From: Andrea Righi <arighi@nvidia.com>
To: Tejun Heo <tj@kernel.org>, David Vernet <void@manifault.com>,
Changwoo Min <changwoo@igalia.com>
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH 2/2] sched_ext/idle: Make scx_bpf_select_cpu_and() usable from any context
Date: Mon, 12 May 2025 17:14:56 +0200 [thread overview]
Message-ID: <20250512151743.42988-3-arighi@nvidia.com> (raw)
In-Reply-To: <20250512151743.42988-1-arighi@nvidia.com>
Allow scx_bpf_select_cpu_and() to be used from any context, not just
from ops.enqueue() or ops.select_cpu().
This enables schedulers, including user-space ones, to implement a
consistent idle CPU selection policy and helps reduce code duplication.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
---
kernel/sched/ext_idle.c | 23 +++++++++++++++++------
1 file changed, 17 insertions(+), 6 deletions(-)
diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
index 6915685cd3d6b..5373132db02b6 100644
--- a/kernel/sched/ext_idle.c
+++ b/kernel/sched/ext_idle.c
@@ -943,10 +943,15 @@ __bpf_kfunc s32 scx_bpf_select_cpu_and(struct task_struct *p, s32 prev_cpu, u64
if (!check_builtin_idle_enabled())
return -EBUSY;
- if (!scx_kf_allowed(SCX_KF_SELECT_CPU | SCX_KF_ENQUEUE))
- return -EPERM;
-
#ifdef CONFIG_SMP
+ /*
+ * If called from an unlocked context, try to acquire
+ * cpus_read_lock() to avoid races with CPU hotplug.
+ */
+ if (scx_kf_allowed_if_unlocked())
+ if (!cpus_read_trylock())
+ return -EBUSY;
+
/*
* This may also be called from ops.enqueue(), so we need to handle
* per-CPU tasks as well. For these tasks, we can skip all idle CPU
@@ -956,10 +961,16 @@ __bpf_kfunc s32 scx_bpf_select_cpu_and(struct task_struct *p, s32 prev_cpu, u64
if (p->nr_cpus_allowed == 1) {
if (cpumask_test_cpu(prev_cpu, cpus_allowed) &&
scx_idle_test_and_clear_cpu(prev_cpu))
- return prev_cpu;
- return -EBUSY;
+ cpu = prev_cpu;
+ else
+ cpu = -EBUSY;
+ goto out_unlock;
}
cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, cpus_allowed, flags);
+
+out_unlock:
+ if (scx_kf_allowed_if_unlocked())
+ cpus_read_unlock();
#else
cpu = -EBUSY;
#endif
@@ -1266,6 +1277,7 @@ BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu_node, KF_RCU)
BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu, KF_RCU)
BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu_node, KF_RCU)
BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu, KF_RCU)
+BTF_ID_FLAGS(func, scx_bpf_select_cpu_and, KF_RCU)
BTF_KFUNCS_END(scx_kfunc_ids_idle)
static const struct btf_kfunc_id_set scx_kfunc_set_idle = {
@@ -1275,7 +1287,6 @@ static const struct btf_kfunc_id_set scx_kfunc_set_idle = {
BTF_KFUNCS_START(scx_kfunc_ids_select_cpu)
BTF_ID_FLAGS(func, scx_bpf_select_cpu_dfl, KF_RCU)
-BTF_ID_FLAGS(func, scx_bpf_select_cpu_and, KF_RCU)
BTF_KFUNCS_END(scx_kfunc_ids_select_cpu)
static const struct btf_kfunc_id_set scx_kfunc_set_select_cpu = {
--
2.49.0
next prev parent reply other threads:[~2025-05-12 15:18 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-12 15:14 [PATCH sched_ext/for-6.16 0/2] sched_ext: Extend usability of scx_bpf_select_cpu_and() Andrea Righi
2025-05-12 15:14 ` [PATCH 1/2] sched_ext: Make scx_kf_allowed_if_unlocked() available outside ext.c Andrea Righi
2025-05-12 15:14 ` Andrea Righi [this message]
2025-05-12 16:58 ` [PATCH 2/2] sched_ext/idle: Make scx_bpf_select_cpu_and() usable from any context David Vernet
2025-05-12 18:29 ` Andrea Righi
2025-05-12 19:07 ` Andrea Righi
2025-05-12 20:00 ` Tejun Heo
2025-05-13 5:23 ` Andrea Righi
2025-05-12 17:19 ` Tejun Heo
2025-05-12 18:35 ` Andrea Righi
2025-05-13 0:46 ` [PATCH sched_ext/for-6.16 0/2] sched_ext: Extend usability of scx_bpf_select_cpu_and() Changwoo Min
2025-05-13 5:48 ` Andrea Righi
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=20250512151743.42988-3-arighi@nvidia.com \
--to=arighi@nvidia.com \
--cc=changwoo@igalia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=tj@kernel.org \
--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
all inboxes | Powered by JetHome®