Re: [PATCH 2/2] Fix two C nits

[Eitan Adler <lists@xxxxxxxxxxxxxx>]

On 18 June 2018 at 02:22, Peter Stephenson <p.stephenson@xxxxxxxxxxx> wrote:
> On Sat, 16 Jun 2018 01:04:27 +0000
> Eitan Adler <lists@xxxxxxxxxxxxxx> wrote:
>> - avoid returning from a function that will never return
>>
>> diff --git a/Src/exec.c b/Src/exec.c
>> index d44527841..b36bcef64 100644
>> --- a/Src/exec.c
>> +++ b/Src/exec.c
>> @@ -4954,7 +4954,6 @@ getpipe(char *cmd, int nullexec)
>>      execode(prog, 0, 1, out ? "outsubst" : "insubst");
>>      cmdpop();
>>      _exit(lastval);
>> -    return 0;
>>  }
>
> I'm not 100% sure about this since you're relying on the compiler
> knowing that _exit won't return.  Probably the majority of compilers
> we're involved with for zsh will work that out, but I'm not sure
> it's actually required by the C standard that they know the function
> doesn't return, is it?  You may know some corner I haven't investigated.

It is only required that any function either return, exit, or loop
forever [0]. While I don't much about non-modern compilers, it
shouldn't affect anything to remove the return. At best, it'll do
nothing. At worst older compilers might incorrectly issue a
diagnostic. IMHO we should defer to modern compilers when it comes to
diagnostics provided the more esoteric options still work.

> I'd have been tempted to add /*NOTREACHED*/ on the next line.


[0] there is some interesting ancient nuance here, but it's not
important for this discussion.




-- 
Eitan Adler