diff --git a/.golangci.yml b/.golangci.yml
index dddccb2f25e39..a4a087333ed3c 100644
--- a/.golangci.yml
+++ b/.golangci.yml
@@ -21,6 +21,8 @@ linters:
     - exportloopref
     - rowserrcheck
     - unconvert
+    - makezero
+
 linters-settings:
   staticcheck:
     checks: ["S1002","S1004","S1007","S1009","S1010","S1012","S1019","S1020","S1021","S1024","S1030","SA2*","SA3*","SA4009","SA5*","SA6000","SA6001","SA6005", "-SA2002"]
@@ -36,3 +38,4 @@ issues:
         - errcheck
         - gosec
         - rowserrcheck
+        - makezero
diff --git a/executor/index_lookup_hash_join.go b/executor/index_lookup_hash_join.go
index b77e446c62104..755c0713f13f8 100644
--- a/executor/index_lookup_hash_join.go
+++ b/executor/index_lookup_hash_join.go
@@ -759,7 +759,7 @@ func (iw *indexHashJoinInnerWorker) doJoinInOrder(ctx context.Context, task *ind
 		}
 	}
 	// TODO: matchedInnerRowPtrs and matchedInnerRows can be moved to inner worker.
-	matchedInnerRows := make([]chunk.Row, len(task.matchedInnerRowPtrs))
+	matchedInnerRows := make([]chunk.Row, 0, len(task.matchedInnerRowPtrs))
 	var hasMatched, hasNull, ok bool
 	for chkIdx, innerRowPtrs4Chk := range task.matchedInnerRowPtrs {
 		for outerRowIdx, innerRowPtrs := range innerRowPtrs4Chk {
diff --git a/executor/slow_query.go b/executor/slow_query.go
index 09f3b2b6df41d..0b57d1fc55e6d 100755
--- a/executor/slow_query.go
+++ b/executor/slow_query.go
@@ -264,7 +264,7 @@ func getOneLine(reader *bufio.Reader) ([]byte, error) {
 	var tempLine []byte
 	for isPrefix {
 		tempLine, isPrefix, err = reader.ReadLine()
-		resByte = append(resByte, tempLine...)
+		resByte = append(resByte, tempLine...) // nozero
 		// Use the max value of max_allowed_packet to check the single line length.
 		if len(resByte) > int(variable.MaxOfMaxAllowedPacket) {
 			return resByte, errors.Errorf("single line length exceeds limit: %v", variable.MaxOfMaxAllowedPacket)
@@ -1050,7 +1050,7 @@ func readLastLines(ctx context.Context, file *os.File, endCursor int64) ([]strin
 		if err != nil {
 			return nil, 0, err
 		}
-		lines = append(chars, lines...)
+		lines = append(chars, lines...) // nozero
 
 		// find first '\n' or '\r'
 		for i := 0; i < len(chars); i++ {
diff --git a/kv/key.go b/kv/key.go
index 191bc65eb7ac3..ae75765535331 100644
--- a/kv/key.go
+++ b/kv/key.go
@@ -59,7 +59,7 @@ func (k Key) PrefixNext() Key {
 	}
 	if i == -1 {
 		copy(buf, k)
-		buf = append(buf, 0)
+		buf = append(buf, 0) // nozero
 	}
 	return buf
 }
diff --git a/planner/cascades/transformation_rules.go b/planner/cascades/transformation_rules.go
index 14cf33a6410d5..6a0d12eb977c0 100644
--- a/planner/cascades/transformation_rules.go
+++ b/planner/cascades/transformation_rules.go
@@ -920,7 +920,7 @@ func (r *pushDownJoin) predicatePushDown(
 			leftCond = append(join.LeftConditions, derivedLeftJoinCond...)
 			join.LeftConditions = nil
 			remainCond = append(expression.ScalarFuncs2Exprs(equalCond), otherCond...)
-			remainCond = append(remainCond, leftPushCond...)
+			remainCond = append(remainCond, leftPushCond...) // nozero
 		} else {
 			remainCond = expression.ExtractFiltersFromDNFs(join.SCtx(), remainCond)
 			// Only derive left where condition, because right where condition cannot be pushed down
@@ -931,7 +931,7 @@ func (r *pushDownJoin) predicatePushDown(
 			rightCond = append(join.RightConditions, derivedRightJoinCond...)
 			join.RightConditions = nil
 			remainCond = append(expression.ScalarFuncs2Exprs(equalCond), otherCond...)
