The goal of this work is to use TTile in Apache TVM.
The work is done on x86 and mppa. Please look at the x86 and mppa branches and in particular the TTile folder.
git clone --recursive [email protected]:S12P/tvm_ttile.git
git submodule init
git submodule update
sudo apt-get update
sudo apt-get install -y python3 python3-dev python3-setuptools gcc libtinfo-dev zlib1g-dev build-essential cmake libedit-dev libxml2-dev
mkdir build
cp cmake/config.cmake build
cd build
cmake ..
make -j4
export TVM_HOME=<PATH_TO_TVM_FOLDER>
export PYTHONPATH=$TVM_HOME/python:${PYTHONPATH}
Please see https://tvm.apache.org/docs/install/from_source.html for more details.
Documentation | Contributors | Community | Release Notes
Apache TVM (incubating) is a compiler stack for deep learning systems. It is designed to close the gap between the productivity-focused deep learning frameworks, and the performance- and efficiency-focused hardware backends. TVM works with deep learning frameworks to provide end to end compilation to different backends.
© Contributors Licensed under an Apache-2.0 license.
TVM adopts apache committer model, we aim to create an open source project that is maintained and owned by the community. Checkout the Contributor Guide
We learned a lot from the following projects when building TVM.
- Halide: Part of TVM's TIR and arithmetic simplification module originates from Halide. We also learned and adapted some part of lowering pipeline from Halide.
- Loopy: use of integer set analysis and its loop transformation primitives.
- Theano: the design inspiration of symbolic scan operator for recurrence.