[llvm] [AArch64] Extend usage of `XAR` instruction for fixed-length operations (PR #139460)

[Rajveer Singh Bharadwaj via llvm-commits]

================
@@ -1,6 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
 ; RUN: llc -mtriple=aarch64 -mattr=+sha3 < %s | FileCheck --check-prefix=SHA3 %s
 ; RUN: llc -mtriple=aarch64 -mattr=-sha3 < %s | FileCheck --check-prefix=NOSHA3 %s
+; RUN: llc -mtriple=aarch64 -mattr=+sve2,sve2-sha3,+sha3 < %s | FileCheck --check-prefix=SVE2SHA3 %s
----------------
Rajveer100 wrote:

I have tested the above 64 bit vector types locally and widening them similar to v1i64 works well, generating the XAR instruction.

Pushing soon, let me know if there's anything else.

https://github.com/llvm/llvm-project/pull/139460