--fp-exp: Better warning message. (#3709)

diff --git a/src/theory/fp/theory_fp.cpp b/src/theory/fp/theory_fp.cpp
index 6a4dc542e788400e0a8c814d8e173c1f9b2aae1f..788545b3c8ece2a9319566a64f0493f81abe1152 100644 (file)
--- a/src/theory/fp/theory_fp.cpp
+++ b/src/theory/fp/theory_fp.cpp
@@ -902,7 +902,8 @@ void TheoryFp::preRegisterTerm(TNode node)
            << sig_sz
            << " is not supported, only Float32 (8/24) or Float64 (11/53) types "
               "are supported in default mode. Try the experimental solver via "
-              "--fp-exp";
+              "--fp-exp. Note: There are known issues with the experimental "
+              "solver, use at your own risk.";
         throw LogicException(ss.str());
       }
     }