🚀 go-pugleaf

On 7/7/2018 1:24 AM, Ahem A Rivet's Shot wrote:
> On Sat, 7 Jul 2018 00:05:12 -0500
> Charles Richmond <nume...@aquaporin4.com> wrote:
> 
>> On 7/6/2018 8:00 PM, Dan Espen wrote:
>>> Charles Richmond <nume...@aquaporin4.com> writes:
>>>
>>>> On 7/6/2018 2:37 PM, Charlie Gibbs wrote:
>>>>> On 2018-07-06, Ahem A Rivet's Shot <ste...@eircom.net> wrote:
>>>>>
>>>>>> On 6 Jul 2018 15:50:16 GMT
>>>>>> Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>>>>>>
>>>>>>> On 2018-07-06, Gareth's Downstairs Computer
>>>>>>> <headstone255.but.n...@yahoo.com> wrote:
>>>>>>>
>>>>>>>> Do incompetent computer programmers become technical authors?
>>>>>>>
>>>>>>> Nah, they go straight into management.
>>>>>>
>>>>>> 	Yep, the sad thing is when it happens to competent
>>>>>> programmers.
>>>>>
>>>>> Which brings to mind that old Unix joke:
>>>>>
>>>>> A programmer went into management and turned out to be quite good at
>>>>> it. One of his old buddies, remembering how much fun he had as a
>>>>> programmer, asked him whether he ever considered moving back.  "No,"
>>>>> the manager said.  "As you know, there is no return for a successful
>>>>> exec()."
>>>>>
>>>> Can an unsuccessful exec() call return???  And where would it return
>>>> to???
>>>
>>> The answer is in the man page:
>>>
>>> RETURN VALUES
>>>     If a function in the exec family returns to the calling process
>>> image, an error has occurred; the return value is -1 and errno is set to
>>>     indicate the error.
>>>     
>>
>> The calling process for exec() must be the child process.  ISTM the
>> child process must signal the parent process and abort... if exec() fails.
> 
> 	There are no parent and child processes with exec(), it's all one
> process. Calling exec() replaces the code in the *current* process with new
> code, if it succeeds it cannot return because the return point no longer
> exists. One obvious failure mode that leaves the original code intact is
> being unable to read the file containing the new code in which case a
> return is trivially possible and expected.
> 
> 	If you want a child you have to fork().
> 

Yes, yes, of course...  the parent and child processes are the result of 
the fork() call.  From this pair of processes, the child will be the one 
to call exec() and over-write the executable with a new binary image. 
Therefore, if that exec fails, it will return to that child process.

Certainly the exec() call has *no* sense of child or parent, and needs 
none.  But when fork() creates the parent-child pair of processes, the 
child will be the one that will then call exec().

-- 
numerist at aquaporin4 dot com