-			remainCond = append(remainCond, rightPushCond...)
+			remainCond = append(remainCond, rightPushCond...) // nozero
 		}
 	default:
 		// TODO: Enhance this rule to deal with Semi/SmiAnti Joins.
diff --git a/planner/core/exhaust_physical_plans.go b/planner/core/exhaust_physical_plans.go
index b6b83e5ce598d..575a177697325 100644
--- a/planner/core/exhaust_physical_plans.go
+++ b/planner/core/exhaust_physical_plans.go
@@ -482,11 +482,11 @@ func (p *LogicalJoin) constructIndexJoin(
 					}
 					outerSchema, innerSchema := p.Children()[outerIdx].Schema(), p.Children()[1-outerIdx].Schema()
 					if outerSchema.Contains(lhs) && innerSchema.Contains(rhs) {
-						outerHashKeys = append(outerHashKeys, lhs)
-						innerHashKeys = append(innerHashKeys, rhs)
+						outerHashKeys = append(outerHashKeys, lhs) // nozero
+						innerHashKeys = append(innerHashKeys, rhs) // nozero
 					} else if innerSchema.Contains(lhs) && outerSchema.Contains(rhs) {
-						outerHashKeys = append(outerHashKeys, rhs)
-						innerHashKeys = append(innerHashKeys, lhs)
+						outerHashKeys = append(outerHashKeys, rhs) // nozero
+						innerHashKeys = append(innerHashKeys, lhs) // nozero
 					}
 					newOtherConds = append(newOtherConds[:i], newOtherConds[i+1:]...)
 				}
@@ -1203,7 +1203,7 @@ func (cwc *ColWithCmpFuncManager) BuildRangesByRow(ctx sessionctx.Context, row c
 		if err != nil {
 			return nil, err
 		}
-		exprs = append(exprs, newExpr)
+		exprs = append(exprs, newExpr) // nozero
 	}
 	ranges, err := ranger.BuildColumnRange(exprs, ctx.GetSessionVars().StmtCtx, cwc.TargetCol.RetType, cwc.colLength)
 	if err != nil {
diff --git a/planner/core/physical_plans.go b/planner/core/physical_plans.go
index 01f5ada33df88..52dd75268fec0 100644
--- a/planner/core/physical_plans.go
+++ b/planner/core/physical_plans.go
@@ -579,13 +579,13 @@ func ExpandVirtualColumn(columns []*model.ColumnInfo, schema *expression.Schema,
 		for _, baseCol := range baseCols {
 			if !schema.Contains(baseCol) {
 				schema.Columns = append(schema.Columns, baseCol)
-				copyColumn = append(copyColumn, FindColumnInfoByID(colsInfo, baseCol.ID))
+				copyColumn = append(copyColumn, FindColumnInfoByID(colsInfo, baseCol.ID)) // nozero
 			}
 		}
 	}
 	if extraColumn != nil {
 		schema.Columns = append(schema.Columns, extraColumn)
-		copyColumn = append(copyColumn, extraColumnModel)
+		copyColumn = append(copyColumn, extraColumnModel) // nozero
 	}
 	return copyColumn
 }
diff --git a/planner/util/path.go b/planner/util/path.go
index 4742a42bc3f79..8c4abd6a94ab5 100644
--- a/planner/util/path.go
+++ b/planner/util/path.go
@@ -103,7 +103,7 @@ func (path *AccessPath) SplitCorColAccessCondFromFilters(ctx sessionctx.Context,
 	}
 	for i, ok := range used {
 		if !ok {
-			remained = append(remained, path.TableFilters[i])
+			remained = append(remained, path.TableFilters[i]) // nozero
 		}
 	}
 	return access, remained
diff --git a/store/gcworker/gc_worker.go b/store/gcworker/gc_worker.go
index ac47930818262..1a8da7c96ad05 100644
--- a/store/gcworker/gc_worker.go
+++ b/store/gcworker/gc_worker.go
@@ -1115,9 +1115,9 @@ retryScanAndResolve:
 			return stat, errors.Errorf("unexpected scanlock error: %s", locksResp)
 		}
 		locksInfo := locksResp.GetLocks()
-		locks := make([]*txnlock.Lock, len(locksInfo))
-		for i := range locksInfo {
-			locks[i] = txnlock.NewLock(locksInfo[i])
+		locks := make([]*txnlock.Lock, 0, len(locksInfo))
+		for _, li := range locksInfo {
+			locks = append(locks, txnlock.NewLock(li))
 		}
 		if w.testingKnobs.scanLocks != nil {
 			locks = append(locks, w.testingKnobs.scanLocks(key, loc.Region.GetID())...)
