Re: GCC-4.0.2 on Cygwin

[Kai Ruottu <karuottu at mbnet dot fi>]

Brian Rose wrote:
> Brian Rose wrote:
>
> > /cygdrive/c/DOCUME~1/brose/LOCALS~1/TEMP/ccMW7YYh.s: Assembler messages:
> > /cygdrive/c/DOCUME~1/brose/LOCALS~1/TEMP/ccMW7YYh.s:31: Error: 
> > operands mismatch -- statement `fmovem.l %fpcr,%d1' ignored
> > /cygdrive/c/DOCUME~1/brose/LOCALS~1/TEMP/ccMW7YYh.s:42: Error: 
> > operands mismatch -- statement `fmovem.l %fpcr,%d1' ignored
>
> OK, I used the --with-as and added a -v to the compiler flags and I am 
> still getting the error. I can see that the compiler is calling the 
> correct assembler (target of m68k-elf).
>
> Is this a bug in the assembler?

 Please RTFM first! Is the opcode valid for m68040?  If so, there is
a bug in the assembler !  If it is not valid, your new gcc-4.x produces
bad code for m68040 !

 I have produced maybe over 20 toolchains for the m68k/cpu32/coldfire
family but cannot say anything before consulting the m68k family 
manuals... I cannot even remember what on earth register this 'fpcr'
is, but my guess could be the 'fpcontrol', my Insight for m68k-elf
shows that in the "Registers Window"... Is there an opcode for doing
this "move data from the 'fpcr' to the 'd1'" or not?

 Maybe yo should try another GCC or another binutils !

 I could try the gcc-4.0.2 sources for 'm68k-elf' on Linux and see
what happens...





------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com