Kates Playground In Natural

launched CUDA CUDA NVidia

recently given the task to use CUDA optimization algorithm consisting of searching the solution space using the "swarm ". The advantage is to use CUDA GPU to perform parallel calculations on a number of variables (at least at the moment so I understand).
I thought that I could complete this project on Ubuntu 8.04.
I went to NVIDIA's website, where at the moment there are drivers in version 185, and CUDA Toolkit 2.2.
Unfortunately, he failed - I could not install the drivers. After every installation X stood up in low-quality graphics. I read on the internet that envyNG install NVIDIA drivers, together with the library libcuda. With this program I managed to install the drivers as the 173rd Then I installed CUDA CUDA Toolkit and SDK. I started to build the sample projects. After make I got the information that you can not find libraries libglut, so I installed it from the repository -
`sudo apt-get install libglut3
libglut3-dev`. Another problem was a design threadMigration - Here is simply changed the name of the makefile renamed to Makefile-and this project can not fire me. In the end, all projects are constructed and go to the directory
/ home / daniel / NVIDIA_CUDA_SDK / bin / linux / release /
. Here
first run. / DeviceQuery
with the following result:

CUDA Device Query (Runtime API) version (CUDART static linking)
There is 1 device supporting CUDA Device 0

"GeForce 8400M GS"
CUDA Capability Major revision number : 1
CUDA Capability Minor revision number: 1
Total Amount of global memory: 267714560 bytes Number of multiprocessors
: 16
Number of cores: 128
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 16384 bytes
Total number of registers available per block: 8192
Warp size: 32
Maximum number of threads per block: 512
Maximum sizes of each dimension of a block: 512 x 512 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 1
Maximum memory pitch: 262144 bytes
Texture alignment: 256 bytes
Clock rate: 0.80 GHz
Concurrent copy and execution: Yes
Run time limit on kernels: No
Integrated: Yes Support
host page-locked memory mapping: Yes
Compute Mode: Default (host multiple threads can use this device Simultaneously)

Test PASSED

Press ENTER to exit .. .


What does it mean that I have a device "CUDA Capable". The next test for the device to communicate with the system:
. / BandwidthTest


Running on ......
device 0: GeForce 8400M GS
Quick Mode Host to Device
Bandwidth for pageable memory
.
Transfer Size (Bytes) Bandwidth (MB / s)
33554432 856.9 Quick Mode

Device to Host Bandwidth for pageable memory .

Transfer Size (Bytes) Bandwidth (MB / s)
33554432 895.4 Quick Mode

Device to Device Bandwidth . Transfer Size (Bytes) Bandwidth (MB / s) 33554432 4214.5 & & & & Test PASSED

Press ENTER to exit ...



So theoretically, all is well, but I can not run any example. The most common mistake is "
cudaSafeCall () API Runtime error in file
, line 51: feature is not yet Implemented.
" It follows from this that you can not CUDA 2.2 council to act on the current drivers and I had to fight with those of luck from NVIDIA.

EDIT:
Newsflash: I managed to install the drivers for XP. Helped select the drivers for notebooks. This allows you to use CUDA version 2.2.