Re: [avr-gcc-list] case statement broken in 4.1.2?

[Juergen Weigert]

> Am Dienstag, 28. November 2006 08:37 schrieb Henning Paul:
> I reproduced the behaviour (if/else works, but switch/case doesn't) 
> with GCC 4.0.2, which _doesn't_ throw a warning. 
> 
> I'm sorry, obviously it isn't GCC's fault.

It may also be an old bug, lurking in our backend code for some time...

> Am Mittwoch, 29. November 2006 00:00 schrieb Eric Weddington:
> > Did you happen to have patches for 4.1.x for ATmega256x support in your
> > toolchain?
> 
> Yes, I made a feature request at Novell's Bugzilla to include the patches 
> from 
> J?rg's FreeBSD CVS. So SuSE's J?rgen Weigert added these to the current 
> development tree (called OpenSUSE Factory). This was the version I used 
> first.

http://ftp.opensuse.org/pub/opensuse/distribution/SL-OSS-factory/inst-source/suse/
        src/cross-avr-gcc-4.1.2_20061115-5.src.rpm
        i586/cross-avr-gcc-4.1.2_20061115-5.i586.rpm
(or better choose a mirror from 
http://en.opensuse.org/Mirrors_Development_Build)


> -mno-tablejump solved all these problems, the firmware is working perfectly, 
> even with GCC 4.1.2.
> 
> Strange, though. Shouldn't happen, should it?

-mno-tablejump effectivly does the same what the preprocessor switch in
Doper does: It converts the switch-cascade into a tree of if-statements.

To me this looks like we have an issue with our machine-description.

        cheers,
                Jw.

-- 
 o \  Juergen Weigert  paint it green! __/ _=======.=======_
<V> | address@hidden       wide open suse_/        _---|____________\/
 \  | 0911 74053-508         (tm)__/          (____/            /\
(/) | __________________________/             _/ \_ vim:set sw=2 wm=8