Errors in v7 and v6

[Seidhberendir]

When I run the command line in v7 I receive the following error:

Traceback (most recent call last):
File "/home/dojo/Desktop/bsgs-main/v7_gpu_trial/bsgs_GPU.py", line 123, in
bsgsgpu = ctypes.CDLL(pathdll)
File "/usr/lib/python3.10/ctypes/init.py", line 374, in init
self._handle = _dlopen(self._name, mode)
OSError: libcudart.so.11.0: cannot open shared object file: No such file or directory

Could not resolve this error so switched to v6, then received the following error:

Traceback (most recent call last):
File "/media/dojo/SATASSD/bsgs-main/v6_dll_bsgs/bsgs_create_bpfile_bloomfile.py", line 98, in
bloom_filter = bytes(b'\x00') * (bloom_bits//8)
MemoryError

I have plenty of memory to create the file (256 GB) using 6 cores. Please assist in resolving this. I am running Linux Ubuntu latest version, 16 GB of RAM, two 1 TB SSDs, and AMD Ryzen CPU with AMD video card. If specific specs are needed, let me know.

Thank you.

[Seidhberendir]

resolved v6 errors and is working properly now. The v7 issue is narrowed down to Cuda GPU issue. Still unresolved. Is there a way to tailor the program and driver to Raytheon GPU rather than Nvidia? This appears to be the error source.

[pixel-wipe]

Got same error with v7 version with cuda 12

while was at cuda 11+ was get another error

which one cuda i must to use?

[mirameshs]

please update the apps with latest version of cuda. we are using apps developed via ice2k on nvidia rtx4090 and need more updates

…

On Mon, Aug 21, 2023 at 2:34 PM pixel-wipe ***@***.***> wrote: Got same error with v7 version with cuda 12 while was at cuda 11+ was get another error which one cuda i must to use? — Reply to this email directly, view it on GitHub <#73 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH4VUUUKP4YNJRKQGDKNCALXWM6D3ANCNFSM6AAAAAAXLNNKTA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[pixel-wipe]

which apps i must update?
all dat was need and dont goes with py by deafault (bit, signal, argparse, numba) was installed today so they are up to date

please update the apps with latest version of cuda. we are using apps developed via ice2k on nvidia rtx4090 and need more updates
…
On Mon, Aug 21, 2023 at 2:34 PM pixel-wipe @.> wrote: Got same error with v7 version with cuda 12 while was at cuda 11+ was get another error which one cuda i must to use? — Reply to this email directly, view it on GitHub <#73 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AH4VUUUKP4YNJRKQGDKNCALXWM6D3ANCNFSM6AAAAAAXLNNKTA . You are receiving this because you are subscribed to this thread.Message ID: @.>

X@X:~/Desktop/bsgs/v7_gpu_trial$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243

X@sX:~/Desktop/bsgs/v7_gpu_trial$ nvidia-smi 
Tue Aug 22 02:37:54 2023       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.86.10              Driver Version: 535.86.10    CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+

[pixel-wipe]

i has checked in system and libcudart.so.11.0 isnt at my system disk

does in cuda 12.0-12.2 it was depricated ?
or which packet need to install to get it?

[hiddenvs]

i has checked in system and libcudart.so.11.0 isnt at my system disk

does in cuda 12.0-12.2 it was depricated ? or which packet need to install to get it?

Install package nvidia-cuda-toolkit f.e. on Debian/Ubuntu sudo apt install nvidia-cuda-toolkit, this will help to mitigate that issue, but this is only the beginning, there are more catches.