[llvm] [LV] Handle FSub Partial Reductions (PR #191186)

[Sander de Smalen via llvm-commits]

================
@@ -0,0 +1,36 @@
+; REQUIRES: asserts
+; RUN: not --crash opt -passes=loop-vectorize -force-vector-interleave=1 -enable-epilogue-vectorization=false -S %s 2>&1 | FileCheck %s --check-prefix=ASSERTION
+; RUN: not --crash opt -passes=loop-vectorize -enable-epilogue-vectorization=false -S %s 2>&1 | FileCheck %s --check-prefix=ASSERTION
+; RUN: not --crash opt -passes=loop-vectorize -force-vector-interleave=1 -vectorizer-maximize-bandwidth -enable-epilogue-vectorization=false -S %s 2>&1 | FileCheck %s --check-prefix=ASSERTION
+
+target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64-none-unknown-elf"
+
----------------
sdesmalen-arm wrote:

Can you add a comment to the test describing why this case is expected to fail?

https://github.com/llvm/llvm-project/pull/191186