- 03 Jan, 2020 1 commit
-
-
Pruthvi authored
* Affine lowering support for 1) GreaterEq 2) LessEq 3) Equal 4) NotEqual * - cast result op from i1 to i8 for comparision operators * Addressed PR comments * Style fix * - style check - use select instead of zero_extendi durinng CompOp lowering * - fix style - use createOneConstant and createZeroConstant helpers in select intrinsic * Use NG_U8_TYPE_ID for BooleanType in ngraph dialect * Diable CE and Softmax unit test in MLIR * - LIT parser test for comparision ops * - Affine dailect LIT tests for Comparision Ops * Address PR feedback * fix typo * - use `cast` to deduce element Type - add more strict type checking to LIT Test * fix CHECK label's for comparision ops * Use UInt8 in verification logic for CMP op's * - use UInt8 for the resultOp verification Logic in CMP op - fix unit test failures
-
- 01 Jan, 2020 1 commit
-
-
Robert Kimball authored
* Update license to new year * Pick up some strays
-
- 23 Dec, 2019 1 commit
-
-
Diego Caballero authored
* [MLIR] Update MLIR/LLVM repo I need it to get MemRef lowering override mechanism (291a309d). * Fix style Co-authored-by: Scott Cyphers <diyessi@users.noreply.github.com>
-
- 17 Dec, 2019 1 commit
-
-
Diego Caballero authored
This PR fixes some of the inconveniences that we had with lit in nGraph and aligns configuration files with MLIR/LLVM. In particular: - `make install` is not needed to run lit test with the latest compiled version of testing tools (ngraph-opt). - LD_LIBRARY_PATH doesn't need to be modified.
-
- 06 Dec, 2019 1 commit
-
-
Diego Caballero authored
* [MLIR] Update MLIR/LLVM repos * Move MLIR/LLVM repos forward This includes fix to affine fusion algorithm. * Fix issues after merge * Fix lit test
-
- 27 Nov, 2019 2 commits
-
-
Nagy Mostafa authored
* Map each ng tensor to a linear buffer and a view * fix comment * Create views only when a value is assigned a buffer id * style * Fix lit test
-
Nagy Mostafa authored
* WIP * WIP * WIP * All ops * Fix layernorm backprop op name * WIP: Adding tests * WIP: Adding LIT parsing/printing tests * WIP * Added LSTM cells. Fixed some ops * All builder tests * PR fixes * Fix spacing. Add missing setter to SpaceToDepth * Update spaceToDepth lit test * PR fixes * Build fix * Another fix * Fixed optional args
-
- 21 Nov, 2019 1 commit
-
-
Gleb Kazantaev authored
* Added U1 precision for binary weights * Handle switch cases with u1 type * Fixed code style * Added convert_to_string support for u1 type * Use real C type for u1 type. Co-Authored-By: Robert Kimball <robert.kimball@intel.com>
-
- 18 Nov, 2019 2 commits
-
-
Nagy Mostafa authored
* WIP * WIP * WIP * WIP * style * WIP * WIP * Add err msg * Fix headers and cleanup
-
Nishant Patel authored
* Fix MLIR build on mac OS * Style * Style
-
- 11 Nov, 2019 1 commit
-
-
Pruthvi authored
* WIP graph pass to lower ngraph to ngraph dialect * resolved compiler errors * - refactor ngraph-dialect to graph pass * - fix compilation issue - unit test passes * - style fix * Addressed PR comments * - move NgDialectConversionPass to anonymous namespace * - use getModule() to access module inside the graph pass - address PR comments * - fix failing unit test case for negative padding - make builder as an object isntead of pointer to an object * Address PR Comments
-
- 04 Nov, 2019 1 commit
-
-
Nagy Mostafa authored
* Init commit to implement interface * Add two op interfaces for v0 and v1. Add a unit-test * Add missing files * Move test to separate file * Add Fused Op interface * Missing files * style * fused ops * Remove V1 ops for now * Added enum attributes. WIP * Completed non-experiemntal non-fused-ops * Add ops_attributes * Minor fixes * Minor fixes * Added enum setting/reading test * style-apply * Added attributes tests * Fix dialect init * style * fix typo * Fix merge errors * Include file with MLIR on
-
- 30 Oct, 2019 1 commit
-
-
Pruthvi authored
* Re-organize files. Create MLIR backend classes * WIP * Refactored. Code compiles * Moved context to Runtime class to outlive compilation and execution * style-apply * Base Runtime class. Few other modifications * Minor fixes * Fixed Runtime::run() to take type-erased pointer * renamed core compiler * rename backend compiler * rename runtime compiler * PR feedback * Fix build fails * - refactor LowerNGDialect - addgraph pass to lower standardDialect to llvm module * Addressed PR Comments * i) Style fix ii) restore deleted file
-
- 29 Oct, 2019 1 commit
-
-
Nagy Mostafa authored
* Re-organize files. Create MLIR backend classes * WIP * Refactored. Code compiles * Moved context to Runtime class to outlive compilation and execution * style-apply * Base Runtime class. Few other modifications * Minor fixes * Fixed Runtime::run() to take type-erased pointer * renamed core compiler * rename backend compiler * rename runtime compiler * PR feedback * Fix build fails
-
- 23 Oct, 2019 1 commit
-
-
Amy Zhuang authored
* [MLIR] Modify mlir subgraph extraction pass. * Address PR feedback. * Modify sub-graph construction algorithm. * Address PR feedback. * Address PR feedback. * Change macro to function.
