[PATCH] D141549: [X86] Add additional tests for ctlz{_zero_undef} to test folding with xor; NFC

Noah Goldstein via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Jan 12 09:32:54 PST 2023


goldstein.w.n added inline comments.


================
Comment at: llvm/test/CodeGen/X86/clz.ll:1784
+
+define i32 @ctlz_xor31_i32_false(i32 %x) {
+; X86-LABEL: ctlz_xor31_i32_false:
----------------
pengfei wrote:
> pengfei wrote:
> > Missing a `ctlz_xor31_i32_ture` unintentionally?
> > Actually, I think it's better if we intersect use `i8_true`, `i16_false` etc. to reduce the size. I'd prefer to use less tests while not compromise the cover rate.
> Please ignore, I found the identical test above.
> Missing a `ctlz_xor31_i32_ture` unintentionally?
> Actually, I think it's better if we intersect use `i8_true`, `i16_false` etc. to reduce the size. I'd prefer to use less tests while not compromise the cover rate.

I intersected `i16` / `i64`. Think worth keeping both cases for `i8` b.c its a special case. LMK if you feel strongly about dropping.



Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D141549/new/

https://reviews.llvm.org/D141549



More information about the llvm-commits mailing list