What Stds Can You Get From Blood

CUDA 2.3 on Ubuntu 9.10 Part. 2

It should check if everything is installed correctly by compiling the sample projects.
shortening the entire process of overcoming the errors that occurred during this command:
1) CUDA compiler does not work with version 4.4. Unfortunately, uninstalling gcc-4.4 was not an option, because I would have to also uninstall newly installed drivers. So I installed gcc and g + + version 4.3.
sudo apt-get install gcc-4.3 g + + -4.3
Then I tried to change the makefile, instead of
CXX: = g + +
CC: = gcc
LINK = g + +-fPIC

have
CXX: = g + + -4.3
CC: = gcc-4.3
LINK: = g + + -4.3-fPIC

Unfortunately, it's possible. One of the members of the Forum NVIDIA came to the conclusion that these variables are overwritten
in any of the other makefile'ów
. I backed the changes and instead changed the link / usr / bin / gcc with gcc-4.4 gcc-4.3.

sudo rm / usr / bin / gcc sudo
ln-s-v gcc-4.3 / usr / bin / gcc
A similar operation conducted also for g + +. I checked whether the link gcc compiler is the correct version:
gcc - version gcc
(Ubuntu 4.3.4-5ubuntu1) 4.3.4
Copyright (C) 2008 Free Software Foundation, Inc..
This is free software, see the source for copying Conditions. There is NO warranty
, not even for MERCHANTABILITY or FITNESS FOR A Particular Purpose. 2) are still required libXmu-dev packages libxi-dev freeglut3 freeglut3-dev. 3) We can now build the project. cd ~ / Projects / NVIDIA_GPU_Computing_SDK / C make If all is well remembered that there should be no compilation errors and you can run a basic test programs for the presence of miracles capable device. Daniel @ daniel
-laptop: ~ / Projects / NVIDIA_GPU_Computing_SDK / C / bin / linux / release $. / DeviceQuery CUDA Device Query (Runtime API) version (CUDART static linking) There is 1 device supporting CUDA Device 0 : "GeForce 8400M GS"
CUDA Driver Version: 2.30 CUDA Runtime Version: 2.30 CUDA Capability Major revision number: 1 Minor CUDA Capability revision number: 1 Total amount of global memory: 267714560 bytes Number of multiprocessors: 2 Number of cores: 16 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: Yes

Integrated: No

Support host page-locked memory mapping: No

Compute mode: Default (multiple host threads can use this device simultaneously)

Test PASSED
oraz

Running on......

device 0:GeForce 8400M GS
Quick Mode
Host to Device Bandwidth for Pageable memory
.

Transfer Size (Bytes) Bandwidth(MB/s) 33554432 1473.1
Quick Mode

Device to Host Bandwidth for pageable memory
.

Transfer Size (Bytes) Bandwidth (MB / s)

33554432 898.1 Quick Mode


Device to Device Bandwidth
.
Transfer Size (Bytes) Bandwidth (MB / s) 33554432 4197.4

& & & & Test PASSED

Woodbury Outlet And Chanel

CUDA 2.3 on Ubuntu 9.10 Part. 1

Another approach to the use of the CUDA on Ubuntu. After the trouble I had with the NVIDIA drivers on Ubuntu 8.04 I decided to move to the latest edition of 9.10 (the flash does not respond to clicks, the computer does not detect that the headphones plugged in - do not hide the fact that a bit of a disappointment). By the way I did upgrade from XP to Windows 7.
latest drivers (version 190) installed with the help of

guide from Ubuntu Geek
. In short:

sudo apt-add-ppa repository: nvidia-vdpau/ppa
sudo apt-key adv - keyserver.ubuntu.com keyserver - recv-keys CEC06767
sudo apt-get update sudo
apt-get install nvidia-190-modaliases nvidia-glx-190 nvidia-settings-190

It should be further activated in the System -> Administration -> Drivers.

Now you will need Cuda Toolkit 2.3 - there is no version for Ubuntu 9.10, so I chose the version for 9.04. Installation instructions smoothly. Then I installed the CUDA SDK in version 2.3 for Ubuntu 9.10.

The next step was to add paths to the wonders toolkit environment variables. In the file ~ /. Bashrc, I added:

if [-d "/ usr / local / cuda"] then
PATH = "/ usr / local / cuda: $ PATH"


fi if [-d "/ usr / local/cuda/lib64 "] then
LD_LIBRARY_PATH =" / usr/local/cuda/lib64: $ LD_LIBRARY_PATH "


fi if [-d" / usr / local / cuda / bin "] then PATH =
"/ usr / local / cuda / bin: $ PATH" fi

first if statement checks whether there is / usr / local / cuda, and if so it adds this path to the PATH system variable. Another f adds the appropriate path to the libraries for 64-bit system. Then added a directory with files implementing in order to use the command nvcc without typing the entire path.
I'm not sure whether to put this script in the right file - if I remember correctly it is called each time the console. It would be an added to it only once at startup, and so have left over. After several starts, PATH looks like this:

daniel @ daniel-laptop: ~ $ printenv PATH

/ usr / local / cuda / bin: / usr / local / cuda: / usr / local / sbin: / usr / local / bin: / usr / sbin: / usr / bin: / sbin: / bin: / usr / games

So far everything went smoothly and the real problems began.