62835a0979
This PR adds an automatic memory alignment mechanism in the Triton runtime. Specifically, the JIT compiler detects the alignment (in bytes) of each pointer argument as well as the largest power of two divisor (between 1 and 16) of each integer argument. Proper .aligned and .multipleof attributes are then added to the Triton-IR on-the-fly for all auto-tunable kernels. There is a cache that remembers all the kernels compiled for each possible configuration. This PR also includes substantial cleaning of the Python API. This adds 2-3us overhead, mostly due to accessing integer #defines from the auto-tuned compilation options. The previous solution was slightly faster but hacky and potentially unsafe, so this is preferred for now.
Triton
This is the development repository of Triton, a language and compiler for writing highly efficient custom Deep-Learning primitives. The aim of Triton is to provide an open-source environment to write fast code at higher productivity than CUDA, but also with higher flexibility than other existing DSLs.
The foundations of this project are described in the following MAPL2019 publication: Triton: An Intermediate Language and Compiler for Tiled Neural Network Computations. Please consider citing us if you use our work!
Installation
You can install the latest release with pip as follows:
sudo apt-get install llvm-10-dev
pip install triton
or the latest development version with:
pip install -e "git+https://github.com/ptillet/triton.git#egg=triton&subdirectory=python"
for the C++ package:
git clone https://github.com/ptillet/triton.git;
cd triton;
mkdir build;
cd build;
cmake ../;
make -j8;