.align problem with i[34]86-*-coff

[Rotter Klaus <Klaus dot Rotter at erl9 dot siemens dot de>]

Hello,

I am doing embedded development with gcc on i386 and i486 based boards.
After I got my things up and I could compile and run programms, I have tried to
clean up my development enviroment.

I used a gcc compiled for i586-pc-cygwin32 and binutils compiled for i386-unkown-coff.
(I did this because the binutils for i586-pc-cygwin32 contained a bug which prevents 
the build of binary output). 

For debugging reasons I have tried to set up a clean enviroment compiling both binutils and
gcc for i386-unkown-aout (2.95.1). Compiling works, but it turned out that the is a bug
with structures. My RTOS didn't allow padding bytes in structures, so I use the flag 
-fpack-struct. This worked fine with i586-pc-cygwin, but with i386-unkown-aout gcc aligns
32-bit ints in structures to a longword boundary (even with -fpack-struct). 

So I tried to set up an enviroment for i386-unkown-coff (both binutils and gcc). This also
won't work because it seems that gcc is configured for .align bits, which means, if gcc wants something to be aligned on a 8 byte boundary, it outputs ".align 3", but 
i386-unkown-coff-as would expect a ".align 8". This happens compiling the
libgcc1.a support library.
 
After that I tried out i[34]86-linux-coff both with gcc2.95.1 and egcs-19992009. The
error remains. In the documentation of as there is also nothing clear about .align with
COFF. It is clear that elf uses .align byte-boundary and aout uses .align bits.

So, I really don't know if the bug belongs to binutils or gcc, this is why I am posing
here. I could get around by unsing my old setup (mixing of both setups), but this 
should be fixed.

	-Klaus

-- 
 Klaus Rotter, Siemens AG,  ATD TD 6 EPG, Postfach 3249, 91020 Erlangen
 Tel.: +49 9131 18-84637, Fax: +49 9131 18-84648 
 mailto:Klaus.Rotter@erl9.siemens.de



-----Ursprüngliche Nachricht-----
Von: Andreas Jaeger [mailto:aj@suse.de]
Gesendet am: Samstag, 9. Oktober 1999 17:48
An: gcc@gcc.gnu.org
Betreff: anoncvs broken?


I tried to update the gcc sources via cvs and got:
cvs [update aborted]: authorization failed: server anoncvs.cygnus.com rejected access
aj@gromit:/usr/src/cvs/gcc > cat CVS/Root 
:pserver:anoncvs@anoncvs.cygnus.com:/cvs/egcs

I didn't change anything on my side - but it suddenly stopped working.
I tried to login again after checking the www pages on CVS read access
but this failed also:

aj@gromit:/usr/src/cvs/gcc > cvs login -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/egcs
(Logging in to anoncvs@anoncvs.cygnus.com)
CVS password: 
cvs [login aborted]: authorization failed: server anoncvs.cygnus.com rejected access

Is anoncvs broken or disabled for now?

Andreas
-- 
 Andreas Jaeger   
  SuSE Labs aj@suse.de	
   private aj@arthur.rhein-neckar.de