Commit Graph

38 Commits

Author SHA1 Message Date
rudy
48bf6e2696 feat(optimizer): report or warn using global p-error 2022-09-12 17:22:38 +02:00
Quentin Bourgerie
30df69366d enhance(python-bindings): Expose the option to set or not the v0 strategy of the optimizer. 2022-09-01 10:37:00 +02:00
Quentin Bourgerie
9257404f5f fix(python-bindings): Support np.array with dtype upt to 64 bits 2022-08-19 14:15:21 +02:00
Andi Drebes
45577fb79e Rebase onto llvm-project f69328049e9e with local changes
This commit rebases the compiler onto commit f69328049e9e from
llvm-project.

Changes:

* Use of the one-shot bufferizer for improved memory management

* A new pass `OneShotBufferizeDPSWrapper` that converts functions
  returning tensors to destination-passing-style as required by the
  one-shot bufferizer

* A new pass `LinalgGenericOpWithTensorsToLoopsPass` that converts
  `linalg.generic` operations with value semantics to loop nests

* Rebase onto a fork of llvm-project at f69328049e9e with local
  modifications to enable bufferization of `linalg.generic` operations
  with value semantics

* Workaround for the absence of type propagation after type conversion
  via extra patterns in all dialect conversion passes

* Printer, parser and verifier definitions moved from inline
  declarations in ODS to the respective source files as required by
  upstream changes

* New tests for functions with a large number of inputs

* Increase the number of allowed task inputs as required by new tests

* Use upstream function `mlir_configure_python_dev_packages()` to
  locate Python development files for compatibility with various CMake
  versions

