Re: [PATCHv2 2/2] gdb: convert sentinel_frame to a frame_info_ptr

[Andrew Burgess <aburgess@redhat.com>]

Simon Marchi <simark@simark.ca> writes:

> On 6/22/26 5:33 PM, Andrew Burgess wrote:
>> @@ -1726,12 +1763,12 @@ get_current_frame (void)
>>    if (get_traceframe_number () < 0)
>>      validate_registers_access ();
>>  
>> -  if (sentinel_frame == NULL)
>> -    sentinel_frame =
>> -      create_sentinel_frame (current_program_space,
>> -			     current_inferior ()->aspace.get (),
>> -			     get_thread_regcache (inferior_thread ()),
>> -			     0, 0).get ();
>> +  if (sentinel_frame == nullptr)
>
> While this works, in the previous patch you have used ".is_null ()".
> I'm pointing it out in case you want to change it, but otherwise it
> doesn't bother me too much.

I actually realised about an hour after posting these patches that my
choice to use is_null was weird.  We mostly compare frame_info_ptr
objects to nullptr throughout GDB, treating them as if they actually
were pointers.  I plan to update the previous patch to use comparison to
nullptr before pushing as I think that is far more common throughout
GDB.

Thanks,
Andrew

>
> Otherwise LGTM.
>
> Approved-By: Simon Marchi <simon.marchi@efficios.com>
>
> Simon