From 950cde13f7ac0d7432392d5f6ae11a00655a82c3 Mon Sep 17 00:00:00 2001 From: lucasdelimanogueira Date: Wed, 22 May 2024 20:59:25 -0300 Subject: [PATCH] small fix sum cuda --- build/cuda.cu.o | Bin 166768 -> 166768 bytes norch/csrc/cuda.cu | 2 +- norch/csrc/tensor.cpp | 1 - norch/libtensor.so | Bin 220544 -> 220544 bytes 4 files changed, 1 insertion(+), 2 deletions(-) diff --git a/build/cuda.cu.o b/build/cuda.cu.o index a60bfa824dbe46bca8440778da4b8210a02983cd..94ea9cd850c1cf122caa0aa21805c86cc4f17426 100644 GIT binary patch delta 23 ecmew`jqAfSu7(!IElje~%qEE`+ZCjl0tx_aTL;7d delta 23 ecmew`jqAfSu7(!IElje~%tjW5+ZCjl0tx_XoCjtA diff --git a/norch/csrc/cuda.cu b/norch/csrc/cuda.cu index 8102e7c..46af278 100644 --- a/norch/csrc/cuda.cu +++ b/norch/csrc/cuda.cu @@ -194,7 +194,7 @@ __host__ void sum_tensor_cuda(Tensor* tensor, float* result_data, int axis) { } cudaDeviceSynchronize(); - + } else { int axis_size = tensor->shape[axis]; diff --git a/norch/csrc/tensor.cpp b/norch/csrc/tensor.cpp index 1f374f8..1fc95e1 100644 --- a/norch/csrc/tensor.cpp +++ b/norch/csrc/tensor.cpp @@ -211,7 +211,6 @@ extern "C" { float* result_data; cudaMalloc((void**)&result_data, size * sizeof(float)); - cudaMemset(result_data, 0, size * sizeof(float)); sum_tensor_cuda(tensor, result_data, axis); if (keepdim) { diff --git a/norch/libtensor.so b/norch/libtensor.so index d5ff2a22ca085e02c213e9795b8508a103ab1fc2..6074e466862319fc5fb5ca661386435d9ccc209e 100755 GIT binary patch delta 25 fcmZqJ&D#J(EsQNpTbL&qFq