[clang] [llvm] [AArch64] Add support for range prefetch intrinsic (PR #170490)

Paul Walker via cfe-commits cfe-commits at lists.llvm.org
Tue Dec 23 07:05:34 PST 2025


================
@@ -96,6 +96,10 @@ TARGET_BUILTIN(__builtin_arm_jcvt, "Zid", "nc", "v8.3a")
 // Prefetch
 BUILTIN(__builtin_arm_prefetch, "vvC*UiUiUiUi", "nc")
 
+// Range Prefetch
+TARGET_HEADER_BUILTIN(__builtin_arm_range_prefetch_x, "vvC*UiUiiUiiz", "n", ARMACLE_H, ALL_LANGUAGES, "")
+TARGET_HEADER_BUILTIN(__builtin_arm_range_prefetch, "vvC*UiUiULLi", "n", ARMACLE_H, ALL_LANGUAGES, "")
----------------
paulwalker-arm wrote:

Should this use `BUILTIN`? because the ACLE intrinsics `__pld_range` and `__pldx_range` are in the header but `__builtin_arm_range_prefetch` and `__builtin_arm_range_prefetch_x` are always available? (e.g. like `__builtin_arm_prefetch`)

https://github.com/llvm/llvm-project/pull/170490


More information about the cfe-commits mailing list