Update CUDA Toolkit version to 11.7 in all projects and docs

[cosmin-bianu]

Fixes #128

[xkszltl]

Any reason why this is still pending?

[cosmin-bianu]

Turns out they don't accept community contributions at the moment as per the README
"At this time we are not accepting contributions from the public, check back here as we evolve our contribution model."

Just clone the forked repo

git clone https://github.com/cosmin-bianu/cuda-samples

Or simply find and replace the version number.

[xkszltl]

@rnertney @Ru7w1k
Saw the 11.8 and 12.0 tag is finally there, great!
Can we also have the skipped 11.7, for the simplicity of automation?
Currently we have to maintain a mapping saying CUDA 11.7 should be routed to 11.6's sample.

[mikechen66]

According to the guideline by cosmin-bianu, I moved Samples to the path '/usr/local/cuda-11.7/'. But I have met the error while I verify the Samples.

1. Locate the current directory

$ cd /usr/local/cuda-11.7/Samples/1_Utilities/deviceQuery

2. make the command

/usr/local/cuda-11.7/Samples/1_Utilities/deviceQuery$ sudo make

usr/local/cuda/bin/nvcc -ccbin g++ -I../../../Common -m64 --threads 0 --std=c++11 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -gencode arch=compute_86,code=compute_86 -o deviceQuery.o -c deviceQuery.cpp
nvcc warning : The 'compute_35', 'compute_37', 'sm_35', and 'sm_37' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
deviceQuery.cpp:34:10: fatal error: helper_cuda.h: No such file or directory
34 | #include <helper_cuda.h>
| ^~~~~~~~~~~~~~~
compilation terminated.
make: *** [Makefile:324: deviceQuery.o] Error 255

Please help indicate what happens with the error.

Notes:

1. If the command "sudo make" make a succcess, it can be verified with the following command './deviceQuery'.

2. Can I construct the the written copy as follows.. Are the following commands correct?

$ cuda-install-samples-11.7.sh ~
$ cd ~/NVIDIA_CUDA-11.7_Samples/5_Simulations/nbody
$ make
$ ./nbody

[mikechen66]

I find out a way to solve the problem. The major problem is that the new Nvidia guidelines are too vague to be understood. I make the following correct procedure to make a success execution.

1. Enter into the following directory

$ cd /usr/local/cuda-11.7

2. Make a new directory

$ mkdir samples

It generates the path as follows.

/cd /usr/local/cuda-11.7/samples

3. Move (or git) cuda-samples to the targeted samples

Download the file, upzip and rename cuda-samples-master as cuda-samples
(Or user can git the file to the above mentioned samples)

$ sudo mv /Downloads/cuda-samples /usr/local/cuda-11.7/samples/

4. Enter into the new directory

$ cd /usr/local/cuda-11.7/samples/cuda-samples/Samples/1_Utilities/deviceQuery

5. Conduct the make command

$ sudo make

6. make the final show

$ ./deviceQuery

./deviceQuery Starting

CUDA Device Query (Runtime API) version (CUDART static linking)

Device 1 CUDA Capable device(s)
Device 0: "NVIDIA GeForce RTX Model Name"
........

Result = PASS

Cheers.

Notes:

I still do not figure out a way to handle the 5_Simulations case. That would be a visual showing and make great sense.