llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT--> @llvm/pr-subscribers-mlir-affine <details> <summary>Changes</summary> In the process of vectorization of the affine loop, the 0 vector size causes the crash with building the invalid AffineForOp. We can catch the case beforehand propagating to the assertion. See: https://github.com/llvm/llvm-project/issues/64262 --- Full diff: https://github.com/llvm/llvm-project/pull/65293.diff 3 Files Affected: - (modified) mlir/include/mlir/Dialect/Affine/Passes.td (+2-1) - (modified) mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp (+5) - (added) mlir/test/Dialect/Affine/SuperVectorize/invalid-zero-size.mlir (+9) ``````````diff diff --git a/mlir/include/mlir/Dialect/Affine/Passes.td b/mlir/include/mlir/Dialect/Affine/Passes.td index 891a6661ca87be7..1036e93a0392409 100644 --- a/mlir/include/mlir/Dialect/Affine/Passes.td +++ b/mlir/include/mlir/Dialect/Affine/Passes.td @@ -349,7 +349,8 @@ def AffineVectorize : Pass<"affine-super-vectorize", "func::FuncOp"> { let dependentDialects = ["vector::VectorDialect"]; let options = [ ListOption<"vectorSizes", "virtual-vector-size", "int64_t", - "Specify an n-D virtual vector size for vectorization">, + "Specify an n-D virtual vector size for vectorization. " + "This must be greater than zero.">, // Optionally, the fixed mapping from loop to fastest varying MemRef // dimension for all the MemRefs within a loop pattern: // the index represents the loop depth, the value represents the k^th diff --git a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp index 85c2602aa266d67..a1d9c7b56c4e154 100644 --- a/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp +++ b/mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp @@ -1732,6 +1732,11 @@ void Vectorize::runOnOperation() { return signalPassFailure(); } + if (llvm::any_of(vectorSizes, [](int64_t size) { return size <= 0; })) { + f.emitError("Vectorization factor must be greater than zero."); + return signalPassFailure(); + } + DenseSet<Operation *> parallelLoops; ReductionLoopMap reductionLoops; diff --git a/mlir/test/Dialect/Affine/SuperVectorize/invalid-zero-size.mlir b/mlir/test/Dialect/Affine/SuperVectorize/invalid-zero-size.mlir new file mode 100644 index 000000000000000..396cc933e25bcf4 --- /dev/null +++ b/mlir/test/Dialect/Affine/SuperVectorize/invalid-zero-size.mlir @@ -0,0 +1,9 @@ +// RUN: mlir-opt %s -verify-diagnostics --affine-super-vectorize=virtual-vector-size=0 + +// expected-error@+1 {{Vectorization factor must be greater than zero}} +func.func @with_zero_vector_size(%arg0: memref<21x12x12xi1>) { + affine.for %arg1 = 0 to 84 step 4294967295 { + } + return +} + `````````` </details> https://github.com/llvm/llvm-project/pull/65293 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
