AMD-Grafikkarten unter Linux: Treiber, OpenCL, BOINC, amdconfig und headless per ssh

MagicEye04 · 09.04.2023

koschi schrieb:
MagicEye04 schrieb:

Ich hatte bei Koschis link gelesen, dass es wohl nur alten Treibern bis 20.40 funktioniert.
Und wenn ich bei AMD nach dem Treiber für den 3200G suche, bekomme ich überhaupt keinen zum Download angeboten, nur Windows-Treiber.

Zum Vergrößern anklicken....

Ich nutze den alten Extrakt auch mit meinem 5700G unter Ubuntu 22.04, das passt.
Nur kann mein Skript keine neueren Treiber zerlegen und die passenden Sachen rauspopeln, da sich die Zusammensetzung und Benennung der Pakete darin geändert hat. Wenn ich mich recht erinnere ist AMD damals aufm ROCm gewechselt und ich hatte ohne neuere GPU (Navi) keine Not mich damit auseinander zu setzen...

Frage zu Deinem Script.
Da sind einige Zeilen drin, wo ein Befehl "ar" nicht gefunden wird. Soll das jeweils "tar" sein?

Code:

root@htpc-A320MH:/home/htpc/Downloads# ./create_amdgpu-ocl-pkg.sh amdgpu-pro-20.40-1147286-ubuntu-20.04.tar.xz
Processing driver archive:
 Extracting amdgpu-pro-20.40-1147286-ubuntu-20.04.tar.xz - done
  Extracting opencl-amdgpu-pro-icd_20.40-1147286_amd64.deb - ./create_amdgpu-ocl-pkg.sh: Zeile 53: ar: Befehl nicht gefunden

    Extracting data.tar.xz - tar (child): data.tar.xz: Funktion open fehlgeschlagen: Datei oder Verzeichnis nicht gefunden
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now

  Extracting opencl-orca-amdgpu-pro-icd_20.40-1147286_amd64.deb - ./create_amdgpu-ocl-pkg.sh: Zeile 60: ar: Befehl nicht gefunden

    Extracting data.tar.xz - tar (child): data.tar.xz: Funktion open fehlgeschlagen: Datei oder Verzeichnis nicht gefunden
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now

./create_amdgpu-ocl-pkg.sh: Zeile 65: cd: opt/amdgpu-pro/lib/x86_64-linux-gnu: Datei oder Verzeichnis nicht gefunden
sed: libamdocl-orca64.so kann nicht gelesen werden: Datei oder Verzeichnis nicht gefunden
  Extracting libdrm-amdgpu-amdgpu1_2.4.100-1147286_amd64.deb - ./create_amdgpu-ocl-pkg.sh: Zeile 75: ar: Befehl nicht gefunden

    Extracting data.tar.xz - tar (child): data.tar.xz: Funktion open fehlgeschlagen: Datei oder Verzeichnis nicht gefunden
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now

Copying files to package directory ./create_amdgpu-ocl-pkg.sh: Zeile 83: cd: opt/amdgpu/lib/x86_64-linux-gnu: Datei oder Verzeichnis nicht gefunden
rm: das Entfernen von 'libdrm_amdgpu.so.1' ist nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für 'libdrm_amdgpu.so.1.0.0' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/opencl/etc' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/opencl/opt/amdgpu-pro/lib/x86_64-linux-gnu/libamdocl64.so' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/opencl/opt/amdgpu-pro/lib/x86_64-linux-gnu/libamdocl-orca64.so' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/opencl/opt/amdgpu-pro/lib/x86_64-linux-gnu/libamdocl12cl64.so' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/libdrm/opt/amdgpu/lib/x86_64-linux-gnu/libdrm_amdgpo.so.1.0.0' nicht möglich: Datei oder Verzeichnis nicht gefunden
mv: Aufruf von stat für '/home/htpc/Downloads/libdrm/opt/amdgpu/lib/x86_64-linux-gnu/libdrm_amdgpo.so.1' nicht möglich: Datei oder Verzeichnis nicht gefunden
done

Building package (allow up to 2min):
dpkg-deb: Paket »amdgpu-ocl« wird in »amdgpu-ocl_20.40-1147286.deb« gebaut.

amdgpu-ocl_20.40-1147286.deb content:
drwxr-xr-x root/root         0 2023-04-09 18:16 ./
drwxr-xr-x root/root         0 2023-04-09 18:16 ./etc/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./opt/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./opt/amdgpu/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./opt/amdgpu/share/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./opt/amdgpu/share/libdrm/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./usr/
drwxr-xr-x root/root         0 2023-04-09 18:16 ./usr/lib/
lrwxrwxrwx root/root         0 2023-04-09 18:16 ./opt/amdgpu/share/libdrm/amdgpu.ids -> /usr/share/libdrm/amdgpu.ids

Use dpkg -i amdgpu-ocl_20.40-1147286.deb to install it.
root@htpc-A320MH:/home/htpc/Downloads#

Doppelposting wurde automatisch zusammengeführt: 09.04.2023

Ich habe gerade noch mal meine anderen Rechner durchsucht und auf Einem habe ich eine ca. 50MB große Datei amdgpu-ocl_20.40-1147286.deb gefunden. Das sieht mir doch stark nach dem .deb Paket aus, was man erzeugen müsste.
Also habe ich das direkt versucht zu installieren.
Geht aber auch nicht.
Vermutlich sind da noch Reste von den anderen beiden Versuchen vorhanden?

koschi · 09.04.2023

ar ist teil von binutils, das muss installiert sein, damit mein Skript den offiziellen Treiber entpacken kann.
Hau mal den libdrm Kram wech und dann nochmal dpkg -i amdgpu-ocl_20.40-1147286.deb, müsste dann hoffentlich gehen.

