[PATCH] D95245: [SVE] Add support for scalable vectorization of loops with int/fast FP reductions

[Kerry McLaughlin via Phabricator via llvm-commits]

kmclaughlin marked an inline comment as not done.
kmclaughlin added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/AArch64/scalable_reductions.ll:223
+  %0 = load float, float* %arrayidx, align 4
+  %add = fadd float %0, %sum.07
+  %iv.next = add nuw nsw i64 %iv, 1
----------------
david-arm wrote:
> I'm a bit surprised this vectorises to be honest, since there is no 'fast' flag here! Perhaps for IEEE math you have to add specific attributes to the function?
I think what happened here is that the hints used to enable vectorization have allowed reordering, similar to using -Ofast. I found this comment at the top of allowReordering() in LoopVectorizationLegality:
```
// When enabling loop hints are provided we allow the vectorizer to change
// the order of operations that is given by the scalar loop. This is not
// enabled by default because can be unsafe or inefficient. For example,
// reordering floating-point operations will change the way round-off
// error accumulates in the loop.
```
This behaviour was queried on the mailing list last year:
https://lists.llvm.org/pipermail/llvm-dev/2020-June/142697.html


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D95245/new/

https://reviews.llvm.org/D95245