Date: Mon, 13 Sep 1999 14:19:20 -0600 From: Nate Williams <nate@mt.sri.com> To: Justin Wells <jread@fever.semiotek.com> Cc: freebsd-java@FreeBSD.ORG Subject: Re: 1.1.8 segfaults Message-ID: <199909132019.OAA23549@mt.sri.com> In-Reply-To: <19990913155143.A1142@fever.semiotek.com> References: <19990903020650.A10276@semiotek.com> <19990913155143.A1142@fever.semiotek.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> JDK 1.1.8 continues to segfault on me. I have been forcded to > rever to 1.1.7; I'm not sure whether this message made it through > last time or not, since nobody has asked me for an example > that demonstrates the crash. It did, but we're hoping that a new release I've about to make (as soon as Motif shows up) will fix these bugs. If things go well, the new release will be out next week or the week after, assuming Motif shows up in a timely manner... Sorry for not getting back to you sooner... Nate > > > On Fri, Sep 03, 1999 at 02:06:50AM -0400, Justin Wells wrote: > > > > The latest version of the JDK 1.1.8 (both ELF and AOUT) segfaults > > on one of my class files, reliably, at the same point. I tried it > > on two different machines: a FreeBSD 3.2 machine with the ELF release; > > and a 2.2.7 machine with the AOUT release installed. It crashes > > on both at the same point. > > > > It works fine under the 1.1.7 JDK, and it works fine under both > > Linux JDK's (IBM and Blackdown). Interestingly enough, though, it > > crashes the Kaffe VM in exactly the same way, at the same line. If > > I remove that line, Kaffe and the 1.1.8 JVM are happy again. > > > > I tried to create a trivial example that replicates the problem, but > > I wasn't able to do it. So I could make a large package including > > the current version of my software available, along with a shell > > script that runs it in the way that shows the problem. Since it's > > a pre-release version in the midst of a major code change, I don't > > want to publish it--but I'll supply it privately to whoever wants > > to try and debug this. > > > > Please e-mail me as justin at the same domain (semiotek.com) and > > ask me to send you a copy. (Replies to this message will wind up > > in my mailing list inbox, which I often mass-delete when I am busy, > > so send it to my regular account instead). > > > > For the curious, it happens during some recursive introspection of a > > hierarchy of classes. The particular Java expression that causes the > > segfault is this: > > > > iterClass.isAssignableFrom(returnType); > > > > in the case that causes the segfault, iterClass is one of my > > own private classes (org.webmacro.util.java2.Iterator) and > > the returntype happens to be java.util.Enumeration. > > > > However, who knows what actually caused the problem. Perhaps > > something odd is happening on the stack. > > > > I compiled with both 'jikes' and with 'javac' and the problem > > results no matter which one is used, so I don't think it would > > involve a compiler bug. > > > > Justin > > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-java" in the body of the message > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-java" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-java" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199909132019.OAA23549>