Building clinfo on Termux Android
truboxl opened this issue · 14 comments
truboxl commented
Hi, this is just an appreciation post for clinfo and
a build instruction for building clinfo on Termux that runs on Android.
It should allow people to investigate OpenCL on 64bit Android devices.
You need to have /system/vendor/lib64/libOpenCL.so present on the device.
If the Android device doesn't have this, it most like doesn't have OpenCL and the build will fail.
Something like cannot find -lOpenCL
I am going to close the issue since its not an issue with clinfo.
Simply install Termux from Google Play or FDroid
Run these commands:
pkg install git make clang -y
cd ~
git clone https://github.com/Oblomov/clinfo
git clone https://github.com/KhronosGroup/OpenCL-Headers
cd ~/clinfo
export LD_LIBRARY_PATH="/system/vendor/lib64:/system/lib64"
export CFLAGS="-I$HOME/OpenCL-Headers"
export LDFLAGS="-L/system/vendor/lib64"
make
./clinfo
Example outputs:
Adreno 505
Number of platforms 1
Platform Name QUALCOMM Snapdragon(TM)
Platform Vendor QUALCOMM
Platform Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch:
Platform Profile FULL_PROFILE
Platform Extensions
Platform Name QUALCOMM Snapdragon(TM)
Number of devices 1
Device Name QUALCOMM Adreno(TM)
Device Vendor QUALCOMM
Device Vendor ID 0xbf4d3c4b
Device Version OpenCL 2.0 Adreno(TM) 505
Driver Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch: Compiler E031.36.02.00
Device OpenCL C Version OpenCL C 2.0 Adreno(TM) 505
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 1
Max clock frequency 1MHz
Device Partition (core)
Max number of sub-devices 1
Supported partition types None
Supported affinity domains (n/a)
Max work item dimensions 3
Max work item sizes 1024x1024x1024
Max work group size 1024
Preferred work group size multiple 1024
Preferred / native vector sizes
char 1 / 1
short 1 / 1
int 1 / 1
long 1 / 0
half 1 / 1 (cl_khr_fp16)
float 1 / 1
double 0 / 0 (n/a)
Half-precision Floating-point support (cl_khr_fp16)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity Yes
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 No
Round to infinity Yes
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Double-precision Floating-point support (n/a)
Address bits 64, Little-Endian
Global memory size 956649472 (912.3MiB)
Error Correction support No
Max memory allocation 239162368 (228.1MiB)
Unified memory for Host and Device Yes
Shared Virtual Memory (SVM) capabilities (core)
Coarse-grained buffer sharing Yes
Fine-grained buffer sharing No
Fine-grained system sharing No
Atomics No
Minimum alignment for any data type 128 bytes
Alignment of base address 1024 bits (128 bytes)
Page size (QCOM) 4096 bytes
External memory padding (QCOM) 0 bytes
Preferred alignment for atomics
SVM 128 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 65536 (64KiB)
Preferred total size of global vars 1048576 (1024KiB)
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 64 bytes
Pitch alignment for 2D image buffers 64 pixels
Max 2D image size 16384x16384 pixels
Max 3D image size 16384x16384x2048 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 4096
Max pipe packet size 1024
Local memory type Local
Local memory size 32768 (32KiB)
Max number of constant args 8
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 1024
Queue properties (on host)
Out-of-order execution Yes
Profiling Yes
Queue properties (on device)
Out-of-order execution Yes
Profiling Yes
Preferred size 655376 (640KiB)
Max size 655376 (640KiB)
Max queues on device 1
Max events on device 1024
Prefer user sync for interop No
Profiling timer resolution 1000ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
printf() buffer size 1048576 (1024KiB)
Built-in kernels (n/a)
Device Extensions cl_khr_3d_image_writes cl_img_egl_image cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_egl_event cl_khr_egl_image cl_khr_fp16 cl_khr_gl_sharing 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_image2d_from_buffer cl_khr_mipmap_image cl_khr_srgb_image_writes cl_khr_subgroups cl_qcom_create_buffer_from_image cl_qcom_ext_host_ptr cl_qcom_ion_host_ptr cl_qcom_perf_hint cl_qcom_read_image_2x2 cl_qcom_android_native_buffer_host_ptr cl_qcom_protected_context cl_qcom_priority_hint cl_qcom_compressed_yuv_image_read cl_qcom_compressed_image
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 [P0]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name QUALCOMM Snapdragon(TM)
Device Name QUALCOMM Adreno(TM)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name QUALCOMM Snapdragon(TM)
Device Name QUALCOMM Adreno(TM)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) Invalid device type for platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