Co-authored-by: Quentin Bourgerie <quentin.bourgerie@zama.ai>
Co-authored-by: Ayoub Benaissa <ayoub.benaissa@zama.ai>
Co-authored-by: Antoniu Pop <antoniu.pop@zama.ai>
2022-06-14 14:35:25 +02:00
youben11
1354759884 chore: rename refs to main branch 2022-06-09 09:19:06 +01:00
youben11
5636342097 feat: add option to enable/disable optimization 2022-06-01 14:35:12 +01:00
Umut
b052157fae refactor: separate runtime context from public arguments 2022-05-30 10:19:14 +03:00
youben11
5f1a539505 feat(python): get path to diff artifacts
- path to client parameters file
- path to shared library
2022-05-09 16:18:21 +01:00
youben11
211241fcb2 refactor: store compil artifacts under a single dir 2022-05-09 16:18:21 +01:00
youben11
f223f02ab7 feat: parameterize artifact generation in lib compilation 2022-05-09 16:18:21 +01:00
rudy
1b70a38754 feat(optimizer): expose the p_error parameter
simplify the handling of options by relying more on CompilationOptions
2022-05-06 13:06:08 +02:00
youben11
843dd0eb5b feat(python): seriliaze client parameters 2022-05-02 10:16:39 +02:00
youben11
e9345b3859 feat: serialize public result w/ python bindings 2022-04-22 15:53:26 +01:00
youben11
0a5881096c feat(python): serialize public arguments 2022-04-15 15:25:25 +01:00
youben11
ec7d1232f1 feat: add runtimeLib link for library compilation
runtime library will be linked to compiled library
2022-04-04 11:52:23 +01:00
youben11
78def04fe5 refactor: rename [jit|library]lambdasupport to [jit|library]support 2022-04-04 09:15:31 +01:00
youben11
999ab4e5ea refactor: recreate the Python API using wrappers
All Cpp objects are now wrapped, and calls are being forwarded after
strict type checking for avoiding weird behaviors.
The same Cpp API is now exposed to Python
2022-04-04 09:15:31 +01:00
youben11
ad79aa627f fix: don't pass ref to runtimeLibPath
We were keeping a reference to the path which led to its use after the
string was freed
2022-03-28 13:42:18 +01:00
Quentin Bourgerie
c70ef1dcda fix(support): Use shared_ptr intead of raw ptr on JitLambdaSupport that allows the JitLambda to be used after that the compilation result is freed 2022-03-24 11:00:36 +01:00
Quentin Bourgerie
1620259807 cleanup(capi/python-bindings): Remove reference to JitCompilerEngine 2022-03-24 11:00:36 +01:00
Quentin Bourgerie
5b83b700d2 enhance(compiler): Expose a compilation options instead of just the funcname 2022-03-24 11:00:36 +01:00
Quentin Bourgerie
8867d313ee feat(python): Expose Jit and Library compiler support 2022-03-24 11:00:36 +01:00
youben11
18f0ad77e7 fix: use new dfr api 2022-03-17 16:15:02 +01:00
youben11
44ebd426f9 feat: setup init/termination of parallel execution in python 2022-03-17 16:15:02 +01:00
youben11
5b37ec640c feat: support parallelization in python 2022-03-17 16:15:02 +01:00
youben11
2009ee1c94 chore: changing ref to repo after its renaming
also formatting and update the check_license script to match that
2022-01-06 12:40:10 +01:00
youben11
f1161f7f6d chore: formatting
Quick fix due to ordering of includes, had to add #include
<mlir/Transforms/DialectConversion.h> to include/concretelang/Conversion/Utils/GenericOpTypeConversionPattern.h
2022-01-03 09:29:19 +01:00
youben11
940cb96be4 chore: rename dialects
HLFHE to FHE
MidLFHE to TFHE
LowLFHE to Concrete
2021-12-29 15:13:34 +01:00
youben11
e73291abdc chore: rename compiler to concrete-compiler
zamalang => concretelang
zamacompiler => concretecompiler
2021-12-29 15:13:34 +01:00
youben11
971cb56182 chore: licensing 2021-12-29 15:13:34 +01:00
youben11
60b2cfd9b7 feat: support more dtype for scalars/tensors
dtype supported now: uint8, uint16, uint32, uint64
2021-12-13 17:40:28 +01:00
Mayeul@Zama
dad4390518 feat(compiler): add a key cache 2021-12-07 12:11:56 +01:00
rudy
975ee86a5e feat(compiler): add --compile option
Resolves #196
2021-11-30 17:42:28 +01:00
youben11
99cce18c6a feat: get RT lib path from py and use as sharedlib
Try to find the runtime library automatically (should only work on
proper installation of the package), and fail silently by not passing
any RT lib. The RT lib can also be specified manually. The RT lib will
be used as a shared library by the JIT compiler.
2021-11-12 15:06:49 +01:00
youben11
59e859177c refactor: replace ExecutionArg by TensorLambdaArg
This add support for tensor inputs from the python bindings
2021-11-08 11:55:02 +01:00
youben11
b501e3d6c0 feat(python): support functions returning tensors 2021-11-08 11:55:02 +01:00
Andi Drebes
1187cfbd62 refactor(compiler): Refactor CompilerEngine and related classes
This commit contains several incremental improvements towards a clear
interface for lambdas:

  - Unification of static and JIT compilation by using the static
    compilation path of `CompilerEngine` within a new subclass
    `JitCompilerEngine`.

  - Clear ownership for compilation artefacts through
    `CompilationContext`, making it impossible to destroy objects used
    directly or indirectly before destruction of their users.

  - Clear interface for lambdas generated by the compiler through
    `JitCompilerEngine::Lambda` with a templated call operator,
    encapsulating otherwise manual orchestration of `CompilerEngine`,
    `JITLambda`, and `CompilerEngine::Argument`.

  - Improved error handling through `llvm::Expected<T>` and proper
    error checking following the conventions for `llvm::Expected<T>`
    and `llvm::Error`.

Co-authored-by: youben11 <ayoub.benaissa@zama.ai>
2021-10-29 13:44:34 +02:00
youben11
2972fa4403 refactor(python): rework the bindings with latest MLIR version
- Go through CAPI for python bindings
- Consuming LLVM errors in CAPI: fixes previous issue which made this
  impossible in the python bindings
2021-10-07 14:38:50 +01:00