Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Karl Pauls
> > Hi folks, > > > > I'd like to start a vote for the 1.4.0 release of the maven-bundle-plugin, > > which now has complete integration of the OBR goals with the bundle > > life-cycle (plus a couple of improvements made possible by this work) > > along with support for attaching bundles with

Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Carsten Ziegeler
Karl Pauls wrote: > > Hi folks, > > I'd like to start a vote for the 1.4.0 release of the maven-bundle-plugin, > which now has complete integration of the OBR goals with the bundle > life-cycle (plus a couple of improvements made possible by this work) > along with support for attaching bu

Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Stuart McCulloch
On 23/02/2008, Stuart McCulloch <[EMAIL PROTECTED]> wrote: > > Hi folks, > > I'd like to start a vote for the 1.4.0 release of the maven-bundle-plugin, > which now has complete integration of the OBR goals with the bundle > life-cycle (plus a couple of improvements made possible by this work) > alo

Performance problems in classloading

2008-02-25 Thread Guillaume Nodet
I'm currently working on ServiceMix 4 which uses Felix.\ I have some really important performance problems in classloading. When loading JBI applications (they have some very specific classloading architecture that must be implemented to be JBI compliant), 95% of the time is spent in the followin

Re: Performance problems in classloading

2008-02-25 Thread Stuart McCulloch
On 25/02/2008, Guillaume Nodet <[EMAIL PROTECTED]> wrote: > > I'm currently working on ServiceMix 4 which uses Felix.\ > > I have some really important performance problems in classloading. > When loading JBI applications (they have some very specific > classloading architecture > that must be impl

Re: Performance problems in classloading

2008-02-25 Thread Guillaume Nodet
Thanks Stuart. The exception was catched and the getMessage method called, so I've hacked a bit more, but the following patch seems to work great for me. Index: src/main/java/org/apache/felix/moduleloader/ModuleImpl.java === --- src/

Re: Performance problems in classloading

2008-02-25 Thread Stuart McCulloch
On 25/02/2008, Guillaume Nodet <[EMAIL PROTECTED]> wrote: > > Thanks Stuart. The exception was catched and the getMessage method > called, so I've > hacked a bit more, but the following patch seems to work great for me. cool, could you raise a JIRA issue and attach the combined patch - thanks I

Re: Performance problems in classloading

2008-02-25 Thread Guillaume Nodet
I suppose we could make the ClassNotFoundException class a static class, give it a transient reference to the IModule and make sure the message is extracted when the Exception is serialized. Thoughts ? On Mon, Feb 25, 2008 at 4:47 PM, Dale Peakall <[EMAIL PROTECTED]> wrote: > These patches do have

Re: Performance problems in classloading

2008-02-25 Thread Dale Peakall
These patches do have the problem that exceptions are supposed to be Serializable - and IModule is not Serializable. I don't know if the information necessary to construct the message is Serializable and can be quickly extracted from the module definition - given the apparent run-time cost of

[jira] Created: (FELIX-500) Performance problems when using the OSGi classloaders in a certain way

2008-02-25 Thread Guillaume Nodet (JIRA)
Performance problems when using the OSGi classloaders in a certain way -- Key: FELIX-500 URL: https://issues.apache.org/jira/browse/FELIX-500 Project: Felix Issue Type: Impr

[jira] Updated: (FELIX-500) Performance problems when using the OSGi classloaders in a certain way

2008-02-25 Thread Guillaume Nodet (JIRA)
[ https://issues.apache.org/jira/browse/FELIX-500?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Guillaume Nodet updated FELIX-500: -- Attachment: FELIX-500.patch Original idea from Stuart. > Performance problems when using the OS

Re: Performance problems in classloading

2008-02-25 Thread Guillaume Nodet
I've raised a JIRA with a patch that should accomodate this problem. I must admit this is not the cleanest thing, but given a 95% performance boost in my case I'd be happy it it is included :-) See https://issues.apache.org/jira/browse/FELIX-500 On Mon, Feb 25, 2008 at 5:05 PM, Guillaume Nodet <[E

Re: Performance problems in classloading

2008-02-25 Thread Karl Pauls
I agree that we will have to fix this. I will look into it. Thanks for the patch. regards, Karl > I've raised a JIRA with a patch that should accomodate this problem. > I must admit this is not the cleanest thing, but given a 95% > performance boost in my case > I'd be happy it it is included

Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Karl Pauls
+1 Looks good to me. There is this build.xml that doesn't have a license header and i'm not sure whether it needs one but that is not a big issue. regards, Karl > Hi folks, > > I'd like to start a vote for the 1.4.0 release of the maven-bundle-plugin, > which now has complete integration of th

Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Stuart McCulloch
On 26/02/2008, Karl Pauls <[EMAIL PROTECTED]> wrote: > > +1 Looks good to me. There is this build.xml that doesn't have a > license header and i'm not sure whether it needs one but that is not a > big issue. yep - I decided not to add a license header to the build.xml file because a) I consider i

Re: [Vote] release maven-bundle-plugin 1.4.0

2008-02-25 Thread Carsten Ziegeler
+1 Carsten Stuart McCulloch wrote: Hi folks, I'd like to start a vote for the 1.4.0 release of the maven-bundle-plugin, which now has complete integration of the OBR goals with the bundle life-cycle (plus a couple of improvements made possible by this work) along with support for attaching bund