-
- 10 Oct, 2019 3 commits
-
-
Nagy Mostafa authored
* Change order of gemm operation * Add Dot lit test * style-apply * Fix captures to avoid escapes on []
-
nmostafa authored
-
nmostafa authored
-
- 09 Oct, 2019 2 commits
-
-
nmostafa authored
-
Diego Caballero authored
-
- 03 Oct, 2019 1 commit
-
-
Diego Caballero authored
-
- 02 Oct, 2019 1 commit
-
-
Diego Caballero authored
An optional optimization level was added to ExecutionEngine, initialized to None by default. This caused fast-isel to be used in LLVM CG, overriding all the flags related to not using fast-isel. In this PR we pass the right optimization level to ExecutionEngine.
-
- 30 Sep, 2019 1 commit
-
-
Diego Caballero authored
* [MLIR] Add support for parsing nGraph tensor type Initial commit that enables nGraph parsing. It's needed for testing. * [MLIR] Enable nGraph dialect in ngraph-opt This PR registers nGraph dialect in ngraph-opt and prepares nGraph lowering pass for LIT testing, fixing all the related issues. Among other things, lowering pass has to be turned into a function pass, dead argument in constructor was removed and `convert-ngraph-to-affine` flag was added. * Fix issue with function name and multiple functions * Extend module_function.mlir lit test * [MLIR] Add support for parsing nGraph element types It introduces initial support for parsing nGraph signed/unsigned integer and floating point data types. * [MLIR] Improve gather op lowering This PR interchanges indices and param loops in gather lowering so that a better memory access patter is generated. Fusion of gather with other ops is also observed with this change.
-
- 28 Sep, 2019 1 commit
-
-
Diego Caballero authored
* [MLIR] Add support for parsing nGraph tensor type Initial commit that enables nGraph parsing. It's needed for testing. * [MLIR] Enable nGraph dialect in ngraph-opt This PR registers nGraph dialect in ngraph-opt and prepares nGraph lowering pass for LIT testing, fixing all the related issues. Among other things, lowering pass has to be turned into a function pass, dead argument in constructor was removed and `convert-ngraph-to-affine` flag was added. * Fix issue with function name and multiple functions * Extend module_function.mlir lit test * Improve module_function.mlir test Remove ngraph to affine dialect conversion since we just need to verify that we can parse and print modules and functions. Add verification for parsing the printed code. * [MLIR] Add support for parsing nGraph element types (#3665) * [MLIR] Add support for parsing nGraph element types It introduces initial support for parsing nGraph signed/unsigned integer and floating point data types. * Improve LIT tests Test parsing and printing of types separately from lowering to affine since these tests will evolve differently, particularly for tensor types. * Missed file I left this file behind in the previous commit
-
- 25 Sep, 2019 1 commit
-
-
Amy Zhuang authored
* [MLIR] Encapsulate internal nodes of CompiledKernel Op to prevent later passes from accessing them. * Fix style and namespace. * Address PR feedback. * Move encapsulate_nodes from pass to CompiledKernel Op. * Address PR feedback. * Follow MLIR naming convention.
