[gambit-list] process-status, close-port, and zombie processes

Lang Martin lang-gb at coptix.com
Wed Oct 3 16:17:00 EDT 2007


On Sep 6, 2007, at 8:06 AM, Marc Feeley wrote:

> To debug this problem I suggest you add a trace in function  
> sigchld_signal_handler in lib/os_io.c.  What interests me is the  
> values returned by the call to waitpid .

I have not had a chance to try this yet, however, it seems as though  
the problem may have gone away with a transition from 4.0.0 --inline- 
jumps --single-host to 4.0.1 --single-host. Does that shed any light  
on the problem?

>   I suspect the status in neither WIFEXITED(status) or WIFSIGNALED 
> (status), so the process is never marked as terminated.
>   Another problem might be that a SIGCHLD signal is never generated  
> for the process.

Am I correct that both of these suggestions presume that the child  
process exits improperly? the child process for this program is  
"convert" from the imagemagick suite, and the program runs properly  
in scsh.

I'm sorry this account is so sloppy, I will return to this problem at  
some point in the future (when I have more time) and try to figure it  
out. In the meantime, think of this as more of a "brain-dump in case  
it's useful" than a "bug report".

Lang



More information about the Gambit-list mailing list