Fix compilation with gcc 11.3 (#192)

triton-inference-server / dali_backend

The Triton backend that allows running GPU-accelerated data pre-processing pipelines implemented in DALI's python API.

MIT License

120 stars 28 forks source link

Closed mc-nv closed 1 year ago

mc-nv commented 1 year ago

updates Catch2 to 2.13.9
adds -Wno-free-nonheap-object compilation flag as the compiler gets confused with SmallVector memory management
initializes dev_id variable