[Mlir-commits] [mlir] [mlir][scf] Clamp normalized loop sizes where they are combined (PR #223256)

[Aviad Cohen]

================
@@ -989,6 +989,21 @@ delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
   return {delinearizedIvs, preservedUsers};
 }
 
+/// Clamp a normalized loop size at zero. `emitNormalizedLoopBounds` returns
+/// `ceilDiv(ub - lb, step)`, which is negative for an empty loop: combining two
+/// such sizes cancels the signs into a positive extent, and a negative value is
+/// not a legal `affine.delinearize_index` basis.
+static Value clampSizeToNonNegative(RewriterBase &rewriter, Location loc,
+                                    Value size) {
+  if (auto cst = getConstantIntValue(size))
+    return *cst >= 0 ? size
+                     : getValueOrCreateConstantIntOp(rewriter, loc,
+                                                     rewriter.getIndexAttr(0));
+  Value zero = arith::ConstantOp::create(rewriter, loc,
+                                         rewriter.getZeroAttr(size.getType()));
+  return arith::MaxSIOp::create(rewriter, loc, size, zero);
----------------
AviadCo wrote:

I am not sure it is o.k. to use zero - what if we had "empty loop" and now zero causes us to run one iteration or more?

https://github.com/llvm/llvm-project/pull/223256