-
- 24 Sep, 2019 2 commits
-
-
Diego Caballero authored
* [MLIR] Add support for parsing nGraph tensor type Initial commit that enables nGraph parsing. It's needed for testing. * Clang format * Fix return in parser
-
Diego Caballero authored
* [MLIR] Add `ngraph` prefix to MLIR flags Some flags collision with some MLIR flags. * [MLIR] Add support for nGraph tensor type in parser Initial commit that enables nGraph parsing. It's needed for testing. * Rename ngraph print flag * Rename ngraph dump mlir flags * Clang format * Revert "[MLIR] Add support for nGraph tensor type in parser" This reverts commit ae371d6a5c8ea590322d5d3b9ba110159d4bf5fa.
-
- 19 Sep, 2019 1 commit
-
-
Diego Caballero authored
* [MLIR] Add flag to dump jitted MLIR to object file This PR adds a flag to be able to dump the jitted MLIR code to an object file and another flag to be able to specify the object file name. * Add curly braces * Fix wrong merge
-
- 18 Sep, 2019 2 commits
-
-
Diego Caballero authored
-
Nagy Mostafa authored
* Enable Padding * Small fix to match MLIR API change * WIP * Subtract pad below from virtual img index * Fixed integer set inequalities * style-apply * remove unneeded padAbove values * Remove dump call * PR feedback * style apply
-
- 17 Sep, 2019 2 commits
-
-
Nagy Mostafa authored
* Re-use memref * add mem opt file * Add missing files * Add flag * Fix style * Small fix * PR fixes. WIP * PR fixes * Fixed conv validation logic. Invoked after ng mem opt pass. Style-apply * style-apply
-
Diego Caballero authored
Replace set with vector container in MLIR sub-graphs for nodes, inputs and outputs. Using a set caused different order of operations in MLIR code across different executions and this led to differences in optimizations.
-
- 16 Sep, 2019 1 commit
-
-
Nagy Mostafa authored
* Fix style in compiler and lowerer files * Fix comment in headers * Revert "Fix comment in headers" This reverts commit d52eed4c1bdf371f3cc7d3f601d9d2b1b0c233e8. * Fix compiler.* header. Fix code style in other files
-
- 10 Sep, 2019 3 commits
-
-
Nagy Mostafa authored
* WIP * Fixes * Increase bitwidth for arg idx attrib * Minor fixes * style-apply
-
Diego Caballero authored
* [MLIR] Enable LIT testing in CMAKE This PR enables LIT testing in nGraph for MLIR Compiler, introduced in PR3523. We can now do `make check-mlir-lit` to run LIT tests. * Address feedback * Revisit PR after cmake clean-up * [MLIR] Introduce ngraph-opt tool To be used, among other things, for LIT testing. `elementwise_binary_ops.mlir` shows a simple LIT test using ngraph-opt. Dummy affine_lowering/lit_test.mlir test that served as initial testing of LIT tool configuration is now removed. * Address feedback
-
Nagy Mostafa authored
* Bump MLIR version and minor fixes in nGraph * advance commit ids * fix commit id
-
- 06 Sep, 2019 1 commit
-
-
Diego Caballero authored
-
- 05 Sep, 2019 1 commit
-
-
Nagy Mostafa authored
* Don't hold to shared pointers after sub-graph extraction pass is done * style
-
- 04 Sep, 2019 2 commits
-
-
Amy Zhuang authored
-
Diego Caballero authored
* [MLIR] Move MLIR compiler files to contrib/mlir/compiler We'll add contrib/mlir/tools directory in subsequent PRs for ngrpah-opt and other tools. * [MLIR] Clean up MLIR related CMake files This PR prepares the introduction of src/contrib/mlir/tools and LIT testing by refactoring and cleaning up the common part with src/contrib/mlir/compiler: - Common code for src/contrib/mlir/tools and src/contrib/mlir/compiler is moved to src/contrib/mlir/CMakeList.txt. - More generic code, common for src/contrib/mlir/* and LIT configuration files, is moved to cmake/external_mlir.cmake with pre-existing similar code. - Redundant environment var LLVM_DIR is removed in favor of LLVM_CMAKE_PATH. - Redundant cmake messages are removed. * Replacing LLVM_CMAKE_PATH with LLVM_DIR
-