[llvm] [AArch64] Limit immediate offsets when folding instructions into addressing modes (PR #67345)

[Momchil Velikov via llvm-commits]

================
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 3
-; RUN: llc -mtriple=aarch64-apple-ios -verify-machineinstrs -global-isel -aarch64-postlegalizer-consecutive-memops=0 < %s | FileCheck %s --check-prefix=CHECK-NO-SPLIT
-; RUN: llc -mtriple=aarch64-apple-ios -verify-machineinstrs -global-isel < %s | FileCheck %s --check-prefix=CHECK-SPLIT
+; RUN: llc -mtriple=aarch64-apple-ios -verify-machineinstrs -global-isel -aarch64-enable-sink-fold=true -aarch64-postlegalizer-consecutive-memops=0 < %s | FileCheck %s --check-prefix=CHECK-NO-SPLIT
----------------
momchil-velikov wrote:

Yes, that's right, for several other tests in the testsuite as well.

https://github.com/llvm/llvm-project/pull/67345