[PATCH] D39906: [InstCombine] Allowing GEP Instructions with loop Invariant operands to combine

[Eli Friedman via Phabricator via llvm-commits]

efriedma added a comment.

> I also don't see where "duplicating code" comes from.

You're inserting an "add" instruction without removing any other instructions; in general, this hurts codesize/performance.  This is why the combine is restricted in the first place.  You might get lucky sometimes and simplify away the extra instruction, but you can't rely on that.

The "reassociation" way to optimize this is to consider the following two instructions:

  %add.ptr = getelementptr inbounds i8, i8* %win, i64 %idx.ext
  %add.ptr2 = getelementptr inbounds i8, i8* %add.ptr, i64 %idx.ext1

These can be reassociated into the following:

  %tmp = getelementptr inbounds i8, i8* %win, i64 %idx.ext1
  %add.ptr2 = getelementptr inbounds i8, i8* %tmp, i64 %idx.ext

The first GEP then gets hoisted out of the loop.


https://reviews.llvm.org/D39906