This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: sh64-elf (SH5) port: directory opcodes


> On Feb  5, 2002, Andrew Cagney <ac131313@cygnus.com> wrote:
> 
> 
>> Why is this necessary?  Shouldn't all disassemblers be included and
>> then, perhaps, just tweak the default arch/mach?
> 
> 
> I suppose it's because there's no sense whatsoever in including the
> SH5 machinery in a sh-elf build.  sh-elf has no support for SH5 in
> binutils and simulator, only sh64-elf does.  SH5 is such a different
> architecture from earlier SH versions that it barely makes sense to
> keep them all in a single bfd_arch.  It only does because one of the
> ISAs supported by SH5 happens to be very similar with that used by
> earlier SHs, so we can reuse a lot of code.


I don't think that is correct.  My understanding of the SH5 simulator 
and the SH5 architecture is they are capable of executing sh2-4 code in 
a compatibility mode.  This is like the P4 can execute 8086 code as it 
is part of IA32.  GDB in turn would most certainly have the capacity to 
debug all of these SH variants (unless that is bfd/opcodes crippled 
their sh support).

Anyway, if you always include support for all the SH variants, does 
anything break?

Andrew


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]