MagicEye04 · 09.04.2023

koschi schrieb:
ar ist teil von binutils, das muss installiert sein

OK, das fehlte wohl noch.
Libdrm stammte dann vermutlich von einem vorherigen Versuch.
Es war eine lange Geburt, aber nun klappt es endlich.

So 09 Apr 2023 21:54:37 CEST | | OpenCL: AMD/ATI GPU 0: AMD Radeon Vega 8 Graphics (driver version 3180.7 (PAL,HSAIL), device version OpenCL 2.0 AMD-APP (3180.7), 1741MB, 1741MB available, 1280 GFLOPS peak)

Ich habe die Schritte mal zusammengefasst, hoffentlich fehlt nicht irgendwas.

Boinc Installation Ubuntu22LTS

Da Boinc auf Ubuntu22 etwas störrisch ist, möchte ich hier mal alles sammeln, was man braucht. Wer ein stabiles boinc System haben will, sollte aber auf jeden Fall vorerst bei Ubuntu 20LTS bleiben. Los geht es mit der neuesten Version des Boinc-Managers 7.20.5, die Ubuntu leider nicht...

forum.planet3dnow.de

Meine Meinung, dass Nvidia bei Ubuntu sehr viel pflegeleichter ist, hat sich irgendwie nur gefestigt.

vnt · 10.04.2023

Hallo MagicEye04,

es freut mich sehr, dass Du OpenCL mit Deinem Ryzen 3 3200G unter Ubuntu 22.04 zum Laufen bekommen hast *greater*

Vielleicht kannst Du gelegentlich hier nochmal die Ausgabe des folgenden Befehles zeigen:

Code:

inxi -Gz

... es würde mich nämlich sehr interessieren, mit welchem Grafiktreiber Du das System jetzt betreibst.

Gruß,
vnt

MagicEye04 · 10.04.2023

Bitteschön:

htpc@htpc-A320MH:~$ inxi -Gz
Graphics:
Device-1: AMD Picasso/Raven 2 [Radeon Vega Series / Radeon Mobile Series]
driver: amdgpu v: kernel
Display: wayland server: X.Org v: 1.22.1.1 with: Xwayland v: 22.1.1
compositor: gnome-shell v: 42.5 driver: gpu: amdgpu
resolution: 3840x2160~60Hz
OpenGL: renderer: AMD Radeon Vega 8 Graphics (raven LLVM 15.0.6 DRM 3.47
5.19.0-38-generic)
v: 4.6 Mesa 22.2.5

Und vielen Dank für Deine Geduld.

Ich hatte übrigens gerade versucht, der iGPU im Bios statt nur 512MB ein ganzes GB zuzuweisen. Das hat Ubuntu mit einem Kernel Panic und "not enough memory" quittiert. Gut, dann muss ich das wohl so lassen. Ist wahrscheinlich sowieso egal, schneller wirds davon auch nicht. Einstein hatte aber mehr als 100% belegt und ich hatte Sorge, dass das irgendwie nicht optimal wäre.

Outlaw9876 · 11.04.2023

Moin!

Ich habe mich auch noch einmal mal an das Thema Linux Mint 21.1 + Boinc + AMD-GPU(s) herangewagt.

Mit der offiziellen Installationsanleitung für Ubuntu hat es nicht geklappt.

Installing BOINC on Ubuntu - BOINC

boinc.berkeley.edu

Auch die Alternative via PPA funktioniert nicht.

boinc development release : Gianfranco Costamagna

launchpad.net

Was allerdings problemlos funktioniert ist die Installation der Flatpak-Version (Flathub).

Install BOINC Manager on Linux | Flathub

Contribute computing power to projects doing research in many scientific areas

flathub.org

Diese lässt sich einfach über die Anwendungsverwaltung von Linux Mint installieren.

Die GPUs sollten auch richtig erkannt sein (3 x RX560 + 1 x RX550 auf X399 Taichi mit TR 1920X):

Number of platforms 1
Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP (3380.4)
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
Platform Extensions function suffix AMD
Platform Host timer resolution 1ns

Platform Name AMD Accelerated Parallel Processing
Number of devices 4
Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:44:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 4294967296 (4GiB)
Global free memory (AMD) 4126696 (3.936GiB) 0 (0B)
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 3650722201 (3.4GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 3285649920 (3.06GiB)
Preferred total size of global vars 4294967296 (4GiB)
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 228170137 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 3650722201 (3.4GiB)
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 3650722201 (3.4GiB)
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) 1681206129957909651ns (Tue Apr 11 11:42:09 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:43:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681206129957909651ns (Tue Apr 11 11:42:09 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:0a:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681206129957909651ns (Tue Apr 11 11:42:09 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name gfx803
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX550/550 Series
Device PCI-e ID (AMD) 0x699f
Device Topology (AMD) PCI-E, 0000:09:00.0
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) 16
SIMD instruction width (AMD) 1
Max clock frequency 1071MHz
Graphics IP (AMD) 8.0
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) 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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681206129957909651ns (Tue Apr 11 11:42:09 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platform
clCreateContext(NULL, ...) [default] No platform
clCreateContext(NULL, ...) [other] Success [AMD]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (4)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
Device Name Baffin
Device Name Baffin
Device Name gfx803
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 (4)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
Device Name Baffin
Device Name Baffin
Device Name gfx803

