[Openmp-commits] [clang] [llvm] [mlir] [openmp] [LoopTiling][Clang][MLIR] Canonical Intra-tile Loops (PR #191114)

[Michael Kruse via Openmp-commits]

================
@@ -56,7 +56,7 @@ extern "C" void foo4() {
 
 
 extern "C" void foo5() {
-#pragma omp for collapse(3)
+#pragma omp for collapse(2)
----------------
Meinersbur wrote:

Why this change? Collapseing grid as well as an intratile loop is exactly what this canonical-intratile loop is for

https://github.com/llvm/llvm-project/pull/191114