Re: [PATCH] gdbsupport: Remove some unnecessary ifdef HAVE_POLL judgments

[Pedro Alves <pedro@palves.net>]

On 2022-05-13 11:09, Youling Tang wrote:
> Hi, Pedro
> 
> On 05/13/2022 05:28 PM, Pedro Alves wrote:
>> On 2022-05-13 08:49, Youling Tang wrote:
>>> By searching event-loop.cc we found that there are only the following 2
>>> places to assign values to use_poll,
>>> $ grep -nr use_poll gdbsupport/event-loop.cc
>>>    88:static unsigned char use_poll = USE_POLL;
>>>    263:    use_poll = 0;
>>>
>>> This USE_POLL is defined as follows,
>>>   #ifdef HAVE_POLL
>>>   #define USE_POLL 1
>>>   #else
>>>   #define USE_POLL 0
>>>   #endif
>>>
>>> So use_poll may be 1 only if HAVE_POLL is defined. Removed "ifdef
>>> HAVE_POLL" judgment in use_poll control block.
>>>
>>> Signed-off-by: Youling Tang <tangyouling@loongson.cn>
>> Maybe I'm missing something, but ISTM this can't possible work on hosts that don'
>>   have poll at all.  Like e.g., mingw:
>>
>>   $ grep POLL *
>>   config.h:/* #undef HAVE_POLL */
>>   config.h:/* #undef HAVE_POLL_H */
>>   config.h:/* #undef HAVE_SYS_POLL_H */
>>
>> Surely they'll fail to compile after the patch?
> You are right my oversight.
> 
> But we can remove these internal_error handling, similar to the following modification:

We can remove _every_ internal_error call in GDB, since by design, they are not supposed
to ever execute, unless we have something wrong.  This is the scenario here, guarding against
someone mistakenly doing that change that would result in use_poll as 1 on a host that doesn't
support poll.  Why change it?