Graphics:
Device-1: AMD Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X]
driver: amdgpu v: kernel
Device-2: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: kernel
Device-3: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: kernel
Device-4: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: kernel
Display: x11 server: X.Org v: 1.21.1.3 driver: X: loaded: amdgpu,ati
unloaded: fbdev,modesetting,vesa gpu: amdgpu resolution: 1: 1920x1080~60Hz
2: 1920x1080~60Hz
OpenGL: renderer: Radeon RX550/550 Series (polaris12 LLVM 15.0.6 DRM 3.47
5.19.0-38-generic)
v: 4.6 Mesa 22.2.5

Allerdings sagt Boinc mir dies:

Bei Einstein und Milkyway werden die WUs geschrottet.

Wo ist der Fehler?

koschi · 11.04.2023

Bei dir stellt Mesa das OpenCL zur Verfügung, nicht amdgpu, daher wird nur device version OpenCL 1.1 angezeigt. Milkyway läuft damit nicht, neuere Einstein auch nicht mehr.

Damit alle Karten auch genutzt werden, musst du in der cc_config.xml noch folgendes setzen:

<use_all_gpus>1</use_all_gpus>

Outlaw9876 · 11.04.2023

Bei der Installation über Flatpak gehört die cc_config.xml übrigens hier hin:

/home/(Benutzername)/.var/app/edu.berkeley.BOINC

Die Ordner nach dem Benutzernamen sind standardmäßig verborgen.

Jetzt sind zwar alle 4 GPUs aktiv, aber noch immer habe ich keinen funktionierenden OpenCL 2.0-Treiber.

Leider habe ich durch mehrere wilde Installationsversuche nun so etwas:

outlaw9876@TR-Mint21:~$ sudo apt-get purge amdgpu-install
[sudo] Passwort für outlaw9876:
Paketlisten werden gelesen… Fertig
Abhängigkeitsbaum wird aufgebaut… Fertig
Statusinformationen werden eingelesen… Fertig
Die folgenden Pakete werden ENTFERNT:
amdgpu-install*
0 aktualisiert, 0 neu installiert, 1 zu entfernen und 0 nicht aktualisiert.
4 nicht vollständig installiert oder entfernt.
Nach dieser Operation werden 0 B Plattenplatz zusätzlich benutzt.
Möchten Sie fortfahren? [J/n] j
amdgpu-dkms (1:5.6.14.224-1147286) wird eingerichtet ...
Removing old amdgpu-5.6.14.224-1147286 DKMS files...
Deleting module amdgpu-5.6.14.224-1147286 completely from the DKMS tree.
Loading new amdgpu-5.6.14.224-1147286 DKMS files...
Building for 5.19.0-38-generic
Building for architecture amd64
Building initial module for 5.19.0-38-generic
Error! Bad return status for module build on kernel: 5.19.0-38-generic (amd64)
Consult /var/lib/dkms/amdgpu/5.6.14.224-1147286/build/make.log for more information.
dpkg: Fehler beim Bearbeiten des Paketes amdgpu-dkms (--configure):
»installiertes amdgpu-dkms-Skript des Paketes post-installation«-Unterprozess gab den Fehlerwert 10 zurück
dpkg: Abhängigkeitsprobleme verhindern Konfiguration von amdgpu:
amdgpu hängt ab von amdgpu-dkms (= 1:5.6.14.224-1147286); aber:
Paket amdgpu-dkms ist noch nicht konfiguriert.

dpkg: Fehler beim Bearbeiten des Paketes amdgpu (--configure):
Abhängigkeitsprobleme - verbleibt unkonfiguriert
dpkg: Abhängigkeitsprobleme verhindern Konfiguration von amdgpu-pro:
amdgpu-pro hängt ab von amdgpu (= 20.40-1147286); aber:
Paket amdgpu ist noch nicht konfiguriert.

dpkg: Fehler beim Bearbeiten des Paketes amdgpu-pro (--configure):
Abhängigkeitsprobleme - verbleibt unkonfiguriert
dpkg: Abhängigkeitsprobleme verhindern Konfiguration von amdgpu-pro-lib32:
amdgpu-pro-lib32 hängt ab von amdgpu (= 20.40-1147286) | amdgpu-hwe (= 20.40-1147286); aber:
Paket amdgpu ist noch nicht konfiguriert.
Paket amdgpu-hwe ist nicht installiert.
amdgpu-pro-lib32 hängt ab von amdgpu-pro (= 20.40-1147286) | amdgpu-pro-hwe (= 20.40-1147286); aber:
Paket amdgpu-pro ist noch nicht konfiguriert.
Paket amdgpu-pro-hwe ist nicht installiert.

dpkg: Fehler beim Bearbeiten des Paketes amdgpu-pro-lib32 (--configure):
Abhängigkeitsprobleme - verbleibt unkonfiguriert
Fehler traten auf beim Bearbeiten von:
amdgpu-dkms
amdgpu
amdgpu-pro
amdgpu-pro-lib32
E: Sub-process /usr/bin/dpkg returned an error code (1)

Bekomme ich das irgendwie wieder geradegebogen, oder hilft nur ne Neuinstallation?

koschi · 11.04.2023

Schmeiß mal alles weg was amdgpu ist, das ggf. hinzugefügte repo auch entfernen.

