# Error at driver init: Call to cuInit results in UNKNOWN\_CUDA\_ERROR (-1)

**URL:** <https://numba.discourse.group/t/error-at-driver-init-call-to-cuinit-results-in-unknown-cuda-error-1/2269>\
**Category:** Community Support\
**Created:** [November 14, 2023, 10:39am UTC](https://numba.discourse.group/t/error-at-driver-init-call-to-cuinit-results-in-unknown-cuda-error-1/2269 "2023-11-14T10:39:24Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![dguhanus](https://yyz2.discourse-cdn.com/free1/user_avatar/numba.discourse.group/dguhanus/32/932_2.png) [@dguhanus](https://numba.discourse.group/u/dguhanus)\
**Post date:** [November 14, 2023, 10:39am UTC](https://numba.discourse.group/t/error-at-driver-init-call-to-cuinit-results-in-unknown-cuda-error-1/2269/1 "2023-11-14T10:39:24Z")

</div>

Hi ,  
I am beginner in CUDA. I want to find polar cordinate for millions of points.  
I am using the below code:  
vectorize([‘float32(float32, float32)’],target=‘cuda’)  
def gpu\_arctan2(y, x):  
theta = math.atan2(y,x)  
return theta  
But it is failing with an error message  
File “/home/g600523/.conda/envs/env\_tf\_gpu/lib/python3.9/site-packages/numba/cuda/cudadrv/driver.py”, line 258, in ensure\_initialized  
raise CudaSupportError(f"Error at driver init: {description}")  
numba.cuda.cudadrv.error.CudaSupportError: Error at driver init: Call to cuInit results in UNKNOWN\_CUDA\_ERROR (-1)

## I am having CUDA version 11.7. Here is the output for numba -s (env\_tf\_gpu) numba -s System info:

**Time Stamp**  
Report started (local time) : 2023-11-14 15:20:34.244004  
UTC start time : 2023-11-14 09:50:34.244030  
Running time (s) : 6.599611

**Hardware Information**  
Machine : x86\_64  
CPU Name : icelake-server  
CPU Count : 64  
Number of accessible CPUs : 32  
List of accessible CPUs cores : 0-63  
CFS Restrictions (CPUs worth of runtime) : None

CPU Features : 64bit adx aes avx avx2  
avx512bitalg avx512bw avx512cd  
avx512dq avx512f avx512ifma  
avx512vbmi avx512vbmi2 avx512vl  
avx512vnni avx512vpopcntdq bmi  
bmi2 clflushopt clwb cmov crc32  
cx16 cx8 f16c fma fsgsbase fxsr  
gfni invpcid lzcnt mmx movbe  
pclmul pconfig pku popcnt prfchw  
rdpid rdrnd rdseed rtm sahf sgx  
sha sse sse2 sse3 sse4.1 sse4.2  
ssse3 vaes vpclmulqdq wbnoinvd  
xsave xsavec xsaveopt xsaves

Memory Total (MB) : 515619  
Memory Available (MB) : 496690

**OS Information**  
Platform Name : Linux-3.10.0-1160.el7.x86\_64-x86\_64-with-glibc2.17  
Platform Release : 3.10.0-1160.el7.x86\_64  
OS Name : Linux  
OS Version : #1 SMP Mon Oct 19 16:18:59 UTC 2020  
OS Specific Version : ?  
Libc Version : glibc 2.17

**Python Information**  
Python Compiler : GCC 11.2.0  
Python Implementation : CPython  
Python Version : 3.9.18  
Python Locale : en\_IN.ISO8859-1

**Numba Toolchain Versions**  
Numba Version : 0.58.1  
llvmlite Version : 0.41.1

**LLVM Information**  
LLVM Version : 14.0.6

**CUDA Information**  
CUDA Device Initialized : False  
CUDA Driver Version : ?  
CUDA Runtime Version : ?  
CUDA NVIDIA Bindings Available : ?  
CUDA NVIDIA Bindings In Use : ?  
CUDA Minor Version Compatibility Available : ?  
CUDA Minor Version Compatibility Needed : ?  
CUDA Minor Version Compatibility In Use : ?  
CUDA Detect Output:  
None  
CUDA Libraries Test Output:  
None

