Nvidia Tesla S2050 Installation Manual page 23

18
04
dRIVER INSTALLATION
2
Make sure that the following locations of the library files and NVIdIA
compiler are included in your $PATH. The .bashrc file may need to be
modified if the $PATH does not point to the correct files.
These are assuming the default locations were used.
PATH=$PATH:/usr/local/cuda/bin
export PATH
Ld_LIbRARY_PATH=/usr/local/cuda/lib
export Ld_LIbRARY_PATH
3
Go to the directory or folder that contains the CUdA SdK samples. In
Linux, the default path is: /root/NVIdIA_CUdA_SdK
4
Compile the SdK samples by typing make make from this working
directory.
5
As root, run the sample file called deviceQuery. The user may need
to change to the directory where the NVIdIA SdK files are located.
Usually these are located at /NVIdIA_CUdA_SdK/bin/linux/release.
for Tesla systems, there are two GPUs per PCI Express connection.
Only two devices (device 0 and device 1) are shown. Output should
look something like the following:
device 0:
Major revision number:
Minor revision number:
Total amount of global memory:
Total amount of constant memory:
Total amount of shared memory per block:
Total number of registers available per block:
Warp size:
Maximum number of threads per block:
Maximum sizes of each dimension of a block:
Maximum sizes of each dimension of a grid:
Maximum memory pitch:
Texture alignment:
Clock rate:
device 1:
Major revision number:
Minor revision number:
Total amount of global memory:
Total amount of constant memory:
Total amount of shared memory per block:
"Tesla 1U GPU"
1
0
4294268245 bytes
65536 bytes
16384 bytes
8192
32
512
512 x 512 x 64
65535 x 65535 x 1
262144 bytes
256 bytes
1440000 kilohertz
"Tesla 1U GPU"
1
0
4294268245 bytes
65536 bytes
16384 bytes