9 messages in net.java.dev.jna.usersRe: [jna-users] jna segfaults on wind...
FromSent OnAttachments
Nick EvgenievJan 31, 2008 12:03 pm 
Timothy WallJan 31, 2008 1:18 pm 
Nick EvgenievFeb 1, 2008 6:29 am 
Timothy WallFeb 1, 2008 6:34 am 
Nick EvgenievFeb 1, 2008 7:07 am 
Timothy WallFeb 1, 2008 7:47 am 
Nick EvgenievFeb 4, 2008 8:14 am 
Timothy WallFeb 4, 2008 8:23 am 
Nick EvgenievFeb 4, 2008 8:37 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:Re: [jna-users] jna segfaults on windows 2003 serverActions...
From:Timothy Wall (twal@dev.java.net)
Date:Feb 4, 2008 8:23:28 am
List:net.java.dev.jna.users

On Feb 4, 2008, at 11:15 AM, Nick Evgeniev wrote:

Hi,

On Feb 1, 2008 6:47 PM, Timothy Wall <twal@dev.java.net> wrote: It looks like SVN rev 310 used plain data memory allocation for callbacks, so that explains the issue. Current SVN rev is 478; 3.0.1 is the official release number. There's a tag and separate download in the project files section.

Definitely upgrade. There have been numerous bug fixes and a few new features since the build.xml version got bumped to 3.0. Especially if you're using stdcall convention callbacks; there was a VM crash fix for those.

I've tried to upgrade to latest jna version, but had to rollback because of the following issue - when I run dll unit tests under new jna I got the following dialog box:

--------------------------- Microsoft Visual C++ Runtime Library

--------------------------- Buffer overrun detected!

Program: C:\java\jdk1.5\jre\bin\java.exe

A buffer overrun has been detected which has corrupted the program's internal state. The program cannot safely continue execution and must now be terminated.

Can you determine which unit test is causing the failure? Which windows version? The VMCrashProtection test intentionally causes an illegal access; try running it independently to see if it's the one causing the fault.

--------------------------- OK

---------------------------

Unfortunately I don't have further information as jvm terminates w/ o any log ^(. So, I just don't know what is the reason for buffer overrun.