Anschließend sollte mein Extrakt reichen für die Karten:

http://kerbodyne.com/boinc/amdgpu-ocl_20.40-1147286.deb

Outlaw9876 · 11.04.2023

Hat leider nicht geklappt.

Irgendwie sieht es noch düstererer aus.
Es wird nur noch OpenCL 1.2 bei den GPUs angezeigt - vorhin war es immerhin 2.0

outlaw9876@TR-Mint21:~$ clinfo
Number of platforms 1
Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP (3180.7)
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
Platform Extensions function suffix AMD
Platform Host timer resolution 1ns

Platform Name AMD Accelerated Parallel Processing
Number of devices 4
Device Name gfx804
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 1.2 AMD-APP (3180.7)
Driver Version 3180.7
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 550 / 550 Series
Device PCI-e ID (AMD) 0
Device Topology (AMD) PCI-E, 0000:09:00.0
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) 16
SIMD instruction width (AMD) 1
Max clock frequency 1071MHz
Graphics IP (AMD) 8.0
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) 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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 1987592192 (1.851GiB)
Global free memory (AMD) 1919872 (1.831GiB) 1549776 (1.478GiB)
Global memory channels (AMD) 2
Global memory banks per channel (AMD) 16
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 1689453363 (1.573GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 2048 bits (256 bytes)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 8
Local memory type Local
Local memory size 32768 (32KiB)
Local memory size per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 1689453363 (1.573GiB)
Preferred constant buffer size (AMD) 16384 (16KiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Prefer user sync for interop Yes
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681239292193171242ns (Tue Apr 11 20:54:52 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 2
Max real-time compute queues (AMD) 0
Max real-time compute units (AMD) 0
SPIR versions 1.2
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 1.2 AMD-APP (3180.7)
Driver Version 3180.7
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0xaed320d0
Device Topology (AMD) PCI-E, 0000:0a:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2133032960 (1.987GiB)
Global free memory (AMD) 2063448 (1.968GiB) 1562736 (1.49GiB)
Global memory channels (AMD) 4
Global memory banks per channel (AMD) 16
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 1813078016 (1.689GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 2048 bits (256 bytes)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 8
Local memory type Local
Local memory size 32768 (32KiB)
Local memory size per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 1813078016 (1.689GiB)
Preferred constant buffer size (AMD) 16384 (16KiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Prefer user sync for interop Yes
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681239292193171242ns (Tue Apr 11 20:54:52 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 2
Max real-time compute queues (AMD) 0
Max real-time compute units (AMD) 2933073360
SPIR versions 1.2
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 1.2 AMD-APP (3180.7)
Driver Version 3180.7
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0
Device Topology (AMD) PCI-E, 0000:43:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2133032960 (1.987GiB)
Global free memory (AMD) 2063448 (1.968GiB) 1562736 (1.49GiB)
Global memory channels (AMD) 4
Global memory banks per channel (AMD) 16
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 1813078016 (1.689GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 2048 bits (256 bytes)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 8
Local memory type Local
Local memory size 32768 (32KiB)
Local memory size per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 1813078016 (1.689GiB)
Preferred constant buffer size (AMD) 16384 (16KiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Prefer user sync for interop Yes
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681239292193171242ns (Tue Apr 11 20:54:52 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 2
Max real-time compute queues (AMD) 0
Max real-time compute units (AMD) 0
SPIR versions 1.2
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 1.2 AMD-APP (3180.7)
Driver Version 3180.7
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0
Device Topology (AMD) PCI-E, 0000:44:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 4250259456 (3.958GiB)
Global free memory (AMD) 4131052 (3.94GiB) 3129120 (2.984GiB)
Global memory channels (AMD) 4
Global memory banks per channel (AMD) 16
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 3612720537 (3.365GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 2048 bits (256 bytes)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 8
Local memory type Local
Local memory size 32768 (32KiB)
Local memory size per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 3612720537 (3.365GiB)
Preferred constant buffer size (AMD) 16384 (16KiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Prefer user sync for interop Yes
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681239292193171242ns (Tue Apr 11 20:54:52 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 2
Max real-time compute queues (AMD) 0
Max real-time compute units (AMD) 0
SPIR versions 1.2
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event

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 gfx804
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (4)
Platform Name AMD Accelerated Parallel Processing
Device Name gfx804
Device Name Baffin
Device Name Baffin
Device Name Baffin
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 (4)
Platform Name AMD Accelerated Parallel Processing
Device Name gfx804
Device Name Baffin
Device Name Baffin
Device Name Baffin

ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.14
ICD loader Profile OpenCL 3.0

Die Ausgabe von inxi -Gz ist komplett identisch wie vorhin.

Laut der Paketverwaltung sind das nun die einzigen Pakete mit Bezug zu amdgpu:

Outlaw9876 · 12.04.2023

Nachdem ich mir Mint komplett zerschossen habe, folgt ein neuer Versuch mit Manjaro.

Über AUR habe ich dies installiert:

[outlaw9876@TR-Manjaro ~]$ clinfo
Number of platforms 1
Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP (3380.4)
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
Platform Extensions function suffix AMD
Platform Host timer resolution 1ns

Platform Name AMD Accelerated Parallel Processing
Number of devices 4
Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:44:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 4294967296 (4GiB)
Global free memory (AMD) 4126696 (3.936GiB) 0 (0B)
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 3650722201 (3.4GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 3285649920 (3.06GiB)
Preferred total size of global vars 4294967296 (4GiB)
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 228170137 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 3650722201 (3.4GiB)
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 3650722201 (3.4GiB)
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) 1681300763155187035ns (Wed Apr 12 13:59:23 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:43:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681300763155187035ns (Wed Apr 12 13:59:23 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name Baffin
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 560 Series
Device PCI-e ID (AMD) 0x67ef
Device Topology (AMD) PCI-E, 0000:0a:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 14
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1226MHz
Graphics IP (AMD) 8.0
Device Partition (core)
Max number of sub-devices 14
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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681300763155187035ns (Wed Apr 12 13:59:23 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name gfx803
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3380.4)
Driver Version 3380.4 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) AMD Radeon RX 550 / 550 Series
Device PCI-e ID (AMD) 0x699f
Device Topology (AMD) PCI-E, 0000:09:00.0
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) 16
SIMD instruction width (AMD) 1
Max clock frequency 1071MHz
Graphics IP (AMD) 8.0
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) 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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 2147483648 (2GiB)
Global free memory (AMD) 2029544 (1.936GiB) 0 (0B)
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 1825361100 (1.7GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 1642824960 (1.53GiB)
Preferred total size of global vars 2147483648 (2GiB)
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 114085068 pixels
Max 1D or 2D image array size 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 1825361100 (1.7GiB)
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 1825361100 (1.7GiB)
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) 1681300763155187035ns (Wed Apr 12 13:59:23 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platform
clCreateContext(NULL, ...) [default] No platform
clCreateContext(NULL, ...) [other] Success [AMD]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (4)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
Device Name Baffin
Device Name Baffin
Device Name gfx803
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 (4)
Platform Name AMD Accelerated Parallel Processing
Device Name Baffin
Device Name Baffin
Device Name Baffin
Device Name gfx803

[outlaw9876@TR-Manjaro ~]$ inxi -Gz
Graphics:
Device-1: AMD Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X]
driver: amdgpu v: kernel
Device-2: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X] driver: amdgpu v: kernel
Device-3: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X] driver: amdgpu v: kernel
Device-4: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X] driver: amdgpu v: kernel
Display: x11 server: X.Org v: 21.1.8 driver: X: loaded: amdgpu
unloaded: modesetting dri: radeonsi gpu: amdgpu resolution:
1: 1920x1080~60Hz 2: 1920x1080~60Hz
API: OpenGL v: 4.6 Mesa 23.0.2 renderer: AMD Radeon RX 550 / 550 Series
(polaris12 LLVM 15.0.7 DRM 3.49 6.2.10-1-MANJARO)

Boinc ist aber immernoch nur mit OpenCL 1.1 via Mesa unterwegs.

koschi · 12.04.2023

Unter Ubuntu brauchte es das Paket boinc-client-opencl damit BOINC die OpenCL devices auch sehen konnte, gibt es sowas auch von Manjaro?

ocl-icd-libopencl1 sorgt dafür dass zB. clinfo die amdgpu OpenCL devices sehen kann, während bei mir mesa-opencl-icd nicht drauf ist, ich daher auch keine Mesa OpenCL devices sehen kann.

Wenn beide icd loader installiert sind sehe ich jede Karte doppelt, die Mesa (Plattform name: Clover) VEGA10 mit OpenCL1.1 und die amdgpu (Plattform name: AMD Accelerated Parallel Processing) VEGA10 mit OpenCL 2.0:

root@frickelbude:~# clinfo
Number of platforms 2
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 21.2.6
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA

Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP (3180.7)
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
Platform Host timer resolution 1ns
Platform Extensions function suffix AMD

Platform Name Clover
Number of devices 2
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
Device Vendor AMD
Device Vendor ID 0x1002
Device Version OpenCL 1.1 Mesa 21.2.6
Driver Version 21.2.6
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Max compute units 56
Max clock frequency 1590MHz
Max work item dimensions 3
Max work item sizes 256x256x256
Max work group size 256
=== CL_PROGRAM_BUILD_LOG ===
fatal error: cannot open file '/usr/local//usr/lib/clc/gfx900-amdgcn-mesa-mesa3d.bc': No such file or directory
Preferred work group size multiple <getWGsizes:1200: create kernel : error -46>
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
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 8589934592 (8GiB)
Error Correction support No
Max memory allocation 6871947673 (6.4GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 32768 bits (4096 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 32768 (32KiB)
Max number of constant args 16
Max constant buffer size 67108864 (64MiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Extensions cl_khr_byte_addressable_store 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_fp64 cl_khr_extended_versioning

Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
Device Vendor AMD
Device Vendor ID 0x1002
Device Version OpenCL 1.1 Mesa 21.2.6
Driver Version 21.2.6
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Max compute units 56
Max clock frequency 1590MHz
Max work item dimensions 3
Max work item sizes 256x256x256
Max work group size 256
=== CL_PROGRAM_BUILD_LOG ===
fatal error: cannot open file '/usr/local//usr/lib/clc/gfx900-amdgcn-mesa-mesa3d.bc': No such file or directory
Preferred work group size multiple <getWGsizes:1200: create kernel : error -46>
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
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 8589934592 (8GiB)
Error Correction support No
Max memory allocation 6871947673 (6.4GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 32768 bits (4096 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 32768 (32KiB)
Max number of constant args 16
Max constant buffer size 67108864 (64MiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Extensions cl_khr_byte_addressable_store 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_fp64 cl_khr_extended_versioning

