[llvm] [TLI] Add glibc 2.35 libmvec vector functions to the x86 table (PR #223817)

[Simon Pilgrim via llvm-commits]

================
@@ -0,0 +1,668 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals
+; RUN: opt -vector-library=LIBMVEC -replace-with-veclib -S < %s | FileCheck %s
+
+; Checks that the x86 libmvec table is reachable from ReplaceWithVeclib, which
+; runs for every target at -O1 and above. The AArch64 equivalent is
+; llvm/test/CodeGen/AArch64/replace-with-veclib-libmvec.ll; x86 had no
+; counterpart. Only functions whose llvm.* intrinsic name is in the table can
+; be replaced here, so this covers the intrinsic rows specifically.
+
+target triple = "x86_64-unknown-linux-gnu"
----------------
RKSimon wrote:

This file should be in llvm\test\Transforms\ReplaceWithVeclib\X86 - no idea why aarch64 put theirs in codegen, but we shouldn't do the same.

https://github.com/llvm/llvm-project/pull/223817