Hi Richard,

Unfortunately I will not be able to bisect the issue - I really don't have the knowledge to do that. 
My best contribution will probably be the testing of a future solution, please let me know if you need any assistance with that.

I really appreciate your work!

Thanks,

Gunnar




> Date: Mon, 15 Aug 2011 22:29:01 +0200
> Subject: Re: [uml-user] java exceptions in UML
> From: richard.weinberger@gmail.com
> To: gunnarlindroth@hotmail.com
> CC: user-mode-linux-user@lists.sourceforge.net
>
> On Mon, Aug 15, 2011 at 5:58 PM, Gunnar Lindroth
> <gunnarlindroth@hotmail.com> wrote:
> > Thank you Richard, that's really great news!
> > My UML has 2GB available, I have assumed that to be enough?
> > I'm sorry if I didn't make myself very clear about the output of the test
> > program.
> > In my UML this test program is not crashing either, but it's sometimes
> > printing "0.75 < 0" which it should _never_ do.
>
> True. :-\
> Your test program prints "0.75 < 0".
>
> I was able to write a stand alone test in c using pthreads.
> There is definitely something broken.
>
> It seems to fail only on newer hosts:
> 2.6.18/i686 UP: OK
> 2.6.18/i686 SMP: OK
> 2.6.34/x86_64 SMP: FAIL
> 2.6.37/x86_64 UP: FAIL
> 2.6.37/x86_64 SMP: FAIL
> 3.0/i686 SMP: FAIL
>
> For all tests I've used 3.1-rc2 as UML kernel.
>
> Something on the host-side broke UML's FPU threading.
> The test program is attached.
> Are you able to bisect the issue?
> If not I'll do it, but it might take some time as I'm currently super busy.
>
> --
> Thanks,
> //richard