[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: powerpc-rtems-4.7-gcc-4.0 patch commited



On Fri, 2005-02-18 at 06:29 -0600, Joel Sherrill  wrote:
> Sergei Organov wrote:
> > Ralf Corsepius <ralf.corsepius@rtems.org> writes:
> > 
> > [...]
> > 
> > 
> >>E.g. the m8260 would be superflous, if it could be removed from cpukit.
> >>The code it uses is _identical_ to the 603e, except of 1 define in
> >>cpukit.
> > 
> > 
> > Well, which one of the following 8 do you mean? Or am I missing
> > something?
> > 
> > cpukit/score/cpu/powerpc/rtems/new-exceptions/cpu.h:180:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > cpukit/score/cpu/powerpc/rtems/new-exceptions/cpu.h:192:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > cpukit/score/cpu/powerpc/rtems/old-exceptions/cpu.h:204:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > cpukit/score/cpu/powerpc/rtems/old-exceptions/cpu.h:216:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > cpukit/score/cpu/powerpc/rtems/score/powerpc.h:211:#elif defined(mpc8260)
> > cpukit/score/cpu/powerpc/rtems/score/powerpc.h:461:#elif defined(mpc8260)
> > cpukit/score/cpu/powerpc/rtems/score/cpu.h:407:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > cpukit/score/cpu/powerpc/rtems/score/cpu.h:436:  || defined(mpc860) || defined(mpc821) || defined(mpc8260))
> > 
> 
> Most of those are related to having configuration info for drivers in 
> libcpu and could easily be defined in libcpu.
> 
> There is a PPC_INTERRUPTS_MAX which still remains to be addressed.
That's the one I was referring to.

I know how to eliminate the others (Introducing an abstract base class
for rtems_cpu_table and to move all CPU-variant specific fields out of
CPUkit - This would be a major redesign of RTEMS), but I have no idea
how to clean up the PPC_INTERRUPTS mess in powerpc.h

> Ralf.. I think the the MIPS defines that in libcpu as a variable used by 
> the cpukit.
Hmm? I don't understand.

Do you mean an extern variable in cpukit to be filled in libcpu?
We already have this at several places. The problem with with powerpc is
the borked design of the way the interrupt defines are being setup in
powerpc.h.

Ralf