[libc-commits] [clang] [lld] [libcxx] [lldb] [mlir] [clang-tools-extra] [compiler-rt] [libc] [llvm] [MLIR][LLVM] Add Continuous Loop Peeling transform to SCF (PR #71555)

[Matthias Springer via libc-commits]

================
@@ -105,6 +106,165 @@ static void specializeForLoopForUnrolling(ForOp op) {
   op.erase();
 }
 
+/// Create a new for loop for the remaining iterations (partiaIteration)
+/// after a for loop has been peeled. This is followed by correcting the
+/// loop bounds for both loops given the index (splitBound) where the
+/// iteration space is to be split up.
+static LogicalResult splitLoopHelper(RewriterBase &b, scf::ForOp forOp,
+                                     scf::ForOp &partialIteration,
+                                     Value &splitBound) {
+  RewriterBase::InsertionGuard guard(b);
+  auto lbInt = getConstantIntValue(forOp.getLowerBound());
+  auto ubInt = getConstantIntValue(forOp.getUpperBound());
+  auto stepInt = getConstantIntValue(forOp.getStep());
+
+  // No specialization necessary if step already divides upper bound evenly.
+  if (lbInt && ubInt && stepInt && (*ubInt - *lbInt) % *stepInt == 0)
+    return failure();
+  // No specialization necessary if step size is 1.
+  if (stepInt == static_cast<int64_t>(1))
+    return failure();
+
+  // Create ForOp for partial iteration.
+  b.setInsertionPointAfter(forOp);
+  partialIteration = cast<scf::ForOp>(b.clone(*forOp.getOperation()));
+  partialIteration.getLowerBoundMutable().assign(splitBound);
+  forOp.replaceAllUsesWith(partialIteration->getResults());
+  partialIteration.getInitArgsMutable().assign(forOp->getResults());
+
+  // Set new upper loop bound.
+  b.updateRootInPlace(
+      forOp, [&]() { forOp.getUpperBoundMutable().assign(splitBound); });
+
+  return success();
+}
+
+/// Convert single-iteration for loop to if-else block.
+static scf::IfOp convertSingleIterFor(RewriterBase &b, scf::ForOp &forOp) {
+  Location loc = forOp->getLoc();
+  IRMapping mapping;
+  mapping.map(forOp.getInductionVar(), forOp.getLowerBound());
+  for (auto [arg, operand] :
+       llvm::zip(forOp.getRegionIterArgs(), forOp.getInitsMutable())) {
----------------
matthias-springer wrote:

nit: `zip_equal`

https://github.com/llvm/llvm-project/pull/71555