https://issues.apache.org/bugzilla/show_bug.cgi?id=52770
Bug #: 52770
Summary: Potential Bug or Inconsistency in
NioBlockingSelector.java
Product: Tomcat 7
Version: trunk
Platform: All
OS/Version: All
Status: NEW
Severity: normal
Priority: P2
Component: Catalina
AssignedTo: dev@tomcat.apache.org
ReportedBy: msrbugzi...@gmail.com
Classification: Unclassified
This is Ken Cheung, a Computer Science M.Phil. student. I observed some
code clones in Tomcat and found inconsistent code:
/tomcat/trunk/java/org/apache/tomcat/util/net/NioBlockingSelector.java
103 try {
104 if ( att.getWriteLatch()==null ||
att.getWriteLatch().getCount()==0) att.startWriteLatch(1);
105 poller.add(att,SelectionKey.OP_WRITE,reference);
106
att.awaitWriteLatch(writeTimeout,TimeUnit.MILLISECONDS);
107 }catch (InterruptedException ignore) {
108 Thread.interrupted();
109 }
110 if ( att.getWriteLatch()!=null &&
att.getWriteLatch().getCount()> 0) {
111 //we got interrupted, but we haven't received
notification from the poller.
112 keycount = 0;
113 }else {
114 //latch countdown has happened
115 keycount = 1;
116 att.resetWriteLatch();
117 }
118
119 if (writeTimeout > 0 && (keycount == 0))
120 timedout = (System.currentTimeMillis() - time) >=
writeTimeout;
/tomcat/trunk/java/org/apache/tomcat/util/net/NioBlockingSelector.java
164 try {
165 if ( att.getReadLatch()==null ||
att.getReadLatch().getCount()==0) att.startReadLatch(1);
166 poller.add(att,SelectionKey.OP_READ, reference);
167 if (readTimeout < 0) {
168 att.awaitReadLatch(Long.MAX_VALUE,
TimeUnit.MILLISECONDS);
169 } else {
170 att.awaitReadLatch(readTimeout,
TimeUnit.MILLISECONDS);
171 }
172 }catch (InterruptedException ignore) {
173 Thread.interrupted();
174 }
175 if ( att.getReadLatch()!=null &&
att.getReadLatch().getCount()> 0) {
176 //we got interrupted, but we haven't received
notification from the poller.
177 keycount = 0;
178 }else {
179 //latch countdown has happened
180 keycount = 1;
181 att.resetReadLatch();
182 }
183 if (readTimeout >= 0 && (keycount == 0))
184 timedout = (System.currentTimeMillis() - time) >=
readTimeout;
Quick description of the inconsistency
Two code snippets are very similar code, but as you see, the first code does
not check "if (readTimeout < 0)" while the second code has the checker.
We thought it could be a potential bug or inconsistency. Hope this helps.
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