Platform Name AMD Accelerated Parallel Processing
Number of devices 2
Device Name gfx900
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3180.7)
Driver Version 3180.7 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX Vega
Device Topology (AMD) PCI-E, 0f:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 56
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1590MHz
Graphics IP (AMD) 9.0
Device Partition (core)
Max number of sub-devices 56
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 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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 8573157376 (7.984GiB)
Global free memory (AMD) 8306688 (7.922GiB)
Global memory channels (AMD) 64
Global memory banks per channel (AMD) 4
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 7059013632 (6.574GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 6353112064 (5.917GiB)
Preferred total size of global vars 8573157376 (7.984GiB)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 2764046336 (2.574GiB)
Local memory type Local
Local memory size 65536 (64KiB)
Local memory syze per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 7059013632 (6.574GiB)
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
P2P devices (AMD) <printDeviceInfo:147: get number of CL_DEVICE_P2P_DEVICES_AMD : error -30>
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681275147279099984ns (Wed Apr 12 06:52:27 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

Device Name gfx900
Device Vendor Advanced Micro Devices, Inc.
Device Vendor ID 0x1002
Device Version OpenCL 2.0 AMD-APP (3180.7)
Driver Version 3180.7 (PAL,HSAIL)
Device OpenCL C Version OpenCL C 2.0
Device Type GPU
Device Board Name (AMD) Radeon RX Vega
Device Topology (AMD) PCI-E, 0c:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 56
SIMD per compute unit (AMD) 4
SIMD width (AMD) 16
SIMD instruction width (AMD) 1
Max clock frequency 1590MHz
Graphics IP (AMD) 9.0
Device Partition (core)
Max number of sub-devices 56
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 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 No
Infinity and NANs No
Round to nearest No
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals No
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 8573157376 (7.984GiB)
Global free memory (AMD) 8306688 (7.922GiB)
Global memory channels (AMD) 64
Global memory banks per channel (AMD) 4
Global memory bank width (AMD) 256 bytes
Error Correction support No
Max memory allocation 7059013632 (6.574GiB)
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 2048 bits (256 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 6353112064 (5.917GiB)
Preferred total size of global vars 8573157376 (7.984GiB)
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 2048 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 2048x2048x2048 pixels
Max number of read image args 128
Max number of write image args 64
Max number of read/write image args 64
Max number of pipe args 16
Max active pipe reservations 16
Max pipe packet size 2764046336 (2.574GiB)
Local memory type Local
Local memory size 65536 (64KiB)
Local memory syze per CU (AMD) 65536 (64KiB)
Local memory banks (AMD) 32
Max number of constant args 8
Max constant buffer size 7059013632 (6.574GiB)
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
P2P devices (AMD) <printDeviceInfo:147: get number of CL_DEVICE_P2P_DEVICES_AMD : error -30>
Profiling timer resolution 1ns
Profiling timer offset since Epoch (AMD) 1681275147279099984ns (Wed Apr 12 06:52:27 2023)
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Thread trace supported (AMD) Yes
Number of async queues (AMD) 4
Max real-time compute queues (AMD) 1
Max real-time compute units (AMD) 0
printf() buffer size 4194304 (4MiB)
Built-in kernels (n/a)
Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA]
clCreateContext(NULL, ...) [default] Success [MESA]
clCreateContext(NULL, ...) [other] Success [AMD]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name Clover
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (2)
Platform Name Clover
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
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 (2)
Platform Name Clover
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)
Device Name Radeon RX Vega (VEGA10, DRM 3.35.0, 5.4.0-146-generic, LLVM 12.0.0)

ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.11
ICD loader Profile OpenCL 2.1

Outlaw9876 · 12.04.2023

Ich habe die Befürchtung, dass es an Flatpak liegt.

Es ist ja schon komisch, dass Boinc unter Mint & Manjaro Mesa 22.3.5 anzeigt, obwohl inxi -Gz bei Mint Mesa 22.2.5 und bei Manjaro Mesa 23.0.2 anzeigt.

Doppelposting wurde automatisch zusammengeführt: 12.04.2023

Jau, lag an Flatpak.

Die Boinc-Version aus dem Manjaro-Softwaretool hat zwar das Problem, dass man den Client extra starten muss, aber OpenCL 2.0 funktioniert.

Einstein läuft auf allen 4 GPUs. *great*

Outlaw9876 · 12.04.2023

Linux macht mich kaputt ......

Nun habe ich es geschafft, dass der Boinc-Client direkt startet.

Nun findet Boinc aber gar keine GPUs mehr. *noahnung*

Seltsamerweise scheint das Boinc-Verzeichnis auch zu wandern.

Bei der Flatpak-Version kann ich das ja noch verstehen - da das ja ne Sandbox ist.

Als es vorhin gefunzt hat (bis auf den sich nicht vebindenden Boinc-Client), lagen die Boinc-Daten direkt im Verzeichnis /bin.

Nun habe ich die Client-Geschichte nach dieser Anleitung behoben:

[GUIDE / FIX] Unable to connect to core client in...

