Orin Shared Memory size documentation

Orin is based on the Ampere architecture, and has compute capability 8.7. The CUDA Toolkit tunig guide for ampere only mentions 8.0 and 8.6, specifically for the shared memory size here. The same is also true for the per-compute-capability feature list here. Table 15 on the same page mentions CC 8.7, with 163KB max Shared Memory per block, similar to CC8.0

However DeviceQuery reports a smaller number (~48KB, see below for reference).

I am wondering what is true.

/usr/local/cuda-11.4/samples/1_Utilities/deviceQuery/deviceQuery Starting...

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

Detected 1 CUDA Capable device(s)

Device 0: "Orin"
  CUDA Driver Version / Runtime Version          11.4 / 11.4
  CUDA Capability Major/Minor version number:    8.7
  Total amount of global memory:                 30623 MBytes (32110198784 bytes)
  (016) Multiprocessors, (128) CUDA Cores/MP:    2048 CUDA Cores
  GPU Max Clock rate:                            1300 MHz (1.30 GHz)
  Memory Clock rate:                             1300 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                 4194304 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        167936 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  1536
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            Yes
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Supports MultiDevice Co-op Kernel Launch:      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 0 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 11.4, CUDA Runtime Version = 11.4, NumDevs = 1
Result = PASS

Running device quer


Thanks for reporting this issue.

We are checking this internally.
Will share more information with you later.


48KB is the default limit for each kernel function.

It is possible to Opt-In to a higher limit for a function by calling cudaFuncSetAttribute.
This is a chip-dependent limit that is present in cudaDevAttrMaxSharedMemoryPerBlockOptin.



1 Like

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.