Author: kkolinko
Date: Tue Nov 16 13:08:15 2010
New Revision: 1035628
URL: http://svn.apache.org/viewvc?rev=1035628&view=rev
Log:
tabs -> 4 spaces
This is to simplify backporting BZ 50228 patch (r1032232) from trunk.
CTR
Modified:
tomcat/tc6.0.x/trunk/java/org/apache/jasper/runtime/PageContex
https://issues.apache.org/bugzilla/show_bug.cgi?id=50279
Summary: Tab police
Product: Taglibs
Version: unspecified
Platform: PC
OS/Version: Windows XP
Status: NEW
Severity: normal
Priority: P2
Component: Un
As well as the tabs, there are broader inconsistencies in the style (e.g.
consistent braces, missing javadoc, and the like) that lead to IDE warnings.
How about running everything through a re-formatter to clean this up? Downside
is that it will make back-patching harder.
+1 from me.
On Nov 16,
I saw the following fragment in ContainerBase.logName():
Lines 1277-1278:
} else if (name.startsWith("##")) {
name = "/" + name;
}
It was added by
http://svn.apache.org/viewvc?view=revision&revision=1029527
Shouldn't it be
name = "/" + name.substring(2);
Anyway,
1) I
On 16 November 2010 15:38, Jeremy Boynes wrote:
> As well as the tabs, there are broader inconsistencies in the style (e.g.
> consistent braces, missing javadoc, and the like) that lead to IDE warnings.
>
> How about running everything through a re-formatter to clean this up?
> Downside is that
Author: markt
Date: Tue Nov 16 16:41:13 2010
New Revision: 1035685
URL: http://svn.apache.org/viewvc?rev=1035685&view=rev
Log:
Avoid NPE when removing sessions
Modified:
tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java
Modified: tomcat/trunk/java/org/apache/catalina/session/Man
Author: markt
Date: Tue Nov 16 16:46:50 2010
New Revision: 1035688
URL: http://svn.apache.org/viewvc?rev=1035688&view=rev
Log:
Add some benchmark code for session creation.
A performance presentation at JavaOne 2010 highlighted Tomcat's session
creation as a synchronisation bottleneck for multipl
Author: markt
Date: Tue Nov 16 17:13:29 2010
New Revision: 1035699
URL: http://svn.apache.org/viewvc?rev=1035699&view=rev
Log:
Session manager performance
Narrow sync - few % improvement
Modified:
tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java
tomcat/trunk/test/org/apache/
Author: markt
Date: Tue Nov 16 17:22:45 2010
New Revision: 1035701
URL: http://svn.apache.org/viewvc?rev=1035701&view=rev
Log:
Session manager performance
Switch to a queue of message digests rather than a single sync'd digest
Small improvement but is part of removing the sync completely that shou
On Tue, 2010-11-16 at 17:13 +, ma...@apache.org wrote:
> URL: http://svn.apache.org/viewvc?rev=1035699&view=rev
> Log:
> Session manager performance
> Narrow sync - few % improvement
> - * 1 thread - 2,422ms
> - * 2 threads - 5,047ms
> - * 4 threads - 13,188ms
> - * 16 thr
On 16/11/2010 15:44, Konstantin Kolinko wrote:
> I saw the following fragment in ContainerBase.logName():
> Lines 1277-1278:
>
> } else if (name.startsWith("##")) {
> name = "/" + name;
> }
>
> It was added by
> http://svn.apache.org/viewvc?view=revision&revision=1029527
>
On 16/11/2010 17:33, Remy Maucherat wrote:
> On Tue, 2010-11-16 at 17:13 +, ma...@apache.org wrote:
>> URL: http://svn.apache.org/viewvc?rev=1035699&view=rev
>> Log:
>> Session manager performance
>> Narrow sync - few % improvement
>> - * 1 thread - 2,422ms
>> - * 2 threads - 5,047
Author: markt
Date: Tue Nov 16 17:48:07 2010
New Revision: 1035709
URL: http://svn.apache.org/viewvc?rev=1035709&view=rev
Log:
Session manager performance
Big performance improvement for Windows. Don't try to create randomIS on every
single session creation call
Modified:
tomcat/trunk/java/
Author: markt
Date: Tue Nov 16 17:50:34 2010
New Revision: 1035711
URL: http://svn.apache.org/viewvc?rev=1035711&view=rev
Log:
Update the metrics - looking a lot better on Windows
Modified:
tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java
Modified: tomcat/trunk/test/org/apache/c
Author: markt
Date: Tue Nov 16 18:07:56 2010
New Revision: 1035720
URL: http://svn.apache.org/viewvc?rev=1035720&view=rev
Log:
Session manager performance
Move sync. No performance change but sync requirements for sub-classes are
clearer.
