Joel Brobecker wrote:
> I think there are two parts to this patch:
> 
...
> 
> This patch is fine, and approved. Please just run it through the testsuite
> before checking in. (can you remember to post a ChangeLog for it as well?)

Ok, I committed this part. I have had weird issues with 
running testsuite, I ended up running all the test 
one-by-one. (tested on linux).


> 
> The mips-tdep part can be treated independently.  Although I don't see
> any problem with it, and you updated it the same way I would have changed
> it, I'm not sure about going against the comment:
> 
>>    /* Map [gdbarch_num_regs .. 2*gdbarch_num_regs) onto the raw registers, 
>>       but then don't make the raw register names visible.  */
> 
> It looks fine to return "0" ... "31" as the names of the raw registers,
> but I'd like someone with more experience with the mips target to confirm
> it.

...

I made all suggested changes. New patch is attached.

Note: I tested this on our gdb based on gdb 6.8 sources, 
using our remote protocol to connect to a MIPS target. The 
changes are the same - the code affected by the patch does 
not differ.
On the HEAD gdb sources I did a compile but I could not do a 
  test on the target.


Thanks,

Aleksandar



* mips-tdep.c (mips_register_name): Handle numeric GPR 
register numbers.
(mips_print_registers_info): Remove gdb_assert.