diff --git a/expression/collation.go b/expression/collation.go
index 4f9fd9a26c8f5..edfba16197895 100644
--- a/expression/collation.go
+++ b/expression/collation.go
@@ -297,7 +297,7 @@ func CheckAndDeriveCollationFromExprs(ctx sessionctx.Context, funcName string, e
 }
 
 func safeConvert(ctx sessionctx.Context, ec *ExprCollation, args ...Expression) bool {
-	enc := charset.FindEncodingUTF8AsNoop(ec.Charset)
+	enc := charset.FindEncodingTakeUTF8AsNoop(ec.Charset)
 	for _, arg := range args {
 		if arg.GetType().Charset == ec.Charset {
 			continue
diff --git a/parser/charset/encoding.go b/parser/charset/encoding.go
index 02f74c1fbad6a..eb9cef46ecc24 100644
--- a/parser/charset/encoding.go
+++ b/parser/charset/encoding.go
@@ -29,9 +29,10 @@ func IsSupportedEncoding(charset string) bool {
 	return ok
 }
 
-// FindEncodingUTF8AsNoop finds the encoding according to charset
-// except that utf-8 is treated as binary encoding.
-func FindEncodingUTF8AsNoop(charset string) Encoding {
+// FindEncodingTakeUTF8AsNoop finds the encoding according to the charset
+// except that utf-8 is treated as no-operation encoding. This is used to
+// reduce the overhead of utf-8 validation in some cases.
+func FindEncodingTakeUTF8AsNoop(charset string) Encoding {
 	enc := FindEncoding(charset)
 	if enc.Tp() == EncodingTpUTF8 {
 		return EncodingBinImpl
diff --git a/parser/yy_parser.go b/parser/yy_parser.go
index a0e373618f26b..5fedd0a1afb48 100644
--- a/parser/yy_parser.go
+++ b/parser/yy_parser.go
@@ -436,6 +436,6 @@ type CharsetClient string
 
 // ApplyOn implements ParseParam interface.
 func (c CharsetClient) ApplyOn(p *Parser) error {
-	p.lexer.encoding = charset.FindEncodingUTF8AsNoop(string(c))
+	p.lexer.encoding = charset.FindEncodingTakeUTF8AsNoop(string(c))
 	return nil
 }
diff --git a/server/util.go b/server/util.go
index 3604c1c779629..8314f7d6cfd05 100644
--- a/server/util.go
+++ b/server/util.go
@@ -298,7 +298,7 @@ type inputDecoder struct {
 
 func newInputDecoder(chs string) *inputDecoder {
 	return &inputDecoder{
-		encoding: charset.FindEncodingUTF8AsNoop(chs),
+		encoding: charset.FindEncodingTakeUTF8AsNoop(chs),
 		buffer:   nil,
 	}
 }
@@ -336,7 +336,7 @@ type resultEncoder struct {
 func newResultEncoder(chs string) *resultEncoder {
 	return &resultEncoder{
 		chsName:  chs,
-		encoding: charset.FindEncodingUTF8AsNoop(chs),
+		encoding: charset.FindEncodingTakeUTF8AsNoop(chs),
 		buffer:   nil,
 		isBinary: chs == charset.CharsetBinary,
 		isNull:   len(chs) == 0,
@@ -353,7 +353,7 @@ func (d *resultEncoder) updateDataEncoding(chsID uint16) {
 	if err != nil {
 		logutil.BgLogger().Warn("unknown charset ID", zap.Error(err))
 	}
-	d.dataEncoding = charset.FindEncodingUTF8AsNoop(chs)
+	d.dataEncoding = charset.FindEncodingTakeUTF8AsNoop(chs)
 	d.dataIsBinary = chsID == mysql.BinaryDefaultCollationID
 }