Re: PATCH: Fix MI stack frame output for synthetic frames

[Andrew Cagney <ac131313@redhat.com>]

> On Fri, Nov 08, 2002 at 05:38:38PM -0500, Andrew Cagney wrote:
> 
> 
>> It should be possible to test the "<function called from gdb>" case. 
>> All the more main-stream architectures now use dummy-frames so any one 
>> of them can be used for the test.
>> 
>> Can you do that?
> 
> 
> OK, I'll write up an MI test case for this and re-post the whole
> patch a bit later (off to help a user just now :-).  I can also 
> test the signal handler frame one a la gdb.base/signals.exp easily
> enough.  

Thanks.

Just note that, unlike all that older code, you can now assume:

- the message always appears (if it doesn't its a bug in that 
architecture not using dummy-frames :-)
- the unwind stops at main (if it doesn't for your target yell and 
someone will fix it :-)

Much easier!
Andrew


> For the tests I added to the Apple gdb, I use the following C file
> to test various cases (the function bar() I used for verifying
> unwindonsignal was working correctly) in MI and CLI mode.