Re: Kaffe on GNU Hurd
Igor Khavkine wrote:
> Also here's some more useful information. After I compiled Kaffe, I ran make
> check in the source tree. IIRC there were 3 tests that failed.
>
> Preempt.java:
> preempting didn't work and the program was deadlocked after
> printing
> started 5 threads
> thread 0
Could you check that SP_OFFSET is 2 in jmp_buf for i386-gnu ? You
could use developers/sp_offset.c to guest that.
>
> StackDump.java:
> from what I could trace, it segfaulted in what seemed to be a
> function produced by the JIT. I haven't tried Kaffe without JIT yet, since
> right now my Hurd box is almost unusable because of file system corruption. I
> need to try a fresh install.
>
> UncaughtException.java:
> I think that was the last one that failed, i didn't get far with this one.
Both use SEGV (aka NullPointerException) but according to Roland McGrath:
> SA_SIGINFO is not currently supported on the Hurd at all.
> However, signal handlers do get passed struct sigcontext *
> as the third argument (like BSD).
your definition of EXCEPTIONPROTO, seem to not be correct:
#define EXCEPTIONPROTO \
int sig, struct sigcontext ctx
BSDI, FreeBSD2, NetBSD1 and OpenBSD have this definition in Kaffe:
#define EXCEPTIONPROTO \
int sig, int code, struct sigcontext* ctx
I will commit this definition (and change of EXCEPTIONFRAME macro).
--
Edouard G. Parmelan
http://egp.free.fr
Reply to: