Re: [PATCH 4/9] run-command: add support for timeout in command finisher

From: Johannes Sixt <j6t@kdbg.org>
Date: May 21, 2026, 07:21 UTC
Message-ID: <b69605a6-e841-47b9-a899-a57e184d3c8b@kdbg.org>
In-Reply-To: <f58c8c522814dce9257f64733e9fbc9bd9f446c0.1779207350.git.siddh.raman.pant@oracle.com>

Am 19.05.26 um 18:30 schrieb Siddh Raman Pant:

Show 5 quoted lines

> A called command may not respond to the initial signal and will get
> stuck in finish_command() -> wait_or_whine().
> 
> So let's add timeout support into the finisher so that if a deadline
> occurs, we can send a force-kill signal.

This is extremely suspicious. A communication protocl with a child program that requires to kill the child looks like a design error. A band-aid like this timeout should not be necessary for a well-behaved child process.

If the (your?) problem is that the child process is actually not well-behaved, then I suggest to use a middle-man as child process that behaves well from the point of view of the git process, but can punish the ill-behaved downstream process when needed.

Please, do not add this infrastructure to core Git, and instead fix the communication protocol.

> 
> The force-kill signal is in the argument because a program may trap a
> signal, so it is the responsibility of caller to pass the correct kill
> signal.

-- Hannes