[llvm] [AArch64] Don't tail call memset if it would convert to a bzero. (PR #98969)

[Eli Friedman via llvm-commits]

================
@@ -0,0 +1,20 @@
+; RUN: llc -o - %s | FileCheck %s
+; RUN: llc -global-isel -o - %s | FileCheck %s
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "arm64-apple-macosx15.0.0"
+
+define ptr @test()  {
+; CHECK-LABEL: test:
+; CHECK-NOT: b _bzero
----------------
efriedma-quic wrote:

I'd prefer a positive check for "bl _bzero",

https://github.com/llvm/llvm-project/pull/98969