Updates to shape functions enabling reuse from MHLO #1918
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ghpvnist
approved these changes
Jan 13, 2024
…operations to MHLO
GleasonK
approved these changes
Jan 18, 2024
sdasgup3
force-pushed
the
reduce-op-migration-to-mhlo
branch
from
c50eaff
to
aafde87
Compare
This was referenced Jan 28, 2024
Merged
stellaraccident
pushed a commit
to llvm/torch-mlir
that referenced
this pull request
Jan 31, 2024
With the recent LLVM integrate and changes from llvm/llvm-project#78260, we hit this build error in Stablehlo (which is quite old). ``` external/stablehlo/stablehlo/transforms/StablehloRefineShapes.cpp:1020:14: error: no member named 'startRootUpdate' in 'mlir::PatternRewriter' rewriter.startRootUpdate(op); ~~~~~~~~ ^ external/stablehlo/stablehlo/transforms/StablehloRefineShapes.cpp:1026:16: error: no member named 'finalizeRootUpdate' in 'mlir::PatternRewriter' rewriter.finalizeRootUpdate(op); ~~~~~~~~ ^ external/stablehlo/stablehlo/transforms/StablehloRefineShapes.cpp:1029:16: error: no member named 'cancelRootUpdate' in 'mlir::PatternRewriter' rewriter.cancelRootUpdate(op); ~~~~~~~~ ^ external/stablehlo/stablehlo/transforms/StablehloRefineShapes.cpp:1108:14: error: no member named 'updateRootInPlace' in 'mlir::PatternRewriter' rewriter.updateRootInPlace(op->getParentOp(), [&]() { return; }); ~~~~~~~~ ^ 4 errors generated. Target @torch-mlir//:torch-mlir-opt failed to build ``` I'm still puzzled as to how this didn't fail with the CMake merge gating CI (do we not test Stablehlo builds/tests?). In any case, bumping our submodule to openxla/stablehlo#1918 fixes it. It exposes a new failing lit test in TorchToStablehlo though, that I have looped stablehlo developers into ([here](https://discord.com/channels/999073994483433573/999074539138990131/1201235845391331419)). ``` bazel run @torch-mlir//test/Conversion:TorchToStablehlo/scatter.mlir.test ...external/torch-mlir/test/Conversion/TorchToStablehlo/scatter.mlir within split at <stdin>:1 offset :33:8: error: unexpected error: Expects non-empty reduction block for type inference %0 = torch.aten.scatter.src %arg0, %int0, %arg1, %arg2 : !torch.vtensor<[?,?],si64>, !torch.int, !torch.vtensor<[?,?],si64>, !torch.vtensor<[?,?],si64> -> !torch.vtensor<[?,?],si64> ^ LLVM ERROR: Failed to infer result type(s). ``` Bazel CI: https://github.com/sjain-stanford/torch-mlir/actions/runs/7732673480/job/21083102228
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The upstream change #1869 in StableHLO updates various API related to shape inference. MHLO shape inference functions in hlo_ops.cc uses those APIs. The PR updates the visibility and signature of those API for a clearer integration.
Specifically, the PR does the followings:
updates
getAccumulatorTypesto return a error status when the input regions is empty: This function is used in type inference of various reduction based operations (eg). This functions enables infering type based on the reduction block of the operation, which is something proposed in RFC. However, there could be instances when type inference can be called with empty region in which case we would like to report a meaningful diagnostic message.Allow
hlo::inferAllReduceOpto accept multiple operands information: In stableHLO,all_reduceop have a single operand (e.g.), whereas in MHLO the op can take multiple operand (e.g.. Thehlo::inferAllReduceOpsignature is updated to accommodate both cases.Remove unused arguments to functions
verifyReduceOpInputsAndInferShapeandinferReduceOp.