After dealing with the craziness I thought it would be a good to have it recorded for posterity. For people that don't consider GUI-less vim to be a user-friendly text-editor. 1. Check the installed packages. You should have boinc from Manjaro's pamac (GUI for XFCE) or pacman (commandline), do...

www.overclock.net

Nun liegen die Boinc-Daten im Verzeichnis /var/lib/boinc

Die alte client_state.xml (wo die GPUs erkannt wurden) habe ich mal kopiert und die Datei in /var/lib/boinc überschrieben.

Boinc und auch den opencl-amd-Treiber habe ich deinstalliert, neu gebootet und beides erneut installiert.

Allerdings findet Boinc weiterhin keine GPU.

Im alten Verzeichnis steht dies in der coproc_info.xml

<coprocs>
<ati_opencl>
<name>AMD Radeon RX 560 Series</name>
<vendor>Advanced Micro Devices, Inc.</vendor>
<vendor_id>4098</vendor_id>
<available>1</available>
<half_fp_config>0</half_fp_config>
<single_fp_config>190</single_fp_config>
<double_fp_config>63</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p </extensions>
<global_mem_size>4294967296</global_mem_size>
<local_mem_size>65536</local_mem_size>
<max_clock_frequency>1226</max_clock_frequency>
<max_compute_units>14</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
<amd_simd_width>16</amd_simd_width>
<amd_simd_instruction_width>1</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 2.1 AMD-APP (3380.4)</opencl_platform_version>
<opencl_device_version>OpenCL 2.0 AMD-APP (3380.4)</opencl_device_version>
<opencl_driver_version>3380.4 (PAL,HSAIL)</opencl_driver_version>
<device_num>0</device_num>
<peak_flops>2196992000000.000000</peak_flops>
<opencl_available_ram>4294967296.000000</opencl_available_ram>
<opencl_device_index>0</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</ati_opencl>
<ati_opencl>
<name>AMD Radeon RX 560 Series</name>
<vendor>Advanced Micro Devices, Inc.</vendor>
<vendor_id>4098</vendor_id>
<available>1</available>
<half_fp_config>0</half_fp_config>
<single_fp_config>190</single_fp_config>
<double_fp_config>63</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p </extensions>
<global_mem_size>2147483648</global_mem_size>
<local_mem_size>65536</local_mem_size>
<max_clock_frequency>1226</max_clock_frequency>
<max_compute_units>14</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
<amd_simd_width>16</amd_simd_width>
<amd_simd_instruction_width>1</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 2.1 AMD-APP (3380.4)</opencl_platform_version>
<opencl_device_version>OpenCL 2.0 AMD-APP (3380.4)</opencl_device_version>
<opencl_driver_version>3380.4 (PAL,HSAIL)</opencl_driver_version>
<device_num>1</device_num>
<peak_flops>2196992000000.000000</peak_flops>
<opencl_available_ram>2147483648.000000</opencl_available_ram>
<opencl_device_index>1</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</ati_opencl>
<ati_opencl>
<name>AMD Radeon RX 560 Series</name>
<vendor>Advanced Micro Devices, Inc.</vendor>
<vendor_id>4098</vendor_id>
<available>1</available>
<half_fp_config>0</half_fp_config>
<single_fp_config>190</single_fp_config>
<double_fp_config>63</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p </extensions>
<global_mem_size>2147483648</global_mem_size>
<local_mem_size>65536</local_mem_size>
<max_clock_frequency>1226</max_clock_frequency>
<max_compute_units>14</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
<amd_simd_width>16</amd_simd_width>
<amd_simd_instruction_width>1</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 2.1 AMD-APP (3380.4)</opencl_platform_version>
<opencl_device_version>OpenCL 2.0 AMD-APP (3380.4)</opencl_device_version>
<opencl_driver_version>3380.4 (PAL,HSAIL)</opencl_driver_version>
<device_num>2</device_num>
<peak_flops>2196992000000.000000</peak_flops>
<opencl_available_ram>2147483648.000000</opencl_available_ram>
<opencl_device_index>2</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</ati_opencl>
<ati_opencl>
<name>AMD Radeon RX 550 / 550 Series</name>
<vendor>Advanced Micro Devices, Inc.</vendor>
<vendor_id>4098</vendor_id>
<available>1</available>
<half_fp_config>0</half_fp_config>
<single_fp_config>190</single_fp_config>
<double_fp_config>63</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_fp64 cl_amd_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_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p </extensions>
<global_mem_size>2147483648</global_mem_size>
<local_mem_size>65536</local_mem_size>
<max_clock_frequency>1071</max_clock_frequency>
<max_compute_units>8</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
<amd_simd_width>16</amd_simd_width>
<amd_simd_instruction_width>1</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 2.1 AMD-APP (3380.4)</opencl_platform_version>
<opencl_device_version>OpenCL 2.0 AMD-APP (3380.4)</opencl_device_version>
<opencl_driver_version>3380.4 (PAL,HSAIL)</opencl_driver_version>
<device_num>3</device_num>
<peak_flops>1096704000000.000000</peak_flops>
<opencl_available_ram>2147483648.000000</opencl_available_ram>
<opencl_device_index>3</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</ati_opencl>
<warning>NVIDIA: libcuda.so: cannot open shared object file: No such file or directory</warning>
<warning>ATI: libaticalrt.so: cannot open shared object file: No such file or directory</warning>
</coprocs>

Im aktuellen Boinc-Verzeichnis steht aber nur dies:

<coprocs>
<warning>NVIDIA: libcuda.so: cannot open shared object file: No such file or directory</warning>
<warning>ATI: libaticalrt.so: cannot open shared object file: No such file or directory</warning>
<warning>OpenCL library present but no OpenCL-capable devices found</warning>
</coprocs>

Diese Datei libaticalrt.so kann ich nirgendwo finden ....

