[Mlir-commits] [mlir] [mlir][scf] Fix front peeling skipped due to float rounding in trip count (PR #231100)

[llvmlistbot at llvm.org]

================
@@ -238,8 +238,13 @@ LogicalResult mlir::scf::peelForLoopFirstIteration(RewriterBase &b, ForOp forOp,
   auto ubInt = getConstantIntValue(forOp.getUpperBound());
   auto stepInt = getConstantIntValue(forOp.getStep());
 
-  // Peeling is not needed if there is one or less iteration.
-  if (lbInt && ubInt && stepInt && ceil(float(*ubInt - *lbInt) / *stepInt) <= 1)
+  // Peeling is not needed if there is at most one iteration. 
+  // Per the scf.for semantics, the trip count is
+  // max(0, ceil((UB - LB) / Step)), so for a positive step there is more than
+  // one iteration exactly when UB - LB > Step. 
+  // Compare the integers directly: going through float is lossy and wrongly skipped
+  // peeling of loops with two iterations.
----------------
StarOne01 wrote:

done

https://github.com/llvm/llvm-project/pull/231100