Wei (Neil) Su
Wei (Neil) Su
Hi, FBGEMM-GPU currently needs to be built with GCC and does not support clang. Can you try installing gcc and g++ and running the setup again?
I think the `cpu_float2half_rn` function is a reference implementation that intentionally implement the algorithm manually. Currently we rely on the compiler to do the optimized CPU float conversion (see line...
Hi @vrnmthr, I would expect the FBGEMM main library to build on ARM sometime in 2024 H2. For now, I would suggest building the cpu-only version of FBGEMM-GPU and using...