**NumPy Information**  
NumPy Version : 1.26.2  
NumPy Supported SIMD features : (‘MMX’, ‘SSE’, ‘SSE2’, ‘SSE3’, ‘SSSE3’, ‘SSE41’, ‘POPCNT’, ‘SSE42’, ‘AVX’, ‘F16C’, ‘FMA3’, ‘AVX2’, ‘AVX512F’, ‘AVX512CD’, ‘AVX512VPOPCNTDQ’, ‘AVX512VL’, ‘AVX512BW’, ‘AVX512DQ’, ‘AVX512VNNI’, ‘AVX512IFMA’, ‘AVX512VBMI’, ‘AVX512VBMI2’, ‘AVX512BITALG’, ‘AVX512\_SKX’, ‘AVX512\_CLX’, ‘AVX512\_CNL’, ‘AVX512\_ICL’)  
NumPy Supported SIMD dispatch : (‘SSSE3’, ‘SSE41’, ‘POPCNT’, ‘SSE42’, ‘AVX’, ‘F16C’, ‘FMA3’, ‘AVX2’, ‘AVX512F’, ‘AVX512CD’, ‘AVX512\_KNL’, ‘AVX512\_KNM’, ‘AVX512\_SKX’, ‘AVX512\_CLX’, ‘AVX512\_CNL’, ‘AVX512\_ICL’)  
NumPy Supported SIMD baseline : (‘SSE’, ‘SSE2’, ‘SSE3’)  
NumPy AVX512\_SKX support detected : True

**SVML Information**  
SVML State, config.USING\_SVML : False  
SVML Library Loaded : False  
llvmlite Using SVML Patched LLVM : True  
SVML Operational : False

**Threading Layer Information**  
TBB Threading Layer Available : False  
±-\> Disabled due to Unknown import problem.  
OpenMP Threading Layer Available : True  
±-\>Vendor: GNU  
Workqueue Threading Layer Available : True  
±-\>Workqueue imported successfully.

**Numba Environment Variable Information**  
None found.

**Conda Information**  
Conda Build : 3.22.0  
Conda Env : 22.9.0  
Conda Platform : linux-64  
Conda Python Version : 3.9.7.final.0  
Conda Root Writable : False

