Hello We have updated our compiler to the Nvidia HPC SDK version 22.9 and we ran into an issue with the offloading. We have a minimal example there hpc-bugs/nvidiagpu.nvc/omp-crash at main · FabioLuporini/hpc-bugs main/nvidiagpu.nvc/omp-crash Contribute to FabioLuporini/hpc-bugs de…

If you add -mp=gpu to your linking, it seems to work #!/bin/bash nvc -g -fPIC -gpu=pinned -mp=gpu -fast -shared omp_crash.c -lm -o omp_crash.so nvc -O3 -g -mp=gpu -Wall main.c -ldl -o main ./main $ sh run.sh Running foo()… DONE!

Hi, unfortunately we don’t have control over the process performing the dynamic loading. In our case, it’s Python :) so we can’t easily do that. I’m guessing we could maybe use LD_PRELOAD but it gets nightmarish… and I’m not even that sure context: this project GitHub - devitocodes/devito: Code gen…

There are three libraries needed. LD_PRELOAD=libacchost.so:libaccdevaux.so:libaccdevice.so ./main Running foo()… DONE!

FWIW, this is a bit painful, for various reasons… installation / configuration. Fine, we have Dockerfiles, but still… maintainance. What if those libs change in the next release, or a new one is added

I would suggest to open a formal bug. If you ldd omp_crash.so, those libraries are there and if you run with LD_DEBUG=libs ./main you can see that there are errors coming from pthreads. 191258: ./main: error: symbol lookup error: undefined symbol: pthread_atfork (fatal) 191258: ./main:…

It looks like the shared library works as expected from Python ( that you indicated was your real use case), at least from this simple example. $ cat pydriver.py from ctypes import cdll lib = cdll.LoadLibrary(‘./omp_crash.so’) lib.foo() $ python pydriver.py Running foo()… DONE!

OMP offloading crash with nvc

Accelerated Computing CUDA CUDA NVCC Compiler

rs277 November 15, 2022, 8:12am 8

Here.

Topic		Replies	Views
Creating a shared library that utilises OpenMP offloading NVHPC 22.5 nvc, nvc++ and nvfortran	4	898	June 23, 2022
Creating a shared library that utilises OpenMP offloading nvc, nvc++ and nvfortran	3	1419	May 25, 2022
Compiler issues when migrating from 20.7 to 22.1 nvc, nvc++ and nvfortran	7	876	January 13, 2022
OpenMP offload w/ CUDA interop: undefined reference to `__fatbinwrap__NV_MODULE_ID' nvc, nvc++ and nvfortran	5	1310	May 22, 2023
Does nvc support GPU offloading with OpenMP nvc, nvc++ and nvfortran	2	1044	December 14, 2020
Cannot dynamically load a shared library containing both OpenACC and CUDA code nvc, nvc++ and nvfortran	8	3026	August 24, 2022
OpenMP offloading with nvc++ compiler wont run on GPU nvc, nvc++ and nvfortran	1	849	August 14, 2023
Nvc++ compilation fails with option -O2 and -mp or -mp=gpu for specific OpenMP-5 offload code nvc, nvc++ and nvfortran	1	59	February 15, 2026
Ncu does not detect kernels, ==ERROR== The application returned an error code (11) Nsight Compute kernel , profiling	5	2335	December 13, 2023
Nvc++ OpenMP error inside llc nvc, nvc++ and nvfortran	5	1278	June 1, 2021

OMP offloading crash with nvc

Related topics