[Inference/Feat] Add convert_fp8 op for fp8 test in the future (#5706)

* add convert_fp8 op for fp8 test in the future * rerun ci
2025-09-23 18:39:56 +00:00 · 2024-05-10 18:39:54 +08:00
parent bfad39357b
commit 50104ab340
5 changed files with 197 additions and 10 deletions
--- a/extensions/pybind/inference/inference_ops_cuda.py
+++ b/extensions/pybind/inference/inference_ops_cuda.py
@@ -17,6 +17,7 @@ class InferenceOpsCudaExtension(_CudaExtension):
                "kernel/cuda/rms_layernorm_kernel.cu",
                "kernel/cuda/get_cos_and_sin_kernel.cu",
                "kernel/cuda/flash_decoding_attention_kernel.cu",
+                "kernel/cuda/convert_fp8_kernel.cu",
            ]
        ] + [self.pybind_abs_path("inference/inference.cpp")]
        return ret