[PATCH] D121319: Tests for D121320

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

I looked at the effect of the 2 proposed patches together, and we should add at least 2 more tests (can be added to this commit or separately):

  define i1 @shifted_mask_testb_lower32(i64 %a) {
    %v0 = and i64 %a, 66846720  ; 0xff << 18
    %v1 = icmp ne i64 %v0, 0
    ret i1 %v1
  }
  
  define i1 @shifted_mask_test_lower32(i64 %a) {
    %v0 = and i64 %a, 131072 ; 0xffff << 1
    %v1 = icmp ne i64 %v0, 0
    ret i1 %v1
  }

The difference for these is that the shifted mask does not extend into the upper 32-bits of the value. We probably want to convert these to use shifts too, and we'll handle that with both patches applied.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D121319/new/

https://reviews.llvm.org/D121319