job-control: mitigate process-kill race

children_kill_cb() is racey. One obvious problem is that
process_close_handles() is *queued* by on_process_exit(), so when
children_kill_cb() is invoked, the dead process might still be in the
`loop->children` list.  If the OS already reclaimed the dead PID, Nvim
may try to SIGKILL it.

Avoid that by checking `proc->status`.

Vim doesn't have this problem because it doesn't attempt to kill
processes that ignored SIGTERM after a timeout.

closes #8269

1 files changed, 3 insertions, 3 deletions

diff --git a/src/nvim/channel.c b/src/nvim/channel.c
index 776e2bfa86..4e6ca8d278 100644
--- a/src/nvim/channel.c
+++ b/src/nvim/channel.c
@@ -658,9 +658,9 @@ static void channel_process_exit_cb(Process *proc, int status, void *data)
     terminal_close(chan->term, msg);
   }
 
-  // if status is -1 the process did not really exit,
-  // we just closed the handle onto a detached process
-  if (status >= 0) {
+  // If process did not exit, we only closed the handle of a detached process.
+  bool exited = (status >= 0);
+  if (exited) {
     process_channel_event(chan, &chan->on_exit, "exit", NULL, 0, status);
   }