Nvidia CUDA Nightmare

[Homi576]

Yep is a nvidia post :mad:

I have been following this guide in order to take advantage of the cuda toolkit but have come across a problem that I am all out ideas trying to overcome. Everything has gone pretty smoothly installing bumblebee etc but when I come to install the cuda toolkit it's just not having it. Below is an output from the terminal installation:


Do you accept the previously read EULA?
accept/decline/quit: accept

You are attempting to install on an unsupported configuration. Do you wish to continue?
(y)es/(n)o [ default is no ]: y

Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 367.48?
(y)es/(n)o/(q)uit: n

Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y

Enter Toolkit Location
[ default is /usr/local/cuda-8.0 ]:

Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y

Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y

Enter CUDA Samples Location
[ default is /root ]:

Installing the CUDA Toolkit in /usr/local/cuda-8.0 ...
Warning: cannot find Toolkit in /usr/local/cuda-8.0. Use --toolkitpath to specify the toolkit location.
Installing the CUDA Samples in /root ...
sh: 1: /usr/local/cuda-8.0/bin/cuda-install-samples-8.0.sh: not found
chown: cannot access '/root/NVIDIA_CUDA-8.0_Samples': No such file or directory

===========
= Summary =
===========

Driver: Not Selected
Toolkit: Installation Failed
Samples: Installation Failed


And this is the cuda_install_log:


Installing the CUDA Toolkit in /usr/local/cuda-8.0 ...
Verifying archive integrity... All good.

Uncompressing NVIDIA CUDA.............................................. .................................................. ...........................$

Can't locate InstallUtils.pm in @INC (you may need to install the InstallUtils module) (@INC contains: /etc/perl /usr/local/lib/x86_64-linux-gnu/per$

BEGIN failed--compilation aborted at ./install-linux.pl line 6.

Verifying archive integrity... All good.

Uncompressing NVIDIA CUDA Samples........................................... .................................................. ......................$

Can't locate InstallUtils.pm in @INC (you may need to install the InstallUtils module) (@INC contains: /etc/perl /usr/local/lib/x86_64-linux-gnu/per$

BEGIN failed--compilation aborted at ./install-sdk-linux.pl line 6.

'uninstall_cuda_8.0.pl' -> '/usr/local/cuda-8.0/bin/uninstall_cuda_8.0.pl'


This issue happens with cuda 8.0 and 7.5, anyone have any suggestions (apart from ditch nvidia ofcourse)? I have been following this guide mostly https://forums.kali.org/showthread.php?21956-%28Guide%29-Installing-Nvidia-Bumblebee-CUDA-for-Optimus-enabled-Laptops

Thanks guys.

[noobiesmall]

Why (https://forums.kali.org/showthread.php?21956-(Guide)-Installing-Nvidia-Bumblebee-CUDA-for-Optimus-enabled-Laptops&p=50488&viewfull=1#post50488) don't you just install the nvidia-cuda-toolkit package from repos ?

[Homi576]

Why (https://forums.kali.org/showthread.php?21956-(Guide)-Installing-Nvidia-Bumblebee-CUDA-for-Optimus-enabled-Laptops&p=50488&viewfull=1#post50488) don't you just install the nvidia-cuda-toolkit package from repos ?

Thank you works great, your guide was really good

[H3xCmd]

Even though 'apt-get install nvidia-cuda-toolkit' (aka cudatoolkit 8.0) does work and installs and may work great, I believe users that wish to utilize pyrit and cpryit_cuda instructions are asking set up a dynamic linker for CUDA

gedit /etc/ld.so.conf
last line add ' /usr/local/cuda-8.0/lib64

when installing from the repo this folder is located somewhere else and unable to add a dynamic link, I searched for lib64 and got 2 hits. one in / 'root' an the other in /usr/lib64 both do not reference nvidia-cuda-toolkit..... so this is in-fact still a nightmare. I have been trying for weeks now to get nvidia optimus (aka intel+nvidia laptop) to work with pyrit. any helpful hints.

thanks.

