[llvm] [hwasan] Optimize outlined memaccess for fixed shadow on Aarch64 (PR #88544)

[Thurston Dang via llvm-commits]

================
@@ -2367,6 +2367,10 @@ def int_hwasan_check_memaccess :
 def int_hwasan_check_memaccess_shortgranules :
   Intrinsic<[], [llvm_ptr_ty, llvm_ptr_ty, llvm_i32_ty],
             [ImmArg<ArgIndex<2>>]>;
+def int_hwasan_check_memaccess_fixedshadow_shortgranules :
----------------
thurstond wrote:

I'll add docs.

The way ASan avoids the need to pass the shadow offset via the intrinsic is by
defining llvm::getAddressSanitizerParams() in lib/Transforms/Instrumentation/AddressSanitizer.cpp. I'll try adapting that to this patch.

https://github.com/llvm/llvm-project/pull/88544