Loading megatron/fused_kernels/layer_norm_cuda_kernel.cu +2 −0 Original line number Diff line number Diff line Loading @@ -645,6 +645,8 @@ void cuComputeGradInput( k_grad_input[l] = static_cast<T>(f_grad_input); } } // prevent race where buf is written again before reads are done __syncthreads(); } } Loading Loading
megatron/fused_kernels/layer_norm_cuda_kernel.cu +2 −0 Original line number Diff line number Diff line Loading @@ -645,6 +645,8 @@ void cuComputeGradInput( k_grad_input[l] = static_cast<T>(f_grad_input); } } // prevent race where buf is written again before reads are done __syncthreads(); } } Loading
