|
|
|
|
|
|
|
|
|
|
xen-devel
Re: [Xen-devel] modify_ldt errors on x86_64
On Monday 14 November 2005 5:59 pm, Nakajima, Jun wrote:
> Karl Rister wrote:
> > On Saturday 12 November 2005 7:06 am, Ian Pratt wrote:
> >>> I am trying to run some Java tests in a dom0 running on
> >>> x86_64 and I am receiving the following error message:
> >>>
> >>> setupLdt: modify_ldt: Invalid argument
> >>>
> >>> After 2 of these the JVM just hangs and I have to kill it.
> >>> This does not happen when booting the systerm without Xen. I
> >>> have tried several different changesets from the past few
> >>> days and have seen the same thing on all of them. My current
> >>> changeset is 7746. Has anyone else seen this?
> >>
> >> No, this is the first report.
> >>
> >> Which version of Java are you using. Have you a simple test case?
> >
> > "java -version" returns the following on baremetal Linux, this simple
> > test is enough to cause the problem in domain0
> >
> > java version "1.4.2"
> > Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.2)
> > Classic VM (build 1.4.2, J2RE 1.4.2 IBM build cxia32142sr1a-20050209
> > (JIT enabled: jitc))
>
> I tested "java -version" on x86_64 xenlinux, and got:
>
> java version "1.4.2-02"
> Java(TM) 2 Runtime Environment, Standard Edition (build
> Blackdown-1.4.2-02)
> Java HotSpot(TM) 64-Bit Server VM (build Blackdown-1.4.2-02, mixed mode)
>
> I downloaded it from ftp.tux.org.
I tried both the 32bit and 64bit Blackdown JVMs as well as a 64bit IBM JVM and
they do not exhibit this problem. It seems to be specific to the 32bit IBM
JVM. I think that I am going to file this as a bug since Xen should not be
affecting a userspace app in this manner.
I should also add that in some instances (mostly on a uniprocessor dom0) this
error renders the box dead and it has to be rebooted through the service
processor.
>
> > An strace on the java process shows these two messages being dumped
> > over and over again in a loop:
> >
> > rt_sigprocmask(SIG_UNBLOCK, [ILL TRAP BUS FPE SEGV], ~[KILL USR2 STOP
> > RTMIN RT_1], 8) = 0
> > --- SIGSEGV (Segmentation fault) @ 0 (0) ---
>
> Jun
> ---
> Intel Open Source Technology Center
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-devel
--
Karl Rister
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
|
|