[PATCH] D106450: [InstCombine] Fold (gep (oneuse(gep Ptr, Idx0)), Idx1) -> (gep Ptr, (add Idx0, Idx1)) (PR51069)

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon created this revision.
RKSimon added reviewers: reames, spatel, lebedev.ri.
Herald added subscribers: dmgreen, hiraditya.
RKSimon requested review of this revision.
Herald added a project: LLVM.

As noticed on D106352 <https://reviews.llvm.org/D106352>, after we've folded "(select C, (gep Ptr, Idx), Ptr) -> (gep Ptr, (select C, Idx, 0))" if the inner Ptr was also a (now one use) gep we could then merge the geps, using the sum of the indices instead.

I've limited this to basic 2-op geps - a more general case further down InstCombinerImpl.visitGetElementPtrInst doesn't have the one-use limitation but only creates a add if it can be created via SimplifyAddInst.

https://alive2.llvm.org/ce/z/f8pLfD  (Thanks Roman!)


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D106450

Files:
  llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
  llvm/test/Transforms/InstCombine/gep-combine-loop-invariant.ll
  llvm/test/Transforms/InstCombine/gep-custom-dl.ll
  llvm/test/Transforms/InstCombine/getelementptr.ll
  llvm/test/Transforms/InstCombine/select-gep.ll
  llvm/test/Transforms/InstCombine/shift.ll
  llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse.ll
  llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse-mask4.ll
  llvm/test/Transforms/LoopVectorize/ARM/mve-reduction-predselect.ll
  llvm/test/Transforms/LoopVectorize/ARM/mve-reductions.ll
  llvm/test/Transforms/LoopVectorize/X86/x86-interleaved-accesses-masked-group.ll
  llvm/test/Transforms/LoopVectorize/consecutive-ptr-uniforms.ll
  llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D106450.360469.patch
Type: text/x-patch
Size: 65106 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210721/905470c5/attachment.bin>