[PATCH] D52289: [ARM] Do not fuse VADD and VMUL on the Cortex-M4 and Cortex-M33

[Dave Green via Phabricator via llvm-commits]

dmgreen added inline comments.


================
Comment at: test/CodeGen/Thumb2/float-intrinsics-float.ll:192
 
 ; FIXME: why does cortex-m4 use vmla, while cortex-a7 uses vmul+vadd?
 ; (these should be equivalent, even the rounding is the same)
----------------
This comment can be removed now?


https://reviews.llvm.org/D52289