Device Info Query with CUDA | Xavier

Device Info Query

In this post, I show how can query to device information with Cuda. In this post, I will use the NVDIA Jetson Xavier Device. However, the below code can run any compatible device with CUDA.

Xavier Specification is:

Device 0Xavier
CUDA Driver Version / Runtime Version10.0 / 10.0
CUDA Capability Major/Minor version number7.2
Total amount of global memory15.32 MBytes (16454500352 bytes)
GPU Clock rate1500 MHz (1.50 GHz)
Memory Clock rate1377 Mhz
Memory Bus Width256-bit
L2 Cache Size524288 bytes
Max Texture Dimension Size (x)1D=(131072)
Max Texture Dimension Size (x,y)2D=(131072,65536)
Max Texture Dimension Size (x,y,z)3D=(16384,16384,16384)
Max Layered Texture Size (dim) x layers (1D)1D=(32768) x 2048
Max Layered Texture Size (dim) x layers (2D)2D=(32768,32768) x 2048
Total amount of constant memory65536 bytes
Total amount of constant memory64.00 KB
Total amount of shared memory per block49152 bytes
Total amount of shared memory per block48.00 KB
Total number of registers available per block65536
Warp size32
Maximum number of threads per multiprocessor2048
Number of multiprocessors4
Maximum number of threads per block1024
Maximum number of warps per multiprocessor64
Maximum sizes of each dimension of a block1024 x 1024 x 64
Maximum sizes of each dimension of a grid2147483647 x 65535 x 65535
Maximum memory pitch2147483647 bytes

Output of code is

Device 0: “Xavier”
CUDA Driver Version / Runtime Version 10.0 / 10.0
CUDA Capability Major/Minor version number: 7.2
Total amount of global memory: 15.32 MBytes (16454500352 bytes)
GPU Clock rate: 1500 MHz (1.50 GHz)
Memory Clock rate: 1377 Mhz
Memory Bus Width: 256-bit
L2 Cache Size: 524288 bytes
Max Texture Dimension Size (x,y,z) 1D=(131072), 2D=(131072,65536), 3D=(16384,16384,16384)
Max Layered Texture Size (dim) x layers 1D=(32768) x 2048, 2D=(32768,32768) x 2048
Total amount of constant memory: 65536 bytes
Total amount of constant memory: 64.00 KB
Total amount of shared memory per block: 49152 bytes
Total amount of shared memory per block: 48.00 KB
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Number of multiprocessors: 4
Maximum number of threads per block: 1024
Maximum number of warps per multiprocessor: 64
Maximum sizes of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 2147483647 x 65535 x 65535
Maximum memory pitch: 2147483647 bytes

Leave a Reply

Your email address will not be published. Required fields are marked *