Platform Name QUALCOMM Snapdragon(TM)
Device Name QUALCOMM Adreno(TM)
Mali-G71
Number of platforms 1
Platform Name ARM Platform
Platform Vendor ARM
Platform Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0###
Platform Profile FULL_PROFILE
Platform Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16
Platform Extensions function suffix ARM
Platform Name ARM Platform
Number of devices 1
Device Name Mali-G71
Device Vendor ARM
Device Vendor ID 0x60a00002
Device Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0###
Driver Version 2.0
Device OpenCL C Version OpenCL C 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0###
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 2
Available core IDs 0-1
Max clock frequency 5MHz
Device Partition (core)
Max number of sub-devices 0
Supported partition types None
Supported affinity domains (n/a)
Max work item dimensions 3
Max work item sizes 384x384x384
Max work group size 384
Preferred work group size multiple 4
Preferred / native vector sizes
char 16 / 4
short 8 / 2
int 4 / 1
long 2 / 1
half 8 / 2 (cl_khr_fp16)
float 4 / 1
double 0 / 0 (n/a)
Half-precision Floating-point support (cl_khr_fp16)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Single-precision Floating-point support (core)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Double-precision Floating-point support (n/a)
Address bits 64, Little-Endian
Global memory size 3815780352 (3.554GiB)
Error Correction support No
Max memory allocation 953945088 (909.8MiB)
Unified memory for Host and Device Yes
Shared Virtual Memory (SVM) capabilities (core)
Coarse-grained buffer sharing Yes
Fine-grained buffer sharing No
Fine-grained system sharing No
Atomics No
Minimum alignment for any data type 128 bytes
Alignment of base address 1024 bits (128 bytes)
Preferred alignment for atomics
SVM 0 bytes
Global 0 bytes
Local 0 bytes
Max size for global variable 65536 (64KiB)
Preferred total size of global vars 0
Global Memory cache type Read/Write
Global Memory cache size 262144 (256KiB)
Global Memory cache line size 64 bytes
Image support Yes
Max number of samplers per kernel 16
Max size for 1D images from buffer 65536 pixels
Max 1D or 2D image array size 2048 images
Base address alignment for 2D image buffers 32 bytes
Pitch alignment for 2D image buffers 64 pixels
Max 2D image size 32768x32768 pixels
Max 3D image size 32768x32768x32768 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 1
Max pipe packet size 1024
Local memory type Global
Local memory size 32768 (32KiB)
Max number of constant args 8
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 1024
Queue properties (on host)
Out-of-order execution Yes
Profiling Yes
Queue properties (on device)
Out-of-order execution Yes
Profiling Yes
Preferred size 2097152 (2MiB)
Max size 16777216 (16MiB)
Max queues on device 1
Max events on device 1024
Prefer user sync for interop No
Profiling timer resolution 1000ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
IL version SPIR-V_1_0
printf() buffer size 1048576 (1024KiB)
Built-in kernels (n/a)
Device Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) ARM Platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [ARM]
clCreateContext(NULL, ...) [default] Success [ARM]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name ARM Platform
Device Name Mali-G71
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name ARM Platform
Device Name Mali-G71
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
Platform Name ARM Platform
Device Name Mali-G71
Oblomov commented
Thank you very much for the information and for sharing the logs. This is actually the first time I've seen the output for some of those extensions, so I'm actually glad it didn't blow up in the first place! Would you mind if I add the instructions to the README?
truboxl commented
You are the author. You call the shots! 😄
Though this is natively building on an Android device. I haven't tried cross compiling yet.
Oblomov commented
I'm trying it now myself on my Android device, and apparently on my system the approach is not sufficient, since there seems to be the library, but additional linkage is necessary (exynos_ion_{alloc,open,close} symbols are missing). I'm guessing that in my case the OpenCL library has a built-in platform that requires the additional symbols in the first place. Interesting. I'm reopening the issue to remind me of this pending issue.
truboxl commented
Hmm, I see. For reference, I am using Nokia 5 (Adreno 505) and Samsung Galaxy A30s (Mali-G71). Both are on Android 9. I am not sure what differs here...
