Re: New function proposal

[David Bateman]

John W. Eaton wrote:
> On 15-Feb-2007, David Bateman wrote:
>
> | I do however see a number of defunct child processes, but that is no
> | different from the old popen2 function..
>
> Should we be calling waitpid for the child processes to clean them up?
>   
This only happens if the execvp in the popen2 call fails, and so the
child process abnormally doesn't exec and we stay in octave. So using a
waitpid in the parent doesn't really make in sense and will mean that
the popen2 function can really return the control to the parent process
for interaction with the child..

I call exit(0) in the child process and it releases the resources, but
the process still exists in the process table as a defunct process for a
failed execvp.. I have no idea why this is the case as I'd have thought
the exit should correctly terminate the child..


> | I think this patch is good to go..
>
> OK, please check it in.
>   
Ok, will do later.

D.

> Thanks,
>
> jwe
>
>   


-- 
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary