From: Tejun Heo <tj@kernel.org>
To: linux-kernel@vger.kernel.org, sched-ext@lists.linux.dev
Cc: void@manifault.com, arighi@nvidia.com, changwoo@igalia.com,
emil@etsalapatis.com, Tejun Heo <tj@kernel.org>
Subject: [PATCH 21/34] sched_ext: When calling ops.dispatch() @prev must be on the same scx_sched
Date: Wed, 4 Mar 2026 12:01:06 -1000 [thread overview]
Message-ID: <20260304220119.4095551-22-tj@kernel.org> (raw)
In-Reply-To: <20260304220119.4095551-1-tj@kernel.org>
The @prev parameter passed into ops.dispatch() is expected to be on the
same sched. Passing in @prev which isn't on the sched can spuriously
trigger failures that can kill the scheduler. Pass in @prev iff it's on
the same sched.
Signed-off-by: Tejun Heo <tj@kernel.org>
---
kernel/sched/ext.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index 0082919e2e8f..1570b6a8158c 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -2392,9 +2392,10 @@ static bool scx_dispatch_sched(struct scx_sched *sch, struct rq *rq,
struct task_struct *prev)
{
struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx);
- bool prev_on_scx = prev->sched_class == &ext_sched_class;
int nr_loops = SCX_DSP_MAX_LOOPS;
s32 cpu = cpu_of(rq);
+ bool prev_on_sch = (prev->sched_class == &ext_sched_class) &&
+ scx_task_on_sched(sch, prev);
if (consume_global_dsq(sch, rq))
return true;
@@ -2418,7 +2419,7 @@ static bool scx_dispatch_sched(struct scx_sched *sch, struct rq *rq,
dspc->nr_tasks = 0;
SCX_CALL_OP(sch, SCX_KF_DISPATCH, dispatch, rq, cpu,
- prev_on_scx ? prev : NULL);
+ prev_on_sch ? prev : NULL);
flush_dispatch_buf(sch, rq);
--
2.53.0
next prev parent reply other threads:[~2026-03-04 22:01 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-04 22:00 [PATCHSET v3 sched_ext/for-7.1] sched_ext: Implement cgroup sub-scheduler support Tejun Heo
2026-03-04 22:00 ` [PATCH 01/34] sched_ext: Implement cgroup subtree iteration for scx_task_iter Tejun Heo
2026-03-04 22:00 ` [PATCH 02/34] sched_ext: Add @kargs to scx_fork() Tejun Heo
2026-03-04 22:00 ` [PATCH 03/34] sched/core: Swap the order between sched_post_fork() and cgroup_post_fork() Tejun Heo
2026-03-06 4:17 ` Tejun Heo
2026-03-06 8:44 ` Peter Zijlstra
2026-03-04 22:00 ` [PATCH 04/34] cgroup: Expose some cgroup helpers Tejun Heo
2026-03-06 4:18 ` Tejun Heo
2026-03-04 22:00 ` [PATCH 05/34] sched_ext: Update p->scx.disallow warning in scx_init_task() Tejun Heo
2026-03-04 22:00 ` [PATCH 06/34] sched_ext: Reorganize enable/disable path for multi-scheduler support Tejun Heo
2026-03-04 22:00 ` [PATCH 07/34] sched_ext: Introduce cgroup sub-sched support Tejun Heo
2026-03-04 22:00 ` [PATCH 08/34] sched_ext: Introduce scx_task_sched[_rcu]() Tejun Heo
2026-03-04 22:00 ` [PATCH 09/34] sched_ext: Introduce scx_prog_sched() Tejun Heo
2026-03-04 22:00 ` [PATCH 10/34] sched_ext: Enforce scheduling authority in dispatch and select_cpu operations Tejun Heo
2026-03-04 22:00 ` [PATCH 11/34] sched_ext: Enforce scheduler ownership when updating slice and dsq_vtime Tejun Heo
2026-03-04 22:00 ` [PATCH 12/34] sched_ext: scx_dsq_move() should validate the task belongs to the right scheduler Tejun Heo
2026-03-04 22:00 ` [PATCH 13/34] sched_ext: Refactor task init/exit helpers Tejun Heo
2026-03-04 22:00 ` [PATCH 14/34] sched_ext: Make scx_prio_less() handle multiple schedulers Tejun Heo
