[clang] Check for comparisons to INF and NaN when in ffast-math mode (PR #76862)

[Zahira Ammarguellat via cfe-commits]

https://github.com/zahiraam created https://github.com/llvm/llvm-project/pull/76862

and generate a warning.

>From 55ee9ac2d36c1485573aa5e291e5867e8f1d5099 Mon Sep 17 00:00:00 2001
From: Ammarguellat <zahira.ammarguellat at intel.com>
Date: Wed, 3 Jan 2024 12:41:44 -0800
Subject: [PATCH] Check for comparisons to INF and NaN when in ffast-math mode
 and generate a warning.

---
 clang/include/clang/Basic/DiagnosticSemaKinds.td | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index e54f969c19039d..1b75ae8f678b68 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -6771,6 +6771,9 @@ def warn_pointer_sub_null_ptr : Warning<
 def warn_floatingpoint_eq : Warning<
   "comparing floating point with == or != is unsafe">,
   InGroup<DiagGroup<"float-equal">>, DefaultIgnore;
+def warn_fast_floatingpoint_eq : Warning<
+  "explicit comparison with %0 in fast floating point mode">,
+  InGroup<TautologicalConstantCompare>;
 
 def err_setting_eval_method_used_in_unsafe_context : Error <
   "%select{'#pragma clang fp eval_method'|option 'ffp-eval-method'}0 cannot be used with "