On 2012-12-25 6:17 PM, David Levine wrote:
> ...
>
>     kill (sm_child, SIGKILL);
>     discard (sm_rfp);
>     discard (sm_wfp);
>
> The discard()'s after killing the child process seem unnecessary to me.

they should be unnecessary, but i'm betting they prevent deadlock in
older pre-Posix systems. ok to nuke 'em as far as i'm concerned --
SIGKILL is not catchable, and Posix requires that processes disconnect
asynchronously from their pipes at death time, even if they were
otherwise blocked in write() waiting for space.


_______________________________________________
Nmh-workers mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/nmh-workers

Reply via email to