diff --git a/csv/src/main/java/com/fasterxml/jackson/dataformat/csv/CsvParser.java b/csv/src/main/java/com/fasterxml/jackson/dataformat/csv/CsvParser.java
index 0ca67c90..279742b9 100644
--- a/csv/src/main/java/com/fasterxml/jackson/dataformat/csv/CsvParser.java
+++ b/csv/src/main/java/com/fasterxml/jackson/dataformat/csv/CsvParser.java
@@ -24,7 +24,10 @@
 public class CsvParser
     extends ParserMinimalBase
 {
-    /**
+    // @since 2.9.9: just to protect against bugs, DoS, limit number of column defs we may read
+    private final static int MAX_COLUMNS = 99999;
+
+            /**
      * Enumeration that defines all togglable features for CSV parsers
      */
     public enum Feature
@@ -351,7 +354,10 @@ private Feature(boolean defaultState) {
     public CsvParser(CsvIOContext ctxt, int stdFeatures, int csvFeatures,
             ObjectCodec codec, Reader reader)
     {
-        super(stdFeatures);    
+        super(stdFeatures);
+        if (reader == null) {
+            throw new IllegalArgumentException("Can not pass `null` as `java.io.Reader` to read from");
+        }
         _objectCodec = codec;
         _textBuffer =  ctxt.csvTextBuffer();
         DupDetector dups = JsonParser.Feature.STRICT_DUPLICATE_DETECTION.enabledIn(stdFeatures)
@@ -720,10 +726,14 @@ protected void _readHeaderLine() throws IOException {
                 if ((name = _reader.nextString()) != null) {
                     _reportError(String.format("Extra header %s", name));
                 }
-            }
-            else {
-                //noinspection StatementWithEmptyBody
-                while (_reader.nextString() != null) { /* does nothing */ }
+            } else {
+                int allowed = MAX_COLUMNS;
+                while (_reader.nextString() != null) {
+                    // If we don't care about validation, just skip. But protect against infinite loop
+                    if (--allowed < 0) {
+                        _reportError("Internal error: skipped "+MAX_COLUMNS+" header columns");
+                    }
+                }
             }
             return;
         }
@@ -731,6 +741,7 @@ protected void _readHeaderLine() throws IOException {
         // either the schema is empty or reorder columns flag is set
         String name;
         CsvSchema.Builder builder = _schema.rebuild().clearColumns();
+        int count = 0;
 
         while ((name = _reader.nextString()) != null) {
             // one more thing: always trim names, regardless of config settings
@@ -743,6 +754,9 @@ protected void _readHeaderLine() throws IOException {
             } else {
                 builder.addColumn(name);
             }
+            if (++count > MAX_COLUMNS) {
+                _reportError("Internal error: reached maximum of "+MAX_COLUMNS+" header columns");
+            }
         }
 
         // Ok: did we get any  columns?
@@ -777,9 +791,8 @@ protected JsonToken _handleStartDoc() throws IOException
             _reader.skipLeadingComments();
         }
         
-        /* Only one real complication, actually; empy documents (zero bytes).
-         * Those have no entries. Should be easy enough to detect like so:
-         */
+        // Only one real complication, actually; empty documents (zero bytes).
+        // Those have no entries. Should be easy enough to detect like so:
         final boolean wrapAsArray = Feature.WRAP_AS_ARRAY.enabledIn(_formatFeatures);
         if (!_reader.hasMoreInput()) {
             _state = STATE_DOC_END;
diff --git a/csv/src/test/java/com/fasterxml/jackson/dataformat/csv/NullReader122Test.java b/csv/src/test/java/com/fasterxml/jackson/dataformat/csv/NullReader122Test.java
new file mode 100644
index 00000000..5e04825d
--- /dev/null
+++ b/csv/src/test/java/com/fasterxml/jackson/dataformat/csv/NullReader122Test.java
@@ -0,0 +1,23 @@
+package com.fasterxml.jackson.dataformat.csv;
+
+import java.io.*;
+import java.util.Map;
+
+import com.fasterxml.jackson.databind.ObjectReader;
+
+public class NullReader122Test extends ModuleTestBase
+{
+    private final CsvMapper MAPPER = mapperForCsv();
+
+    // for [dataformats-text#122]: passing `null` Reader leads to infinite loop
+    public void testEmptyStream() throws Exception {
+        CsvSchema columns = CsvSchema.emptySchema().withHeader().withColumnSeparator(';');
+        ObjectReader r = MAPPER.readerFor(Map.class).with(columns);
+        try {
+            /*Object ob =*/ r.readValue((Reader) null);
+            fail("Should not pass");
+        } catch (IllegalArgumentException e) {
+            verifyException(e, "Can not pass `null`");
+        }
+    }
+}
diff --git a/release-notes/VERSION-2.x b/release-notes/VERSION-2.x
index 86957721..1d7a60d9 100644
--- a/release-notes/VERSION-2.x
+++ b/release-notes/VERSION-2.x
@@ -10,8 +10,10 @@ Modules:
 
 2.9.9 (not yet released)
 
-#63: `null` Object Id serialized as anchor for YAML
+#63: (yaml) `null` Object Id serialized as anchor for YAML
  (reported by jflefebvre06@github)
+#122: (csv) `readValues(null)` causes infinite loop
+ (reported by andyeko@github) 
 
 2.9.8 (15-Dec-2018)