Still looking to remove sync completely.
Modified:
to
On 16 November 2010 18:07, wrote:
> Author: markt
> Date: Tue Nov 16 18:07:56 2010
> New Revision: 1035720
>
> URL: http://svn.apache.org/viewvc?rev=1035720&view=rev
> Log:
> Session manager performance
> Move sync. No performance change but sync requirements for sub-classes are
> clearer.
> Sti
On 16/11/2010 18:35, sebb wrote:
> On 16 November 2010 18:07, wrote:
>> Author: markt
>> Date: Tue Nov 16 18:07:56 2010
>> New Revision: 1035720
>>
>> URL: http://svn.apache.org/viewvc?rev=1035720&view=rev
>> Log:
>> Session manager performance
>> Move sync. No performance change but sync require
Author: markt
Date: Tue Nov 16 19:39:07 2010
New Revision: 1035759
URL: http://svn.apache.org/viewvc?rev=1035759&view=rev
Log:
Additional results for reference
Modified:
tomcat/trunk/test/org/apache/catalina/session/Benchmarks.java
Modified: tomcat/trunk/test/org/apache/catalina/session/Benc
4-core OSX (for comparison with Winders)
Did half a dozen runs in Eclipse, for just this unit test.
for testManagerBaseGenerateSessionId()
1 thread -~500ms
2 threads - ~1200ms
4 threads - ~2500ms
16 threads - ~1ms
for testManagerBaseCreateSession()
1 thread -~700ms
2 th
2010/11/16 Mark Thomas :
> On 16/11/2010 15:44, Konstantin Kolinko wrote:
>> I saw the following fragment in ContainerBase.logName():
>> Lines 1277-1278:
>>
>> } else if (name.startsWith("##")) {
>> name = "/" + name;
>> }
>>
>> It was added by
>> http://svn.apache.org/viewvc
https://issues.apache.org/bugzilla/show_bug.cgi?id=50282
Summary: Reference from
javax.security.auth.login.LoginContext.contextClassLoa
der initalized in static code block prevents an
undeployed Web application from being garba
What's the formatter you plan to use? Is it public and other committer
obeyed? Otherwise, it will mess up again in future..
Anyway, +1 for tab replacement with 4 spaces.
-Rex
2010/11/16 Jeremy Boynes
> As well as the tabs, there are broader inconsistencies in the style (e.g.
> consistent brace
+1 for reformatting.
I've lived with the terrible code style in taglibs for years because I
felt that reformatting it just for me was over the top. Now there are
more eyes on the code; PLEASE MAKE IT READABLE! :)
I'm assuming it's a one-time only reformat to get us onto something sane.
Hen
On T
https://issues.apache.org/bugzilla/show_bug.cgi?id=50279
--- Comment #1 from Jeremy Boynes 2010-11-17 02:05:43 EST
---
Created an attachment (id=26304)
--> (https://issues.apache.org/bugzilla/attachment.cgi?id=26304)
IDEA reformat of impl module
Results of reformatting impl using IDEA
Added T
For a one-time reformat I'd probably go with the one in IDEA followed by manual
review for where it messed up line breaks etc.
For ongoing checks, we could add a checkstyle report to the build using the
same ruleset as Tomcat (or a stricter subset if desired).
The results would look something l
25 matches
Mail list logo
