How to enable DLA?

Autonomous Vehicles DRIVE AGX Orin DRIVE AGX Orin General
1

Please provide the following info (tick the boxes after creating this topic):
Software Version
DRIVE OS 6.0.10.0
[1] DRIVE OS 6.0.8.1
DRIVE OS 6.0.6
DRIVE OS 6.0.5
DRIVE OS 6.0.4 (rev. 1)
DRIVE OS 6.0.4 SDK
other

Target Operating System
[1] Linux
QNX
other

Hardware Platform
DRIVE AGX Orin Developer Kit (940-63710-0010-300)
DRIVE AGX Orin Developer Kit (940-63710-0010-200)
DRIVE AGX Orin Developer Kit (940-63710-0010-100)
DRIVE AGX Orin Developer Kit (940-63710-0010-D00)
DRIVE AGX Orin Developer Kit (940-63710-0010-C00)
[1] DRIVE AGX Orin Developer Kit (not sure its number)
other

SDK Manager Version
2.1.0
[1] other

Host Machine Version
native Ubuntu Linux 20.04 Host installed with SDK Manager
native Ubuntu Linux 20.04 Host installed with DRIVE OS Docker Containers
native Ubuntu Linux 18.04 Host installed with DRIVE OS Docker Containers
other

Issue Description
Hi team, I have a little confusion about DLA.

  1. How to enable DLA in sample dnn tensor?
    I executed this code but i am getting error as “terminate called recursively”.
    Please help
3

Dear @akshay.tupkar ,
May I know the used command?
Can you try ./sample_dnn_tensor --cudla=1 --offscreen=1 --dla-engine=0 ?

4

Dear @SivaRamaKrishnaNV
I used the command
./sample_dnn_tensor --input-type=video --video=video_0.h264 --tensorRT_model=tensorRT_model.dla.bin --cudla=1 --offscreen=0 --dla-engine=0

The program executes successfully, but i noticed the video gets slowed down and the speed of inference is quite low. Why is this happening? I am attaching the video output.

5

Dear @akshay.tupkar ,
Note that iGPU is powerful compared to DLA. Also, FYI, When a model is executed on DLA, if there is a non supported layer, it gets executed on GPU. In such cases, some intermediate data transfers can effect overall performance.

2 Likes
6

Okay got it @SivaRamaKrishnaNV
Thanks for support

1 Like