diff --git a/contrib/benchmark/README.enwiki b/contrib/benchmark/README.enwiki
index f9d49300e87..62999495908 100644
--- a/contrib/benchmark/README.enwiki
+++ b/contrib/benchmark/README.enwiki
@@ -20,3 +20,50 @@ After that, ant enwiki should process the data set and run a load
 test. Ant targets get-enwiki, expand-enwiki, and extract-enwiki can
 also be used to download, decompress, and extract (to individual files
 in work/enwiki) the dataset, respectively.
+
+NOTE: This bug in Xerces:
+
+  https://issues.apache.org/jira/browse/XERCESJ-1257
+
+which is still present as of 2.9.1, causes an exception like this when
+processing Wikipedia's XML:
+
+Caused by: org.apache.xerces.impl.io.MalformedByteSequenceException: Invalid byte 2 of 4-byte UTF-8 sequence.
+	at org.apache.xerces.util.ErrorHandlerWrapper.createSAXParseException(Unknown Source)
+	at org.apache.xerces.util.ErrorHandlerWrapper.fatalError(Unknown Source)
+	at org.apache.xerces.impl.XMLErrorReporter.reportError(Unknown Source)
+	at org.apache.xerces.impl.XMLErrorReporter.reportError(Unknown Source)
+	at org.apache.xerces.impl.XMLDocumentFragmentScannerImpl$FragmentContentDispatcher.dispatch(Unknown Source)
+	at org.apache.xerces.impl.XMLDocumentFragmentScannerImpl.scanDocument(Unknown Source)
+	at org.apache.xerces.parsers.XML11Configuration.parse(Unknown Source)
+	at org.apache.xerces.parsers.XML11Configuration.parse(Unknown Source)
+	at org.apache.xerces.parsers.XMLParser.parse(Unknown Source)
+	at org.apache.xerces.parsers.AbstractSAXParser.parse(Unknown Source)
+	at org.apache.lucene.benchmark.byTask.feeds.EnwikiDocMaker$Parser.run(EnwikiDocMaker.java:77)
+	... 1 more
+
+The original poster in the Xerces bug provided this patch:
+
+--- UTF8Reader.java	2006-11-23 00:36:53.000000000 +0100
++++ /home/rainman/lucene/xerces-2_9_0/src/org/apache/xerces/impl/io/UTF8Reader.java	2008-04-04 00:40:58.000000000 +0200
+@@ -534,6 +534,16 @@
+                     invalidByte(4, 4, b2);
+                 }
+ 
++                // check if output buffer is large enough to hold 2 surrogate chars
++                if( out + 1 >= offset + length ){
++                    fBuffer[0] = (byte)b0;
++                    fBuffer[1] = (byte)b1;
++                    fBuffer[2] = (byte)b2;
++                    fBuffer[3] = (byte)b3;
++                    fOffset = 4;
++                    return out - offset;
++		}
++
+                 // decode bytes into surrogate characters
+                 int uuuuu = ((b0 << 2) & 0x001C) | ((b1 >> 4) & 0x0003);
+                 if (uuuuu > 0x10) {
+
+which I've applied to Xerces 2.9.1 sources, and committed under
+lib/xerces-2.9.1-patched-XERCESJ-1257.jar.  Once XERCESJ-1257 is fixed
+we can upgrade to a standard Xerces release.
diff --git a/contrib/benchmark/build.xml b/contrib/benchmark/build.xml
index 05174e962d6..25209865db2 100644
--- a/contrib/benchmark/build.xml
+++ b/contrib/benchmark/build.xml
@@ -104,7 +104,7 @@
     <property name="collections.jar" value="commons-collections-3.1.jar"/>
     <property name="logging.jar" value="commons-logging-1.0.4.jar"/>
     <property name="bean-utils.jar" value="commons-beanutils-1.7.0.jar"/>
-    <property name="xercesImpl.jar" value="xerces-2.9.0.jar"/>
+    <property name="xercesImpl.jar" value="xerces-2.9.1-patched-XERCESJ-1257.jar"/>
     <property name="xml-apis.jar" value="xml-apis-2.9.0.jar"/>
 
     <path id="classpath">
diff --git a/contrib/benchmark/lib/xerces-2.9.0.jar b/contrib/benchmark/lib/xerces-2.9.0.jar
deleted file mode 100644
index b190fd95304..00000000000
--- a/contrib/benchmark/lib/xerces-2.9.0.jar
+++ /dev/null
@@ -1,2 +0,0 @@
-AnyObjectId[50df93072e50efffced25bc814f0f20426a6b69c] was removed in git history.
-Apache SVN contains full history.
\ No newline at end of file
diff --git a/contrib/benchmark/lib/xerces-2.9.1-patched-XERCESJ-1257.jar b/contrib/benchmark/lib/xerces-2.9.1-patched-XERCESJ-1257.jar
new file mode 100644
index 00000000000..e1665118f98
--- /dev/null
+++ b/contrib/benchmark/lib/xerces-2.9.1-patched-XERCESJ-1257.jar
@@ -0,0 +1,2 @@
+AnyObjectId[ea000b3b79ec201e637841bebce16cf004231096] was removed in git history.
+Apache SVN contains full history.
\ No newline at end of file