Benson Ma

Hi @jeanschmidt It looks iike some of the runners are not available. Since the PR was made 3 months ago Is it still relevant, or should we close it out?

Hi @cyyever sorry for the delay, there have been some internal issues preventing the PR from being imported into our internal systems, but we are working to import and land...

> Hmm, 5.6 test is failing: `Build x86 Linux Wheels / pytorch/FBGEMM / manywheel-py3_8-rocm5_6` > > ``` > /__w/FBGEMM/FBGEMM/pytorch/FBGEMM/fbgemm_gpu/src/split_embeddings_utils.hip:27:10: fatal error: 'rocm-core/rocm_version.h' file not found > #include > ``` >...

> @q10 Do you know how to disable it? Is it something @liligwu can just remove 5.6 from one of the GHA yml files? It'd be good to disable this...

> HI @q10 looks good. Can you please provide a reference how you use this workflow ? and some testing plan ? @atalman https://github.com/pytorch/FBGEMM/pull/3890 shows an example of how I...

Hi @ywangwxd `batched_unary_embeddings_test.py` is known to be broken at the moment (see https://github.com/pytorch/FBGEMM/issues/1559) and we haven't had the bandwidth to fix it yet. It is currently ignored in the CI...

Hi @ywangwxd Generally, the installation of FBGEMM can be validated by running `python -c "import fbgemm_gpu"` after install. Our CI already runs the test suites for each PR for the...

Hi @ywangwxd This error usually indicates that you're running on a CUDA hardware model for which we did not compile the FBGEMM code for. We generally compile FBGEMM for SM...

@cyyever looks good to me, hopefully it can be landed soon

Hi @zhuhaozhe could you let us know what is the PyTorch version you are using? It is likely that you have a non-nightly version of PyTorch installed. We will make...