Oblomov commented
I'm testing on a Moto One Vision. Apparently the OpenCL library tries to link with libion_exynos.so, which is not present in the system (only libion.so is)
Oblomov commented
Scratch that, it is present, but I have manually link it, before linking libOpenCL.so. And to make things worse, apparently rpath doesn't work as expected.
truboxl commented
Maybe the order in LD_LIBRARY_PATH should be changed? I honestly dunno :(
Oblomov commented
Nah, your approach is sane. It's the stupid Android dynamic linker at fault here, it ignores rpath. I finally got it working. Now I'll try to make something more general, hopefully you can be my test subject 8-)
Oblomov commented
I've pushed some changes. Now cloning KhronosGroup/OpenCL-Headers in a directory parallel to the clinfo clone (like you propose) should allow building clinfo with a simple make OS=Android, without any need for compilation-specific variable overrides (at least on my machine!).
Running still requires the LD_LIBRARY_PATH because the Bionic runtime linker doesn't use DT_RPATH nor DT_RUNPATH, and the vendor egl library path may need to be included. This is all documented in the README.
I'd appreciate if you could test the latest master and verify. Feel free to reopen or open a new issue for further improvements.
Saikatsaha1996 commented
Hi, this is just an appreciation post for clinfo and
a build instruction for building clinfo on Termux that runs on Android.
It should allow people to investigate OpenCL on 64bit Android devices.You need to have
/system/vendor/lib64/libOpenCL.sopresent on the device.
If the Android device doesn't have this, it most like doesn't have OpenCL and the build will fail.
Something likecannot find -lOpenCLI am going to close the issue since its not an issue with clinfo.
Simply install Termux from Google Play or FDroid
Run these commands:pkg install git make clang -y cd ~ git clone https://github.com/Oblomov/clinfo git clone https://github.com/KhronosGroup/OpenCL-Headers cd ~/clinfo export LD_LIBRARY_PATH="/system/vendor/lib64:/system/lib64" export CFLAGS="-I$HOME/OpenCL-Headers" export LDFLAGS="-L/system/vendor/lib64" make ./clinfoExample outputs:
Adreno 505Number of platforms 1 Platform Name QUALCOMM Snapdragon(TM) Platform Vendor QUALCOMM Platform Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch: Platform Profile FULL_PROFILE Platform Extensions Platform Name QUALCOMM Snapdragon(TM) Number of devices 1 Device Name QUALCOMM Adreno(TM) Device Vendor QUALCOMM Device Vendor ID 0xbf4d3c4b Device Version OpenCL 2.0 Adreno(TM) 505 Driver Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch: Compiler E031.36.02.00 Device OpenCL C Version OpenCL C 2.0 Adreno(TM) 505 Device Type GPU Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 1 Max clock frequency 1MHz Device Partition (core) Max number of sub-devices 1 Supported partition types None Supported affinity domains (n/a) Max work item dimensions 3 Max work item sizes 1024x1024x1024 Max work group size 1024 Preferred work group size multiple 1024 Preferred / native vector sizes char 1 / 1 short 1 / 1 int 1 / 1 long 1 / 0 half 1 / 1 (cl_khr_fp16) float 1 / 1 double 0 / 0 (n/a) Half-precision Floating-point support (cl_khr_fp16) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero No Round to infinity Yes 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 No Round to infinity Yes IEEE754-2008 fused multiply-add No Support is emulated in software No Correctly-rounded divide and sqrt operations No Double-precision Floating-point support (n/a) Address bits 64, Little-Endian Global memory size 956649472 (912.3MiB) Error Correction support No Max memory allocation 239162368 (228.1MiB) Unified memory for Host and Device Yes Shared Virtual Memory (SVM) capabilities (core) Coarse-grained buffer sharing Yes Fine-grained buffer sharing No Fine-grained system sharing No Atomics No Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Page size (QCOM) 4096 bytes External memory padding (QCOM) 0 bytes Preferred alignment for atomics SVM 128 bytes Global 0 bytes Local 0 bytes Max size for global variable 65536 (64KiB) Preferred total size of global vars 1048576 (1024KiB) 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 64 bytes Pitch alignment for 2D image buffers 64 pixels Max 2D image size 16384x16384 pixels Max 3D image size 16384x16384x2048 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 4096 Max pipe packet size 1024 Local memory type Local Local memory size 32768 (32KiB) Max number of constant args 8 Max constant buffer size 65536 (64KiB) Max size of kernel argument 1024 Queue properties (on host) Out-of-order execution Yes Profiling Yes Queue properties (on device) Out-of-order execution Yes Profiling Yes Preferred size 655376 (640KiB) Max size 655376 (640KiB) Max queues on device 1 Max events on device 1024 Prefer