DLA submit failed when using enqueueV2. TRT7.1.3

Autonomous Machines Jetson & Embedded Systems Jetson AGX Xavier
, ,
1

When we are inferring our unpublic model using enqueueV2 function on DLACore0, we got the err below:

NVMEDIA_DLA : 1928, ERROR: Submit failed.
../rtExt/dla/native/dlaUtils.cpp (194) - DLA Error in submit: 7 (Failure to submit program to DLA engine.)
FAILED_EXECUTION: std::exception
NVMEDIA_DLA :  885, ERROR: runtime registerEvent failed. err: 0x4.
NVMEDIA_DLA : 1849, ERROR: RequestSubmitEvents failed. status: 0x7.
../rtExt/dla/native/dlaUtils.cpp (194) - DLA Error in submit: 7 (Failure to submit program to DLA engine.)
FAILED_EXECUTION: std::exception
NVMEDIA_DLA :  885, ERROR: runtime registerEvent failed. err: 0x4.
NVMEDIA_DLA : 1849, ERROR: RequestSubmitEvents failed. status: 0x7.
../rtExt/dla/native/dlaUtils.cpp (194) - DLA Error in submit: 7 (Failure to submit program to DLA engine.)
FAILED_EXECUTION: std::exception
...

After we change the function to executeV2, the model is inferring with no problem.
Is there some differents between the double functions in the low-level cause the error?

3

Hi,

The main difference between execute() and executeV2() is the batchsize.
You can find a similar issue below:
https://github.com/NVIDIA/TensorRT/issues/489#issuecomment-612742601

Thanks.

4

Thanks for your reply.
But the functions witch i’m puzzling are enqueueV2 and executeV2, and got the error by enqueueV2.

5

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

Hi,

Sorry about that.
To check executeV2 and enqueueV2 issue, would you mind to share the model and reproducible source with us?

Thanks.