[llvm] [PowerPC] Implement a more efficient memcmp in cases where the length is known. (PR #158657)

[via llvm-commits]

================
@@ -156,7 +127,7 @@ entry:
 define signext i32 @test4(ptr nocapture readonly %buffer1, ptr nocapture readonly %buffer2)  {
 
 entry:
-  %call = tail call signext i32 @memcmp(ptr %buffer1, ptr %buffer2, i64 65)
+  %call = tail call signext i32 @memcmp(ptr %buffer1, ptr %buffer2, i64 165)
----------------
RolandF77 wrote:

Wouldn't 129 be the equivalent?

https://github.com/llvm/llvm-project/pull/158657