Retype stack slots that cache a pointer to an allocation moved to the stack When LowerAllocAddr moves an allocation back to the stack, a pointer derived from it may have been stored into a reverse-pass cache, for example a phi that cannot be recomputed. RecursivelyReplaceAddressSpace only handled stores through the moved pointer, so such a cache store failed with "Illegal address space propagation". Casting the stack pointer back to a tracked address space would put a stack address where the GC looks for objects. Instead, when the slot is an alloca that only holds this value (plus constant stores), replace it with an alloca of the new pointer type and follow each load from it as a value derived from the moved allocation. This is what broke Enzyme.jl's Flux LSTM/GRU integration tests on Julia 1.11. It replaces the earlier enzyme_fromstack guard on the null-check exception from #3266, which is reverted here. Assisted-by: Claude Code (Opus 5.5)
Enzyme is a plugin that performs automatic differentiation (AD) of statically analyzable LLVM and MLIR.
Enzyme can be used by calling __enzyme_autodiff on a function to be differentiated as shown below. Running the Enzyme transformation pass then replaces the call to __enzyme_autodiff with the gradient of its first argument.
double foo(double); double grad_foo(double x) { return __enzyme_autodiff(foo, x); }
Enzyme is highly-efficient and its ability to perform AD on optimized code allows Enzyme to meet or exceed the performance of state-of-the-art AD tools.
Detailed information on installing and using Enzyme can be found on our website: https://enzyme.mit.edu.
A short example of how to install Enzyme is below:
cd /path/to/Enzyme/enzyme mkdir build && cd build cmake -G Ninja .. -DLLVM_DIR=/path/to/llvm/lib/cmake/llvm -DLLVM_EXTERNAL_LIT=/path/to/lit/lit.py ninja
Or, install Enzyme using a package manager:
brew install enzyme
spack install enzyme
nix-shell -p enzyme
To get involved or if you have questions, please join our mailing list.
If using this code in an academic setting, please cite the following three papers (first for Enzyme as a whole, second for GPU+optimizations, and third for AD of all other parallel programs (OpenMP, MPI, Julia Tasks, etc.)):
@inproceedings{NEURIPS2020_9332c513,
author = {Moses, William and Churavy, Valentin},
booktitle = {Advances in Neural Information Processing Systems},
editor = {H. Larochelle and M. Ranzato and R. Hadsell and M. F. Balcan and H. Lin},
pages = {12472--12485},
publisher = {Curran Associates, Inc.},
title = {Instead of Rewriting Foreign Code for Machine Learning, Automatically Synthesize Fast Gradients},
url = {https://proceedings.neurips.cc/paper/2020/file/9332c513ef44b682e9347822c2e457ac-Paper.pdf},
volume = {33},
year = {2020}
}
@inproceedings{10.1145/3458817.3476165,
author = {Moses, William S. and Churavy, Valentin and Paehler, Ludger and H\"{u}ckelheim, Jan and Narayanan, Sri Hari Krishna and Schanen, Michel and Doerfert, Johannes},
title = {Reverse-Mode Automatic Differentiation and Optimization of GPU Kernels via Enzyme},
year = {2021},
isbn = {9781450384421},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3458817.3476165},
doi = {10.1145/3458817.3476165},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
articleno = {61},
numpages = {16},
keywords = {CUDA, LLVM, ROCm, HPC, AD, GPU, automatic differentiation},
location = {St. Louis, Missouri},
series = {SC '21}
}
@inproceedings{10.5555/3571885.3571964,
author = {Moses, William S. and Narayanan, Sri Hari Krishna and Paehler, Ludger and Churavy, Valentin and Schanen, Michel and H\"{u}ckelheim, Jan and Doerfert, Johannes and Hovland, Paul},
title = {Scalable Automatic Differentiation of Multiple Parallel Paradigms through Compiler Augmentation},
year = {2022},
isbn = {9784665454445},
publisher = {IEEE Press},
booktitle = {Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis},
articleno = {60},
numpages = {18},
keywords = {automatic differentiation, tasks, OpenMP, compiler, Julia, parallel, Enzyme, C++, RAJA, hybrid parallelization, MPI, distributed, LLVM},
location = {Dallas, Texas},
series = {SC '22}
}
Julia bindings, Rust bindings, and Fortran bindings are available for Enzyme.