🐛 Bug

Memory usage increasing steadily when repeatedly calling forward.

To Reproduce

import torch
import torch.nn as nn
import torch.nn.functional as F

class Net(torch.jit.ScriptModule):

    def __init__(self):
        super(Net, self).__init__()

        self.conv = nn.Conv2d(1, 1, 1)

        self.lin = nn.Linear(100, 1)

    @torch.jit.script_method
    def forward(self, x):
        x = F.relu(self.conv(x))
        x = x.view(x.shape[0], -1)
        x = F.relu(self.lin(x))

        return x

m = Net()

x = torch.ones((1, 1, 10, 10))

for i in range(1000000):
    m(x)

Expected behavior

Environment

PyTorch version: 1.1.0
Is debug build: No
CUDA used to build PyTorch: None

OS: Ubuntu 18.04.1 LTS
GCC version: (Ubuntu 7.3.0-16ubuntu3) 7.3.0
CMake version: version 3.10.2

Python version: 3.6
Is CUDA available: No
CUDA runtime version: No CUDA
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA

Versions of relevant libraries:
[pip3] numpy==1.14.2
[pip3] numpydoc==0.7.0
[pip3] torch==1.1.0
[pip3] torchvision==0.3.0
[conda] blas 1.0 mkl
[conda] mkl 2018.0.1 h19d6760_4
[conda] mkl-dnn 0.14 2 intel
[conda] mkl-service 1.1.2 py36h17a0993_4
[conda] pytorch-cpu 1.1.0 py3.6_cpu_0 pytorch
[conda] torchvision-cpu 0.3.0 py36_cuNone_1 pytorch

Additional context

Used memory_profiler to record the memory usage by mprof run script.py. Results are shown below.