[hausoo]

Even though 'apt-get install nvidia-cuda-toolkit' (aka cudatoolkit 8.0) does work and installs and may work great, I believe users that wish to utilize pyrit and cpryit_cuda instructions are asking set up a dynamic linker for CUDA

gedit /etc/ld.so.conf
last line add ' /usr/local/cuda-8.0/lib64

when installing from the repo this folder is located somewhere else and unable to add a dynamic link, I searched for lib64 and got 2 hits. one in / 'root' an the other in /usr/lib64 both do not reference nvidia-cuda-toolkit..... so this is in-fact still a nightmare. I have been trying for weeks now to get nvidia optimus (aka intel+nvidia laptop) to work with pyrit. any helpful hints.

thanks.

hello
/usr/lib/nvidia-cuda-toolkit
type nvcc -V in console to see if the cuda compiler is configured correctly

~# nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Sun_Sep__4_22:14:01_CDT_2016
Cuda compilation tools, release 8.0, V8.0.44

get pyrit from here : https://github.com/JPaulMora/Pyrit
read here :https://github.com/JPaulMora/Pyrit/wiki :and here to make pyrit work with cuda : https://github.com/JPaulMora/Pyrit/wiki/GPU-Acceleration

~# pyrit list_cores
Pyrit 0.5.1 (C) 2008-2011 Lukas Lueg - 2015 John Mora
https://github.com/JPaulMora/Pyrit
This code is distributed under the GNU General Public License v3+

The following cores seem available...
#1: 'CPU-Core (SSE2/AES)'
#2: 'CPU-Core (SSE2/AES)'
#3: 'CPU-Core (SSE2/AES)'
#4: 'CPU-Core (SSE2/AES)'
#5: 'CPU-Core (SSE2/AES)'
#6: 'CPU-Core (SSE2/AES)'

The following CUDA GPUs seem aviable...
#1: 'CUDA-Device #1 'GeForce GTX 1060 3GB''

[itanon]

Hi, I'm having problem with cuda installation on Kali 2016.2 live USB on my laptop... I need to do this in order to run pyrit with cuda(cpyrit)...I have an old Nvidia GPU 330m....All the repos installation I tried wasn't good since the apt-get, download the lasts version of proprietry driver and toolkit that not support my gpu configuration...So I thought to download the cuda compatible cuda toolkit(6.5 for Ubuntu 10.4 64bit) .Run file and execute it from terminal...The problem I get is that, once the installation start it abort giving me the error: "not enough space on /tmp partition" (this seems to be widely common problem in Linux environment, but I don't get the solution anywhere)....So here's mine doubts:
1) is there any solution to avoid the unavailability of free space in /tmp directory?
2) I've red that some cuda installation need to consider the bumblebee configuration...Is it a must step(I did not considered in my installation process) or is possible that use cuda's power ignoring it?
3)in case I will include bumblebee support in my cuda installation, what will be the advantages? (More compute power due to the support of Intel's integrated GPU combined with Nvidia's GPU?)
I'm not an expert so I may be wrong on some aspects, thanks for the help.

[H3xCmd]

hello
/usr/lib/nvidia-cuda-toolkit
type nvcc -V in console to see if the cuda compiler is configured correctly

~# nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Sun_Sep__4_22:14:01_CDT_2016
Cuda compilation tools, release 8.0, V8.0.44

~# pyrit list_cores
Pyrit 0.5.1 (C) 2008-2011 Lukas Lueg - 2015 John Mora
https://github.com/JPaulMora/Pyrit
This code is distributed under the GNU General Public License v3+



root@Debian:~\% optirun nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Sun_Sep__4_22:14:01_CDT_2016
Cuda compilation tools, release 8.0, V8.0.44

root@Debian:~\% optirun pyrit list_cores
Pyrit 0.5.0 (C) 2008-2011 Lukas Lueg - 2015 John Mora
https://github.com/JPaulMora/Pyrit
This code is distributed under the GNU General Public License v3+

The following cores seem available...
#1: 'CPU-Core (SSE2/AES)'
#2: 'CPU-Core (SSE2/AES)'
#3: 'CPU-Core (SSE2/AES)'
#4: 'CPU-Core (SSE2/AES)'

root@Debian:~\% optirun nvidia-smi
ERROR: ld.so: object 'libdlfaker.so' from LD_PRELOAD cannot be preloaded (wrong ELF class: ELFCLASS64): ignored.
ERROR: ld.so: object 'libvglfaker.so' from LD_PRELOAD cannot be preloaded (wrong ELF class: ELFCLASS64): ignored.
Sat Mar 18 22:44:54 2017
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 375.39 Driver Version: 375.39 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+================= =====+======================|
| 0 GeForce GT 720M Off | 0000:03:00.0 N/A | N/A |
| N/A 45C P0 N/A / N/A | 4MiB / 1985MiB | N/A Default |
+-------------------------------+----------------------+----------------------+


So the problem is I do not get what you get

The following CUDA GPUs seem aviable...
#1: 'CUDA-Device #1 'GeForce GTX 1060 3GB''

[hausoo]

to enable GPU computing, you must edit your config file located at ~/.pyrit/config