История изменений
Исправление Kroz, (текущая версия) :
Так?
P. S. Кстати, ffmpeg, не смотря на то, что он показывает наличие OpenCL, так и не смог использовать ни одного фильтра с OpenCL. Но это будет тема моего следующего топика :)
$ ffmpeg -hwaccels
...
Hardware acceleration methods:
cuda
opencl
vulkan
$ ls -1 /etc/OpenCL/vendors/*.icd
/etc/OpenCL/vendors/amdocl64.icd
/etc/OpenCL/vendors/nvidia.icd
$ cat opencl_test.py
import pyopencl
# Show devices
platforms = pyopencl.get_platforms()
for platform in platforms:
for device in platform.get_devices():
print(f"Platform: {platform.name}, Device: {device.name}")
$ python3 opencl_test.py
Platform: AMD Accelerated Parallel Processing, Device: gfx90c:xnack-
Platform: NVIDIA CUDA, Device: NVIDIA GeForce RTX 3050 Ti Laptop GPU
$ clinfo
Number of platforms 2
Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP.dbg (3614.0)
Platform Profile FULL_PROFILE
...
Platform Name NVIDIA CUDA
Platform Vendor NVIDIA Corporation
Platform Version OpenCL 3.0 CUDA 12.4.131
Platform Profile FULL_PROFILE
...
Platform Name AMD Accelerated Parallel Processing
Number of devices 1
Device Name gfx90c:xnack-
Device Version OpenCL 2.0
Driver Version 3614.0 (HSA1.1,LC)
Device Type GPU
Device Board Name (AMD) AMD Radeon Graphics
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 7
Max clock frequency 1800MHz
...
Platform Name NVIDIA CUDA
Number of devices 1
Device Name NVIDIA GeForce RTX 3050 Ti Laptop GPU
Device Version OpenCL 3.0 CUDA
Driver Version 550.135
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 20
Max clock frequency 1485MHz
Compute Capability (NV) 8.6
Global memory size 4084400128 (3.804GiB)
Max memory allocation 1021100032 (973.8MiB)
Исходная версия Kroz, :
Так?
P. S. Кстати, ffmpeg, не смотря на то, что он показывает наличие OpenCL, так и не смог использовать ни одного фильтра с OpenCL…
$ ffmpeg -hwaccels
...
Hardware acceleration methods:
cuda
opencl
vulkan
$ ls -1 /etc/OpenCL/vendors/*.icd
/etc/OpenCL/vendors/amdocl64.icd
/etc/OpenCL/vendors/nvidia.icd
$ cat opencl_test.py
import pyopencl
# Show devices
platforms = pyopencl.get_platforms()
for platform in platforms:
for device in platform.get_devices():
print(f"Platform: {platform.name}, Device: {device.name}")
$ python3 opencl_test.py
Platform: AMD Accelerated Parallel Processing, Device: gfx90c:xnack-
Platform: NVIDIA CUDA, Device: NVIDIA GeForce RTX 3050 Ti Laptop GPU
$ clinfo
Number of platforms 2
Platform Name AMD Accelerated Parallel Processing
Platform Vendor Advanced Micro Devices, Inc.
Platform Version OpenCL 2.1 AMD-APP.dbg (3614.0)
Platform Profile FULL_PROFILE
...
Platform Name NVIDIA CUDA
Platform Vendor NVIDIA Corporation
Platform Version OpenCL 3.0 CUDA 12.4.131
Platform Profile FULL_PROFILE
...
Platform Name AMD Accelerated Parallel Processing
Number of devices 1
Device Name gfx90c:xnack-
Device Version OpenCL 2.0
Driver Version 3614.0 (HSA1.1,LC)
Device Type GPU
Device Board Name (AMD) AMD Radeon Graphics
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 7
Max clock frequency 1800MHz
...
Platform Name NVIDIA CUDA
Number of devices 1
Device Name NVIDIA GeForce RTX 3050 Ti Laptop GPU
Device Version OpenCL 3.0 CUDA
Driver Version 550.135
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 20
Max clock frequency 1485MHz
Compute Capability (NV) 8.6
Global memory size 4084400128 (3.804GiB)
Max memory allocation 1021100032 (973.8MiB)