user sync for interop No Profiling timer resolution 1000ns Execution capabilities Run OpenCL kernels Yes Run native kernels No printf() buffer size 1048576 (1024KiB) Built-in kernels (n/a) Device Extensions cl_khr_3d_image_writes cl_img_egl_image cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_egl_event cl_khr_egl_image cl_khr_fp16 cl_khr_gl_sharing 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_image2d_from_buffer cl_khr_mipmap_image cl_khr_srgb_image_writes cl_khr_subgroups cl_qcom_create_buffer_from_image cl_qcom_ext_host_ptr cl_qcom_ion_host_ptr cl_qcom_perf_hint cl_qcom_read_image_2x2 cl_qcom_android_native_buffer_host_ptr cl_qcom_protected_context cl_qcom_priority_hint cl_qcom_compressed_yuv_image_read cl_qcom_compressed_image 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 [P0] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM) clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM) clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) Invalid device type for platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM)Mali-G71
Number of platforms 1 Platform Name ARM Platform Platform Vendor ARM Platform Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Platform Profile FULL_PROFILE Platform Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16 Platform Extensions function suffix ARM Platform Name ARM Platform Number of devices 1 Device Name Mali-G71 Device Vendor ARM Device Vendor ID 0x60a00002 Device Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Driver Version 2.0 Device OpenCL C Version OpenCL C 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Device Type GPU Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 2 Available core IDs 0-1 Max clock frequency 5MHz Device Partition (core) Max number of sub-devices 0 Supported partition types None Supported affinity domains (n/a) Max work item dimensions 3 Max work item sizes 384x384x384 Max work group size 384 Preferred work group size multiple 4 Preferred / native vector sizes char 16 / 4 short 8 / 2 int 4 / 1 long 2 / 1 half 8 / 2 (cl_khr_fp16) float 4 / 1 double 0 / 0 (n/a) Half-precision Floating-point support (cl_khr_fp16) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Single-precision Floating-point support (core) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations No Double-precision Floating-point support (n/a) Address bits 64, Little-Endian Global memory size 3815780352 (3.554GiB) Error Correction support No Max memory allocation 953945088 (909.8MiB) Unified memory for Host and Device Yes Shared Virtual Memory (SVM) capabilities (core) Coarse-grained buffer sharing Yes Fine-grained buffer sharing No Fine-grained system sharing No Atomics No Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Preferred alignment for atomics SVM 0 bytes Global 0 bytes Local 0 bytes Max size for global variable 65536 (64KiB) Preferred total size of global vars 0 Global Memory cache type Read/Write Global Memory cache size 262144 (256KiB) Global Memory cache line size 64 bytes Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 65536 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 32 bytes Pitch alignment for 2D image buffers 64 pixels Max 2D image size 32768x32768 pixels Max 3D image size 32768x32768x32768 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 1 Max pipe packet size 1024 Local memory type Global Local memory size 32768 (32KiB) Max number of constant args 8 Max constant buffer size 65536 (64KiB) Max size of kernel argument 1024 Queue properties (on host) Out-of-order execution Yes Profiling Yes Queue properties (on device) Out-of-order execution Yes Profiling Yes Preferred size 2097152 (2MiB) Max size 16777216 (16MiB) Max queues on device 1 Max events on device 1024 Prefer user sync for interop No Profiling timer resolution 1000ns Execution capabilities Run OpenCL kernels Yes Run native kernels No IL version SPIR-V_1_0 printf() buffer size 1048576 (1024KiB) Built-in kernels (n/a) Device Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16 NULL platform behavior clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) ARM Platform clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [ARM] clCreateContext(NULL, ...) [default] Success [ARM] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name ARM Platform Device Name Mali-G71 clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name ARM Platform Device Name Mali-G71 clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name ARM Platform Device Name Mali-G71
Every thing is good thank you for your information..
But why showing OpenCL gpu frequency only 1 MHZ..?
I have also same my gpu Adreno 650 .. why showing OpenCL gpu frequency only 1 MHZ ?.. how can I increase it ? Can you please help me?
truboxl commented
@Saikatsaha1996
It's a bug in Adreno GPU driver and it affected not just yours...
