Re: Patch to avoid SegFault in natClass.cc

[Martin Kahlert <martin dot kahlert at infineon dot com>]

On Fri, Jul 13, 2001 at 02:14:24AM -0700, Alexandre Petit-Bianco wrote:
> 
> Martin Kahlert writes:
> 
> > The compiler generates a direct call to _Jv_IsInstanceOf, it does
> > *not* call java::lang::Class::isInstance, which would have called
> > _Jv_InitClass first.
> 
> Do you you guys think it should be changed? (I which way it probably
> wouldn't go in 3.01 but only the trunk.)

Hi!
I assume, you have the most knowledge of natClass & Co:
I wanted to produce a small test for my problem, 
but i can't reproduce it with my small chunks
of code, only my big application gives the error.
Can you tell me, what i have to do in order to get an uninitialized
class for an instanceof test?
I wildly mixed abstract classes with .class calls but i had no success.

Is there a deeper reason at all, why the classloader should be allowed to return
an uninitialized class? If not, the _Jv_InitClass would be superflous.

Thanks,
Martin.

PS: I CC'ed java, since the audience there seems to be larger,
    and this message is not directly about a patch.

-- 
The early bird catches the worm. If you want something else for       
breakfast, get up later.