[PATCH] D99481: [InstCombine] Fix miscompile on GEP+load to icmp fold (PR45210)

[Juneyoung Lee via Phabricator via llvm-commits]

aqjune added inline comments.


================
Comment at: llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp:289
+          IdxBeforeSext->getType()->getPrimitiveSizeInBits().getFixedSize() >=
+          PtrSize;
+    else
----------------
This is to avoid creating redundant expressions, right?
Could you elaborate a bit more with an example?


================
Comment at: llvm/test/Transforms/InstCombine/load-cmp.ll:60
 ; CHECK-LABEL: @test1_noinbounds_as1(
-; CHECK-NEXT:    [[TMP1:%.*]] = trunc i32 [[X:%.*]] to i16
-; CHECK-NEXT:    [[R:%.*]] = icmp eq i16 [[TMP1]], 9
+; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[X:%.*]], 32767
+; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[TMP1]], 9
----------------
Could you leave a comment explaining why this is 32767?
Same for test10_struct_arr_noinbounds case below


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D99481/new/

https://reviews.llvm.org/D99481