Hat wer ne Idee?

koschi · 12.04.2023

Ist boinc bei dir in den Gruppen video und render?

Outlaw9876 · 12.04.2023

Oh Ha .....

Keine Ahnung, aber ich habe Manjaro runtergehauen und jetzt Ubuntu installiert ....

Schauen wir mal mal wie weit ich hier komme ....

herby44 · 13.04.2023

Ich habe vorhin mal Mint 21 nicht Mint 21.1 auf einem B450er mit einem R7 5700G installiert, das System aktualisiert
neu gestartet danach Boinc installiert die app_config angepasst und den User zu der Gruppe boinc hinzugefügt dann
ab und wieder angemeldet danach den Workaround für Ubuntu 22.04 angewendet dann ab und wieder angemeldet
das OCL Paket von @koschi installiert Neustart und die IGP macht Milkyway, danach auf Mint 21.1 aktualisiert und
Neustart, die IGP ist auch mit OCL 2.0 angegeben ich hatte auch mal kurz eine RX 470 drauf gesteckt die lief mit OCL
1.2 hat aber auch ein paar Milkyway gerechnet und eine Einstein.

Outlaw9876 · 13.04.2023

Jo, Ubuntu läuft mit dem Treiberpaket von koschi.

Boinc zeigt auch nur OpenCL 1.2 an obwohl die GraKas 2.0 können, aber Einstein läuft. *greater*

Outlaw9876 · 13.04.2023

Ubuntu ist nun auch wieder Geschichte ... *chatt*

Boinc hat manchmal nach einem Neustart den Boinc-Client nicht gestartet.

Teamviewer ist leider auch manchmal abgeschmiert (Teamviewer mag Wayland wohl nicht).

Besonders toll war, das beim Zugriff via Teamviewer Ubuntu jedes Mal die Berechtigung auf die 2 Monitore erteilt werden musste. Macht Sinn ... *kopfkratz

__

Nun läuft wieder Mint 21.1.

Boinc habe ich gem. der Anleitung von MagicEye04 problemlos installieren können.

Boinc Installation Ubuntu22LTS

Da Boinc auf Ubuntu22 etwas störrisch ist, möchte ich hier mal alles sammeln, was man braucht. Wer ein stabiles boinc System haben will, sollte aber auf jeden Fall vorerst bei Ubuntu 20LTS bleiben. Los geht es mit der neuesten Version des Boinc-Managers 7.20.5, die Ubuntu leider nicht...

forum.planet3dnow.de

Die 4 GraKas habe ich mit einer Kombination aus den dort beschriebenen Anleitungen 1 + 2 und der darauf folgenden Installation des akuellen AMD-Treibers mit dem Workstation Use Case zum Laufen gebracht.

Installing or Uninstalling the AMDGPU stack — amdgpu graphics and compute stack unknown-build documentation

amdgpu-install --usecase=workstation -y --vulkan=pro --opencl=rocr,legacy

Die GraKas werden mit OpenCL 2.0 bei clinfo angezeigt.

Interessant ist auch der Treiber laut inxi -Gz:

Graphics:
Device-1: AMD Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X]
driver: amdgpu v: 5.18.13
Device-2: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: 5.18.13
Device-3: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: 5.18.13
Device-4: AMD Baffin [Radeon RX 460/560D / Pro
450/455/460/555/555X/560/560X]
driver: amdgpu v: 5.18.13
Display: x11 server: X.Org v: 1.21.1.3 driver: X: loaded: amdgpu,ati
unloaded: fbdev,modesetting,vesa gpu: amdgpu resolution: 1: 1920x1080~60Hz
2: 1920x1080~60Hz
OpenGL: renderer: Radeon RX550/550 Series
v: 4.6.0 Core Profile Context 19.30.230130

Nix mehr von Mesa zu sehen.

Bei Boinc läuft gerade Einstein.

----

Ich habe fertig! *elch2*

Eigentlich hätte ich hier noch mindestens 2 Rechner mit Mint 20.3 + AMD-GPUs, die auch noch auf Mint 21.1 aktualisiert werden müssten ....
Das spare ich mir aber wohl auf für ein anderes Mal ... oder Leben. *elch*

Mike · 05.08.2023

Nach dem letzten Solus update habe ich jetzt Kernel 6.3.12.244 mit Mesa 23.1.5 drauf.
Das System meldet OpenCL 3.0.
Boinc erkennt die GPU AMD RX 6600XT.
Muss man da auch die group video hinzu fügen und wenn ja wie ?

AMD-Grafikkarten unter Linux: Treiber, OpenCL, BOINC, amdconfig und headless per ssh

Grand Admiral Special

Anhänge

Grand Admiral Special

Grand Admiral Special

Fleet Captain Special

Grand Admiral Special

Admiral Special

Grand Admiral Special

Admiral Special

Grand Admiral Special

Admiral Special

Admiral Special

Grand Admiral Special

Admiral Special

Admiral Special

Grand Admiral Special

Admiral Special

Vice Admiral Special

Admiral Special

Admiral Special

Grand Admiral Special

Ähnliche Themen

Aktuelle Aktionen

Wichtige Links