Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [aspectj-dev] VerifyError in around advice during load time weaving

Hi Andy,
 
The -XnoInline option did not work. The user cannot change the JVM level. The offending code is a small part of a large application. The user is trying to isolate the problem. If he is successful in doing that I will raise a bug.
I was just wondering if you had seen this 'Accessing value from uninitialized register ' error when using AspectJ. I couldn't find any in searching the web.
 
Thanks,
Choudary.
On Thu, May 6, 2010 at 12:37 PM, Andy Clement <andrew.clement@xxxxxxxxx> wrote:
Hi,

Are you able to raise a bug and include the offending code?

You could try running with -XnoInline to see if that makes a
difference.  Or on a different JVM level.

Andy

On 6 May 2010 05:21, Choudary Kothapalli <choudary.kothapalli@xxxxxxxxx> wrote:
> Hi,
>
> My tool MaintainJ captures the runtime call trace using the load time
> weaving mechanism. I use around advice to capture the call trace. It has
> been working well, but one of our users is getting VerifyError on some of
> the classes in his application.
>
> The JRE is Sun JRE- 1.6.  build 19.
> Both 1.6.8 and 1.6.4 of aspectjweaver.jar are giving the problem.
>
> The abstract aspect provided in a jar is overridden by the pointcut specific
> to the application. It is a pretty simple aspect that captures all execution
> join points. I removed all logic in the advice and was just printing the
> thisJoinPoint and still getting the same error. The error looks as below.
>
> Caused by: java.lang.VerifyError: (class: clientPackage/clientClass2,
> method: init$_aroundBody10 signature:
> (LclientPackage/clientClass2;Ljava/lang/String;ZLorg/aspectj/lang/JoinPoint;)V)
> Accessing value from uninitialized register 4
> at clientPackage.clientClass1.clientMethod(clientClass1.java:70)
>
> I checked the code beaing woven and there is nothing weird about it. Let me
> know if you need any other details to figure this out.
>
> Thanks,
> Choudary Kothapalli.
>
> _______________________________________________
> aspectj-dev mailing list
> aspectj-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/aspectj-dev
>
>
_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/aspectj-dev


Back to the top