Others also reported the same problem
https://stackoverflow.com/questions/48028559/snapdragon-820-adreno-530-doesnt-return-correct-device-info
even mine (lost the phone) also reported as 1MHz
3rd party app "OpenCL Z" also reports as 1MHz
So there's really nothing we can do about it except for Qualcomm themselves to fix it
I don't think it affects the performance, just cosmetic that you can dismiss easily
Oblomov commented
What you are reporting has nothing to do with clinfo, please report it to the appropriate project.
Saikatsaha1996 commented
Hi, this is just an appreciation post for clinfo and a build instruction for building clinfo on Termux that runs on Android. It should allow people to investigate OpenCL on 64bit Android devices.
You need to have
/system/vendor/lib64/libOpenCL.sopresent on the device. If the Android device doesn't have this, it most like doesn't have OpenCL and the build will fail. Something likecannot find -lOpenCLI am going to close the issue since its not an issue with clinfo.
Simply install Termux from Google Play or FDroid Run these commands:
pkg install git make clang -y cd ~ git clone https://github.com/Oblomov/clinfo git clone https://github.com/KhronosGroup/OpenCL-Headers cd ~/clinfo export LD_LIBRARY_PATH="/system/vendor/lib64:/system/lib64" export CFLAGS="-I$HOME/OpenCL-Headers" export LDFLAGS="-L/system/vendor/lib64" make ./clinfoExample outputs: Adreno 505
Number of platforms 1 Platform Name QUALCOMM Snapdragon(TM) Platform Vendor QUALCOMM Platform Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch: Platform Profile FULL_PROFILE Platform Extensions Platform Name QUALCOMM Snapdragon(TM) Number of devices 1 Device Name QUALCOMM Adreno(TM) Device Vendor QUALCOMM Device Vendor ID 0xbf4d3c4b Device Version OpenCL 2.0 Adreno(TM) 505 Driver Version OpenCL 2.0 QUALCOMM build: commit #a7f70d7 changeid #I06947f1a28 Date: 11/19/18 Mon Local Branch: Remote Branch: Compiler E031.36.02.00 Device OpenCL C Version OpenCL C 2.0 Adreno(TM) 505 Device Type GPU Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 1 Max clock frequency 1MHz Device Partition (core) Max number of sub-devices 1 Supported partition types None Supported affinity domains (n/a) Max work item dimensions 3 Max work item sizes 1024x1024x1024 Max work group size 1024 Preferred work group size multiple 1024 Preferred / native vector sizes char 1 / 1 short 1 / 1 int 1 / 1 long 1 / 0 half 1 / 1 (cl_khr_fp16) float 1 / 1 double 0 / 0 (n/a) Half-precision Floating-point support (cl_khr_fp16) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero No Round to infinity Yes 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 No Round to infinity Yes IEEE754-2008 fused multiply-add No Support is emulated in software No Correctly-rounded divide and sqrt operations No Double-precision Floating-point support (n/a) Address bits 64, Little-Endian Global memory size 956649472 (912.3MiB) Error Correction support No Max memory allocation 239162368 (228.1MiB) Unified memory for Host and Device Yes Shared Virtual Memory (SVM) capabilities (core) Coarse-grained buffer sharing Yes Fine-grained buffer sharing No Fine-grained system sharing No Atomics No Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Page size (QCOM) 4096 bytes External memory padding (QCOM) 0 bytes Preferred alignment for atomics SVM 128 bytes Global 0 bytes Local 0 bytes Max size for global variable 65536 (64KiB) Preferred total size of global vars 1048576 (1024KiB) 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 64 bytes Pitch alignment for 2D image buffers 64 pixels Max 2D image size 16384x16384 pixels Max 3D image size 16384x16384x2048 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 4096 Max pipe packet size 1024 Local memory type Local Local memory size 32768 (32KiB) Max number of constant args 8 Max constant buffer size 65536 (64KiB) Max size of kernel argument 1024 Queue properties (on host) Out-of-order execution Yes Profiling Yes Queue properties (on device) Out-of-order execution Yes Profiling Yes Preferred size 655376 (640KiB) Max size 655376 (640KiB) Max queues on device 1 Max events on device 1024 Prefer user sync for interop No Profiling timer resolution 1000ns Execution capabilities Run OpenCL kernels Yes Run native kernels No printf() buffer size 1048576 (1024KiB) Built-in kernels (n/a) Device Extensions cl_khr_3d_image_writes cl_img_egl_image cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_egl_event cl_khr_egl_image cl_khr_fp16 cl_khr_gl_sharing 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_image2d_from_buffer cl_khr_mipmap_image cl_khr_srgb_image_writes cl_khr_subgroups cl_qcom_create_buffer_from_image cl_qcom_ext_host_ptr cl_qcom_ion_host_ptr cl_qcom_perf_hint cl_qcom_read_image_2x2 cl_qcom_android_native_buffer_host_ptr cl_qcom_protected_context cl_qcom_priority_hint cl_qcom_compressed_yuv_image_read cl_qcom_compressed_image 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 [P0] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM) clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM) clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) Invalid device type for platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name QUALCOMM Snapdragon(TM) Device Name QUALCOMM Adreno(TM)Mali-G71
