test(compiler): Add check for omp.parallel after lowering a parallelizable FHELinalg op

This test ensures that at least one parallel region is generated for
an FHELinalg operation that is guaranteed to result in a parallel loop
when `concretecompiler` is invoked with `--parallelize`.
This commit is contained in:
Andi Drebes
2023-02-09 11:19:04 +01:00
committed by Quentin Bourgerie
parent fc02bb5cd8
commit 5d3af16617

View File

@@ -0,0 +1,10 @@
// RUN: concretecompiler %s --action=dump-llvm-dialect --parallelize 2>&1| FileCheck %s
// Check that at some point the compilation pipeline generates a parallel region
// CHECK: omp.parallel
func.func @apply_lookup_table(%arg0: tensor<2x3x4x!FHE.eint<2>>) -> tensor<2x3x4x!FHE.eint<2>> {
%arg1 = arith.constant dense<"0x0000000000000000000000000000000100000000000000020000000000000003"> : tensor<4xi64>
%1 = "FHELinalg.apply_lookup_table"(%arg0, %arg1): (tensor<2x3x4x!FHE.eint<2>>, tensor<4xi64>) -> (tensor<2x3x4x!FHE.eint<2>>)
return %1: tensor<2x3x4x!FHE.eint<2>>
}