You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@commons.apache.org by "Matt Whitlock (JIRA)" <ji...@apache.org> on 2012/04/29 22:03:52 UTC
[jira] [Created] (OGNL-49) "Double-checked locking" anti-pattern in
OgnlRuntime
Matt Whitlock created OGNL-49:
---------------------------------
Summary: "Double-checked locking" anti-pattern in OgnlRuntime
Key: OGNL-49
URL: https://issues.apache.org/jira/browse/OGNL-49
Project: Commons OGNL
Issue Type: Bug
Environment: java version "1.6.0_31"
Java(TM) SE Runtime Environment (build 1.6.0_31-b04)
Java HotSpot(TM) 64-Bit Server VM (build 20.6-b01, mixed mode)
Reporter: Matt Whitlock
Priority: Critical
Attachments: 0001-ognl-avoid-double-checked-locking-anti-pattern-in-Og.patch
OgnlRuntime in OGNL 3.0.5 has several occurrences of the "double-checked locking" anti-pattern. This has already manifested in our product in the form of:
ognl.MethodFailedException: Method "xxx" failed for object yyy [java.lang.NoSuchMethodException: x.y.z.C.getFoo(java.lang.String)]
But method x.y.z.C.getFoo(String) does exist, and in fact the same OGNL expression evaluates correctly when it is tried again subsequently.
There are 11 occurrences of double-checked locking in OgnlRuntime that must be corrected. The simplest approach is to remove the unsynchronized accesses to the guarded structures. A better approach would be to use ReadWriteLocks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (OGNL-49) "Double-checked locking" anti-pattern
in OgnlRuntime
Posted by "Simone Tripodi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/OGNL-49?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13293921#comment-13293921 ]
Simone Tripodi commented on OGNL-49:
------------------------------------
while applying the patch, the {{patch}} command refuses 16 changes, maybe because of the {{OgnlRuntime}} class evolvement - can you please double check?
TIA,
-Simo
> "Double-checked locking" anti-pattern in OgnlRuntime
> ----------------------------------------------------
>
> Key: OGNL-49
> URL: https://issues.apache.org/jira/browse/OGNL-49
> Project: Commons OGNL
> Issue Type: Bug
> Environment: java version "1.6.0_31"
> Java(TM) SE Runtime Environment (build 1.6.0_31-b04)
> Java HotSpot(TM) 64-Bit Server VM (build 20.6-b01, mixed mode)
> Reporter: Matt Whitlock
> Assignee: Simone Tripodi
> Priority: Critical
> Attachments: 0001-ognl-avoid-double-checked-locking-anti-pattern-in-Og.patch
>
>
> OgnlRuntime in OGNL 3.0.5 has several occurrences of the "double-checked locking" anti-pattern. This has already manifested in our product in the form of:
> ognl.MethodFailedException: Method "xxx" failed for object yyy [java.lang.NoSuchMethodException: x.y.z.C.getFoo(java.lang.String)]
> But method x.y.z.C.getFoo(String) does exist, and in fact the same OGNL expression evaluates correctly when it is tried again subsequently.
> There are 11 occurrences of double-checked locking in OgnlRuntime that must be corrected. The simplest approach is to remove the unsynchronized accesses to the guarded structures. A better approach would be to use ReadWriteLocks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (OGNL-49) "Double-checked locking" anti-pattern in
OgnlRuntime
Posted by "Matt Whitlock (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/OGNL-49?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Matt Whitlock updated OGNL-49:
------------------------------
Attachment: 0001-ognl-avoid-double-checked-locking-anti-pattern-in-Og.patch
Here is a patch that applies the simplest possible fix.
> "Double-checked locking" anti-pattern in OgnlRuntime
> ----------------------------------------------------
>
> Key: OGNL-49
> URL: https://issues.apache.org/jira/browse/OGNL-49
> Project: Commons OGNL
> Issue Type: Bug
> Environment: java version "1.6.0_31"
> Java(TM) SE Runtime Environment (build 1.6.0_31-b04)
> Java HotSpot(TM) 64-Bit Server VM (build 20.6-b01, mixed mode)
> Reporter: Matt Whitlock
> Priority: Critical
> Attachments: 0001-ognl-avoid-double-checked-locking-anti-pattern-in-Og.patch
>
>
> OgnlRuntime in OGNL 3.0.5 has several occurrences of the "double-checked locking" anti-pattern. This has already manifested in our product in the form of:
> ognl.MethodFailedException: Method "xxx" failed for object yyy [java.lang.NoSuchMethodException: x.y.z.C.getFoo(java.lang.String)]
> But method x.y.z.C.getFoo(String) does exist, and in fact the same OGNL expression evaluates correctly when it is tried again subsequently.
> There are 11 occurrences of double-checked locking in OgnlRuntime that must be corrected. The simplest approach is to remove the unsynchronized accesses to the guarded structures. A better approach would be to use ReadWriteLocks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (OGNL-49) "Double-checked locking" anti-pattern
in OgnlRuntime
Posted by "Matt Whitlock (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/OGNL-49?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13295402#comment-13295402 ]
Matt Whitlock commented on OGNL-49:
-----------------------------------
The patch was based on OGNL 3.0.5. Indeed there has been a significant reorganization in trunk since 3.0.5, so the patch no longer applies at all. Unfortunately, trunk still exhibits the double-checked locking anti-pattern. I'll submit another patch when I can get around to it.
> "Double-checked locking" anti-pattern in OgnlRuntime
> ----------------------------------------------------
>
> Key: OGNL-49
> URL: https://issues.apache.org/jira/browse/OGNL-49
> Project: Commons OGNL
> Issue Type: Bug
> Environment: java version "1.6.0_31"
> Java(TM) SE Runtime Environment (build 1.6.0_31-b04)
> Java HotSpot(TM) 64-Bit Server VM (build 20.6-b01, mixed mode)
> Reporter: Matt Whitlock
> Assignee: Simone Tripodi
> Priority: Critical
> Attachments: 0001-ognl-avoid-double-checked-locking-anti-pattern-in-Og.patch
>
>
> OgnlRuntime in OGNL 3.0.5 has several occurrences of the "double-checked locking" anti-pattern. This has already manifested in our product in the form of:
> ognl.MethodFailedException: Method "xxx" failed for object yyy [java.lang.NoSuchMethodException: x.y.z.C.getFoo(java.lang.String)]
> But method x.y.z.C.getFoo(String) does exist, and in fact the same OGNL expression evaluates correctly when it is tried again subsequently.
> There are 11 occurrences of double-checked locking in OgnlRuntime that must be corrected. The simplest approach is to remove the unsynchronized accesses to the guarded structures. A better approach would be to use ReadWriteLocks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira