Help requested with installing the CUDA environment

Latest response

Hey everyone. 

 

 I'm stepping up from Fedora to RHEL6.2 and I really appreciate the solid foundation of RHEL6

 

 I am also unsure of how to go about doing some things.

 

 I am following the https://access.redhat.com/knowledge/solutions/64300 inorder to install this CUDA so I can begin to learn about programming with the graphics card.

 

 I followed along and have hit a snag.  Perhaps there is a simple thing to do but I will do much better asking those in the know.

 

At this point something is wrong and I am not sure what to do.

 

 /tmp/NVIDIA*.run --kernel-source-path /usr/src/kernels/`rpm -qa kernel --queryformat "%{PROVIDEVERSION}.%{ARCH}\n" | tail -1` --no-questions --ui=none --no-questions --accept-license
bash: /tmp/NVIDIA-Linux-x86_64-275.19.run: Permission denied
 

 This is the install parrt of the "Nvidia Download and install Nvidia Graphics drivers:"

 

 

 I appreciate the help. 

Un
Log in to join the conversation

Responses

rhn-support-mflitter's picture Red Hat Guru 1638 points
12 March 2012 12:44 PM

Hi Ernst,

 

Can you check if the script is executable please.

 

do an "ls -l" on the file and look for rwxrwxrw- on the left hand side (you might see rw-rw---)

 

if there is no x in column 3 add one with:

 

chmod u+x /tmp/<filename>

 

whe you get more comfortable with permissions, you will probably find it easier you use the number form of the chmod command, but for now the above is enough.

 

If it is executable and still does not work, can you check to see if selinux is enforcing:

 

getenforce

 

if it is, please try

 

setenforce 0

 

and then run the script again.  The selinux test is only for if we find the script is not executable.