[08/23/2022-11:24:58] [TRT] [E] C:\source\rtSafe\safeRuntime.cpp (32) - Cuda Error in nvinfer1::internal::DefaultAllocator::free: 1 (invalid argument)

[jiaqizhang123-stack]

Hello, when I use tensorrt7.2.3.4, the errors were occured [08/23/2022-11:24:58] [TRT] [E] C:\source\rtSafe\safeRuntime.cpp (32) - Cuda Error in nvinfer1::internal::DefaultAllocator::free: 1 (invalid argument) I don't know why, thanks for your answer

[Henry-Avery]

I have also meet this problem，maybe you can check your cuda version

[zerollzeng]

@oxana-nvidia ^ ^

[oxana-nvidia]

@Henry-Avery @jiaqizhang123-stack Could you please share more details about your use case? And a repro if possible. Not clear what you are trying to do. TRT 7.2 is not safety certified version and should not be used for safety applications. But your code fails when trying to use safety runtime. You should be using TRT 6.5 for safety applications or move to a newer version TRT 8.4.11 https://developer.nvidia.com/docs/drive/drive-os/latest/tensorrt/index.html

[ttyio]

closing since no activity for more than 14 days, please reopen if you still have question, thanks!