Re: [PATCH v7 3/4] Share fork_inferior et al with gdbserver

[Pedro Alves <palves@redhat.com>]

On 06/07/2017 10:41 PM, Sergio Durigan Junior wrote:
> On Wednesday, June 07 2017, I wrote:
> 
>> On Wednesday, June 07 2017, Pedro Alves wrote:
>>> I think we could get rid of the new common-inferior.h
>>> header too, but I'll leave that as is, for now at least.
>>
>> And move get_exec_{wrapper,file} to fork-inferior.h and fork-child.c,
>> you mean?
> 
> Probably not, because get_exec_file is used by Windows targets as well.
> I'll wait for your reply before submitting the next patch.
> 

Considering exec_file for instance:  fork_inferior calls
get_exec_file if exec_file is NULL.  But, we could just
pass down the right exec_file to fork_inferior in the first
place.  We could do that by e.g., adding fork_child
wrapper functions to both gdb and gdbserver's fork-child.c that
would call the shared fork_inferior function, and then adjust
targets to call that instead.  For the exec wrapper, we could
add a new "const char *exec_wrapper" parameter to fork_inferior.

Since it's arguable which approach is better in this case,
I'm totally fine with leaving things as is.

I notice now that get_exec_file is still declared
in gdbcore.h, however.

Thanks,
Pedro Alves