Number of platforms 1 Platform Name ARM Platform Platform Vendor ARM Platform Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Platform Profile FULL_PROFILE Platform Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16 Platform Extensions function suffix ARM Platform Name ARM Platform Number of devices 1 Device Name Mali-G71 Device Vendor ARM Device Vendor ID 0x60a00002 Device Version OpenCL 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Driver Version 2.0 Device OpenCL C Version OpenCL C 2.0 v1.r16p0-01rel0.###other-sha0123456789ABCDEF0### Device Type GPU Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 2 Available core IDs 0-1 Max clock frequency 5MHz Device Partition (core) Max number of sub-devices 0 Supported partition types None Supported affinity domains (n/a) Max work item dimensions 3 Max work item sizes 384x384x384 Max work group size 384 Preferred work group size multiple 4 Preferred / native vector sizes char 16 / 4 short 8 / 2 int 4 / 1 long 2 / 1 half 8 / 2 (cl_khr_fp16) float 4 / 1 double 0 / 0 (n/a) Half-precision Floating-point support (cl_khr_fp16) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Single-precision Floating-point support (core) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations No Double-precision Floating-point support (n/a) Address bits 64, Little-Endian Global memory size 3815780352 (3.554GiB) Error Correction support No Max memory allocation 953945088 (909.8MiB) Unified memory for Host and Device Yes Shared Virtual Memory (SVM) capabilities (core) Coarse-grained buffer sharing Yes Fine-grained buffer sharing No Fine-grained system sharing No Atomics No Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Preferred alignment for atomics SVM 0 bytes Global 0 bytes Local 0 bytes Max size for global variable 65536 (64KiB) Preferred total size of global vars 0 Global Memory cache type Read/Write Global Memory cache size 262144 (256KiB) Global Memory cache line size 64 bytes Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 65536 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 32 bytes Pitch alignment for 2D image buffers 64 pixels Max 2D image size 32768x32768 pixels Max 3D image size 32768x32768x32768 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 1 Max pipe packet size 1024 Local memory type Global Local memory size 32768 (32KiB) Max number of constant args 8 Max constant buffer size 65536 (64KiB) Max size of kernel argument 1024 Queue properties (on host) Out-of-order execution Yes Profiling Yes Queue properties (on device) Out-of-order execution Yes Profiling Yes Preferred size 2097152 (2MiB) Max size 16777216 (16MiB) Max queues on device 1 Max events on device 1024 Prefer user sync for interop No Profiling timer resolution 1000ns Execution capabilities Run OpenCL kernels Yes Run native kernels No IL version SPIR-V_1_0 printf() buffer size 1048576 (1024KiB) Built-in kernels (n/a) Device Extensions 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_byte_addressable_store cl_khr_3d_image_writes cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp16 cl_khr_icd cl_khr_egl_image cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_il_program cl_khr_create_command_queue cl_arm_core_id cl_arm_printf cl_arm_thread_limit_hint cl_arm_non_uniform_work_group_size cl_arm_import_memory cl_arm_import_memory_dma_buf cl_arm_import_memory_host cl_arm_integer_dot_product_int8 cl_arm_integer_dot_product_accumulate_int8 cl_arm_integer_dot_product_accumulate_int16 NULL platform behavior clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) ARM Platform clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [ARM] clCreateContext(NULL, ...) [default] Success [ARM] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name ARM Platform Device Name Mali-G71 clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name ARM Platform Device Name Mali-G71 clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name ARM Platform Device Name Mali-G71
Hello dear how you complete for Arm-Mali GPU?
My GPU Arm-Mali G-610 MC6
OpenCL not reporting in termux..
Can you help please..
But don't know only Android application working for OpenCL..
truboxl commented
@Oblomov I think you can lock this issue to avoid further spamming. Thanks.
@Saikatsaha1996 Please report your issue at https://github.com/termux/termux-packages/issues