Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
sched_ext: Improve comment on idle_sched_class exception in scx_task_iter_next_locked()
scx_task_iter_next_locked() skips tasks whose sched_class is
idle_sched_class. While it has a short comment explaining why it's testing
the sched_class directly isntead of using is_idle_task(), the comment
doesn't sufficiently explain what's going on and why. Improve the comment.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: David Vernet <void@manifault.com>
+23
-2
+23
-2
kernel/sched/ext.c
+23
-2
kernel/sched/ext.c
···
1252
1252
1253
1253
while ((p = scx_task_iter_next(iter))) {
1254
1254
/*
1255
-
* is_idle_task() tests %PF_IDLE which may not be set for CPUs
1256
-
* which haven't yet been onlined. Test sched_class directly.
1255
+
* scx_task_iter is used to prepare and move tasks into SCX
1256
+
* while loading the BPF scheduler and vice-versa while
1257
+
* unloading. The init_tasks ("swappers") should be excluded
1258
+
* from the iteration because:
1259
+
*
1260
+
* - It's unsafe to use __setschduler_prio() on an init_task to
1261
+
* determine the sched_class to use as it won't preserve its
1262
+
* idle_sched_class.
1263
+
*
1264
+
* - ops.init/exit_task() can easily be confused if called with
1265
+
* init_tasks as they, e.g., share PID 0.
1266
+
*
1267
+
* As init_tasks are never scheduled through SCX, they can be
1268
+
* skipped safely. Note that is_idle_task() which tests %PF_IDLE
1269
+
* doesn't work here:
1270
+
*
1271
+
* - %PF_IDLE may not be set for an init_task whose CPU hasn't
1272
+
* yet been onlined.
1273
+
*
1274
+
* - %PF_IDLE can be set on tasks that are not init_tasks. See
1275
+
* play_idle_precise() used by CONFIG_IDLE_INJECT.
1276
+
*
1277
+
* Test for idle_sched_class as only init_tasks are on it.
1257
1278
*/
1258
1279
if (p->sched_class != &idle_sched_class)
1259
1280
break;