2026-03-04 22:01 ` [PATCH 15/34] sched_ext: Move default slice to per-scheduler field Tejun Heo
2026-03-04 22:01 ` [PATCH 16/34] sched_ext: Move aborting flag " Tejun Heo
2026-03-04 22:01 ` [PATCH 17/34] sched_ext: Move bypass_dsq into scx_sched_pcpu Tejun Heo
2026-03-04 22:01 ` [PATCH 18/34] sched_ext: Move bypass state into scx_sched Tejun Heo
2026-03-04 22:01 ` [PATCH 19/34] sched_ext: Prepare bypass mode for hierarchical operation Tejun Heo
2026-03-04 22:01 ` [PATCH 20/34] sched_ext: Factor out scx_dispatch_sched() Tejun Heo
2026-03-04 22:01 ` Tejun Heo [this message]
2026-03-04 22:01 ` [PATCH 22/34] sched_ext: Separate bypass dispatch enabling from bypass depth tracking Tejun Heo
2026-03-04 22:01 ` [PATCH 23/34] sched_ext: Implement hierarchical bypass mode Tejun Heo
2026-03-06 7:03 ` Andrea Righi
2026-03-06 7:23 ` Andrea Righi
2026-03-06 17:39 ` [PATCH v2 " Tejun Heo
2026-03-04 22:01 ` [PATCH 24/34] sched_ext: Dispatch from all scx_sched instances Tejun Heo
2026-03-04 22:01 ` [PATCH 25/34] sched_ext: Move scx_dsp_ctx and scx_dsp_max_batch into scx_sched Tejun Heo
2026-03-04 22:01 ` [PATCH 26/34] sched_ext: Make watchdog sub-sched aware Tejun Heo
2026-03-04 22:01 ` [PATCH 27/34] sched_ext: Convert scx_dump_state() spinlock to raw spinlock Tejun Heo
2026-03-04 22:01 ` [PATCH 28/34] sched_ext: Support dumping multiple schedulers and add scheduler identification Tejun Heo
2026-03-04 22:01 ` [PATCH 29/34] sched_ext: Implement cgroup sub-sched enabling and disabling Tejun Heo
2026-03-06 9:41 ` Cheng-Yang Chou
2026-03-06 17:39 ` [PATCH v2 " Tejun Heo
2026-03-04 22:01 ` [PATCH 30/34] sched_ext: Add scx_sched back pointer to scx_sched_pcpu Tejun Heo
2026-03-04 22:01 ` [PATCH 31/34] sched_ext: Make scx_bpf_reenqueue_local() sub-sched aware Tejun Heo
2026-03-04 22:01 ` [PATCH 32/34] sched_ext: Factor out scx_link_sched() and scx_unlink_sched() Tejun Heo
2026-03-04 22:01 ` [PATCH 33/34] sched_ext: Add rhashtable lookup for sub-schedulers Tejun Heo
2026-03-04 22:01 ` [PATCH 34/34] sched_ext: Add basic building blocks for nested sub-scheduler dispatching Tejun Heo
2026-03-06 4:09 ` [PATCHSET v3 sched_ext/for-7.1] sched_ext: Implement cgroup sub-scheduler support Tejun Heo
2026-03-06 4:17 ` Tejun Heo
2026-03-06 7:29 ` Andrea Righi
2026-03-06 18:14 ` Tejun Heo
-- strict thread matches above, loose matches on Subject: below --
2026-02-25 5:01 [PATCHSET v2 " Tejun Heo
2026-02-25 5:01 ` [PATCH 21/34] sched_ext: When calling ops.dispatch() @prev must be on the same scx_sched Tejun Heo
2026-02-25 5:00 [PATCHSET v2 sched_ext/for-7.1] sched_ext: Implement cgroup sub-scheduler support Tejun Heo
2026-02-25 5:00 ` [PATCH 21/34] sched_ext: When calling ops.dispatch() @prev must be on the same scx_sched Tejun Heo
2026-01-21 23:11 [PATCHSET v1 sched_ext/for-6.20] sched_ext: Implement cgroup sub-scheduler support Tejun Heo
2026-01-21 23:11 ` [PATCH 21/34] sched_ext: When calling ops.dispatch() @prev must be on the same scx_sched Tejun Heo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260304220119.4095551-22-tj@kernel.org \
--to=tj@kernel.org \
--cc=arighi@nvidia.com \
--cc=changwoo@igalia.com \
--cc=emil@etsalapatis.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sched-ext@lists.linux.dev \
--cc=void@manifault.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox