Re: [wanted] Test of GDB modifying a value in a register

[Andrew Cagney <ac131313@redhat.com>]

> Andrew Cagney wrote:
> 
>> 
>> Hello,
>> 
>> In doing a further cleanup of the frame code, I discovered that the
>> current testsuite isn't testing the codepaths that write a value in a
>> register.
>> 
>> That is, things like:
>> 
>> (gdb) set $pc = 10
>> register int i;
>> (gdb) set i = 10;
>> 
>> Anyone interested in comming up with, or have a test case?
> 
> 
> 
> Well, $pc is the only one I can think of that we can count on
> being defined on all targets -- and based on some past conversations
> with you, I'm wondering if even that is a safe assumption?
> 
> Basically, we have almost no tests of register functionality, 
> and yes it would be good if we did.

Already done :-)  GCC, at least takes sufficient notice of:
 >> register int i;
 >> (gdb) set i = 10;
for it to be possible to come up with useful tests (sufficiently useful 
to find wierd bugs! :-).  See:

http://sources.redhat.com/ml/gdb-patches/2002-11/msg00019.html

Andrew