Use this tag to ask questions about the overall oneAPI programming model which provides unified developer tools that can be used across multiple hardware platforms, including a range of performance libraries spanning several workload domains. The libraries include functions custom-coded for each target architecture so the same function call delivers optimized performance across supported architectures.
Questions tagged [intel-oneapi]
220 questions
0
votes
0 answers
How to launch without segfault an Intel python3.9 shell on macOS 13.5 Ventura
On MacOS Venture 13.5, I have installed Intel OneAPI Base and HPC toolkit and also Intel distribution of python3.9 ( python 3.9.7).
I have a strange behavior : if I execute a Python script, everything works fine. But from another side, impossible to…
guizmo133
- 11
- 3
0
votes
0 answers
How to use kernel_bundle with custom kernels in SYCL?
I'm trying to get a handle on SYCL's kernel_bundle functionality. My target is to use it as a db storage for my AOT precompiled custom kernels. The desirable result is to precompile these kernels into a stand alone library for other libraries to…
agoraphobi
- 1
- 1
0
votes
0 answers
Linker issue with oneAPI if I use the -ipo flag
I'm using oneAPI (intel's compiler) and ran into issues using the -ipo flag when compiling. The setup:
my_lib project: static lib
my_test: executable that links to my_lib
Files:
my_lib:
main.cpp
int foo() {
return 0; …
theNotSoPro
- 308
- 3
- 16
0
votes
0 answers
How to know the maximum local memory size that can be used in FPGA kernel (using oneAPI SYCL or OpenCL)?
There is a big difference between the document provided through the vendor document and the program output, and I don't quite understand how the local memory is used in the FPGA.
The FPGA I am using is Intel Stratix 10 GX 2800 and according to the…
Ypro
- 1
- 1
0
votes
0 answers
ifort: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.21' not found (required by ifort)
I am unable to remove the following error as required package is not available in the package manager yum for centos:
ifort: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.21' not found (required by ifort)
can anyone help me about it?
I try to find…
0
votes
1 answer
dpcc code for fpga emulation using jupyter notebooks
I was performing emulation of hough transform for FPGA's on Jupyterlab but then all i had to do was run the cells:
I get the following error ->
u196294 is performing Hough Transform compilation emulation notebook.
icpx: warning: use of 'dpcpp' is…
0
votes
0 answers
Unable to build a project on WSL2 using Visual Studio (VS) 2019 with OneAPI compilers
I am using VS 2019 with WSL2 to build a cross-platform app. I use OneAPI tools. I could successfully compile on Windows, but not on WSL2. The reason is VS does not recognize 'mpiifort' as it doesn't read the default bash environment. How to solve…
nhm
- 104
- 7
0
votes
1 answer
How to use Intel VTunes to detect where does UPI flows comes from?
In memory access evaluation module, there exists a panel called platform diafram, which shows the UPI Utilization outgoing. I wish to ask that whetther it is possiable for me to find out where thos UPI flows comes from?
It would be so great if I…
Shadow_visual
- 19
- 2
0
votes
1 answer
How to install onAPI/tbb on alpine
I'm currently running alpine on a docker, and I've tried installing the tbb library of intel's oneAPI via pip and also tried making it from source, and none of the those did work, is there any way to do that? if so how?
Kami SM
- 29
- 6
0
votes
1 answer
MPI_Scatterv from Intel MPI (mpiifort) using MPI data types is much slower (23 times) compared to flattening array and scattering. Why it could be?
ifort version: (IFORT) 2021.8.0 20221119
OS: WSL Ubuntu 20.04LTS
I have a (1000x1000x1000) 3D array to distribute among procs. In the first approach, I flatten the array and then distribute arrays among procs and it takes about 7.86sec
In the…
nhm
- 104
- 7
0
votes
1 answer
Unable to compile mkl_dfti.f90
I am trying to use the dft examples provided with MKL. I am trying to compile the program basic_dp_complex_dft_1d.f90, which uses MKL_DFTI module. Since Intel doesn't provide with pre-compiled modules, I'm trying tom compile the mkl_dfti.f90 in the…
0
votes
1 answer
gdb-oneapi "No debugging symbols found in "
I don't know how to add debugging symbols to my compiled C code on windows. My attempt so far is below.
My test code:
#include
#include
int main()
{
printf("Hello world!\n");
return EXIT_SUCCESS;
}
My compilation…
DJames
- 571
- 3
- 17
0
votes
0 answers
SYCL Explicit Data Movement from Image to Host
I'm having trouble trying to figure out how to do explicit data transfer to/from host when using sycl::image. I'm familiar with sycl::handler.copy when using buffers:
float* host_ptr;
sycl::buffer buf;
sycl::queue&…
Biggy Smith
- 910
- 7
- 14
0
votes
0 answers
Installing Intel OneAPI Basekit in Windows Container
I'm trying to install an Intel OneAPI BaseKit inside of a windows container using docker for windows. However, when I try to install (inside the dockerfile or directly in the windows container), I get the following error:
cmd /S /C…
0
votes
1 answer
How to calculate FFT forward and backward multiple times in fast frame rate (about 700 fps) using Intel MKL?
We have a fast camera (>= 700 fps), we need to calculate FFT on images and do some action. However, the FFT works for some frames and then crashes. The problem doesn't appear when I use slow frame rate like 10 fps.
To further isolate the problem, I…
Harsh M
- 625
- 2
- 11
- 25