**Installed Packages**  
\_libgcc\_mutex 0.1 main  
\_openmp\_mutex 5.1 1\_gnu  
absl-py 2.0.0 pypi\_0 pypi  
astunparse 1.6.3 pypi\_0 pypi  
ca-certificates 2023.08.22 h06a4308\_0  
cachetools 5.3.2 pypi\_0 pypi  
certifi 2023.7.22 pypi\_0 pypi  
charset-normalizer 3.3.2 pypi\_0 pypi  
contourpy 1.2.0 pypi\_0 pypi  
cycler 0.12.1 pypi\_0 pypi  
flatbuffers 23.5.26 pypi\_0 pypi  
fonttools 4.44.0 pypi\_0 pypi  
gast 0.5.4 pypi\_0 pypi  
google-auth-oauthlib 1.0.0 pypi\_0 pypi  
google-pasta 0.2.0 pypi\_0 pypi  
grpcio 1.59.2 pypi\_0 pypi  
h5py 3.10.0 pypi\_0 pypi  
idna 3.4 pypi\_0 pypi  
importlib-metadata 6.8.0 pypi\_0 pypi  
importlib-resources 6.1.1 pypi\_0 pypi  
keras 2.14.0 pypi\_0 pypi  
kiwisolver 1.4.5 pypi\_0 pypi  
ld\_impl\_linux-64 2.38 h1181459\_1  
libclang 16.0.6 pypi\_0 pypi  
libffi 3.4.4 h6a678d5\_0  
libgcc-ng 11.2.0 h1234567\_1  
libgomp 11.2.0 h1234567\_1  
libstdcxx-ng 11.2.0 h1234567\_1  
llvmlite 0.41.1 pypi\_0 pypi  
markdown 3.5.1 pypi\_0 pypi  
markupsafe 2.1.3 pypi\_0 pypi  
matplotlib 3.8.1 pypi\_0 pypi  
ml-dtypes 0.2.0 pypi\_0 pypi  
ncurses 6.4 h6a678d5\_0  
numba 0.58.1 pypi\_0 pypi  
numpy 1.26.2 pypi\_0 pypi  
nvidia-cublas-cu11 11.11.3.6 pypi\_0 pypi  
nvidia-cuda-cupti-cu11 11.8.87 pypi\_0 pypi  
nvidia-cuda-nvcc-cu11 11.8.89 pypi\_0 pypi  
nvidia-cuda-runtime-cu11 11.8.89 pypi\_0 pypi  
nvidia-cudnn-cu11 8.7.0.84 pypi\_0 pypi  
nvidia-cufft-cu11 10.9.0.58 pypi\_0 pypi  
nvidia-curand-cu11 10.3.0.86 pypi\_0 pypi  
nvidia-cusolver-cu11 11.4.1.48 pypi\_0 pypi  
nvidia-cusparse-cu11 11.7.5.86 pypi\_0 pypi  
nvidia-nccl-cu11 2.16.5 pypi\_0 pypi  
oauthlib 3.2.2 pypi\_0 pypi  
openssl 3.0.12 h7f8727e\_0  
opt-einsum 3.3.0 pypi\_0 pypi  
packaging 23.2 pypi\_0 pypi  
pillow 10.1.0 pypi\_0 pypi  
pip 23.3 py39h06a4308\_0  
protobuf 4.25.0 pypi\_0 pypi  
pyasn1 0.5.0 pypi\_0 pypi  
pyasn1-modules 0.3.0 pypi\_0 pypi  
pyparsing 3.1.1 pypi\_0 pypi  
python 3.9.18 h955ad1f\_0  
python-dateutil 2.8.2 pypi\_0 pypi  
readline 8.2 h5eee18b\_0  
requests 2.31.0 pypi\_0 pypi  
requests-oauthlib 1.3.1 pypi\_0 pypi  
rsa 4.9 pypi\_0 pypi  
setuptools 68.0.0 py39h06a4308\_0  
six 1.16.0 pypi\_0 pypi  
sqlite 3.41.2 h5eee18b\_0  
tensorboard 2.14.1 pypi\_0 pypi  
tensorboard-data-server 0.7.2 pypi\_0 pypi  
tensorflow 2.14.0 pypi\_0 pypi  
tensorflow-estimator 2.14.0 pypi\_0 pypi  
tensorflow-io-gcs-filesystem 0.34.0 pypi\_0 pypi  
tensorrt 8.5.3.1 pypi\_0 pypi  
termcolor 2.3.0 pypi\_0 pypi  
tk 8.6.12 h1ccaba5\_0  
typing-extensions 4.8.0 pypi\_0 pypi  
tzdata 2023c h04d1e81\_0  
urllib3 2.1.0 pypi\_0 pypi  
werkzeug 3.0.1 pypi\_0 pypi  
wheel 0.41.2 py39h06a4308\_0  
wrapt 1.14.1 pypi\_0 pypi  
xz 5.4.2 h5eee18b\_0  
zipp 3.17.0 pypi\_0 pypi  
zlib 1.2.13 h5eee18b\_0

No errors reported.

## **Warning log** Warning (cuda): CUDA device initialisation problem. Message:Error at driver init: Call to cuInit results in UNKNOWN\_CUDA\_ERROR (-1) Exception class: \<class ‘numba.cuda.cudadrv.error.CudaSupportError’\> Warning (psutil): psutil cannot be imported. For more accuracy, consider installing it.

If requested, please copy and paste the information between  
the dashed (----) lines, or from a given specific section as  
appropriate.

# ============================================================= IMPORTANT: Please ensure that you are happy with sharing the contents of the information present, any information that you wish to keep private you should remove before sharing.

(env\_tf\_gpu)

Surprisingly, if I use,  
jit(target\_backend=‘cuda’) it is working for a simple function like addition, but even jit is also failing for finding Polar cordinate.

---

<div class="post-metadata">

**Author:** ![esc](https://yyz2.discourse-cdn.com/free1/user_avatar/numba.discourse.group/esc/32/49_2.png) [@esc](https://numba.discourse.group/u/esc)\
**Post date:** [November 21, 2023, 12:44pm UTC](https://numba.discourse.group/t/error-at-driver-init-call-to-cuinit-results-in-unknown-cuda-error-1/2269/2 "2023-11-21T12:44:51Z")

</div>

@dguhanus thank you for your post. May I suggest to reformat the post to use synatx highlighting. You can find more information about it here:

> **[An Unofficial Discourse User Reference Guide](https://forum.digikey.com/t/an-unofficial-discourse-user-reference-guide/1125#heading--CodeFormatting)**
>
> This is a “quick” how-to guide for using Discourse forum/community software. It contains a majority of the common features avialble to users/content editors, but may not have complete coverage of the featureset. If anyone sees anything to add or...