parallel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Alternate termination sequence option --term-seq


From: Ole Tange
Subject: Re: Alternate termination sequence option --term-seq
Date: Tue, 2 Jun 2015 23:46:51 +0200

On Mon, Jun 1, 2015 at 10:22 AM, Rasmus Villemoes <rv@rasmusvillemoes.dk> wrote:
> On Tue, May 26 2015, Ole Tange <ole@tange.dk> wrote:
>
>>   * giving a process its own process group also gives it a new
>> terminal, and thus Ctrl-C does not kill the children and it breaks
>> --tty. That means Ctrl-C will have to be propagated by Parallel.
>
> Well, first of all, I think it would be reasonable to start by handling
> the common case first, so one could disable all process group logic in
> the --tty case. Also, the --tty case is not where I'd expect rogue trees
> of subprocesses to be spawned.

The current git code wraps every command with a perl script that
simply does 'setpgrp' and execs the real command. By my measurements
it cost 0.3 ms per job, so acceptable for all longer running jobs, but
bad news for the jobs in the ms range.

It fucks up --tty somewhat, but passes the rest of the testsuite.

Please give it a spin and let me know if something else broke.
Especially if you can make it kill an unrelated process group.


/Ole



reply via email to

[Prev in Thread] Current Thread [Next in Thread]