On Tue, 2006-04-18 at 10:25 -0700, David Daney wrote: > GCC bug 26858 (http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26858) may > also effect other JVMs that run on linux *and* use SIGSEGV to generate > NullPointerExceptions. I don't know if such JVMs exist. But if they > do, it might be worthwhile for their maintainers to run the test case in > the PR to see what happens. > > If GCJ is the only thing effected, I apologize for the noise. Hi! Yes, we do. The testcase works on my local linux system, but I think we have the same problem in cases where memory is mapped to low addresses. I really see a big problem here in our JIT for unresolved classes. TWISTI
