We do need to send a SIGQUIT to a process/thread, even if it is running
It could be stuck in a syscall, and not abort on its own.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
diff --git a/fio.c b/fio.c
index 60a2a1b..c769445 100644
--- a/fio.c
+++ b/fio.c
@@ -69,10 +69,7 @@
for_each_td(td, i) {
if (group_id == TERMINATE_ALL || groupid == td->groupid) {
- /*
- * if the thread is running, just let it exit
- */
- if (td->runstate < TD_RUNNING)
+ if (td->runstate <= TD_RUNNING)
kill(td->pid, SIGQUIT);
td->terminate = 1;
td->o.start_delay = 0;