From d20d338a3a3a9323330100a5e913e76ed3217514 Mon Sep 17 00:00:00 2001
From: Trustin Lee <trustin@gmail.com>
Date: Fri, 28 Nov 2008 14:47:49 +0000
Subject: [PATCH] Clarification on the overhead of throwing an error

---
 .../jboss/netty/handler/codec/replay/ReplayingDecoder.java   | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/src/main/java/org/jboss/netty/handler/codec/replay/ReplayingDecoder.java b/src/main/java/org/jboss/netty/handler/codec/replay/ReplayingDecoder.java
index 941d74d802..604f6eafc6 100644
--- a/src/main/java/org/jboss/netty/handler/codec/replay/ReplayingDecoder.java
+++ b/src/main/java/org/jboss/netty/handler/codec/replay/ReplayingDecoder.java
@@ -99,6 +99,11 @@ import org.jboss.netty.handler.codec.frame.FrameDecoder;
  * back to the 'initial' position (i.e. the beginning of the buffer) and call
  * the {@code decode(..)} method again when more data is received into the
  * buffer.
+ * <p>
+ * Please note that the overhead of throwing an {@link Error} is minimal unlike
+ * throwing a new {@link Exception} in an ordinary way. {@link ReplayingDecoder}
+ * reuses the same {@link Error} instance so that it does not need to fill its
+ * stack trace, which takes most of {@link Exception} initialization time.
  *
  * <h3>Limitations</h3>
  * <p>