendor                               OCL Icd free softwarens
  ICD loader Version                              2.3.1ns
  ICD loader Profile                              OpenCL 3.0ns
# rocminfo 
ROCk module is NOT loaded, possibly no GPU devices
That I could expect in view of number of devices above
Comment 26 provides a good test AFAICS, so OK for me.

Whiteboard: (none) => MGA9-64-OK
CC: (none) => herman.viaene

Like katnatek, my AMD GPUs are all too old, and about 99% (or more) of this is over my head, anyway. I'm inclined to agree with Herman that the test of comment 26 looks to be enough, but I just don't know.

Seeking Dave Hodgins' advice here before validating, if he's willing.

CC: (none) => andrewsfarm

Used userdrake to add my self to the groups after installing. Note that the
video group is a system group, so requires unselecting the option to filter
system users/groups.

On my desktop, installs cleanly. Added myself to the groups, logged out/in.
# lspcidrake -v|grep Card
Card:ATI Radeon HD 5000 to HD 6300 (radeon): Advanced Micro Devices, Inc. [AMD/ATI]|Cedar [Radeon HD 5000/6000/7350/8350 Series] [DISPLAY_VGA] (vendor:1002 device:68f9 subv:1043 subd:03ca)
From clinfo ...
  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 0

On my laptop, installs cleanly.
$ lspcidrake -v|grep Card
Card:ATI Volcanic Islands and later (amdgpu): Advanced Micro Devices, Inc. [AMD/ATI]|Renoir [DISPLAY_VGA] (vendor:1002 device:1636 subv:1043 subd:1e21) (rev: c6)
Card:NVIDIA GeForce 745 series and later: NVIDIA Corporation|TU106M [GeForce RTX 2060 Mobile] [DISPLAY_VGA] (vendor:10de device:1f15 subv:1043 subd:1e21) (rev: a1)

$ clinfo|grep ^Num
Number of platforms                               2
Number of devices                                 1
Number of devices                                 1

$ darktable-cltest|grep -e DEVICE: -e 'PLATFORM NAME'
   DEVICE:                   0: 'NVIDIA GeForce RTX 2060'
   PLATFORM NAME & VENDOR:   NVIDIA CUDA, NVIDIA Corporation
   DEFAULT DEVICE:           NO
   DEVICE:                   1: 'gfx90c:xnack-'
   PLATFORM NAME & VENDOR:   AMD Accelerated Parallel Processing, Advanced Micro Devices, Inc.
   DEFAULT DEVICE:           NO

Will use the system for a day or so to see if it generates any problems.

No regression's noticed. Validating the update.

Keywords: (none) => validated_update

An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGAA-2024-0001.html

Status: NEW => RESOLVED
Resolution: (none) => FIXED