Clean up llama csrc code base

Summary: - Clean up quantization ops - Clean up GQA ops - Change some APIs Reviewed By: sryap, jiawenliu64 Differential Revision: D56781779 fbshipit-source-id: ffddc66ceee5b0939f2b5155c1f69b076b381df5
pytorch · May 5, 2024 · bc4e9c3 · bc4e9c3
1 parent a40afe5
commit bc4e9c3
Showing 1 changed file with 0 additions and 10 deletions.
diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/cublas_utils.h b/fbgemm_gpu/experimental/gen_ai/src/quantize/cublas_utils.h
@@ -20,13 +20,3 @@ inline void checkCublasStatus(cublasStatus_t status) {
     throw std::logic_error("cuBLAS API failed");
   }
 }
-
-inline void checkCudaStatus(cudaError_t status) {
-  if (status != cudaSuccess) {
-    printf(
-        "cuda API failed with status %d: %s\n",
-        status,
-        cudaGetErrorString(status));
-    throw std::logic_error("cuda API failed");
-  }
-}