[PATCH] D100383: [LSR] Fix for pre-indexed generated constant offset

[Sjoerd Meijer via Phabricator via llvm-commits]

SjoerdMeijer added inline comments.


================
Comment at: llvm/test/Transforms/LoopStrengthReduce/AMDGPU/atomics.ll:95
-; OPT: %scevgep4 = getelementptr i32, i32 addrspace(3)* %lsr.iv2, i32 16383
-; OPT: %tmp4 = call i32 @llvm.amdgcn.atomic.inc.i32.p3i32(i32 addrspace(3)* %scevgep4, i32 undef, i32 0, i32 0, i1 false)
-; OPT: %tmp7 = call i32 @llvm.amdgcn.atomic.inc.i32.p3i32(i32 addrspace(3)* %lsr.iv1, i32 undef, i32 0, i32 0, i1 false)
----------------
What happened to these atomics? I think this is crucial for these tests, and we still need to match them. 
Same for the atomic reads/writes in tests above and also below here.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D100383/new/

https://reviews.llvm.org/D100383