[llvm] [llubi] Implement experimental vector intrinsics (PR #206899)

[Antonio Frighetto via llvm-commits]

================
@@ -0,0 +1,120 @@
+; NOTE: Assertions have been autogenerated by utils/update_llubi_test_checks.py UTC_ARGS: --version 6
+; RUN: llubi --verbose < %s 2>&1 | FileCheck %s
+
+define void @main() {
+  %cttz_some = call i8 @llvm.experimental.cttz.elts.i8.v4i1(<4 x i1> <i1 false, i1 false, i1 true, i1 false>, i1 true)
+  %cttz_all = call i8 @llvm.experimental.cttz.elts.i8.v4i1(<4 x i1> zeroinitializer, i1 false)
+  %cttz_poison = call i8 @llvm.experimental.cttz.elts.i8.v4i1(<4 x i1> zeroinitializer, i1 true)
+
+  %evl_zero = call i32 @llvm.experimental.get.vector.length.i32(i32 0, i32 4, i1 false)
+  %evl_short = call i32 @llvm.experimental.get.vector.length.i32(i32 3, i32 4, i1 false)
+  %evl_full = call i32 @llvm.experimental.get.vector.length.i32(i32 9, i32 4, i1 false)
+  %evl_poison = call i32 @llvm.experimental.get.vector.length.i64(i64 poison, i32 4, i1 false)
----------------
antoniofrighetto wrote:

Should we also test the vscale variant?

https://github.com/llvm/llvm-project/pull/206899