[PATCH] D86418: [AVR] Improve inline rotate/shift expansions

[Ayke via Phabricator via llvm-commits]

aykevl added a comment.

I have tested this locally with my compiler-rt test setup. All tests still pass with this patch applied while the binary size is slightly reduced (when optimizing for size).



================
Comment at: llvm/test/CodeGen/AVR/rot.ll:16
+  ; CHECK-NEXT: dec r22
+  ; CHECK-NEXT: brpl .LBB0_1
 
----------------
Note: this doesn't decrease binary size because LLVM duplicates the loop check (`dec` and `br`). When optimizing for size, this would be reduced by one instruction, such as in `shift_i8_i8_size` of shift.ll.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D86418/new/

https://reviews.llvm.org/D86418