diff --git a/store/mockstore/unistore/tikv/mvcc/tikv.go b/store/mockstore/unistore/tikv/mvcc/tikv.go
index ef12e2f9eadb1..7b8a52dfd9159 100644
--- a/store/mockstore/unistore/tikv/mvcc/tikv.go
+++ b/store/mockstore/unistore/tikv/mvcc/tikv.go
@@ -70,8 +70,8 @@ const (
 // EncodeWriteCFValue accepts a write cf parameters and return the encoded bytes data.
 // Just like the tikv encoding form. See tikv/src/storage/mvcc/write.rs for more detail.
 func EncodeWriteCFValue(t WriteType, startTs uint64, shortVal []byte) []byte {
-	data := make([]byte, 1)
-	data[0] = t
+	data := make([]byte, 0)
+	data = append(data, t)
 	data = codec.EncodeUvarint(data, startTs)
 	if len(shortVal) != 0 {
 		data = append(data, byte(shortValuePrefix), byte(len(shortVal)))
@@ -82,16 +82,16 @@ func EncodeWriteCFValue(t WriteType, startTs uint64, shortVal []byte) []byte {
 
 // EncodeLockCFValue encodes the mvcc lock and returns putLock value and putDefault value if exists.
 func EncodeLockCFValue(lock *Lock) ([]byte, []byte) {
-	data := make([]byte, 1)
+	data := make([]byte, 0)
 	switch lock.Op {
 	case byte(kvrpcpb.Op_Put):
-		data[0] = LockTypePut
+		data = append(data, LockTypePut)
 	case byte(kvrpcpb.Op_Del):
-		data[0] = LockTypeDelete
+		data = append(data, LockTypeDelete)
 	case byte(kvrpcpb.Op_Lock):
-		data[0] = LockTypeLock
+		data = append(data, LockTypeLock)
 	case byte(kvrpcpb.Op_PessimisticLock):
-		data[0] = LockTypePessimistic
+		data = append(data, LockTypePessimistic)
 	default:
 		panic("invalid lock op")
 	}
diff --git a/tablecodec/tablecodec.go b/tablecodec/tablecodec.go
index 48b4eda5410ee..f610e09104572 100644
--- a/tablecodec/tablecodec.go
+++ b/tablecodec/tablecodec.go
@@ -1159,7 +1159,8 @@ func TryGetCommonPkColumnRestoredIds(tbl *model.TableInfo) []int64 {
 
 // GenIndexValueForClusteredIndexVersion1 generates the index value for the clustered index with version 1(New in v5.0.0).
 func GenIndexValueForClusteredIndexVersion1(sc *stmtctx.StatementContext, tblInfo *model.TableInfo, idxInfo *model.IndexInfo, IdxValNeedRestoredData bool, distinct bool, untouched bool, indexedValues []types.Datum, h kv.Handle, partitionID int64, handleRestoredData []types.Datum) ([]byte, error) {
-	idxVal := make([]byte, 1)
+	idxVal := make([]byte, 0)
+	idxVal = append(idxVal, 0)
 	tailLen := 0
 	// Version info.
 	idxVal = append(idxVal, IndexVersionFlag)
@@ -1216,7 +1217,8 @@ func GenIndexValueForClusteredIndexVersion1(sc *stmtctx.StatementContext, tblInf
 
 // genIndexValueVersion0 create index value for both local and global index.
 func genIndexValueVersion0(sc *stmtctx.StatementContext, tblInfo *model.TableInfo, idxInfo *model.IndexInfo, IdxValNeedRestoredData bool, distinct bool, untouched bool, indexedValues []types.Datum, h kv.Handle, partitionID int64) ([]byte, error) {
-	idxVal := make([]byte, 1)
+	idxVal := make([]byte, 0)
+	idxVal = append(idxVal, 0)
 	newEncode := false
 	tailLen := 0
 	if !h.IsInt() && distinct {