GPU is not used on Jetson TX2

[ingenieroariel]

My machine has 8GB of RAM and CUDA 8 on Ubuntu 16.04 and I've set the following env vars:

LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/usr/lib/jvm/java-8-openjdk-arm64/jre/lib/aarch64/server:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/usr/local/mapd-deps/lib:$LD_LIBRARY_PATH

When running mapd the following error appears in the console:

E0511 11:10:42.262192 16533 MapDHandler.cpp:282] No GPUs detected, falling back to CPU mode

Note: The web based query editor / chart generator is amazing. Here is a screenshot with uname -a and the warning I see:

[ingenieroariel]

Here is what deviceQuery[1] from CUDA sample says:

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

Detected 1 CUDA Capable device(s)

Device 0: "GP10B"
  CUDA Driver Version / Runtime Version          8.5 / 8.0
  CUDA Capability Major/Minor version number:    6.2
  Total amount of global memory:                 7854 MBytes (8235356160 bytes)
  ( 2) Multiprocessors, (128) CUDA Cores/MP:     256 CUDA Cores
  GPU Max Clock rate:                            1301 MHz (1.30 GHz)
  Memory Clock rate:                             13 Mhz
  Memory Bus Width:                              64-bit
  L2 Cache Size:                                 524288 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 number of registers available per block: 32768
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  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 1 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 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 = 8.5, CUDA Runtime Version = 8.0, NumDevs = 1, Device0 = GP10B
Result = PASS

[1]
https://gist.github.com/ingenieroariel/c4e8e1299be58a5b852d91d85ba7da24

[andrewseidl]

The most likely cause is that the GPU detection in startmapd isn't properly handling the Jetson. This detection is actually not required anymore now that mapd_server can handle it, so I've removed it.

Could you do a git pull to pick up the new version of startmapd to see if that helps? Commit f694c47.

---

A second potential cause is not having permission to access the GPU devices, which I've seen happen in older versions of L4T if not using the default user. This is probably not it in your case since the samples work and you're using the default nvidia user.

[ingenieroariel]

I get no warnings now. Thanks!